University of Southern California Dissertations and Theses

Multi-scale biomimetic structure fabrication based on immersed surface accumulation

(USC Thesis Other)

Multi-scale biomimetic structure fabrication based on immersed surface accumulation

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

MULTI-SCALE BIOMIMETIC STRUCTURE FABRICATION
BASED ON IMMERSED SURFACE ACCUMULATION

By:

Xiangjia Li

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of
the Requirements for the Degree
DOCTOR OF PHILOSOPHY

(INDUSTRIAL AND SYSTEMS ENGINEERING)

August 2019

Copyright 2019 Xiangjia Li

i

Acknowledgements
In my Ph.D. study, I would like to show my respect and gratefulness to my advisor Dr.
Yong Chen. Professor Chen conducted research on the development of novel additive
manufacturing process, which enables future product design and manufacturing. It’s my
great pleasure to be the student of professor Chen. He opened the door of the brand-new
research field for me and lead me into the world of 3D printing. In the past 5 years,
professor Chen made constructive comments and suggestions on my study plan, research,
and teaching practices. Professor gave me a great deal of freedom to pick the research
topics and allow me to learn cutting edge stuff in different fields, such as biomedical
engineering, computer science, and material science. Even he is very busy with his work,
he still spent time each week to have a brainstorm with me. His passion for teaching and
research was an inspiration for me and set an excellent example for my future faculty career.
Besides, I hope to express my gratitude to my collaborator Dr. Yang Chai for the
generous support on my interdisciplinary research. Professor Chai is the expert in the field
of craniofacial molecular biology, and we worked together on the project “suture and bone
regeneration for craniosynostosis and long bone graft”. He supervised my research in this
area and stimulated me the interest in the field of 3D printing of biomedical devices.
I also would like to thank Dr. Berok Khoshnevis, who taught me how to integrate the
inventive thinking with my research project. He helped me to establish an attitude and the

ii
skills to creativity, and he shared his experience in the process of creative and novel ideas
into the real new products and systems development. Under his instruction, I was more
familiar with the patent application, prototyping, product design and evaluation, and
product commercialization.
What’s more, I wish to take this chance to express my thanks to all my collaborators.
Firstly, I want to thank Dr. Yuan Yuan, Zoe Johnson, and Dr. Yuxing Guo in the Center
for Craniofacial Molecular Biology of University of Southern California for the bone
regeneration project. Then I would like to express my gratefulness to Dr. Yang Yang, Dr.
Xuan Song, Dr. Chi Zhou, and Dr. Zeyu Chen, who worked with me for the biomimetic
additive manufacturing. Meanwhile, I would like to show my deep thanks to Dr. Yayue
Pan for the contribution in the continuous printing with mask video projection. Moreover,
I would like to show my thanks to all my labmates, including Dr. Yuen-Shan Leung, Dr.
Tsz-Ho Kwok, Dr. Kai Xu, Dr Dongping Deng, Huachao Mao, Jie Jin, Yang Xu, and Han
Xu in the Center for Advanced Manufacturing, and all my mentored students (Haozhe zi,
Xuan Zheng, Benshuai Xie, Yue Wang, Jonghan Lim, Zixuan Huang, Qiyuan Zhang,
Luyang Liu, Ming Chu, Fuyue Lei, Chaoran Lu, Hanyu Zhao, Weitong Shan, Siyang Hao,
Yiyu Chen, Steven Li, Maggie Shi, and Agustin Sanchez). I am so grateful to have
everyone accompany in my Ph.D. study, and make my campus life more colorful.
My work was partially supported by the National Science Foundation (NSF) grant, the
National Institute of Health (NIH) grant, the Alfred Mann Institute (AMI) for Biomedical
Engineering in University of Southern California (USC), and USC Stevens Institute

iii
Technology Advancement Grant. These funding support assured the successful proceeding
of my research work of novel advanced manufacturing processes development, with
emphasis on manipulating and mimicking the intrinsically multi-scale, multi-material, and
multi- functional structures in nature.
I would like to show my appreciation to my screening committee, qualify committee
and thesis committee: Dr. John Gunnar Carlsson, Dr. Qiang Huang, Dr. Berok Khoshnevis,
Dr. Qifa Zhou, Dr. Wei Wu, Dr. Stephen Lu.
Finally, I want to say thanks to my husband Mr. Xianli Luo. Without the support of
my husband, I can’t stick to my research. I will always remember he accompany when I
conduct my experiments in countless days and nights. He always trusts me and believes
that I can make it. He always stands by my side and encourages me to overcome any
difficulties when I feel depressed. I feel so grateful for all the efforts you put into our family
and your greatest love for me in the past ten years. Besides, I hope to express my gratitude
to my grandparents, my parents, my parents in law and all my other relatives. Thanks for
the understanding and wish all of you have good health in the future.

iv

Table of Contents
Acknowledgements .................................................................................................................... i
Table of Contents ..................................................................................................................... iv
List of Tables ........................................................................................................................... ix
List of Figures ............................................................................................................................x
Abstract ....................................................................................................................................xv
Chapter 1 Biomimetic Fabrication by using Additive Manufacturing ......................................1
1.1 Additive manufacturing and its status ............................................................................. 1
1.1.1 Principle of additive manufacturing..........................................................................1
1.1.2 Current additive manufacturing processes ................................................................2
1.1.3 Vat Photopolymerization ..........................................................................................6
1.1.3.1 Stereolithography .............................................................................................. 9
1.1.3.2 MIP-SL ............................................................................................................. 9
1.2 AM of biomimetic structures and materials and research opportunity ......................... 14
1.2.1 Status of AM of biomimetic structures and materials and related problems ..........17
1.2.2 Motivation of biomimetic AM tool.........................................................................22
1.2.3 Challenge of biomimetic AM tool ..........................................................................25
1.3 Research focus in the dissertation ................................................................................. 30
1.3.1 The principle goal, research questions and hypotheses ..........................................30
1.3.2 Validation design and planning ..............................................................................33
1.3.3 Contributions from this dissertation .......................................................................34
1.4 Outline of dissertation ................................................................................................... 36
Chapter 2 Literature Review: AM Process Development for Biomimetic Fabrication ...........38
2.1 Multi-scale fabrication based on single AM process .................................................... 40

v
2.1.1 Direct ink writing ....................................................................................................40
2.1.2 Photopolymerization based printing process ..........................................................43
2.1.3 Inkjet Printing .........................................................................................................45
2.2 Multi-scale fabrication based on hybrid AM process ................................................... 46
2.2.1 Photopolymerization based hybrid process ............................................................47
2.2.2 Extrusion based hybrid process ..............................................................................48
2.3 Summary ....................................................................................................................... 50
Chapter 3 Immersed Surface Accumulation: Microscale Surface Texture Fabrication ..........51
3.1 Principle of Immersed Surface Accumulation .............................................................. 51
3.2 Prototype machine design ............................................................................................. 53
3.3 Optical image system .................................................................................................... 54
3.4 Light guide tool development ....................................................................................... 57
3.5 Calibration of light intensity ......................................................................................... 60
3.6 Process planning ........................................................................................................... 63
3.7 Hardware and software development ............................................................................ 64
3.8 Summary ....................................................................................................................... 66
Chapter 4 Process modeling: Parameter Optimization of Immersed Surface Accumulation ..67
4.1 Light curing performance ............................................................................................. 67
4.1.1 horizontal plane .......................................................................................................67
4.1.2 Vertical plane ..........................................................................................................69
4.2 Material attachment ...................................................................................................... 70
4.3 Material filling .............................................................................................................. 72
4.4 Moving speed ................................................................................................................ 77
4.4.1 Layer based ISA process .........................................................................................77
4.4.2 Continuous ISA process ..........................................................................................79

vi
4.5 Fabrication capability identification ............................................................................. 81
4.7 Surface quality .............................................................................................................. 81
4.8 Summary ....................................................................................................................... 82
Chapter 5 Microscale Biomimetic Texture Fabrication by using LISA ..................................83
5.1 Overview ....................................................................................................................... 83
5.2 Background introduction .............................................................................................. 83
5.3 Material and methods .................................................................................................... 85
5.3.1 Water collection experimental set-up .....................................................................85
5.3.2 Preparation of polymer/MWCNTs composite resin ...............................................85
5.3.3 Surface roughness ...................................................................................................86
5.4 Design of cactus inspired water collection structure .................................................... 88
5.5 Fabrication of cactus inspired water collection structure by LISA............................... 92
5.6 Experiment results and analysis .................................................................................... 94
5.6.1 Motion of water droplet ..........................................................................................94
5.6.2 The effect of moist airflow direction ......................................................................95
5.6.3 Water collection of multi-directional bionic spines ................................................96
5.6.4 Water collection of optimized multiple bionic spines ............................................97
5.7 Summary ....................................................................................................................... 99
Chapter 6 Multi-scale Biomimetic Structure Fabrication by using CISA .............................101
6.1 Overview ..................................................................................................................... 101
6.2 Background introduction ............................................................................................ 101
6.3 Design of biomimetic superhydrophobic and superoleophilic multi-scale structure .. 104
6.4 Material development ................................................................................................. 109
6.5 Fabrication of biomimetic superhydrophobic and superoleophilic multi-scale
structure by CISA ....................................................................................................... 113
6.6 Experiment results and analysis .................................................................................. 115

vii
6.6.1 Oil/water separation test .......................................................................................115
6.6.2 Oil remove test ......................................................................................................117
6.6.3 Droplet manipulations ...........................................................................................120
6.7 Summary ..................................................................................................................... 122
Chapter 7 ISA based Hybrid Process Investigation for Biomimetic Hierarchical
Structure Fabrication .............................................................................................124
7.1 Overview ..................................................................................................................... 124
7.2 Background introduction ............................................................................................ 125
7.3 Multi-scale hybrid process development .................................................................... 128
7.3.1 Macroscale & mesoscale MIP-SL ........................................................................128
7.3.2 Microscale ISA .....................................................................................................130
7.3.3 Nanoscale TPP ......................................................................................................132
7.3.4 Integration of MIP-SL, ISA and TPP ...................................................................134
7.3.4.1 Transition region optimization ...................................................................... 134
7.3.4.2 Vision assisted alignment ............................................................................. 136
7.3.4.3 Initial position identification ......................................................................... 139
7.4 Case study I: macro & micro scale biomimetic cell culture microenvironment
fabrication ................................................................................................................... 141
7.4.1 Design of biomimetic cell culture microenvironment ..........................................141
7.4.2 Material development ...........................................................................................142
7.4.3 Fabrication of biomimetic cell culture microenvironment ...................................142
7.4.3.1 Process planning ........................................................................................... 143
7.4.3.2 Fabrication results and analysis .................................................................... 143
7.4.4 Case study I summary ...........................................................................................147
7.5 Case study II: bioinspired optical filter with hierarchical structures ranging
from macroscale to nanoscale ..................................................................................... 147
7.5.1 Bioinspired design of hierarchical optical structure .............................................147

viii
7.5.2 Material development ...........................................................................................148
7.5.3 Fabrication of biomimetic optical filter by hybrid process ...................................149
7.5.3.1 Process planning ........................................................................................... 149
7.5.3.2 Process parameter optimization .................................................................... 149
7.5.3.3 Fabrication results ......................................................................................... 152
7.5.4 Case study II summary..........................................................................................154
7.6 Summary ..................................................................................................................... 154
Chapter 8 Conclusions and Recommendations ......................................................................155
8.1 Answering the research questions ............................................................................... 155
8.2 Achievements and research contributions .................................................................. 158
8.3 Limitation and future work ......................................................................................... 163
8.3.1 Novel additive manufacturing process development ............................................164
8.3.1.1 Multi-directional layer-less 3D printing ....................................................... 164
8.3.1.2 Physical field assisted multi-material printing .............................................. 164
8.3.2 Bioinspired functional material fabrication via ISA based nanocomposite
printing ..................................................................................................................165
8.3.2.1 3D-printing of bioinspired hydrodynamic functional surface ...................... 165
8.3.2.2 Multi-scale printing of bioinspired hierarchical structures ........................... 166
8.3.3 3D printing medical devices for healthcare ..........................................................166
8.3.3.1 3D printing functional scaffold for tissue regeneration ................................ 166
8.3.3.2 3D printing cell culture microenvironment................................................... 167
Reference ...............................................................................................................................168

ix
List of Tables
Table 1 The overlook of current additive manufacturing processes ............................................... 5
Table 2 The overlook of MIP-SL based on key factors ................................................................ 12
Table 3 The distribution of hypotheses and validation in the dissertation ................................... 33
Table 4 The validation design and planning in the dissertation .................................................... 34
Table 5 The mechanical strength of material fabricated by using different Kz ............................ 80

x
List of Figures
Figure 1 The principle of additive manufacturing .......................................................................... 1
Figure 2 Different types of material fabricated by AM processes. ................................................. 2
Figure 3 Current AM processes developed to fabricate different types of material ....................... 3
Figure 4 The main applications of current AM technologies ......................................................... 4
Figure 5 The material selection based on the performance of material .......................................... 7
Figure 6 An illustration of principle of photopolymerization based AM process ........................ 10
Figure 7 An illustration of functional multiscale structures in nature .......................................... 15
Figure 8 Critical issues in AM of biomimetic structures and materials ....................................... 17
Figure 9 An illustration of the layer-based AM processes and related stair-stepping effect ........ 19
Figure 10 An illustration of the fabrication scale of photopolymerization based AM processes. 21
Figure 11 Function modification by adding bioinspired microstructures. .................................... 23
Figure 12 An illustration of the timeline for biomimetic functional devices development and
my research focus. ........................................................................................................ 24
Figure 13 The light distortion in the microscale MIP-SL ............................................................. 27
Figure 14 An illustration of curing effect of photopolymer in vertical direction ......................... 28
Figure 15 An illustration of curing effect of photopolymer in horizontal direction ..................... 29

xi
Figure 16 Outline of dissertation .................................................................................................. 37
Figure 17 An illustration of status of AM of bioinspired material and structures ........................ 39
Figure 18 AM of multi-scale biomimetic vascular channel by using direct ink writing .............. 41
Figure 19 AM of multi-scale biomimetic design by using direct ink writing .............................. 43
Figure 20 AM of multi-scale structures by using Vat polymerization based printing method..... 44
Figure 21 AM of multi-scale structures by using Inkjet printing based method .......................... 46
Figure 22 Multi-scale structure fabricated by photopolymerization based hybrid process. ......... 48
Figure 23 Hybrid process by integrating extrusion based printing process with other process ... 49
Figure 24 A schematic illustration of the CNC accumulation system .......................................... 51
Figure 25 A schematic illustration of Immersed Surface Accumulation printing system ............ 54
Figure 26 The layout of the immersed surface accumulation imaging system............................. 56
Figure 27 The design of light guide tools ..................................................................................... 57
Figure 28 An illustration of the fabrication process using light guide tools in ISA process ........ 59
Figure 29 The light intensity distribution of 2D patterned light beam ......................................... 60
Figure 30 A fabrication result of micro pillar array by using 2D patterned light beam with
adjusted light intensity .................................................................................................. 62
Figure 31 Flow chart of process planning for ISA process .......................................................... 63

xii
Figure 32 The hardware and software development in ISA process. ........................................... 65
Figure 33 The exposure study of liquid resin ............................................................................... 68
Figure 34 The curing depth control of liquid resin ....................................................................... 70
Figure 35 The illustration of separation force and attachment force of the new cured layer ....... 72
Figure 36 The schematic diagram of liquid filling progress of ISA process ................................ 74
Figure 37 The valid fabrication section area δ with self-refilling resin using different moving
speed of platform .......................................................................................................... 76
Figure 38 The moving speed setting in ISA process. ................................................................... 78
Figure 39 The fabrication of circular cone by using layer based ISA printing process. ............... 80
Figure 40 The cones fabricated by layer based and continuous ISA printing. ............................. 81
Figure 41 Appearance, microscale structures, and water collect performance of cactus spines. . 84
Figure 42 The schematic illustration of the set-up used to measure water collection efficiency. 85
Figure 43 The water collection efficiency of surface with and without bionic spines made
by different material.. ................................................................................................... 87
Figure 44 The water collection of bionic spines with different angle .......................................... 90
Figure 45 Design of array of bionic spines ................................................................................... 91
Figure 46 Fabrication of bionic spines array by layer based ISA process .................................... 93
Figure 47 The movement of water droplet condensed from humidity environment .................... 94

xiii
Figure 48 The fog collection efficiency of bionic spine array under moist airflow from
different directions ........................................................................................................ 95
Figure 49 The water collection of bionic spines with multiple branches ..................................... 96
Figure 50 The simulation of air flow around each bionic spine with different arrangements ...... 98
Figure 51 The water collection of bionic spines with optimized arrangement ............................. 99
Figure 52 Biomimetic superhydrophobic and superlipophilic structure..................................... 103
Figure 53 The wettability of bionic eggbeater shaped structure with different designs ............. 105
Figure 54 The simulation of relative stress of the superhydrophobic eggbeater shaped
structure designed with different number of hair. ....................................................... 108
Figure 55 The air/solid ration of eggbeater shaped structures with different height... ............... 109
Figure 56 The hydrophilic and lipophilic property of material. ................................................. 111
Figure 57 The surface roughness effect on hydrophobic and oleophobic property of
designed material ........................................................................................................ 112
Figure 58 The fabrication of eggbeater shaped structure by CISA process ............................... 114
Figure 59 The oil separation of 3D-printed eggbeater shaped structure at different conditions. 117
Figure 60 The absorption of different oils by the 3D-printed eggbeater shaped structure . 118
Figure 61 Oil absorption comparison of the 3D-printed eggbeater shaped structures with
different height.. .......................................................................................................... 119
Figure 62 The droplet manipulation by 3D-printed eggbeater shaped structure ....................... 120

xiv
Figure 63 An illustration of hierarchical structure in nature and the design & process planning
of multiscale hierarchical printing .............................................................................. 126
Figure 64 Schematic diagram and physical prototype of the mesoscale MIP-SL ...................... 130
Figure 65 The diagram of the ISA process.. ............................................................................... 131
Figure 66 Experimental setup and test result in TPP.. ................................................................ 133
Figure 67 An illustration of the transition process between multiple processes. ........................ 135
Figure 68 An illustration of the design of vision system in ISA and TPP processes ................. 137
Figure 69 An illustration of the alignment in ISA and TPP processes ....................................... 138
Figure 70 An illustration of positioning the light guide tool ...................................................... 140
Figure 71 The CAD model of biomimetic microscale structures for long-term cell culture ...... 141
Figure 72 The process planning diagram of the ISA based hybrid process for the fabrication
of 3D cell culture microenvironment. ......................................................................... 143
Figure 73 The design and fabrication of 3D cell culture microenvironment using the
integrated MIP-SL and ISA process. .......................................................................... 145
Figure 74 The scaffolds with 3D microscale concave textures are printed using our integrated
MIP-SL and ISA processes ......................................................................................... 146
Figure 75 The design of bioinspired optical filter with hierarchical structures .......................... 148
Figure 76 The process parameter optimization in photopolymerization based AM process ...... 151
Figure 77 The preliminary printing result of optical filter with hierarchical structures ranging
from macroscale to nanoscale .................................................................................... 153

xv

Abstract
Nature provides us a vast of promising functional material systems consisted of hierarchical
structures, where complex geometry and dimensional discrepancy across sections are present.
Such hierarchical structures possess functionalities that inspired various biomimetic applications
but are overall difficult to be built with traditional manufacturing process primarily due to the
limitation of manufacturing capability set by conventional machine tools. Nowadays, additive
manufacturing (AM) methods are commonly deployed to tackle with three-dimensional objects
with complex structure by gradually accumulating material layer by layer. However, the
hierarchical structure measured from macroscale to nanoscale still raises a challenge to the current
commercial AM process, whose manufacturing capability is intrinsically specified within a certain
scope. It is desirable to develop a new AM process with the capability to fabricate biomimetic
functional material, and scalability to accommodate hierarchical structures at multi-scale. The
major thrust of this research, therefore, is to investigate and develop novel printing approaches and
strategies for manipulating and mimicking the intrinsically multi-scale, multi-material, and
multifunctional structures in nature.
In this research work, a novel microscale AM processes called immersed surface accumulation
(ISA) is first developed to fabricate functional cactus inspired multi-scale structure. In this process,
a special optical system is developed with a surface-based light guide tool to deliver light beam,
and the calibration of light intensity is further conducted to obtain 2 dimensional (2D) patterned
light beam with uniform light distribution. In order to fabricate cactus inspired microstructure, a
light guide tool is immersed inside a tank filled with liquid based composite material, and material
is gradually accumulated on the flat surface with the light exposure. Moreover, the system design

xvi
and nonlinear exposure settings were proposed and implemented to fabricate cactus inspired
microstructure arrays. Water collection of artificial three dimensional (3D) printed microstructure
array is studied and high-efficiency water collection device is further developed and tested.
To investigate the high-efficient fabrication of multi-scale bio-mimic superhydrophobic and
superoleophilic structures, continuous immersed surface accumulation printing (CISA) approach
was developed. In order to achieve continuous accumulation of material in printing direction, the
self-filling and curing performance of material is investigated and analyzed. In addition, the
material refilling of CISA process is discussed to get the good surface quality of bioinspired
microstructures, and the mechanical performance of the printed parts with different speed is
investigated. The composite material composed of photocurable material and multi-walled carbon
nanotubes (MWCNTs) is developed, and special bio-inspired superhydrophobic and
superoleophilic microstructures are designed and fabricated by the proposed ISA process. Droplet
manipulation and Oil/water separation/removing are demonstrated respectively to show the
superhydrophobicity and superoleophilicity of 3D printed bio-inspired microstructures.
Based on the above macroscale and microscale AM processes, the integrated
photopolymerization based hybrid process is developed to investigate the fabrication of multi-
scale biomimetic texture. The macro- and mesoscale mask image projection based
stereolithography is introduced and the corresponding process parameters are studied and
optimized. Then the integration of the microscale ISA process is discussed to achieve high-
resolution microscale fabrication results. A multi-scale 3D printing method is presented by
integrating the ISA and mask image projection based stereolithography (MIP-SL) processes. To
achieve high- resolution fabrication, process planning, material curing performance, and the
printing parameters are discussed. To enable cell culture study, multi-scale cell culture

xvii
environment with multiple bio-inspired textures are fabricated with bio-compatible material based
on developed multi-scale AM process. The experimental results verify the efficiency and accuracy
of the developed multi-scale 3D printing method on the fabrication of multi-scale functional
structures.
To fabricate bio-inspired hierarchical structure with macro-, micro- and nanoscale features, a
3D hierarchical printing approach with an investigation on integrating multiple printing processes
is proposed. Firstly, special hierarchical design and process planning towards integrating multiple
printing processes are demonstrated. Then sub-processes including macroscale mask image
projection based stereolithography, microscale ISA, and nanoscale two photon polymerization are
developed, and the key process parameters for each sub-process, e.g., the curing characteristics of
the material, are optimized to improve the fabrication quality of each process. To address the
challenge of the integration of multiple processes, a vision assisted approach is introduced, and the
alignment of each sub process for the transition is introduced and developed. Throughout this
research work, a systematic study for multi-scale AM development including the material
development, biomimetic structure design, the process development in multiple manufacturing
scales is presented in order to achieve certain functionalities inspired by nature.
1

Chapter 1 Biomimetic Fabrication by using Additive
Manufacturing
1.1 Additive manufacturing and its status
1.1.1 Principle of additive manufacturing
Additive manufacturing (AM) shows advantages to build three dimensional (3D) objects with
complex geometric shape over the traditional manufacturing technologies [1]. Due to the unique
fabrication principle, AM plays an important role in the fabrication of free form objects with
complex geometric shape [2, 3]. Over the past thirty years, the manufacturing community benefits
a lot from additive manufacturing, thanks to its unique capability of fabricating complex geometry
from its 3D computer-aided design (CAD) models. Basically, the principle of additive
manufacturing can be described as the following steps (refer to Figure 1). Firstly, a set of two
dimensional (2D) layers are generated by slicing the 3D digital model of designed part, and then
one layer of material will be totally solidified at one time or gradually built into the 2D pattern [4].
By progressively accumulating material layer by layer, the 3D object will be formed into the
desired shape (refer to Figure 1) [5].

Figure 1 The principle of additive manufacturing

2
1.1.2 Current additive manufacturing processes
In 1986, Chuck Hull invited the first 3D printing technology called stereolithography (SL), and
based on it the first prototype machine was built to fabricate the polymer-based object [6]. After
that, a lot of efforts have been made to develop new additive manufacturing process, and a large
range of material including metal [7], plastic [8], composite [9], ceramic [10], and glass [11] can
be fabricated by current advanced AM technologies (refer to Figure 2). Due to the unique
fabrication mechanism, AM has become an effective way to fabricate specific materials, which are
impossible to be processed into 3D shape by using traditional manufacturing methods [12].

Figure 2 Different types of material fabricated by AM processes. (a) Plastic fabricated by binder
jetting; (b) bio-material fabricated by SL;(c) glass fabricated by extrusion based process
(copyright from [11]); (d) ceramic fabricated by slurry based SL; and (e) metal fabricated by
commercial machine from EOS.
For example, aerogel is one kind of porous and ultralight material with low thermal
conductivity, and due to its attractive functional performance, it can be used for a variety of
applications. However, there is a big challenge to form the aerogels into 3D shape, and the
fabrication limitation become a major obstacle that restrict its use for wider application. Recently,
researchers successfully fabricated the aerogel by using extrusion-based 3D printing process and
further developed a lightweight and high compressible material based on 3D shaped aerogel [12,

3
13]. Another good example is organ printing, where bioactive living micro-tissues are
accumulated layer by layer to construct artificial organ or 3D macro-tissues [14, 15]. Inkjet
printing and direct ink writing have been used to build this kind of artificial organ, and artificial
organ, such as heart, kidney, liver structures, etc., has been printed with the similar appearance [14,
15]. But the functionality and bioactivity of 3D printed artificial organ are still not competitive as
the natural one, and more effort must be put in this field to move it forward [14, 15].

Figure 3 Current AM processes developed to fabricate different types of material. (Image from
3D printing industry)
To achieve the fabrication of different type of material, a large number of AM processes are
developed based on different principles. According to different principles, most common AM
processes are stereolithography (SL) [16], Inkjet printing [17], selective laser sintering/melting
(SLS/M) [18], binder jetting [19], fused deposition modeling (FDM) [20], selective deposition
lamination (SDL) [21], and direct ink writing (DIW) [22]. Currently most of AM processes are

4
fabricated the object in a layer manner, which the material is accumulated layer by layer to form
the 3D shape [2, 16, 18]. How the material is accumulated and what type of material is processes
are the main differences between each AM process (refer to Figure 3). Generally, most of AM
process is developed to fabricate one type of material, and each AM process has its unique
technical characteristics and fabrication limitations (refer to Table1). Proper AM technology can
be selected to build the object based on the material and product development requirements [1].

Figure 4 The main applications of current AM technologies
Since AM methods convert 3D fabrication into the accumulation of layer based material, it
enables the fabrication of structures with certain material and geometric complexity, which is
better than the other manufacturing processes [23]. Compared with traditional manufacturing
processes, AM technologies present significant advantages with respect to design freedom,
assembly complexity, customization, material diversity, as well as fabrication cost and efficiency
[24, 25]. These advantages facilitate AM to become one of the most popular commercial

5
technologies (refer to Figure 4) [33], which have been widely used in the variety of fields [23],
such as biomedical engineering [26], automobile [27], electronics [28], food [29], aerospace [30],
sustainable energy [31], and civil engineering [32], etc. Besides that, reinforced mechanics, shape-
changing, interfacial, optical as well as thermal devices can be fabricated and further developed
by AM technologies, which create new opportunities for manipulating the intrinsically multi-scale,
multi-material and multi-functional devices in future [34].
Table 1 The overlook of current additive manufacturing processes
Name Material Resolution Speed Cost Application
SL [16, 35]
Polymer/Ceramic
slurry/Composite material
5 m - 500 m 1-100 m/s

Biomedical,
Electronics,
Sustainable-
energy,
optical, etc.
FDM [17]
Polymer/polymer based
composite material
100 m - 500 m 1-10 m/s

Toy, Food,
Prototype,
etc.
Binder jetting
[19]
Ceramic/polymer/metal
powder
50 m - 500 m 1-10 m/s

Biomedical,
Aerospace,
etc.
Inkjet printing
[20]
Polymer/Polymer based
composite material
1 m - 500 m 10-50 m/s

Biomedical,
Electronics,
Sustainable-
energy,
optical, etc.
SLS/SLM [18] Metal powder 50 m - 500 m 1-10 m/s

Biomedical,
Aerospace,
etc.
SDL [21] Metal or paper sheet 50 m - 500 m 1-5 m/s

Toy, Food,
Prototype,
etc.
DIW [22]
Polymer/ composite
material
30 m - 500 m 1-10 m/s

Biomedical,
Electronics,
Sustainable-
energy, etc.

6
1.1.3 Vat Photopolymerization
With the development of AM processes, the revolution of AM has brought us possibilities in
designing and building complex free-from surface model fast and accurately [23]. Various AM
processed are developed to solve the challenge of 3D fabrication, and a large range of material can
be formed into 3D shape by using AM technologies [23, 34]. AM is widely applied in every area
of social life, such as aerospace, biomedical engineering, civil engineering, electronics, etc. [23,
34]. Vat photopolymerization shows unique superiority in manufacture industry among all the AM
processes [37]. It is developed based on one type of chemical reaction called photopolymerization,
which the photocurable polymer will be crosslinked under the initiation of light exposure [38]. a
linear or crosslinked polymer structures are formed in the photopolymerization by using light to
initiate and propagate [39]. The stereolithography is further developed based on
photopolymerization, where photopolymer is solidified and accumulated into 3D shape after
exposure to the light [38]. The photopolymer, composed of monomers, oligomers, and
photoinitiator, undergoes the process called photocuring, where oligomer are cross-linked to form
a network polymer after receiving enough energy from light exposure [40]. The crosslinking of
the photocurable polymer mainly occurs between the carbon double bonds in acrylates or
methacrylate [39].
Large range of material can be fabricated by using the vat photopolymerization [37]. Generally,
the material used in the AM process can be divided into categories of plastic, glass, ceramic, rubber,
metal and composite material as shown in Figure 5a [1]. To print non-curable material, currently
the most common way is to firstly mix the non-curable material with photopolymer, and then build
the mixture into 3D shape by using vat photopolymerization [41]. Later the photopolymer material
is removed by other post-processing methods like chemical dissolution [42], high-temperature

7
sintering [43]. The material selection of functional material and photopolymer, which likely be
added to assist the curing process, is vital in the AM of bioinspired functional material and
structures.
(a)
(b)
Figure 5 The material selection based on the performance of material. (a)The biomaterial can be
used in vat photopolymerization; and (b) The optical performance of material can be used in vat
photopolymerization.

8
The material selection in vat photopolymerization depends on the specific functions of the
intended applications [34]. For instance, the scaffold is designed and fabricated to support the cell
attachment and growth in tissue regeneration. In general, the material of scaffold should be
biocompatible, exhibiting appropriate mechanical properties, superior degradation kinetics, and
tissue biomimicry [35]. Therefore, the material of scaffold for cell culture and tissue regeneration
should be not only photocurable but also biocompatible if we choose vat photopolymerization as
the printing process [35]. Many types of materials, including natural and synthetic materials, have
been investigated for biomedical purposes, while the low photosensitivity, insufficient mechanical
properties of material have been major barriers for the AM of bioinspired tissue implant in the
biomedical field by using vat photopolymerization [44].
Another good example is the material selection for constructing complex optical component
by using vat photopolymerization. When we select the material for the optical application, a lot of
optical specifications of material have to be considered, including refractive index, Abbe value,
reflectivity, and transparency (refer to Figure 5b) [34]. With the development of AM technologies,
different types of optical material, such as glass, optical crystal, plastics and semiconductor
materials can be fabricated by using AM technologies [34]. By using vat photopolymerization,
optical polymer can be formed into 3D shape with complex geometric design, which is impossible
to fabricate by the traditional molding process [45]. A various of optical polymer, such as
polycarbonates, acrylics, and polystyrenes, can be printed by using vat photopolymerization
process [46]. For example, by mixing optical crystal and semiconductor materials with
photocurable polymer, the composite also can be successfully cured and further be accumulated
into complex 3D biomimetic shape [47, 48]. Overall, for all aforementioned material selection
cases in the vat photopolymerization of functional devices, the material selection should be

9
conducted based on the performance of material in all major areas of interest and the
manufacturing capability [45-48].
1.1.3.1 Stereolithography
In stereolithography, the photocurable polymer can be selectively cured into a solid object by
using laser-based or digital light processing based process [40]. There are two typical processes:
laser based stereolithography and mask image projection based stereolithography. In the laser-
based stereolithography, the tank is filled with photocurable polymer, and the laser, which goes
through the optical components, illuminates the surface of liquid polymer and cures the material
along the movement of laser beam (refer to Figure 6) [38]. 3D object is firstly sliced into 2D layers,
and the tool path of laser is further planned to cover all the area inside each 2D layer. Thus, the
laser dot has to go through all the internal area of the building part, which will cost pretty long
time to fabricate the part [49]. The length of the tool path depends on the dimension of the laser
dot, and the tool path increases dramatically with the decrease of laser dot in term of macroscale
fabrication [50]. In mask image projection based stereolithography (MIP-SL), the whole layer of
the photocurable polymer is selectively cured with one-time exposure of light beam [51, 52].
Compared with laser-based stereolithography, MIP-SL shows time efficiency in building the 3D
object [53]. MIP-SL is one branch of vat polymerization, and basic information, including the
principle, development, optimization, and prospects will be introduced in following section.
1.1.3.2 MIP-SL
In MIP-SL, the light reflected by the digital micromirror device (DMD) illuminates the surface
of photocurable resin as shown in Figure 6c [54]. After curing one layer, Z stage moves up/down
with the distance equal to the layer thickness, and then a new layer will flow back to the light

10
projection area. To speed up the material refilling and reduce the separation force, two-way
movement is developed in MIP-SL [53]. The hardware of MIP-SL includes optical imaging system,
material feeding system, and three axis motion system [58]. In optical imaging system, the
computer communicates with the controller to adjust each angle of the micro mirror in the DMD
and further control the light intensity of each pixel. The dimension of focusing light beam is
determined by the optical design. By changing the combination of the convex lenses, the projection
light beam can be reduced or enlarged into different size [55]. By using the different optical design,
the dimension of features printed by MIP-SL ranges from 5 μm to 200 mm [55-58]. By controlling
the energy model of each pixel in the projection image, MIP-SL can achieve high-resolution
fabrication with sharp features [58, 59]. During the fabrication process, the projection image
planning [60-62], light energy control [58, 59], resin flow [63], and separate force [53, 64, 65] are
main factors that enhance the fabrication quality and speed of MIP-SL.

Figure 6 An illustration of principle of photopolymerization based AM process. (a) The principle
of photopolymerization; (b)laser based SL process with top-down frame; and (c) MIP-SL process
with bottom-up frame.

11
In MIP-SL, there are two types of light sources: light-emitting diode (LED), and lamp. The
type of photocurable polymer, which can be fabricated in MIP-SL, is decided by the wavelength
of light source [35, 55]. LED is easy to turn on/off by using a relay, and it eliminates the extra
exposure in the interval period between fabrication of adjacent layers when the Z stage moves up
or down. Other advantages of LED are narrow wavelength, low price, excellent focusing quality
and high switching rate. Although UV LEDs has high energy power, in fact, only 25% of energy
was used for the printing process [66]. The lamp is also widely used in MIP-SL, and it requires a
filter to get the suitable wavelength light for the curing process due to the broad bandwidth of light
emitted from the lamp [16]. Compared with laser and LED, the light intensity of the lamp is much
higher than other kinds of light sources, so that it costs much shorter time to curing the same
material [56].
In the optical image system, the light emitting from the light source generates the divergence
when it travels [58, 67]. To reduce the divergence effect of light, the collimating lens is necessary
to be mounted in front of the light source, and the energy of light emitting from the light source
like LED and lamp will be converged [67]. The projection image is modified several times larger
or smaller by the special designed optical system which is consist of a set of optical lens. Based
on the dimension of projection image size, the MIP-SL can be divided into 4 categories:
macroscale MIP-SL [61], mesoscale MIP-SL [68], microscale MIP-SL [66], and multi-scale MIP-
SL process [69]. Table 2 shows key indicators of MIP-SL processes developed in the past thirty
years.

12
Table 2 The overlook of MIP-SL based on key factors
Literature
Light
source
Mask Image size Resolution
Layer
thickness
Speed
Monneret et al
1999, 2001 [71, 72]
Visible light LCD 3-10 mm
2
2 m 10 m 60s/layer
Bertsch et al.
2001 [73]
UV light DMD 3 x 2 mm
2
10 m 5 m 18s/layer
I.B.Park et al. 2010
[74]
Lamp (UV) DMD Not reported Not reported 50 m 180s/layer
A.S. Limaye et al.
2007[75]
Lamp (UV) DMD 1.1 x 1.8
mm
2

6 m 100 m 90s/layer
Young Myoung Ha
et al. 2007[76]
Lamp (UV) DMD 2 x 2 mm
2
2 m 5 m 100s/layer
J. W. Choi et al.
2006[66]
Lamp (UV) DMD Not reported 5 m 30 m 15s/layer
L.H. Han et al.
2008[77]
UV LED LCOS 10 x 10 mm
2
10 m 1 m 10s/layer
C. Xia et al.
2009[78]
UV LED LCOS 3 x 2 mm
2
5 m 1 m <1s/layer
C. Sun et al.2005
[70]
Lamp (UV) DMD 7 x 5 mm
2
7.1 m 5 m-20 m Not
reported
S. Xuan et al.
2017[79]
Lamp
(Visible)
DMD 16 x 12 mm
2
100 m 50 m 15s/layer
Y. Pan et
al.2017[56]
Lamp
(Visible)
DMD 12.7 x 8
mm
2

80 m 100 m 2s-6s/layer
C. Zhou et
al.2009[58]
Lamp
(Visible)
DMD 216 x 162.6
mm
2

210 m 100 m 20s/layer
Shih Hsuan Chiu et
al. 2008[80]
Lamp (UV) DMD 27in - 300in 669 m-7.44 mm 2mm 15.5s/layer

After light goes through the lens system, there would be distortion and aberration in the
focusing image [67]. To improve the focusing quality of light beam, some type of lenses, such as
bi-convex lens, have to be used in MIP-SL [8]. Besides, automatic focusing adjustment system is

13
added by using the pellicle beam splitter to capture the focusing image [70]. The position of the
tank will be finely adjusted by analyzing the focusing image [67].
In MIP-SL the direction of light can be changed by a mirror. Based on the direction of
projection light, MIP-SL can be divided into the bottom-up based MIP-SL [68] and the top-down
based MIP-SL [67]. As shown in Figure 6c, the light shoot from the bottom and goes through the
bottom surface of resin tank in the bottom-up based MIP-SL. After the finish of one layer’s
fabrication, the platform, which is used to attach the building part, will move up. A thin channel is
formed between the platform and transparent glass, and the new layer of resin will fill inside the
thin channel [63]. Since the focusing image is located on the surface of transparent resin tank, and
the newly cured layer will be attached on the bottom surface of the building part, which is bonded
on the surface of platform (refer to Figure 6c) [61]. After the finish of one layer’s fabrication, the
platform will move down with one layer distance for the next layer fabrication in the top-down
based MIP-SL [57]. Therefore, both the whole printed part and the platform have to be merged
inside the resin tank. In that case, the light illuminates from the top side and will be focused on the
top surface of liquid resin [66].
Each type of layout has strengths and drawbacks. In top-down based MIP-SL, since there is no
constraint surface of liquid flow, the flatness of the cured layer is not easy to control (refer to
Figure 6b) [69]. What’s more, the whole platform and pre-cured portion have to be merged inside
the resin tank. Therefore, the height of resin is at least the same height as the building part, and it
will assume a large amount of resin [81]. While, top-down based MIP-SL process shows
advantages in the quality of focusing image. Because the light is directly focused on the top surface
of the resin in top-down based MIP-SL, and there is no the distortion and refraction of focusing
image [81]. The focusing quality of projection image is a little bit better than the one generated by

14
bottom-up based MIP-SL [58]. This is because the light goes through the transparent glass, and
refraction of transparent glass generates the light distortion in the bottom-up based MIP-SL [56].
Especially for the micro-stereolithography, the distortion of focusing image seriously affect the
fabrication quality and accuracy [67]. As shown in Figure 6c, since it is not necessary to immerse
the whole printed part inside the resin tank, the bottom-up based MIP-SL consumes significantly
less resin compared with the top-down based MIP-SL [36]. Furthermore, it is easy to control the
flatness and thickness of each layer in the bottom-up based MIP-SL, because the platform and
surface of transparent tank generated a uniformed gap after each layers’ fabrication. Based on that,
it requires less settling time for the resin to flat, so that it is not necessary to add additional blade
for the material recoating, which is integrant in the top-down based MIP-SL [36, 53]. However,
due to the large attaching force between the cured layer and the top surface of transparent glass,
there is a big challenge to separate the cured layer from the transparent glass in the bottom-up
based MIP-SL [53, 54]. To solve this problem, one thin layer of polydimethylsiloxane (PDMS) or
Teflon film is coated on the top surface of transparent glass [53]. Because of the low oxygen
permission of glass and PDMS, oxygen inhibition occurs in the upper surface of PDMS, and
generates a very thin liquid layer of non photocurable resin, which is beneficial for the reduction
of separation force in the bottom-up based MIP-SL [36, 53].
1.2 AM of biomimetic structures and materials and research opportunity
Many material systems from nature exhibit outstanding properties not found in artificial or
synthetic systems, since natural structures have evolved over millions of years surviving rigorous
environmental conditions [82]. The exceptional performances of natural material, such as the
structural color of butterfly, the drag force reduction of fish scales, the superhydrophobic effect of

15
rose petal, and water collection of desert beetles, benefits from hierarchical structures spanning
over a large range of scales from macro scale to nanoscale (refer to Figure 7) [83, 84]. Such multi-
scale structures in nature provide inspiration for composite material design and show the promising
multifunctional application in mechanical, optical, thermal, and electrical fields [85-89]. Nature
creatures and structures possess comprehensive complexity in geometry, hierarchy, and material,
setting challenges for mimicking, reproduction of functionality, and particularly the manufacturing
with traditional approaches [83].

Figure 7 An illustration of functional multiscale structures in nature
To be specific, a biological material system in nature is not limited to be a single function. For
example, the microstructure of the butterfly wing cannot only generate gorgeous structural color
but also show de-wettability due to superhydrophobicity [90]. Similarly, spider silk, which is made
of protein fiber, shows impressive mechanical performance and it also can collect water [91, 92].

16
The spider silk is stronger and tougher than steel, and this promising mechanical is attributed to
special spun pattern [91]. Meanwhile, there are microscale puff and joint on the surface of spider
silk, and these microscale structures promote the water collection and transportation of spider silk
[92]. Learning from examples, biological structures in the material system show multiple functions
due to multi-scale structures.
Biological functional structures require multiple categories of material to be formed, and each
type of material plays an important role in functional performance [34]. The limpet teeth exhibit
extreme mechanical strength with a distinctive multi-material structure, which is a soft protein
with a high-volume fraction of goethite nanofibers reinforced composite [100]. Due to this unique
material distribution, microscale limpet teeth show better mechanical performance than spider silk,
which is initially highest tensile strength among all existing biological material [100]. Nacre has
high strength and toughness due to the brick-and-mortar like structure, which is constituted by
hexagonal platelets of calcium carbonate and elastic biopolymers [101]. This special multi-
material arrangement of nacre inhibits transverse crack propagation [102]. Different type of
material introduces quite different functions, and the natural multi-material systems give us
inspiration and design principles for multi-functional innovative applications in various areas [23].
The biological functional material system is composed of hierarchical structures ranging from
macroscale to nanoscale. For instance, the gecko adhesive foot possesses macro to nanoscale
structures with dramatically descending dimension from the finger to the hair tip, featuring
functional adhesion force that inspires applications such as attach force sensor [88, 103]. Another
good example is the fly eyes. Fly eyes have unique optical systems, which can observe 10 times
the large field of view than other creatures [104]. This optical functional performance is attributed
to multi-scale compound eyes, which is consisted of micro and nanoscale features [104]. When

17
we look at the biological material system, such as shark skin, gecko feet, fly eyes, and butterfly
wing, etc., specific functional structures, including micro and nanoscale features, grow on
macroscale substrates, [83], and such natural hierarchical structures play important role in the
functional performance. With increasing on the design of devices for special functionalities,
mimicking the biological multi-scale structures has become necessary and important.

Figure 8 Critical issues in AM of biomimetic structures and materials
1.2.1 Status of AM of biomimetic structures and materials and related problems
There are three main issues in AM of bioinspired functional material. First, the biological
material system in nature is not limited to be a single function, and it has functional complexity
[105]. Second, biological functional structures are based on multiple material systems, which
require multiple categories of material to be formed [100]. Third biological material system is
composed of hierarchical structures ranging from macroscale to nanoscale, usually requires

18
manufacturing capabilities at multiple levels [89]. However, the status of AM technologies shows
insufficient capabilities toward functional, material and hierarchical complexities (refer to Figure
8). For example, most of the current am processes are focused on geometry centered design. Also,
they show limitation in material selection and only a few types of material can be fabricated by
AM technologies. Meanwhile, the current AM technologies only can achieve the fabrication at a
small range of scale fabrication. While the natural system is usually composed of multi-scale
structures.
As stated, recent progress on the applications of multi-scale AM technologies shows 3D
printing as a tool to validate the biomimetic design with hierarchical multi-scale structures [23,
34]. However, AM approaches shows limited printing capability, with compromised performance
in printing accuracy and printing scale. Therefore, we can identify two problems in the field of
biomimetic AM.
(1) The fabrication accuracy of AM cannot meet the requirements of biomimetic design
Biomimetic design can be applied in the different fields, including mechanics, electronics, civil
engineering, biomedical engineering, aerospace, etc. [106]. The evolution of AM technologies
provides more space for designer to develop the functional devices with the structures close to the
biological feature [34]. In the phase of biomimetic design, functionality always be put in the first
place so that the structure is relatively complex, which brings the difficulties for the manufacturing.
This is because the functional biological structures have complex 3D features, which are hard for
AM technologies to accurately produce the complex 3D shape as same as the natural structures.
As discussed in Section 1.1, most of additive manufacturing processes are layer based approach
[1-3]. To fabricate 3D object, the digital model is firstly sliced in to a set of 2D patterns, and the

19
material is deposited layer by layer based on the sliced 2D patterns [4]. Due to this unique
manufacturing principle, the surface of printed part reveals a corrugated appearance (refer to
Figure 9), which is known as stair-effect. Since the surface quality of the building part determines
the final functional performance [107, 108], this kind of uneven features have negative effect on
the performance of 3D printed part in the different applications. For example, the compound eyes
inspired optical components fabricated by 3D printing, cannot compete with the natural one due
to the stair-effect. This is because the unsmooth surface decreases the contrast and image sharpness
of 3D printed optical components [109]. Similarly, the artificial blood vessel can’t reach the same
function as natural one because the unsmooth surface increases the drag force [110]. Learning
from examples, we can know that the fabrication accuracy of AM plays important roles in the
biomimetic applications, the fabrication accuracy of layer based AM restricts and impacts the
development of biomimetic applications.

Figure 9 An illustration of the layer-based AM process and related stair-stepping effect.

20
(2) The fabrication scale of AM cannot meet the requirements of biomimetic design
The current AM can only achieve the fabrication of object with a single scale size with the
ratio of fabrication extent to fabrication resolution at around 1000 [59]. For example, in laser-
based SL, microscale structures can be fabricated by controlling the beam size of laser projection.
However, the smallest feature fabricated with a microscale laser beam can only achieve 10 - 50
μm, and it is difficult to fabricate smaller structure due to the limitation of optical components
[111]. Meanwhile, it is not time efficient to build large scale structure with microscale laser beam
because the laser beam has to move over all the inner area of the building part [111]. The smaller
beam size is advantageous for the microscale details fabrication, but it controversially cost longer
time for the fabrication of macroscale object. In contrast, MIP- SL can accomplish high-resolution
fabrication within a short amount of time due to the process planning, where one full layer of
material can be solidified with single light exposure. Since the DMD has millions of micro mirrors
and each represents one pixel in the focusing image, the projected mask image can achieve high-
resolution fabrication [66, 67]. Using different optical designs, the projection image can have
different dimensions, ranging from several millimeters to several hundreds of millimeters [60-70].
Due to fixed optical design and limited interchangeability, the machine’s fabrication scale is
intrinsically defined by configuration and restricted to a certain range. Although superior to laser-
based printing process, the varying scale of fabrication with single MIP-SL is still costly to
implement, and the tradeoff exists between overall building size and smallest printing capability
[67]. However, multi-scale hierarchical structures in nature contain features from macroscale to
nanoscale, where the ratio of the size span to the smallest geometry features could be in the range
of 10
8
- 10
9
[84]. It is pretty difficulty to fabricate multi-scale biomimetic design by using current
AM (refer to Figure 10) [34].

21

Figure 10 An illustration of the fabrication scale of photopolymerization based AM processes.
Thus, we believe the current AM process can’t meet the requirements of fabrication of
biomimetic design with multi-scale structures. It is necessary to develop multi-scale manufacturing
tool for the fabrication of biomimetic design. The key question in this dissertation is:
How to develop a high accurate photopolymerization based AM process to reproduce multi-
scale biomimetic design for functional applications.
To achieve fabrication of multi-scale biomimetic design, a feasible solution is to leverage
multiple AM manufacturing processes through synthesizing and integration. Following this idea,
two problems need to be solved in sequence to answer the main question in this dissertation:1)
how to develop a high accurate AM tool to fabricate microscale features? and 2) how to reproduce
multi-scale biomimetic structures by using developed AM tool ？To find the solutions of these two
problems, the principle goal, research opportunities and hypotheses will be discussed in section
1.3.

22
1.2.2 Motivation of biomimetic AM tool
As stated, the fabrication of multifunctional structures become possible through innovation
and integrations of AM technologies at different scales, and inspires advances in material
development and functional design in different fields [23]. Due to the development of research on
biological structures, design and fabrication of biomimetic system for biological functional
integration have become increasingly significant [34]. The development demand of functional
system drives the need for engineers to design and manufacture biomimetic multi-scale structures.
According to the functionality requirement, the original surface property of the existing model can
be modified by growing specific biomimetic multi-scale structures on its surface (refer to Figure
11) [112]. Fabrication of multi-scale 3D bio-inspired features on a pre-existing physical object
may have immense applications since bio-inspired structures exhibit inherent multifunctional
integration (refer to Figure 11), such as self-cleaning, drag reduction, water collection, strength
enforcement, antireflection, and oil removing, which are of commercial interest. The surface
modification by adding biomimetic structure is desire and sometimes is critical to achieve certain
functionalities.
To achieve the biological function through modification with bio-inspired multiscale structures
is a complex process and comprehensive study [83, 86]. In a typical biomimetic application, how
to mimic the structure in nature, enabling it performs the similar function, turns to be a critical
problem in the beginning stage. Basically, it is necessary to reveal and understand the scientific
theory behind the functional appearances of multi-structures in nature (refer to Figure 7). Based
on the biological study, the artificial multi-scale structure can be designed with optimization for
superior functions. For example, the surface wettability is changed from hydrophobic to super-
hydrophobic, when the surface is decorated with bionic microscale pillar array inspired by the rose

23
petal [114], and the detail of rose petal inspired structure can be further optimized in all scales to
satisfy functional requirements [23, 114].

Figure 11 Function modification by adding bioinspired microstructures. (a) Lotus leaf inspired
self-cleaning microstructures; (b) shark skin inspired drag force reduction;(c) cactus inspired
water-collection microstructures;(d) gecko feet inspired adhesive microstructure;(e) fly eye
inspired antireflective microstructures; and (f) salvinia molesta inspired oil cleaning
microstructure.
After that, the manufacturing of biomimetic design is the second stage in the whole
development process. Not only the design of biomimetic structure is important, but also the
material is critical to achieving a certain function. The biological system performs the attractive
function together with the help of the unique 3D shape and material composition [113]. For
functional consequences, the material selection and preparation is a vital question to answer ahead
of the fabrication. In the AM of biomimetic design, it includes material development,

24
manufacturing process development, manufacturing process planning, and manufacturing. A
rational design, proper material, and the reproducible construction are essential to a development
of biomimetic application, which requires the integration of researches in multiple fields [23].
Among all the research fields, the work in this dissertation will focused on microscale AM
process development, process modelling and parameter optimization, which enables to build
microscale biomimetic structures around the inserted macroscale object, and facilitates the hybrid
process development for the fabrication of multi-scale structures ranging from macroscale to
nanoscale (refer to Figure 12).

Figure 12 An illustration of the timeline for biomimetic functional devices development and my
research focus.

25
1.2.3 Challenge of biomimetic AM tool
While how to fabrication multi-scale biomimetic features on the pre-existing model is a big
challenge for biomimetic AM tool. To build around insert, we can get inspirations from traditional
manufacturing methods. Computer numerical control (CNC) machining is subtractive
manufacturing to cut the material following the tool path. In our previous work, inspired by
traditional CNC machining, a CNC accumulation fabrication method was developed, which the
material can be accumulated under light exposure during the movement of accumulation [115,
116]. Both point-based and line-based (CNC) accumulation methods are developed to build
macroscale features on the surface of a pre-existing object [115-118]. For the point based CNC
accumulation method, the point-shaped light beam is used to add parts on the surface of the pre-
existing object, and the surface quality of part printed by CNC accumulation is much smoother
than other layer based AM approaches [117, 118]. For the line based CNC accumulation, the light
tool can sweep the line shaped light beam to build mesoscale features on the surface of immersed
object [117]. However, both methods are time-consuming to fabricate features with fine details,
since the light beam has to go through all the area, where the liquid resin need to be solidified.
Besides, the multi-scale biomimetic structures are hard to be reproduced by using both above
methods due to the limitation of fabrication resolution. Meanwhile, there is difficulty to implement
building around inserts by using most of the existing AM processes, which has the flexibility to
print high-resolution microscale features [42, 67]. To fabricate high-resolution biomimetic
structures on the surface of the macroscale substrate, we can develop surface based CNC
accumulation process by using 2D patterned light beam. To accomplish this method, two
challenges of surface based CNC accumulation are figured out in this dissertation.

26
1) The stereolithographic exposure of 2D patterned light beam have impact on the
fabrication quality of biomimetic design
In the photopolymerization based AM process, liquid resin is selectively cured by the exposure
of point-shaped laser or 2D patterned projection light, and 3D shape is formed by stacking the
cured material layer by layer [58]. Therefore, the resolution of printed parts mainly depends on the
dimension of a laser beam or the resolution of 2D patterned light beam, and the curing performance
of photosensitive material with the controllable exposure of laser or 2D patterned light beam [50,
58, 67, 119]. For the physical optical system, light distortion, commonly existing in
photopolymerization based AM process, has remarkable impacts on the AM process accuracy in
the horizontal plane. Especially for microscale AM process, the distortion of light generates extra
over-cured features, which changes the original desired shape (refer to Figure 13). One question
that has baffled the surface based CNC accumulation is how to reduce the influence of light
distortion on the printing process, and to generate high resolution 2D patented light beam [67].
In the photopolymerization based AM process, liquid photopolymer is solidified by the
exposure to the electromagnetic radiation over the practical range of wavelengths including visible
and ultraviolet (UV) [38]. Typically, the photocurable polymer is consisted of monomers,
oligomers, stabilizers, and photoinitiator [38, 39]. Monomers molecules are reacted together to
form complex polymer chains or three-dimensional networks, which is initiated by the absorption
of visible or ultraviolet light [68]. Because the photopolymerization process is inherently
anisotropic, the curing performance of resin in the vertical and horizontal plane showed different
characteristic [40].

27

Figure 13 The light distortion in the microscale MIP-SL. (a) The schematic diagram of the MIP-
SL process; (b) the CAD model of a mesoscale part with 100µm pillar array; (c) focusing image
of the microscale MIP-SL with the projection area; (d) the fabrication result of microscale features
using the microscale MIP-SL process.
As shown in Figure 14, the light penetrates the photocurable material, and the crosslinking of
photocurable polymer will occur when the polymer receives enough energy [40]. The curing depth
Cd of photocurable polymer is determined by how deep the light can penetrate through the
photocurable polymer [119-121]. If the penetration depth of light is smaller than the layer thickness,
the new cured material cannot attach on the surface of previous layer. If the penetration depth of
light is much larger than the layer thickness, the penetrated light will solidify the photocurable
polymer, which is located at the previous layer (refer to Figure14). The Cd of commercial
photopolymer, for example, SI500, E-glass (purchased from EnvisionTEC.inc), G+ (purchased
from Maker Juicy) used in conventional SL process is in the range of 75 μm-150 μm. The cure
depth of material is suitable for macroscale fabrication with a strong bonding force. However, in

28
microscale MIP-SL, where the layer thickness only spans from 20 μm to 50 μm for microscale
features fabrication, the material with large light penetration is usually prone to have over cure at
pre-building regions in Z direction (refer to Figure 14). Besides, the cure depth changes
dramatically with slight modification of the exposure energy [120]. In order to achieve high
accuracy, the material should be developed based on the printing requirements and the cure
performance of material in vertical direction should be well studied.

Figure 14 An illustration of curing effect of photopolymer in vertical direction (a) the cure depth
of material equals the layer thickness; (b) the light penetrates the previous cured area and
generates over-curing features in Z direction; and (c) the light penetrates the previous cured area
and block the hole in Z direction.
Meanwhile, the cure performance of photopolymer has an impact on the resolution of 3D
printing in the horizontal plane (refer to Figure 15) [38, 40]. The light energy distribution of each
pixel in the projection light follows Gaussian function, and the dimension of a cured feature in a
horizontal direction is equal to the width of light beam, of which the energy is bigger than the
critical value Ec [58, 61]. Thus, the light beam size determines the smallest feature, which can be
fabricated in the photopolymerization based AM process. The critical value of

29
photopolymerization has relation with the concentration of light absorber and photoinitiator [38].
Using the same light beam, the width of a polymerized feature turns to be slightly larger by adding
more photoinitiator. Conversely, larger concentration of light absorber reduces the reactivity of
the resin, and further increases the threshold energy, which is necessary for the
photopolymerization process. As the result, the width of a polymerized feature turns to be slightly
smaller. Therefore, the curing properties of material in vertical direction is another key factor,
which determines the fabrication accuracy of surface based CNC accumulation process.

Figure 15 An illustration of curing effect of photopolymer in horizontal direction. (a) the channel
can be fabricated accurately without over-curing features in horizontal direction;(b) the channel
was blocked due to light accumulation of neighbor pixels; and (c) the printed channel turned to
be bigger due to the weak light energy.
2) Biomimetic design contains hierarchical structures ranging from macroscale to
Nanoscale
Bioinspired hierarchical structures, ranging from macroscale to nanoscale, is necessary to
achieve certain functionalities. A lot of AM tool are developed to fabricate structures at specific
range of scale. The review of current progress of multi-scale AM will be discussed in Chapter 2.

30
The challenges in hierarchical structures fabrication is lack of biomimetic AM tool, which can act
as a bridge to connect macroscale fabrication and nanoscale fabrication.
Therefore, the research presented in this dissertation is conducted based on light energy control
and hybrid manufacturing. Based on the light energy control, the fabrication quality of biomimetic
design will be studied. Based on hybrid manufacturing, multi-scale fabrication process will be
developed.
1.3 Research focus in the dissertation
1.3.1 The principle goal, research questions and hypotheses
The research goal of this work is the development of multi-scale photopolymerization based
additive manufacturing processes with integration of bioinspired design methodology and
functional programmable material to achieve the investigation of biomimetic functional
applications. The statement of primary research questions is listed below:
Corresponding to the primary research questions, two following hypotheses are put forward.
Each hypothesis describes the sub goals of this dissertation and is investigated in this research.
Q 1: How to fabricate high accurate microscale features by using photopolymerization
based microscale AM process?
Q 2: How to reproduce multi-scale biomimetic design for functional applications?
Hypothesis 1. Microscale surface texture can be fabricated by a novel process called
Immersed Surface Accumulation (ISA).
Hypothesis 2. The hybrid process by integrating ISA and other manufacturing
processes can fabricate multi-scale functional biomimetic structures.

31
There are several challenges in each research problem. Based on the challenges discussed in
Section1.2.3, we divided each research questions into three sub research questions. To solve each
question, we come up with relative hypothesis. The first question is related to high accurate
microscale fabrication, and the sub problems and corresponding hypothesis are listed below:

The second hypothesis is related to the multi-scale fabrication for biomimetic design. As
discussed in section 1.2.3, the integration of a group of photopolymerization processes, whose
capability ranges from microscale to nanoscale, will be used to address the dilemma between
building efficiency, accuracy, and capability. Furthermore, based on the ISA process, which
features the capability of building microscale detail on the pre-printed object and providing
Q 1.1: How to improve the surface quality of printed microscale features by using ISA
process?
Q 1.2: How to fabricate microscale structures with complex geometry by using ISA
process?
Q 1.3: How should composite material be prepared for the microscale fabrication of
ISA to achieve nanoscale functionalities?

Hypothesis 1.1: ISA process can provide continuous fabrication for biomimetic
functional design.
Hypothesis 1.2: ISA process can fabricate microscale features with complex geometry
by controlling light energy distribution with special designed 2D patterned light beam.
Hypothesis 1.3: Carbon Nanotube based composite material can be fabricated by ISA
process at microscale level and achieve nanoscale functional textures at the same time.

32
structural support for nanoscale fabrication, the integration researches are conducted for building
multi-scale hierarchical biomimetic functional structures with high surface quality in short amount
of time. The integration of multiple photopolymerization processes raises challenges over multiple
areas and puts forward problems to address Each sub research questions and the corresponding
hypothesis are list below:

We introduced each sub question in different chapter and talk about corresponding hypothesis
and validation plan. Table 3 showed the relationship between research hypotheses, validation, and
the context of dissertation. The development of ISA process was described in Chapter3, and the
parameter optimization of ISA process was introduced in Chapter4. Chapter 5 showed one test
case of microscale biomimetic structure fabrication by using layer based ISA. Chapter 6 showed a
case study of multi-scale biomimetic structure fabrication by continuous ISA. The hybrid process
development by integrating ISA process with other scale AM process was presented in Chapter 7.
Two cases were demonstrated by using the proposed hybrid process in Chapter 7.
Q 2.1: How to accurately build microscale textures on the surface of macroscale
substrate?
Q 2.2: How to add nanoscale features on the surface of microscale substrate fabricated
by ISA process?
Q 2.3: How to solve the transition problem between two different processes?

Hypothesis 2.1: ISA process can integrate with macroscale MIP-SL process.
Hypothesis 2.2: ISA process can integrate with nanoscale TPP process.
Hypothesis 2.3: Geometry optimization of multi-scale biomimetic design can provide
the support of printed features at different scale.

33
Table 3 The distribution of hypotheses and validation in the dissertation
Hypothesis1
Microscale features fabrication
Hypothesis2
Multi-scale features fabrication
Question Q1.1 Q1.2 Q1.3 Q2.1 Q2.2 Q2.3
Hypothesis H1.1 H1.2 H1.3 H2.1 H2.2 H2.3
Description § 3.1, § 3.6
§ 4.3, § 4.4
§3.5, § 3.6
§ 4.1, § 4.5
§ 5.3, § 6.4 § 3.3, § 7.3 § 3.3, § 7.3 § 3.3, § 7.3
Validation § 4.7, §5.6,
§7.5
§ 6.5 § 5.6, § 6.6 § 7.4 § 7.5 § 7.4, § 7.5

1.3.2 Validation design and planning
There are two type of validation methods: qualitative validation and quantitative validation
[122]. In this dissertation, both theoretical and empirical analysis methods were used to validate
hypotheses. In the first problem, a novel microscale ISA process was developed to fabricate
microscale features. The fabrication capability of ISA process was validated qualitatively. To
improve the fabrication quality, mathematic models of main fabrication factors were fitted based
on both theoretical and empirical analysis. In the second problem, ISA process was integrated with
other AM process to fabricate multi-scale structures. Firstly, empirical validation was conducted
to prove the fabrication capability of proposed hybrid process. After that, theoretical analysis was
applied to construct the mathematical model, which could be used to solve the transition problem
in the integration of multiple process. Table 4 shows the detail of validation design and planning
in this dissertation.

34
Table 4 The validation design and planning in the dissertation
Hypothesis1
Microscale features fabrication
Hypothesis2
Multi-scale features fabrication
Validation
Theoretical
qualitative
validation
Theoretical
quantitative
validation
Empirical
qualitative
validation
Empirical
quantitative
validation
Theoretical
qualitative
validation
Theoretical
quantitative
validation
Empirical
qualitative
validation
Empirical
quantitative
validation
Chapter3

Chapter4

Chapter5

Chapter6

Chapter7

1.3.3 Contributions from this dissertation
AM process provides a potential solution to fabricate multi-scale biomimetic structure and
materials. A lot interesting application have been demonstrated in the field of biomimetic
fabrication by using AM technologies [34]. However, it is still big challenges to reproduce
biomimetic structures and material by using current existing AM processes, because AM
technologies still have some shortcomings, such the surface quality, fabrication accuracy,
fabrication scale capability, etc. As a novel biomimetic AM tool, ISA was developed to fabricate
microscale features, and it has the capability to build biomimetic microscale features with 3D
complex geometric shape on the surface of macroscale substrate. What’s more, due to flexible
light guide tool, it is easy to integrate with other manufacturing technologies, such as MIP-SL [36],
two photon polymerization (TPP) [123]. It provides a bridge that connect the macroscale

35
manufacturing technologies with nanoscale manufacturing technologies by using microscale
fabrication. The contributions of this dissertation are listed below:
Contribution related to microscale fabrication:
 Develop a novel microscale printing process, enabling to build microscale biomimetic
structures on the surface of the macroscale substrate with flexible light guide tool;
 Generate the energy distribution model of the 2D patterned light beam for the radical
polymerization in microscale fabrication;
 Experimentally model the curing performance of composite material with the exposure
of 2D patterned light beam and develop the appropriate photocurable material for
different functionalities;
 Experimentally study physical models of fluid flow in the printing process and optimized
the process parameters;
 Developed ISA based continuous microscale printing process to achieve smooth surface
quality.
Contribution related to multi-scale biomimetic design fabrication:
 Develop algorithm to design transition region to support the structures fabricated by the
manufacturing process at different scale;
 Experimentally explore the integration of ISA with other photopolymerization based
printing process by using vision assisted system;
 Investigate the integrated multi-scale manufacturing process solution for the fabrication
of biomimetic functional material and structures, and formulate quantitative models to
characterize and optimize the manufacturing process parameters.

36
1.4 Outline of dissertation
Due to the credit of unique fabrication capability, recent advances in AM technology have
shown progressive achievements in different fields of applications, including the fabrication of
nature-inspired, functional material and structures [124]. As various applications implied, each
type of AM processes is built to work with its most appropriate fabrication scope in which the
optimized resolution is applied. It is preferred to fabricate features at different scale with AM
process whose scope best suits the purposive fabrication. Therefore, for fabricating biomimetic
functional materials and structures at multi-scale, the development of the AM process requires
both process development for specified scale and systematic integration of multiple appropriate
processes [125]. Such integration of these scale-exclusive fabrication methods, in turn, demands
the compatibility of each sub-process, and necessary process optimization during integration
between different methods. The aim of this research work is to systematically study
photopolymerization based AM processes ranging from macroscale to nanoscale and then develop
new fabrication approach named as ISA to solve the challenges in multi-scale AM fabrication of
bio-inspired functional material and structures which have multiple applications in different fields.
Figure16 shows the outline of this dissertation. The recent progress in AM of biomimetic
structure and material is firstly introduced in the Chapter 1. The foundations required for this
dissertation are presented in Section 1.1. The research opportunity in the field of AM of biomimetic
structures is discussed in Section 1.2. It also shows why this dissertation is conducted. The
motivation and challenges of biomimetic AM also are presented in that section. Based on the
discussion, the research focus of this dissertation is put forward, and the principle goal, research
questions, and corresponding hypotheses is stated in detail. The validation design and planning is

37
described, and the contribution of this dissertation is also summarized in Section 1.3. The status
of AM process development in the biomimetic multi-scale fabrication is reviewed in Chapter 2.

Figure 16 Outline of dissertation
All the information related to the ISA process development is introduced in Chapter 3. It
includes prototype machine design and development, optical system design, and process planning.
Chapter 4 elaborates on the process modeling and parameter optimization of the ISA process.
Chapter 5 and Chapter 6 present a case study of the layer based ISA and the continuous ISA,
respectively. How to select material and fabricate microstructures by using newly developed ISA
processes and how to achieve the high accurate reproduce of multi-scale biomimetic functional
material and structures are discussed in both Chapters. Chapter 5 introduces cactus spines inspired
high efficient water collection surface. A superhydrophobic and superoleophilic surface decorated
with salvinia molesta inspired eggbeater shaped structure is introduced in Chapter 6. Chapter 7
discusses the ISA-based hybrid process development, and two case studies of the hybrid process
are demonstrated in this chapter. In Chapter 8, all the research questions raised from Section 1.3
are answered with a discussion of the findings in this dissertation. The achievements and
contributions including engineering achievements and scientific findings are explained.

38

Chapter 2 Literature Review: AM Process Development for
Biomimetic Fabrication
During the past thirty years, AM technologies have been highly developed for the purpose
of fabrication of 3D objects with complex geometric shape, and in most of AM processes material
is deposited in a layer manner [126]. Compared with traditional manufacturing methods, additive
manufacturing is more flexible to build the 3D object with biomimetic design [126, 127]. Due to
the advantages, such as, simple operation, high performance and low cost, AM technologies have
been extensively applied in the fields of bioinspired material and structure fabrication, including
mechanics, shape changing, electrics, optics, and interface (refer to Figure 17) [23, 34].
With increasing researches on the design of multi-scale biomimetic structures for unique
functionalities, it is necessary to develop multi-scale fabrication process for the fabrication of
biomimetic design [23]. Taking a look at the functional biological material system, some common
features can be figured out. They are all composed of multi-scale and multi-material structures,
and the special functionality is determined by the unique multi-scale features and material
distribution [23]. However, most of the current AM technologies are developed to fabricate single
material, and only certain range of scale can be fabricated by current AM technologies [23, 34].
Moreover, functional biological material system, such as shark skin, adhesive gecko feet, large
field viewed fly eyes and superhydrophobic butterfly wing, etc. are composed of functional
hierarchical features ranging from macroscale to nanoscale (refer to Figure6) [23, 83, 84]. In the
past, a molding process was applied to fabricate biological structures with simplified design due
to the limitation of fabrication capabilities. Therefore, the performance of fabricated part with
single scale features cannot be competitive with the natural one [128, 129]. Due to the advantages

39
of AM, AM provides a potential solution of the fabrication of multi-scale biomimetic design [23,
34].

Figure 17 An illustration of status of AM of bioinspired material and structures [34].
To solve the challenges of multi-scale biomimetic structure fabrication, the development of
novel multi-scale AM processes draws more attention in recent research [23]. The straight forward
solution is to integrate multiple AM processes, and each of AM process is focused on the
fabrication of features at one specific range of scale [130]. By using this integration method,
different sections of the multi-scale design are built with optimized process [130]. It shows more
flexibilities and higher fabrication capabilities than the single printing process [23]. However, how
to integrate multiple types of AM process to achieve the fabrication of object with multi-scale
bioinspired structures is still a significant challenge. Another solution is to develop one single AM
process with the multi-scale fabrication capabilities. This solution provides more flexibilities for

40
the design of bioinspired structures, and it simultaneously eliminates the transition problem
between different AM processes by using the hybrid process.
2.1 Multi-scale fabrication based on single AM process
Recent progress on multi-scale AM technologies facilitates the performance study of multi-
scale biomimetic structures. Most of functional biological structures is made by functional
composite material [34]. Three types of AM processes are widely used to form functional
composite into multi-scale shape. They are direct ink writing based multi-scale AM process,
photopolymerization based multi-scale AM process, and inkjet printing based multi-scale AM
process [23]. In this section, multi-scale fabrication by using single AM process will be discussed.
2.1.1 Direct ink writing
In direct ink writing, the functional composite can be continuously extruded from the nozzle
tips, and the material is accumulated to form 3D structures followed by the tool path of nozzle tip.
By using microscale nozzle tips, multi-scale features can be fabricated by direct ink writing as
shown in Figure 18 and 19. The fabrication resolution of direct ink writing is determined by
multiple parameters, such as the moving speed of nozzle, the material feeding speed and the
dimension of nozzle [131-136]. For example, the dimension of printed line can be reduced by
increasing the moving speed of nozzle or reducing the material feeding speed [131]. The smallest
feature that can be printed by direct ink writing is determined by the size of nozzle, and the length
of tool path is also related to the size of nozzle. Due to the special printing capability, direct ink
writing is widely used in the fabrication of multi-scale biomimetic structures, e.g., microvascular
network, artificial organ, optical components [137, 138].

41
For example, because of the lack of vascular perfusion, it is challenging to print large tissues
with a complex geometric shape. Direct ink writing based process provides a solution to fabricate
the fluidic vascular channel, and 3D printed two fluidic vessels formed a microvascular bed for
the cell growth study [136]. As shown in Figure 18, a microscale nozzle tip is immersed inside the
tank filled with high viscous gel, and the biocompatible hydrogel was injected from the nozzle tip
[133]. The nozzle tip moved along the tool path generated based on the 3D geometry of vascular
structures, and multi-scale vascular network can be fabricated after support gel turned to be liquid
with the change of temperature from room temperature to 0 degree [134]. Based on this method,
multi-scale complex structures can be printed, and the dimension of printed features ranges from
100 μm to 10 mm [131-135].

Figure 18 AM of multi-scale biomimetic vascular channel by using direct ink writing. (a) 3D
printed microvascular networks [133]; and (b) multi-scale fluid channel fabricated by gel printing
[135].
What’ more, a multi-scale biomimetic spiders spin can be fabricated by using direct ink
writing (refer to Figure 19a) [139]. The 3D printed webs made by the artificial spin shows

42
promising characteristics of mechanical performance such as high strength, elasticity, and graceful
failure [139]. Meanwhile, direct ink writing provides a fabrication tool of functional composite
material to develop advanced optical devices with multi-scale biomimetic structures and. For
example, bioinspired multicolor shifting pattern was printed by using colloidal photonic crystals
(refer to Figure 19b). The special mesoporous colloidal nanoparticle ink was developed to achieve
the vapor-responsive color shift, and the size and mesoporous proportion of nanoparticles were
adjusted to change the color of printed pattern under different vapor conditions [140]. Not only
multi-scale vascular structures can be fabricated by using direct ink writing, but also the multi-
scale artificial organ can be fabricated by using direct ink writing. To fabricate multi-scale artificial
organ, a special nozzle was developed by using a bundle of capillaries, and each capillary
connected with its own material supply (refer to Figure 19c) [141]. Different types of cell were
cultured on these artificial organs, which were fabricated by direct ink writing [141].

43
Figure 19 AM of multi-scale biomimetic design by using direct ink writing. (a) 3D printed multi-
scale spider web [139]; (b) 3D printed vapor-responsive colloidal photonic crystal patterns [140];
and (c) machine development of the digitally tunable continuous multi-material printing and 3D
printed artificial organ [141].
2.1.2 Photopolymerization based printing process
In direct ink writing, high-precision motion control of the printing nozzle is necessary to
fabricate multi-scale structures with smooth surface and accurate dimensions. In vat
photopolymerization based printing approaches, microscale features can be printed by using high-
resolution light exposure [16, 68]. However, as we discussed in Section 1.2, the resolution of
printed objects is restricted to the physical dimension of DMD chip or laser beam, which is a
bottleneck of vat photopolymerization in the field of multi-scale fabrication. To solve this problem,
one solution is to generate large area projection image by jointing multiple DMD chips. However,
the light intensity of light beam generated by this approach is not uniform, and it is also hard to
adjust the focus when multiple DMD chips work simultaneously.
Designing a special optical system is another way to achieve the multi-scale fabrication by
using MIP-SL process. The 2D patterned light beam can move in the X-Y plane, so that the large-
scale structure can be fabricated in a sequence using the same high-resolution light beam.
Specifically, the light source is mounted on the moving stage, and each time the light is moved
sideways [69, 142]. This approach meet the problem of image planning and overlapping between
two adjoining printed sections [142, 143]. To solve the overlapping problem in this method, the
projection image can move sideways with only one pixel distance, and each portion of printed part
requires light illumination with several times [143]. Figure 20a shows a scanning projection

44
printing system, which is developed based on the moving light approach. By using scanning MIP-
SL, 5 cm large area can be fabricated with microscale features at 10 μm [144].

Figure 20 AM of multi-scale structures by using Vat polymerization based printing method. (a)
The scanning projection printing system and the side view of printed multi-scale lattice structures
[144]; and (b) a schematic diagram of shaped beam based multi-scale printing process [50].

45
Pan etc. developed a multi-scale printing process, which the laser spot can dynamically
change the focus. By using different size of projection image, the structure with a large range of
scale can be fabricated [65]. Besides, the optical filter with high-contrast gratings can be used to
adjust the dimension of laser dot, and different scale of features area able to be fabricated by the
corresponding laser beam [50, 111]. In this method, the large-scale laser dot is used to fabricate
the inner section, and the boundary features is fabricated by using laser with smaller dimension.
By using this approach, the fine detail of designed object can be produced. Figure 20b shows the
process planning of shaped beam based multi-scale printing process [111]. To further improve the
printing speed and resolution, the large size laser can be replaced by the DMD based projection
light.
2.1.3 Inkjet Printing
Inkjet printing is one type of AM technologies that the droplets of ink is propelled onto the
substrates and formed into 3D shape. The shape of droplet determines the smallest features that
can be printed by Inkjet printing, and the shape of droplet can be precisely controlled by the shape
deformation of piezo plate [20, 145]. Due to the process character, the inkjet printing is widely
used to fabricate electronic devices and optical devices [20, 48]. For example, multi-scale micro
lens array was fabricated by using Inkjet printing as shown in Figure21a [146]. The geometrical
shape of deposited droplet can be controlled by modifying the wettability of surface. Inkjet printing
shows advantages in reproducibility, and the deviation of printed structures in Z direction is less
than 1 μm [146]. Without additional process, microscale array can be fabricated by using this high-
resolution Inkjet printing [146]. Unlike the photopolymerization based approach, which only
photocurable material can be printed, large range of material can be printed by Inkjet printing. For
example, photonic crystal can be fabricated by Inkjet printing, and the shape of printed micro

46
droplet is modulated by changing the contract angle of subtracts and the concentration of particles
in the print ink (refer to Figure 21b) [48]. Moreover, multi-scale micro lens array was fabricated
on the curved substrates by Inkjet printing as shown in Figure 21b [48].

Figure 21 AM of multi-scale structures by using Inkjet printing based method. (a) Multi-scale
micro lens arrays fabricated on flat surface by inkjet printing [146]; and (b) wide viewing angle
displays fabricated by Inkjet printing [48].
2.2 Multi-scale fabrication based on hybrid AM process
To fabricate multi-scale structures, researchers also developed hybrid process by integrating
multiple processes, including AM process [147], subtractive manufacturing methods [148], and
Nanotechnologies [149]. Both multi-material and multi-scale structures can be easily fabricated
by the hybrid process. In this section, AM based multi-scale hybrid process development for
different applications will be discussed.

47
2.2.1 Photopolymerization based hybrid process
In the photopolymerization based printing process, the photocurable polymer will be
selectively cured by projecting the light. It is easy to combine photopolymerization based printing
process with other manufacturing approaches. For example, multi-jet modeling and microscale
MIP-SL were integrated to be a hybrid process to fabricate multi-scale swirling flow coaxial
phacoemulsifier sleeve (refer to Figure 22a) [130]. Because of special fabrication capability, such
as fabrication quality, geometry complexity, photopolymerization based multi-scale process is
used to fabricate the biomimetic design that are hard or even impossible to implement by other
AM processes. For example, eagle eye inspired objectives lens was fabricated by using
photopolymerization based AM process, and was decorated onto a complementary metal-oxide-
semiconductor (CMOS) image sensor (refer to Figure 22b), which is fabricated by traditional
nanotechnology [109]. A full field of view of 70° was achieved by this bioinspired optical system
[109, 150]. Similarly, two-photon direct laser writing was developed to fabricate microscale and
nanoscale bionic lens array, and the field of view of 80° can be accomplished by printed micro-
lens, of which the diameter is only100 μm. This printing process has the capability to integrate
with other printing process to fabricate multi-scale biomimetic design. For example, 3D scaffolds
were fabricated by using the hybrid process which combines direct laser writing with fused
filament fabrication [151]. As shown in Figure 22c, the scaffold was firstly fabricated by using
fused filament fabrication process also known as FDM, and then the printed scaffold is immersed
into a monomer solution to build the inner structures by using direct laser writing [151]. This
proposed hybrid process can be applied in the fields such as microelectromechanical system,
microfluidic, tissue engineering [151].

48

Figure 22 Multi-scale structure fabricated by photopolymerization based hybrid process. (a)
Multi-scale swirling flow coaxial phacoemulsifier sleeve fabricated by hybrid process [130]; (b)
eagle eyes inspired imaging system with lens array fabricated by multi-scale process [109]; and
(c) PLA based scaffold fabricated by thermal extrusion printing with direct laser writing [151].
2.2.2 Extrusion based hybrid process
To fabricate scaffold with multi-scale structures, extrusion based AM process is combined with
other manufacturing technologies, such as freezing dying [152], wet spinning [153], melt
electrospinning [154], and (d) electrohydrodynamic (EHD)-jet process [155]. The key scaffold
architecture can be fabricated by using extrusion based printing process, and the microstructures
can be obtained by other manufacturing equipment. Scaffold with both compartmented and

49
integrated architectures can be fabricated by using extrusion based hybrid process. The bonding
between each material fabricated by different process is strong enough to support the whole
structures, and the porosity of fabricated scaffold can reach 60 % - 98 % [152-155]. Although an
impressive scaffold was fabricated by the extrusion based hybrid process, it is hard to fabricate the
multi-scale biomimetic design with complex geometric structures, due to the fabrication limitation
of extrusion based approach [156].

Figure 23 Hybrid process by integrating extrusion based printing process with different process.
(a) Freezing dying [152]; (b) wet spinning [153]; (c) melt electrospinning [154]; and (d)
electrohydrodynamic (EHD)-jet process [155].

50
2.3 Summary
A lot of research have been conducted to develop new AM process to solve the challenges of
multi-scale fabrication. One direction of multi-scale fabrication is based on the single AM process.
Direct ink writing, photopolymerization based printing, and Inkjet printing based multi-scale
printing process were introduced in this section. Another way to fabricate multi-scale structures is
based on hybrid process, which is developed by combining more than one manufacturing process.
In hybrid process, both photopolymerization based AM process and extrusion based AM process
are widely integrated with other manufacturing process. The structures fabricated by hybrid
process can obtain the complex geometric shape as same as the single process, but the integration
of different manufacturing process limits the advancement of the hybrid process.
Overall, the promising functional material system consisted of hierarchical structures, where
significant discrepancies in dimension was present among sections. Such a hierarchical structure
is difficult to build by traditional manufacturing process due to the limitation of manufacturing
capability. Nowadays, three-dimensional objects with complex structures can be printed by
additive manufacturing (AM), which material is gradually accumulated layer by layer. However,
the hierarchical structure measured from macroscale to nanoscale also raises the challenge to single
AM process, whose manufacturing capability is intrinsically specified within a certain scope. It is
desired to develop one multi-scale 3D printing process narrowing the gap between scales of feature,
with the capability to fabricate hierarchical structures.
51

Chapter 3 Immersed Surface Accumulation: Microscale Surface
Texture Fabrication
3.1 Principle of Immersed Surface Accumulation
Computer numerical controlled (CNC) machining in one type of subtractive manufacturing
method, which the material is removed following the movement of cutting tool. Conversely, the
material could be accumulated along the moving of special tool. Inspired by that, the point-based
and line-based CNC accumulation methods have been developed [115–118]. The basic idea is
shown in Figure 24.
In point based CNC accumulation, the LED as light source is mounted on one side of optical
fiber, and another side of optical fiber is inserted inside the liquid tank filled with photocurable
polymer [107]. The light is projected on the top surface of optical fiber so that the photocurable
polymer is cured along the movement of optical fiber. The optical fiber can be mounted on the 5-
axis stage, enabling building the features around the insert [107].

Figure 24 A schematic illustration of the CNC accumulation system [107]

52
As we discussed in section1.2.3, it is difficult to build the microscale structure with fine detail
by point based and line based CNC accumulation processes. Meanwhile, the existing AM
processes that can fabricate high-resolution microscale features usually lack the flexibility of
building around inserts [157]. The two-photon polymerization (TPP) process is a submicron
printing process that can fabricate complex 3D microstructures with no topological constraints
[123]. However, only certain type of resin can be used in the TPP process and the fabrication
equipment of TPP is very expensive [158]. Besides, it is also time-consuming to build microscale
structure array in a large area [159]. To fabricate bioinspired surface for hydrophilic and
hydrophobic study, the molding process is widely used due to the simple operation and low cost
[134]. However, the shape of structures fabricated by the molding process is restricted to the
demolding process, and the structures with over-hang or hollow features cannot be fabricated by
molding process [160]. The electrospinning method has also been used to fabricate microscale
biomimetic structures. However, the printed structures are mostly mesh shaped, and it shows
limitation in the geometric complexity of the fabricated part [161]. In addition, the laser-based 3D
microscale printing technology has previously been used to build the artificial shark skin structure
on the surface of plate [86]. Comparing with above microscale fabrication processes, microscale
ISA shows advantages in microscale fabrication. In ISA, layers of material are selectively
deposited with 2D patterned light exposure; in addition, the light intensity of each pixel can be
controlled by a DMD, which enables the high-resolution fabrication [60-62]. DMD based optical
imaging system provides the possibility to print the microscale objects with fine detail [57, 119,
143]. Hence the surface based CNC accumulation (ISA) is developed by DMD based optical
imaging system [162].

53
We extend the point-based and line-based CNC accumulation processes to the surface-based
accumulation process and name it the immersed surface accumulation (ISA) process [162]. Instead
of a single point spot or a line beam, high-resolution 2D patterned light beam was used to
accumulate materials on the surface of an inserted object surrounded with the photocurable
material [163]. If the light can be guided on the surface of object, where we want to add microscale
features, and continuously project the light, the microscale 3D features can be fabricated by the
surfaced-based CNC accumulation process. Furthermore, combining both dynamically controlled
light beam with the multi-axis movement of light guide tool, the surfaced-based CNC
accumulation process can build high-resolution microscale features on the surface of a free formed
object [163]. To achieve the idea of surfaced-based CNC accumulation, immersed surface
accumulation (ISA) was developed.
In this chapter, the porotype machine design of ISA was put forward. The optical image system
was developed to generate a set of high resolution 2D patterned light beams. Similar to the widely-
used CNC machining process, the light guide tools were designed with various shapes and sizes.
To achieve uniformed light distribution, a calibration algorithm was figured out based on the light
energy model. The process planning of layer based ISA and continuous ISA was developed, and
the hardware and software was presented at the end of this chapter.
3.2 Prototype machine design
The prototype machine design of ISA is shown in Figure 25, and it contains three modules:
the light beam generation module; vision assisted alignment module; and multi-axis motion
module. In light beam generation module, a set of light guide tools are designed by combining an
objective lens with optical fiber or acrylic rod and the light beam can reach 2.5 μm per pixel.

54
Different 2D patterned light beams can be dynamically projected during the movement of the light
guide tool to accumulate material into desired 3D shapes. The light guide tool can flexibly add the
microscale features on the inner or outer surface of pre-existing object [112].

Figure 25 A schematic illustration of Immersed Surface Accumulation printing system
3.3 Optical image system
Similar to the point-based CNC accumulation process [115], the ISA process uses a light guide
tool to cure features in liquid resin, and material is accumulated along the moving direction of the
light guide tool [163]. In comparison, the point-based CNC accumulation cannot fabricate
microscale features since it uses accumulation tools whose sizes are over 0.3 mm [116]; in the ISA
process, a light guide tool with DMD-based imaging system is used as the accumulation tool,
which is immersed in liquid resin to build microscale features on a given object. In addition, a tool
changing system is developed to change the accumulation tools that are used to build 3D features

55
on the top or side surfaces of an inserted object. Either the light guide tool or the pre-existing object
can be moved in the fabrication process [163]. In the prototype machine, a bottom-up frame is
used, so that less resin can be used in the fabrication of microscale features.
The optical system of the ISA process contains a bulb, a DMD chip, and a set of lenses. The
power of the light bulb used in our experimental system is 2500 lm. Its illumination beam is first
collimated by a series of lenses (refer to Figure 26). The light source is considered as a point source
and the light from the bulb is collimated before reaching the DMD chip. An illustration diagram
of the accumulation tool with the DMD-based optical system is shown in Figure 26. The DMD
chip can be controlled to individually rotate each of its micromirrors ±10–12◦ to set its state on or
off [164]. In the on the state, the light from the bulb, which is reflected by the micromirror onto
the lens, make the pixel appear bright on the top surface of the light guide tool. In the off state, the
light from the bulb, which is directed by the micromirror elsewhere, make the pixel in the light
beam turns to be dark [163]. The brightness of each pixel in the light beam can be controlled by
the angle of the micromirrors. An achromatic doublets lens (purchased from Thorlabs, Inc.), with
focus fad = 150 mm is used to converge the light beam and to reduce the light distortion [162]. A
filter lens (purchased from Thorlabs, Inc.) is used to block the light with a wavelength other than
405 nm. The collimated light beam then goes through a 4× objective lens with the focus distance
f = 15 mm. The desired dimension of the projection image is determined by the objective lens
(purchased from Thorlabs, Inc.), and the distance between the objective lens and the collimated
lens is not fixed based on the optical design (refer to Figure 26). The dimension of the accumulation
light beam with 4X objective lens used in our system is 3.67 × 2.75 mm. Since the resolution of
the DMD chip is 1920 × 1080, the resolution of the light beam in our system setup is 2.5 μm per
pixel.

56
To dynamically control light beam to fabricate microscale features, the light beam on the tip
of the accumulation tool needs to be focused. However, since the accumulation image is rather
small, the focusing detail cannot be directly observed. Consequently, a vision system is integrated
into the optical design in order to observe the controlled light beam. This vision system contains a
45:55 pellicle beam splitter (purchased from Thorlabs, Inc.), a convex lens (purchased from
Thorlabs, Inc.), and a complementary metal-oxide-semiconductor (CMOS) camera. The light
beam with a 2D microscale pattern goes back to the objective lens, which is reflected by the pellicle
beam-splitter. A CMOS camera is mounted behind the convex lens so that the light beam reflected
by the beam-splitter can go into the convex lens and is finally captured by the CMOS camera. The
captured image is displayed on a computer monitor with 50× magnifications. Based on it, the
accumulation image can be observed to accurately adjust its focus.

Figure 26 The layout of the immersed surface accumulation imaging system

57
3.4 Light guide tool development
In the point-based CNC accumulation process, the accumulation tool is made up of a small
optical fiber that can transmit light from a light-emitting diode (LED) [115, 116]. The liquid resin
can be solidified and accumulated by moving the light guide tool on an object surface [115, 116].
In the ISA process, a light guide tool is also needed to transmit a patterned light beam. We construct
the light guide tool using two types of material, flexible optical fiber, and a rigid acrylic rod [162].

Figure 27 The design of light guide tools. (a) An optical acrylic rod based light guide tool; (b) an
optical fiber with multi-cores based light guide tool; (c) 2D patterned light beam generated by the
optical acrylic rod based light guide tool; and (d) 2D patterned light beam generated by the optical
fiber with multi-cores based light guide tool - the dimension of each core inside the used fiber is
80 m.
(1) One type of light guide tool is developed based on an optical acrylic rod as shown in Figure

58
27a. Since we use transparent optical acrylic rod as the light guide tool, the light which went
through the objective lens transmitted into acrylic rod and directly focused on the top surface of
acrylic rod (refer to Figure 27c). The height of the acrylic rod needs to be adjusted based on the
focal distance of the objective lens to get focused image [162]. The highest resolution of the light
guide tool using the acrylic rod is determined by the DMD-based optical system. A drawback of
the acrylic rod is rigid and it is designed with a fixed length, which is hard to move around. Hence,
for the light guide tool based on the acrylic rob, a multi-axis motion system is used to move the
inserted object instead of the light guide tool.
(2) Because the acrylic rod is rigid, it is hard to use it to add the microscale features on the
surface of pre-existing object with complex geometric shape. The optical fiber is flexible and can
easily mount it to a motion system to position the 2D patterned light beam in different directions
and locations. Furthermore, there is no length constraint for optical fiber to transmit the input light.
Hence, the optical fiber based light guide tool was developed, which enables the DMD-based
optical system to be located anywhere away from the inserted object. For the light guide tool made
by optical fiber, the optical fiber is multi-cores and as shown in Figure 27b there is missing portions
of focusing image due to multi-cores. When the light transmits from the one core in the optical
fiber, it will keep the features. If the light just go through the hollow section, the light cannot reach
the terminal portion of the part. Therefore, there is less influence of focusing quality of 2D
patterned light beam using acrylic rob compared with multi-core optical fiber. The minimum
feature fabricated by this process using acrylic rod is 8 μm. The minimum feature fabricated by
the multi-core optical fiber is 15 μm, since the core size of the optical fiber we use is 80 μm. It can
know that the more precisely microscale features can be fabricated with optical fiber made by
much smaller core dimension [163].

59
The layout of the light guide tool based on the acrylic rod and the optical fiber are shown in
Figure27 a and b, respectively. We mount one side of the optical fiber or acrylic rod on the bottom
of the objective lens with a designed distance, and coated a thin layer of Polydimethylsiloxane
(PDMS) on another side of the optical fiber or acrylic rod is coated with. Because the PDMS can
help the separation of the cured resin from the optical fiber [115]. Thus, the light reflected by the
DMD chip goes through the objective lens and the light guide tool, and generate focused light
beam on the top surface of the light guide tool (refer to Figure 27c and d). Taking advantage of the
vision alignment system, we can adjust the distance between the light guide tool and the objective
lens so that focused light beam can be generated on the top surface of the light guide tool.

Figure 28 An illustration of the fabrication process using light guide tools in ISA process. (a)
Building around inserts using a light guide tool with a straight head tip; and (b) adding microscale
features on a side surface using a light guide tool with 90° tip.
In the ISA process, we choose light guide tool according to the geometric shape of the inserted
object. For example, a light guide tool with a straight head or with a 90 degree tip angle can be
used in fabricating microscale features on surfaces with different curvature (refer to Figure 28). In
Traditional CNC machining system, an automatic tool changing system can be used to change the
machine tools based on the fabrication requirements. Similarly, a light guide tool changing system
was developed for the ISA process to address the needs of building features on different object

60
surfaces. The light guide tools with different head angles and tip shapes can be developed based
on the fabrication requirements [162].
3.5 Calibration of light intensity
The light intensity of a 2D patterned light beam generated by optical imaging system is not
uniform. To illustrate the non-uniformed light distribution, a test pattern, which is used to build
microscale pillar array, is shown in Figure 29a. Under the exposure of the light beam generated by
this test pattern, some portions of the micro-pillars are over-cured, while other portions cannot be
fabricated due to the non-uniformed light intensity distribution (refer to Figure 29c). Such non-
uniformed light intensity distribution of 2D patterned light beam significantly makes impact on
the printing accuracy of ISA process.

Figure 29 The light intensity distribution of 2D patterned light beam. (a) Test pattern sent to DMD
with gray scale level of 255; (b) the focusing image captured by a CMOS camera; (c) the fabricates
result by the tested light beam with an exposure time of 0.8s; (d) the light intensity distribution

61
model of the 2D patterned light beam; (e) the gray scale database of 2D pattern; and (f) the gray
scale level distribution of 2D pattern.
To address this problem, a light intensity adjustment approach was developed, in which the
light intensity of 2D patterned light beam was calibrated and accordingly gray scale values were
adjusted to reach the same light intensity. We assumed that the variation of the light intensity of
neighboring pixels within 30 pixels can be negligible. Firstly, 2D patterned light beam was divided
into a set of blocks. Each block is a square area with 25 x 25 pixels. We set the same grayscale
value for all the pixels in each block, and their grayscale values are initially set at 255. A CMOS
camera is used to capture the focusing image of such a patterned light beam (refer to Figure 29b),
which contains three information components - red, green, and blue values. To calculate the light
intensity from these three colors, we converted the color image into a gray scale G (i, j) by using
the weighted average method (See in Eq. 1). The calculated grayscale levels of the image represent
the light intensity of pixels in the 2D patterned light beam [58, 61]. After identifying the effective
exposure area, the light intensity L (i, j) of each effective block can be calculated (refer to Eq. 2).
The original light intensity distribution of the 2D patterned light beam is shown in Figure 29d.
𝐺 (𝑖 , 𝑗 ) = 0.30𝑅 (𝑖 , 𝑗 ) + 0.59𝐺 (𝑖 , 𝑗 ) + 0.11𝐵 (𝑖 , 𝑗 ) (1)
𝐿 (𝑖 , 𝑗 ) =
∑ 𝐺 (𝑚 ,𝑛 )
𝑚 =25×(2𝑖 +1)𝑛 =25×(2𝑗 +1)
𝑚 =50𝑖 ;𝑛 =50𝑗 25×25
(2)
Based on the light intensity distribution of the 2D patterned light beam, the grayscale values
of the input image were modified; at the same time, the visual detection system was used to
dynamically capture the focusing images of the projection image with different level of gray scale.
For each patterned light beam, the light intensity of each pixel was calculated based on the

62
grayscale levels of the captured image. Furthermore, we used different gray scale levels in the 2D
patterned light beam to build micro-pillars to verify the adjustment. The mathematical model was
established to fit the relation between the light intensity and the grayscale level of the pixels in the
2D patterned light beam. Based on the light intensity models of each block, the gray scale level H
(i, j) of each block can be adjusted to reach the same light intensity value Lmin. Finally, the light
intensity distribution of the whole 2D patterned light beam can reach the same level after several
iterations.

Figure 30 A fabrication result of micro pillar array by using 2D patterned light beam with adjusted
light intensity. (a) The 2D patterned image generated based on the light intensity database to
fabricate micro-pillars array; (b) the focusing image of the 2D patterned light beam captured by
a CMOS camera; and (c) the micro-pillar array fabricated by the 2D patterned light beam.
Using the adjusted gray scale level H (i, j) derived from the light intensity model for each pixel,
the grayscale distribution database for the whole 2D patterned light beam was established based
on the bilinear interpolation (refer to Figure 29 e-f). We successfully fabricated uniform microscale
pillar array at the size of 150  150 m based on the 2D patterned light beam with adjusted light
intensity. The microscope image of the fabricated micro-pillars is shown in Figure 30c. The same
mask image planning method was used to adjust the light intensity distribution for building
microscale features in Section 5.5 and 6.5.

63

Figure 31 Flow chart of process planning for ISA process
3.6 Process planning
To fabricate the microscale features on the surface of pre-existing object, the initial position of
platform will be identified. Based on the material curing performance, the fabrication parameter is
set in the software and the detail will be shown in Chapter 4. A series of 2D patterned images are
generated by slicing the digital model with special thickness. Basically, the slicing thickness of
layer based ISA process is 25 μm, and it will be reduced to only 5 μm when the part is printed by
using continuous ISA process. After that, the gray scale of each pixel in the 2D patterned image
will be adjusted to get the uniform light intensity. The duration of each light beam is modulated
based on the dimension of 2D patterned light beam, which will be discussed in Chapter 4. For

64
layer based ISA, the distance between the platform and light guide tool will be increased after the
fabrication of one layer and at the same time the light is shut off. The light only exposures the
fabrication area after the material flows back to the fabrication area. The above process is repeated
until the fabrication of last layer is finished. For the continuous ISA, the light continuously
illuminate the photocurable resin by the light guide tool with the increasing of the distance between
the light guide tool and building part. The projection light will be shut off until the movement of
the platform stops. The moving speed of light guide tool or the platform will be discussed in
Chapter 4. The detail of process planning of ISA is shown in Figure 31.
3.7 Hardware and software development
To demonstrate the fabrication capability of the ISA, a prototype machine was built as shown
in Figure 32, which has the aforementioned optical imaging system, the vision assisted alignment
system, and the multi-axis motion system. In our motion system, a high-performance 6-axis motion
control board with 28 bidirectional I/O pins from Dynomotion Inc. (Calabasas, CA) was used to
drive the linear and rotation stages [162]. Three precise linear stages were used to drive the inserted
object in the X, Y, and Z axes, and a rotation stages was combined to change the angle of the pre-
existing object so that features can be added from different orientations (Figure 32a). A software
testbed for ISA process has been developed using the C++ language with Microsoft Visual Studio
C++ Compiler. The testbed integrated the mask image planning, the light beam generation, and
the motion control modules [162]. It also synchronized the light beam projection with the
movement of the inserted object. The graphical user interface of the developed software system
for the ISA set-up is shown in Figure 32b.

65
(a)
(b)
Figure 32 The hardware and software development in ISA process. (a) The physical setup of the
surface-based CNC accumulation system; and (b) the graphical user interface of the developed
software system.

66
3.8 Summary
This chapter presents the work related to the design and development of immersed surface
accumulation (ISA) process, including the basic principle, prototype design, light guide tool
development, process planning, and hardware and software implementation. To propagate the
high-resolution 2D pattern light beam, light guide tool as a novel print head was accomplished
with acrylic and optical fiber. A high-speed and resolution 3D texture-printing machine by
integrating accumulation imaging system, multi-axis motion, and light guide tool has been
developed. To uniform the light intensity distribution of light beam, the energy model of light
beam was studied and, the mathematic relation of the gray scale and light intensity was discussed.
After that, the process planning for layer and continuous ISA was showed respectively. At the end,
the hardware and software were developed to demonstrate the fabrication capability of ISA.
67

Chapter 4 Process modeling: Parameter Optimization of
Immersed Surface Accumulation
In the ISA process, the light guide tool is immersed inside the liquid resin, and the 2D patterned
light beam is focused on the surface of light guide tool. The liquid resin is solidified at the place
where the light guide tool goes. For an input CAD model of 3D microscale features, a set of 2D
patterned images are firstly generated by slicing the CAD model of building part. Then the pre-
existing object will be moved to the initial place, and the moving speed of the pre-existing object
depends on the curing characteristic of liquid resin. To achieve high accuracy of fabrication, the
curing performance of resin under the exposure of light guide tool is studied in this chapter. The
material filling and moving speed of pre-existing object are optimized. The scaling capability of
the developed ISA process is demonstrated and the surface quality is investigated in this chapter.
4.1 Light curing performance
4.1.1 horizontal plane
The light distribution of each pixel in the 2D patterned light beam follows Gaussian function
[61] (refer to Figure 33a). The light intensity of neighboring pixels overlaps and convolutes into
accumulated light intensity as shown in Figure 33b [58, 61]. The phase of resin is changed from
liquid to solid after the resin receives enough energy from the light exposure [39]. The energy E
is the production of exposure time t and light intensity I.
𝐸 = 𝐼𝑡 (3)

68
The light intensity varies with the dimension of light beam. Based on the Eq.3, the light energy
absorbed by the resin exposure by the light beam with larger exposure area will be higher than the
one with smaller exposure area. Consequently, the resin exposure by the light beam with larger
exposure could be easily over-cured compared to the portion with a smaller exposure area. The
light intensity of light beam is fixed, so the exposure time needs to be set based on the exposure
area of the 2D patterned light beam. A set of experiments have been performed, in which the
microscale cone arrays was printed by using different dimensional 2D patterned light beam (refer
to Figure 33c). The exposure time increases non-lineally with the decrease of the exposure area in
the 2D patterned light beam. The test parameters and the fabrication results are shown in Figure
33d. With the increasing of exposure area of the light beam, the exposure time required for the
accumulation process will be reduced (refer to Figure 33d).

Figure 33 The exposure study of liquid resin. (a) An illustration of Gaussian function for the light
energy distribution of a pixel in the projection image; (b) the overlapping effect of light energy in

69
neighboring pixels; (c) the 2D patterned images used to fabricate the micro features with different
exposure areas; and (d) the experimental results of exposure time versus exposure area in 2D
patterned light beam.
4.1.2 Vertical plane
The moving speed of the light guide tool is determined by the cure depth of the photocurable
resin. That is, the light guide tool needs to maintain an appropriate gap between the pre-existing
object to make the newly cured resin attach to the surface of pre-cured object. A series of
experiments have been performed to study the relationship between the cure depth and the energy
power of an input 2D patterned light beam. Based on the polymerization principle [165], the
classical Beer Lambert’s law of the light propagation shows that the cure depth of material can be
represented as Eq. (4).
𝐶 𝑑 = 𝐷 𝑝 𝑙𝑛 (
𝐸 𝑚𝑎𝑥 𝐸 𝐶 ) (4)
When we fabricate microscale features by using the ISA process, the light guild tool
continuously moves away. The distance between the light guide tool and the surface of the pre-
existing object should be less than the cure depth of the resin. The light intensity of the light beam
with full scale, of which the dimension is 3.67 mm 2.75 mm, is 28 mw/cm
2
. The design of
experiments (DOE) was applied to study the cure depth of the photocurable resin, and the overhang
features was built as shown in Figure 34a. The liquid resin used in our experiments was E-glass
3sp purchased from EnvisionTEC Inc. (Dearborn, MI). Different percentages of multi-wall carbon
nanotubes (MWCNTs) that serves as the light absorber were added inside the photocurable
polymer. The fabrication results of test overhanging features are shown in Figure 34c. By
controlling the light intensity of 3D patterned light beam with different gray scale values, the cure

70
depth of liquid resin could be adjusted from 35 m to 350 m (refer to Figure 34d). An illustration
of the relationship between the input light energy and the related cure depth is shown in Figure
34d.

Figure 34 The curing depth control of liquid resin. (a) The schematic design of eggbeater shaped
structure with elliptical shape ring; (b) the fabrication result of artificial eggbeater shaped
structures; (c) experimental results of the curing depth with different percentages of light absorber;
and (d) experimental results of curing depth, grayscale levels of input light, and different light
absorbers.
4.2 Material attachment
Liquid resin is cured under the exposure of the light beam transmitted by the light guide tool
during the ISA process, and the newly cured resin is deposited on the previously cured resin portion
instead of on the light guide tool. In the previous study, the relation between the attaching force

71
and the separation force of the cured resin was established [115-117, 162]. Suppose the attaching
force between the cured resin and the pre-existing object or the previously cured resin is FM, and
the separation force between the cured resin and the PDMS film is FT.
If FM > FT during the whole building process, the newly cured resin will always attach to the
inserted object; thus, the building process will be successful.
If FT > FM , the new cured layer will be damaged and attach to the surface of light guide tool;
thus the building part will fail.
The main separation FT is in the Z-direction when the moving direction of light guide tool is
Z direction and separation force occurs when the platform was pulled up from the PDMS film
coated on the light guide tool. The separation FT is the function of section area and the elasticity
of the PDMS film [53]. Initially, there is ~ 2.5 μm non-polymerization liquid area near the PDMS
film. During the separation, the pressure difference is generated by the air pressure and the weight
of liquid resin, and the separation force is generated by the movement of liquid resin which is
driven by this pressure difference. We simplified the problem and the separation force can be
expressed:
𝐹 𝑇 = 𝐹 𝑣 = ∫(𝜇 𝜕𝑣
𝜕𝑥
)𝑑𝑧 (5)
where 𝜇 is viscosity of liquid resin, 𝑣 is the speed of liquid resin.
The attaching force FM is determined by the material property and the contact area of fabricated
parts, and it can be calculated by the below equation [54]:
𝐹 𝑀 = 𝜎 𝑏 𝑤𝑙 (6)

72
where w is the width of printed part,
b
 is the flexural strength which is 65MPa for EnvisionTEC
SI500, l is length of the printed part [54].
To keep a new cured layer attach on the previous built part, the separation force FT in the
pulling-up process should be less than the bonding force FM of the new cured layer and the pre-
cured portion (refer to Figure35). Otherwise, the cured layer will be detached and material cannot
be accumulated to 3D shape. The separation force increase dramatically with the increasing of
section area A and compared with bending force, the increasing was much faster. Therefore, it is
much easy to separate the microscale structure compare with large scale structure [53].

Figure 35 The illustration of separation force and attachment force of the new cured layer.
4.3 Material filling
In ISA process, since we coated the PDMS film on the surface of a glass tank, the certain
amount of oxygen permeated through the glass and PDMS, and oxygen uniformly mixed with
resin at the bottom of the tank. The resin with a mixture of oxygen cannot be cured under the light
exposure due to the oxygen inhibition [54]. Due to the 2.5 μm thin non-polymerization layer
between PDMS and curing part, the channel was generated for liquid resin to flow [54]. Therefore,
oxygen inhibition layer provided the possibility to fabricate the model continuously with sufficient

73
resin [53, 54]. The relationship between flow filling time t, the size of the cross section of part l
and gap distance h was firstly analyzed (refer to Figure 36). We supposed the surface of platform
and PDMS was parallel. The assumption was set that the resin flow can be treated as Hele-Shaw
flow [166, 167]. The liquid flow below the cured part was considered fully developed and without
time disturbance. The liquid resin was incompressible liquid flow, which satisfied the following
equation:
𝑣 𝑟 =
∆𝑝 𝜇𝐿
[(
ℎ
2
)
2
− (𝑧 −
ℎ
2
)
2
] (7)
where ∆𝑃 is the local pressure difference, Z is the distance in the perpendicular direction, 𝜇 is the
viscosity of the liquid resin, L is the size of cured object, and 𝑣 𝑟 is the velocity in the direction of
𝑥 𝑖 .
For layer based ISA, the platform moves up with one layer thickness and the new resin will
flow back to the gap between light guide tool and the surface of pre-cured part. The time t of the
liquid resin refiling was predicted based the size of cured object L and the flow speed of liquid
resin 𝑣 𝑟 as shown in following equation:
𝑡 =
𝐿 𝑣 𝑟 (8)
Substituting Eq.7 into Eq.8, the time t can be represented as:
𝑡 =
𝜇 𝐿 2
Δ𝑝 [(
ℎ
2
)
2
−(𝑧 −
ℎ
2
)
2
]
(9)

74

Figure 36 The schematic diagram of liquid filling progress of ISA process
For the continuous ISA process, the refilling resin will be cured during the movement of platform.
Suppose standard atmospheric pressure is 101325 Pa, the height of resin H is 20 mm, h is 2.5 µm,
and viscosity of resin 𝜇 is 180 cP at 30 °C and the density of resin 𝜌 is 1.10 g/cm
3
. The speed of
filling resin was calculated, so that the resin refilling speed driven by gravity and air pressure was
only 0.57 mm/s. The flow of liquid resin under the cured part with square section area L
2
can be
predicted as:
𝑄 = ∫ 𝑑𝑦 ∫
∆𝑝 𝜇 𝑒 𝑘 𝐿 [(
ℎ
2
)
2
− (𝑧 −
ℎ
2
)
2
]
ℎ
0
𝐿 0
𝑑𝑧 (10)
However, in the layer-less additive manufacturing processes with continuous video projection,
the liquid is exposed to the light during its filling process. Hence, the viscosity increases due to
the photopolymerization and the refilling speed will decrease. The constant viscosity assumption
in Eq. (10). will be violated. Due to the polymerization reaction, the viscosity increases
exponentially with monomer conversion [165, 168]. To illustrate the filling dynamics in the photo
polymerization process, we used a simplified exponential function to describe the viscosity profile
of the liquid in the gap (refer to Eq.11):
𝜂 𝑡 = 𝜂 0
× 𝑒 𝑘𝑡
(11)

75
where 𝜂 𝑡 and 𝜂 0
are the viscosities of resin at time t and time 0, respectively, k is a constant. If the
liquid doesn’t absorb light energy, k is zero, otherwise, it is a nonzero constant.
The refilled liquid resin will be cured along with the movement of curing platform with the
speed 𝑉 𝑧 . In a unit time, the newly cured resin is equal to the amount of resin that reaches the
fabrication area. To maintain a sufficient resin refilling, the refilling resin flow volume should
satisfy:
4𝑄𝑡 = 2𝐿 2
𝑉 𝑧 𝑡 (12)
Substituting Eq.10 into Eq.12, the valid side length Lr of the cured part built with sufficient
resin refilling can be calculated with given Vz as:
𝐿 𝑟 = (
2∆𝑝 ℎ
3
3𝜇 𝑒 𝑘 𝑉 𝑧 )
1
2
(13)
Based on the former calculation, we figured out that the self-filling speed of resin driven by
gravity and air pressure is so slow that only microscale features could be continuously fabricated.
To verify this assumption, we did the experiment to identify the relation between the valid
fabrication area δ and the continuous printing speed Vz (refer to Figure 37). We fabricated a series
of cubes with different section areas ranging from 0.01 mm
2
to 4mm
2
using different printing speed
Vz. We considered that the printed model was failed when there were air bubbles or shadows inside
the cross-section area of the printed part, and if there were no bubbles and no intersections shadow
of resin inside the printed part, we considered it was valid continuous fabrication result by using
CISA process. Based on experimental results, we further figured out critical values of valid
continuous fabrication capabilities using different printing speed Vz from 30 μm/s to 150 μm/s only
with self-refilling resin driven by air pressure and resin gravity (refer to Figure 37).

76

Figure 37 The valid fabrication section area δ with self-refilling resin using different moving speed
of platform
Under the boundary blue line shown in Figure 37, all of the models located in the valid area
could be continuous printed using directly CISA process with self-refilling resin. Based on results
shown in Figure 37, we knew that the valid fabrication area δ turns to be bigger as the continuous
printing speed Vz decreases. When the cross-section area A is larger than the valid fabrication area
δ with certain printing speed Vz, we would find out bubbles existing in the middle area of the built
part because fresh liquid could not be able to fill the gap completely in time only by gravity and
air pressure. In that situation, the continuous printing process requires more waiting time for the
resin to refill to the central fabrication area, which can be achieved by slowing down the printing
speed Vz. To fabricate the cross-section area A, which is larger than 2 mm, the printing speed had
to be reduced to only around 5 μm/s. Therefore, to achieve a fast-continuous fabrication using

77
CISA process, it is desired to instantly fill the sufficient resin to the fabrication area during the
printing process. The continuous fabrication of large area could not only depend on the self-
refilling flow of the resin pushed by the gravity of the resin and air pressure.
4.4 Moving speed
4.4.1 Layer based ISA process
In the ISA process, the printed part moves along with the platform in the Z direction, and the
shear force between the liquid resin and cured part was increased with the increasing of the moving
speed of the platform. If the speed is faster than the threshold, the microscale part will be damaged
during the movement of the platform. To figure out the appropriate moving speed of the platform,
we conducted the experiments that a series of rods were fabricated using different moving speeds.
The side length of rod b was from 10 μm to 160 μm, and L was the height of rod, which was equal
to 2.5mm. Figure 38a shows the fabrication result of rod by using different speed.
The curing features need to overcome the drag force Fd, which is determined by the viscosity
of the liquid resin and the moving speed v of the pre-existing object. For a moving speed v, the
drag force could be calculated as the Eq.14:
𝐹 𝑑 = 𝐶𝑉 (14)
where 𝑉 is the flow velocity of the object, 𝐶 is the drag coefficient, which depends on the contact
area of beam 4 L*b, and the viscosity 𝜇 of liquid resin.
Meanwhile, the attachment bonding force depends on the material property and the section
area of the rod b × b (refer to Figure 38). The bonding force, which is equal to the breaking point
of the rod can be calculated as:

78
𝐹 𝑝 = 𝜎 𝑡 × 𝑏 2
(15)
Where σt is the tensile strength of the material, and it is 78.1MPa for SI500 (purchased from
EnvisionTEC).
Based on the Eq.14, the drag force turns to be bigger with the increasing of the moving speed
or the length of the printed beam. ISA process is focused on microscale fabrication, and the moving
speed should be set at a proper value. The contact area of microscale rod is much smaller than the
one of macroscale rod, with the decreasing of contact area, the valid speeding setting area turns to
be smaller. Based on Eq.14 and 15, the moving speed should be set at the value smaller than Vmin:
𝑉 𝑚𝑖𝑛 = 𝑘 𝜎𝑡 ×𝑏 4𝐿 𝜇

(16)

Figure 38 The moving speed setting in ISA process. (a) CAD of test model; (b) the illustration of
the forces of rob during the fabrication; (c) the fabrication results of robs with the moving speed

79
of platform 100 𝜇 m/s and some robs were failed; and (d) the fabrication results of robs with the
moving speed of platform 30 𝜇 m/s.
4.4.2 Continuous ISA process
The material is accumulated during the movement of the light guide tool. If the moving speed
of the light tool or the pre-existing object is too fast, the newly cured resin portion may not be able
to attach to the previously cured resin portion. Hence the continuously moving speed of the light
guide tool needs to be set based on the cure depth as tested before. However, when the continuous
printing speed is faster than the cure depth of photocurable material, there are several gaps existing
on the surface of the built model. Because the light energy is not enough for fully
photopolymerization of resin with such fast speed. A series of experiments were conducted to
figure out the best continuous printing speed Vz to get the best surface quality (refer to Eq.17), and
finally the 𝑘 𝑧 was set in the range 0.9-1.2 depends on the mechanical strength requirement of built
part.
𝑉 𝑍 = 𝑘 𝑧 𝐶 𝑑 (17)
The mechanical strength of the polymer has a relation with the crosslinking level of the
polymer. The mechanical strength of printed part turned to be bigger with smaller value setting
because the mechanical strength of printed part was increased with higher crosslink level of the
polymer after receiving more light exposure with slower continuous printing speed. Based on the
material continuous printing property and energy distribution of projection light, we further
controlled the cure depth of material in our process by adding the photoinitiator or light absorber,
which make it possible to achieve a large range of mechanical strength of built part with a
continuous printing speed.

80
Table 5 The mechanical strength of material fabricated by using different Kz

Figure 39 The fabrication of circular cone by using layer based ISA printing process
Kz Young’s modulus (E) Mpa
0.9 81 - 108
1 58 - 72
1.1 38 - 45
1.2 5 - 7

81
4.5 Fabrication capability identification
To verify the fabrication capability of our newly developed ISA printing process, we used a
layer-based approach to fabricate the circular cone with gradually decreasing dimensions. Firstly,
we sliced the 2.5 mm circular cone into a series of 2D layers with a thickness of 25 μm. Then we
applied the non-linear exposure setting algorithm to fabricate each layer of this test part. To make
sure the part wasn’t damaged during the movement of the platform, we set the moving speed of
the platform is 30 μm/s. As shown in Figure 39, the smallest tip of the cone we can fabricate was
8 μm, which was cured by using the 2D patterned mask image consisted of 4 pixels. The exposure
time to cure this sharp point was the 40 s and the whole printed cone is shown in Figure 39.

Figure 40 The cones fabricated by layer based and continuous ISA printing. (a) the cone fabricated
by the layer-based SL process; and (b) the cone fabricated by the surfaced-based CNC
accumulation process (with the moving speed of 50 m/s).
4.7 Surface quality
The fabricated microscale cones by the layer-based ISA and the CISA processes are shown in
Figure 40a and Figure 40b, respectively. The surface of the cones fabricated by the continuous ISA

82
accumulation process, which has no obvious stair-stepping effect on their side surface, is much
smoother than the layer-based SL process. We measured the surface roughness by using Image J.
The surface of cone printed by layer based ISA had uniformed stairs between each layer and the
dimension of stair was around 25 μm.
4.8 Summary
In this chapter, process modeling was conducted to optimize the parameter setting in the ISA
process. In order to achieve high resolution fabrication by using ISA process, the light curing
performance of 2D patterned light beam was studied. The material attachment was discussed so
that microscale textures can be added on the surface of pre-existing object. To build multi-scale
structures, the material filling under the light guide tool was studied. Based on the material filling
model, the moving speed setting for both layer based ISA and continuous ISA process was
discussed. The fabrication capability and surface quality were identified by using experimental
method. The ISA process shows significant strengths on fabricating microscale features on
macroscale object surfaces. The ISA process showed the capability to build around inserts and
fabricate multi-scale geometric features, which could enable various applications in different fields.
83

Chapter 5 Microscale Biomimetic Texture Fabrication by using
LISA
5.1 Overview
As a unique plant, cactus can survive in extremely aridness area rigorous for living creatures,
due to the fact that its spine can collect water effectively even in the harsh environment, and each
biological spine contains microscale grooves and oriented barbs. This microscale feature can
accelerate the expansion of moist air and the following movement of the collected water droplet.
For high-efficiency water collection and transportation, we present a multi-scale cactus spine
inspired structure fabricated by the layer based ISA process.
5.2 Background introduction
Water is the headspring of life. However, due to the worsening environmental problem, more
and more places are experiencing a severe water shortage. To survive in an extremely arid region,
creatures in nature have evolved a special mechanism to collect water from the outside
environment. For example, one kind of beetles can survive in the desert due to the special
microscale textures on its back. The textures divide the back into the hydrophobic region and a
hydrophilic region. The droplet condenses in the flat hydrophobic region and can be quickly
collected in the hydrophilic region. When the drop is big enough, the water droplet will roll off
from the hydrophilic region [169]. Inspired by that, researchers developed a bilayer membrane
with hydrophobic and hydrophilic materials [170]. Similarly, spider silks show efficient water
collection with special microscale structures of spindle-knots and joints, which generate a surface

84
energy gradient due to the different Laplace pressure [171, 172]. These features enlighten
researchers to developed artificial fibers by using dip-coating and co-axial electrospinning process
[171, 172]. The fabricated artificial spider silks with spider silk inspired structures showed the
ability to gather water [171, 172]. Different from other plants living in humid conditions, cactus
can live in the desert for quite a long time. Some scholars studied this unique phenomenon of
cactus and finally, discover that high efficient water collection results in the specially designed
spines. With the apex angle of 10 degree -15 degree, the spines generate the large Laplace pressure
difference for water droplet collection and movement. What’s more, the microscale grooves can
guide the droplet rolling down to the bottom trichrome area (refer to Figure 41) [173]. Even though
the cactus has outstanding water collection performance, the cactus inspired artificial water
collection device is hard to fabricate using traditional manufacturing methods, because of the
complex geometric shape of cactus spines [173-179]. Therefore, we used our layer based ISA
(LISA) process to fabricate cactus inspired multi-scale structures for the study of high efficient
water collection.

Figure 41 Appearance, microscale structures, and water collect performance of cactus spines

85
5.3 Material and methods
5.3.1 Water collection experimental set-up
We developed a set-up to observe water collection rate (refer to Figure 42). In this set-up, we
mounted the printed structure on the electrical balance, and the microscope was used to record the
water collection process of the printed surface with and without bioinspired microstructures. In
this confined space, the humidifier was used to generate moist air, and the constant moist air was
blew inside the electrical balance. The humidity was controlled based on the indicator of the
hygrometer. By reading the weight, we evaluated the water collection rate of the inserted object.

Figure 42 The schematic illustration of the set-up used to measure water collection efficiency
5.3.2 Preparation of polymer/MWCNTs composite resin
The photocurable polymer resin E-glass was purchased from EnvisionTEC (Dearborn, MI) and
used as received. 3% (w/v) multiwall carbon Nanotubes (MWCNTs) (length (1-5 μm) and outer
diameter (5-15 nm)), purchased from Bucky USA. Inc) was mixed with photocurable polymer E-

86
glass. The mixture was stirred for 2h by using a magnetic stirrer and then the mixture was put
inside the ultrasonic bath for 30 min. 0.05% Oil red dye (purchased from Sigma Aldrich) was
added inside the E-glass to reduce the cure depth. All photocurable material were degassed in the
vacuum before usages.
5.3.3 Surface roughness
Figure 43 showed the effect of surface roughness of bionic spine on the water collection
efficiency. After adding multi-wall carbon Nano-tubes (MWCNTs), the water collection rate of
the bionic spines was improved three times compared with the one made only by pure polymer.
This is because water is easy to condensate from moist airflow with high roughness surface [176],
and the surface roughness of bionic spines was increased due to the addition of multi-wall carbon
Nano-tubes (MWCNTs) (refer to Figure 43d). Meanwhile, since the roughness of surface was
increased by adding MWCNTs, the contact angle (θW) of water on the flat surface made by E-
glass/MWCNTs was smaller than the one made by pure E-glass, which resulted in the
improvement of material hydrophilicity. In previous studies, scholars found that hydrophobic
materials were able to speed up the water collection by reducing the speed of water re-evaporation
and making small water drop together fast into a stream to flow down [177]. To further improve
the water collection efficiency of the bionic spine made by E-glass/MWCNTs, the surface
wettability of the bionic spine was modified by using vapor-phase process. In a vapor-phase
process, the superhydrophobic monomolecular layer was coated on the surface of the hydrophilic
bionic spine made by E-glass/MWCNTs [178], but the monomolecular coating had no impact on
the roughness of surface due to the thin layer thickness of the coating (refer to Figure 43e). For the
water collection of the bionic spine, it was shown that superhydrophobic coating enhanced the
further water condensation and transportation (Figure 43f). Specifically, the water collection mass

87
rate of the bionic spine with superhydrophobic coating was increased four times compared with
the water collected by the original hydrophilic bionic spine (refer to Figure 43g). The comparison
denoted the necessity of hydrophobic coating to the bionic spine for high efficient water collection.

Figure 43 The water collection efficiency of surface with and without bionic spines made by
different material. Contact angles of water and SEM images of flat surface made with (a) E-glass
and (b) E-glass/MWCNTs; Water collection and SEM images of bionic spine array made with (c)
E-glass, (d) E-glass/MWCNTs and (e) E-glass/MWCNTs after superhydrophobic coating (with
magnified image show the rough surface); (f) The record of water collection process of flat surface
with/ without bionic spine array in 4mins; and (g) water mass collected by flat surface with/
without bionic spine array.

88
Our water collection methodology is solely using the difference in capillary forces acting on
the two phases between liquid (water), vapor and solid. The material we used showed the
hydrophilic property, and the surface with the Nano-coating showed the superhydrophobic
property. As shown in the Figure 43, the hydrophilic property was improved by adding MWCNTs,
which further increased the roughness of the surface. Superhydrophobic surfaces had an attractive
ability to make small water drop together into a stream. To further improve the water collection
efficiency, we coated a nanoscale hydrophobic film on the surface of bionic spines. Compared
with the non-coating spines, the water collection rate of coated bionic spines was increased double
as the one without coating (refer to Figure 43). Compared with the bionic spins made by polymer,
the spines printed with MWCNTs/polymer showed more efficient water collection functionality.
5.4 Design of cactus inspired water collection structure
According to Kalvin equation and Laplace equation, the pressure difference ∆𝑃 of the water
drop can be calculated as [172]:
∆𝑃 =
2𝜎 𝑤 𝑟 (18)
where 𝜎 𝑤 is the surface tension of water, and r is the radius of the water droplet.
The critical coagulate radius 𝑟 𝑐 of water can be define as below:
𝑟 𝑐 =
2𝜎 𝑤 𝑉 𝑚𝑙
𝑅𝑇𝑙𝑛 (
𝑃 𝑃 𝑠 )
(19)
where 𝑉 𝑚𝑙
is the molar volume of water, 𝑃 𝑠 is the saturated vapor pressure of water on a level
surface, T is the temperature.

89
Based on the Eq.19, water vapor coagulation is easier to occur at the place designed with a
high radius of curvature. If the angle of bionic spines is not small enough, the water droplet cannot
be collected. However, it is hard to achieve the highest vapor coagulate theoretically. Assume that
the surface of each bionic spine is smooth, the pressure difference can be represented by the
following formula [179]:
∆𝑃 𝑐𝑢𝑟𝑣𝑎𝑡𝑢𝑟𝑒 = − ∫
2𝜎 𝑤 (𝑅 +𝑅 0
)
2
𝐿 ∗𝑡𝑎𝑛 𝛼 2
0
𝑠𝑖𝑛 𝛼 2
𝑑𝑧 (20)
where, ∆P curvature is the gradient of Laplace pressure, L is the height of bionic spine, and R0 is the
radii of the bionic spine and the collected water droplet, α is the angle of the bionic spine, and dz
is the integral variable of the bionic spine.
The pressure driven the collected water droplet moves from top tip to the bottom can be
calculated by the below equation:
∆𝑃 𝑐𝑢𝑟𝑣𝑎𝑡𝑢𝑟𝑒 =
2𝜎 𝑤 𝑠𝑖𝑛
𝛼 2

[(𝐿 −2𝑅 0
)tan
𝛼 2
+𝑅 0
]
2
−
2𝜎 𝑤 𝑠𝑖𝑛
𝛼 2

(𝐿𝑡𝑎𝑛
𝛼 2
+𝑅 0
)
2
(21)
However, this is the situation when the bionic spine places horizontally. In actual conditions,
the bionic spine is placed vertically or multi-directionally. we need to consider whether the gravity
of water droplet may make the impact on the water collection. We assumed the angle between the
axis of the bionic spine and the ground as β, and the gravity 𝐺 of water droplet along the bionic
spine can be defined as below:
𝐺 =
2
3
𝜋 𝑟 3
𝜌𝑔𝑠𝑖𝑛 (𝛽 −
𝛼 2
) (22)
where, 𝜌 is the density of water drop, 𝑔 is the gravitational acceleration.

90
Therefore, after considering the gravity, the total pressure difference ∆𝑃 ′
can be recalculated as
below:
∆𝑃 ′
= ∆𝑃 𝑐𝑢𝑟𝑣𝑎𝑡𝑢𝑟𝑒 +
2
3
𝜋𝑟𝜌𝑔𝑠𝑖𝑛 (𝛽 −
𝛼 2
) (23)
We used Matlab to simulate the relationship of driven pressures and the apex angle of bionic
spines, and the result is shown in Figure 44c, From the observation of the simulation curve, the
optimized apex angle of the bionic spine decreases with the increasing of the height. To verify our
assumption, we conducted the experiment to test the water collection efficiency of our printed
bionic spines with different angles.

Figure 44 The water collection of bionic spines with different angle. (a) The water collection of
bionic spines at different time; (b) the water collection rate of bionic spines with different angle;
and (c) the relation of pressure difference and the angle of bionic spine.

91
We put the printed bionic spines in an environment with constant humidity, which is 90%.
Figure 44a showed the water collection results of bionic spines with different apex angles at the
beginning of 4 mins. From the experiment result we can know that the water was gradually
condensed at the tip of each bionic spines, and the water droplet turned to be bigger until it rolled
down along the bionic spines, we measured the weight of the water collected by our printed bionic
spines and found that the bionic spines with 10° collected the maximum weight of water compared
with other bionic spines at each time. The water collection rate of bionic spines with 10° reached
125 μg/s, which was the 1.5 times of the weight of water collected by the bionic spines with 30°.

Figure 45 Design of array of bionic spines. (a-d) The design model and fabrication results of
bionic spines with angles ranging from 10° to 40°; (e-h) the design model of bionic spines with
multiple branches; and (i-h) the fabrication results of multiple branches of bionic spines.

92
To figure out the best angle of bionic spines for high efficiency water collection, we designed
the apex angle of each bionic spine ranges from 10° to 40°. Using LISA printing process, we
successfully fabricated the bionic spines with the height of 1.5 mm, and based on the observation
of SEM image, the top tip was only 8 μm, which achieved large curvature. The fabrication results
of bionic cactus with different apex angle are shown in Figure 45 (a-d). Since the cactus spines in
nature have multiple branches for each bunch, we mimicked the natural cactus spines, and designed
each bunch of bionic spines with multiple directional branches. The gap with each bunch was
550 μm and the design CAD model of bionic cactus with different number of bionic spines are
shown in Figure 45 (e-h). The side views of built bionic spines with multiple branches are shown
in Figure 45 (i-l).
5.5 Fabrication of cactus inspired water collection structure by LISA
The surface roughness of bionic spines improved the water collection efficiency, so we applied
a layer based ISA printing strategy to build the bionic spines. The fog collection performance of
microscale bionic spine was determined by its shape and size (refer to Section 5.4). Three
dimensional digital models of bionic spine array with different shape and arrangement were
created by using Solidworks. Then each model was sliced into a set of 2D patterns, and the slicing
layer thickness was equal to 35μm (Figure 46a). The light intensity of projection beam generated
by our optical system was calibrated, and the gray scale level of each pixel in the 2D patterned
grayscale images was adjusted to achieve uniform light intensity distribution (Figure 46c). After
curing one layer of material, the previous cured section moved away from the light guide tool
(refer to Figure 46b). The distance between the light guide tool and previous cured section was 35
μm to make sure the following cured layers can stick onto the previously built layers. To generate

93
the sharp corner of artificial cactus, of which the dimension was smaller than 8 μm, the exposure
time of each layer was set based on the study shown in Section 4.1 [162]. To prohibit the over cure
features in fabrication direction, the cure-depth of material we used was controlled by adding
MWCNTs or oil red dye [162]. From the microscopy images, we obviously observed the stair
effect existing on the surface of printed bionic spines by using LISA process.

Figure 46 Fabrication of bionic spines array by layer based ISA process. (a) The slicing
process of CAD model of bionic spines array;(b) the fabrication of bionic spines by using layer
based ISA process; and (d) the optimized projection images to fabricate bionic spines array with
uniformed light beam.

94
5.6 Experiment results and analysis
5.6.1 Motion of water droplet
In nature, due to the gradient Laplace pressure, the condensed water droplet always moves
along with the spines to the bottom no matter how the spine is arranged. To test the water droplet
movement of the printed spines, we tilt the printed spines for 90, 135 degrees and even upside
down for 180 degrees. As shown in Figure 47, the water droplet moved along the printed spines
even the spines were put upside down. This is because even the Laplace pressure was reduced with
the increasing of the diameter of spines, the difference of Laplace pressure was much larger than
the gravity of water droplet. Driven by the pressure difference ∆𝑃 ′
, the collected water droplet
moved along the bionic spine to the root. Meanwhile, small droplets in neighboring region firstly
converged into a big droplet, and the converged droplet would not move to the root of spines
immediately until the water liquid grew big enough. This unique phenomenon enabled the bionic
spine to efficiently collect water in multiple directions.

Figure 47 The movement of water droplet condensed from humidity environment

95
5.6.2 The effect of moist airflow direction
The water collection efficiency of bionic spines array has a relation with the direction of
airflow, because the velocity of moist airflow around each bionic spine depends on the direction
the moist airflow [179]. The greater the contact area between the moist air and bionic spines, the
more fog in the moist airflow can be condensed into water. For example, for tetragonal arranged
bionic spines array, the moist airflow, which blew in the horizontal direction, was blocked by
windward side spines, and the fog stream became hard to exposure the following spines in the
moist airflow blowing direction, resulting in the reduction of water collection of following bionic
spines.

Figure 48 The fog collection efficiency of bionic spine array under moist airflow from different
directions
As shown in Figure 48, water was condensed on the surface of each bionic spine (blue arrow)
after 1 min when the moist airflow blew from the top side, while the condensed water droplet only
appeared at the front windward spines (blue arrow) when the airflow blew from the side at the

96
same time. After 4 mins, it can be observed that the water droplets condensed by bionic spines
under moist airflow from the top were uniformed attached on the surface of each bionic spine, but
the water droplets condensed by bionic spine under moist airflow from side decreased successively
in the blowing direction of the moist airflow.

Figure 49 The water collection of bionic spines with multiple branches. (a) The water collection
of bionic spines at different time; (b) the weight of water collected by bionic spines at different
time; and (c) the water collection rate of bionic spines with multiple branches.
5.6.3 Water collection of multi-directional bionic spines
After we verified that placing direction had no significant influence on the water collection of
bionic spines (refer to Section5.6.1), we further tested the water collection performance of bionic
spines with multiple branches. In this experiment, we changed the number of spines on each bunch
ranging from 1 to 4. We put our printed bionic spines in the environment with the constant
humidity, and the continuous moisture air flow blew from the top. As shown in Figure 49, the
more branches of bionic spines, more water droplets were condensed at each time. Compared with

97
the measurement results, the weight of water collected by bionic spines with four branches was 4
times larger than the one of the bionic spine with only single branch. The weight of water collected
by bionic spines with four branches was 75 mg after 4mins, and the water collection rate reached
312.5 μg/s.
5.6.4 Water collection of optimized multiple bionic spines
From the above experimental results, the arrangement of bionic spines severely impacted water
collection efficiency of bionic spines. To demonstrate and characterize the ability to achieve high
efficiency of water collection by controlling the arrangement of bionic spines array, we simulated
the fog stream among each bionic spine by using COMSOL software. Velocities of airflow around
tetragonal arranged bionic spines were simulated under the airflow blowing from horizontal
direction (Figure 50a and b). When each spine was tetragonal arranged in the bionic spines array,
the airflow was blocked by the first column of windward spines, resulting in the velocity of airflow
around the following spines was sharply reduced. Even the distance between each spine was
adjusted 4 times larger than the original value, the velocity of airflow around each bionic spine
was still not uniformed distributed among the tetragonal arranged bionic spines array (Figure 50c).
In nature, spines hexagonally arranged on the surface of the cactus stem as shown in Figure
50c. Inspired by the natural cactus spiny stem, bionic spines were arranged in a hexagonal pattern,
and the simulation of velocity distribution around each hexagonally arranged bionic spine was
conducted. The airflow uniformly went through each bionic spine in the hexagon patterned array
with quite high speed (refer to Figure 50d). After optimizing the arrangement of bionic spines, the
velocity of airflow around each bionic spine was increased more than 5 times of original design,
which the bionic spines were arranged in a tetragonal pattern.

98

Figure 50 The simulation of air flow around each bionic spine with different arrangements. (a)
The 2D simulation of flow velocity distribution around tetragonal arranged bionic spines under
air flow in horizontal direction; (b) The 2D simulation of flow velocity distribution around the
tetragonal arranged bionic spines with larger interval; (c) The picture of real cactus with
hexagonally arranged spine’ bundles; and (d) The 2D simulation of flow velocity distribution
around the hexagonal arranged bionic spines under air flow in horizontal direction.
After we optimized the branches positions of each bionic spine, we further arranged the
direction of bionic spines in the hexagon pattern. We optimized the arrangement with three
different design (refer to Figure 51a). By using LISA process, we fabricated the bionic spines with
the different designs, and the SEM image showed that the bionic spines array can be fabricated
with high accuracy. We put 3D printed bionic spines array into the humidity environment and then
test the water collection of our 3D printed bionic spines array. As shown in Figure 51b, the water

99
collection rate of bionic spines array with optimized 3 pattern was increased 6 times bigger than
the result of the original design (Figure 51b). the water collection rate of bionic spines array with
super hydrophobic coating was increased 5 times bigger than the result of the one without coating.
Overall, we made a conclusion that the 3D printed bionic spines with optimized design can
dramatically improve the water collection efficiency, and the high-efficiency water collection
bionic spines can be applied in different fields in future.

Figure 51 The water collection of bionic spines with optimized arrangement. (a) The water
collection of bionic spines with different arrangements; and (b) the water collection rate of bionic
spines with different arrangements.
5.7 Summary
In this chapter, we firstly applied a 3D-printing approach named as LISA to fabricate high-
resolution microscale bionic spines on a flat surface in multiple directions. Inspired by cactus, we
designed multi-scale bionic spines with different apex angles ranging from 10 ° to 50 °, and studied
the water drop growth rate and drop motion velocity of our printed bionic spines to optimize the

100
apex angle of bionic spines for high efficiency of water collection. Multi-wall Carbon Nanotubes
was added in the hydrophilic photocurable polymer to be the material of our bionic spines. To
further accelerate the water growth rate of printed spines, we coated nanoscale superhydrophobic
coating on the outer surface of our printed bionic spines. Furthermore, we printed the bionic spines
at multi-directions and quantified the water collection capacity of multi-directional bionic spines.
The influence of different design layout of bionic spines array over the efficiency of water
collection was investigated under different flow directions, and based on the theory of
aerodynamics, a special patterned array with artificial multi-directional spines was further
designed and tested. The 3D-printed surface with our optimized biomimetic multi-directional
spines showed prospective applications in new energy-efficient water collection and water
transportation.
101

Chapter 6 Multi-Scale Biomimetic Structure Fabrication by using
CISA
6.1 Overview
The plant leaves show the superhydrophobic property due to the multi-scale eggbeater shaped
structure, and the eggbeater shaped structure can grip the water droplet with the nanoscale
hydrophilic tips. Because the unique superhydrophobicity, this multi-scale eggbeater shaped
structure has potential applications in self-cleaning, micro-droplet transportation, anti-reflection,
and oil/water separation [185]. However, it is hard to full scale reproduce this multi-scale eggbeater
shaped structure by using current AM process. Therefore, in this section, we applied continuous
immersed surface accumulation (CISA) method to build a novel reusable superhydrophobic and
lipophilic surface inspired by the natural eggbeater shaped structure. The geometric shape design
and optimization will be presented in Section 6.3, and special composite material was developed
to enhance the functional performance of 3D printed eggbeater shaped structures. Based on the 3D
printed superhydrophobic and oleophilic surface decorated with bionic eggbeater shaped structures
several interesting applications, such oil/water separation, oil cleaning, droplet manipulation, were
demonstrated in Section6.6.
6.2 Background introduction
As we know, water is the source of the universe, and it is vital for maintaining human life and the
development of societies. However, water pollution is a major global problem, and has become more
urgent. The main water contamination comes from insoluble organic and inorganic macromolecule

102
compound, such as oils, pitch; or soluble substances like dyes. The demand of the development of
energy efficient materials that can clean the oil pollution increased theses years. Recently, industries
generate large volumes of wastewater, and typically oily wastewater are most common pollution
sources among all the water pollution contaminants. Currently, the industrial wastewater with high
concentrations of oil pollutants is processed by distillation, electrophoresis, pressure filtration, dissolved
air flotation, or centrifugation [180-182]. However, there are many challenges of oil cleaning by using
current approach, including large quantity of operation, complex equipment, energy consuming, and
high cost with low efficiency.
A lot of research have been conducted to separate oil/water mixture by designing and fabricating
micro- and nanostructures with special wettability. To achieve high efficiency oil water separation,
polymer-coated, hydrogel-coated and surface-modified meshes and polymer membrane are developed
recently [180-182]. Externally high pressure or high voltage electric field are continuously required,
when the membrane based oil/water separation devices are used to separate the oil from the oil/water
mixture [183, 184]. It is necessary to develop an environmental friendly and energy free oil/water
separation material that can be used for multiple time even in harsh environments. Recently, it is a highly
promising method to solve the oil/water separation problem by learning form the biological material
system in nature.
Plant leaves showed self-cleaning due to the superhydrophobic and superoleophilic multi-scale
structures [83, 86]. Inspired by that, different designs of oil/water separation devices, such as mesh
matrix, porous foams or membranes, are coating with superhydrophobic and superoleophilic materials
[180 - 182]. Based on this special wettability, the oil can be separated from oil/water mixture and further
absorbed by the superhydrophobic and superoleophilic materials with large capillary force, meanwhile
the water droplet will keep the sphere shape due to the superhydrophobic property [180 - 184]. However,

103
the efficiency of oil separation using this method is limited to the oil absorption capacity of
subtracts. To solve this problem, we developed superhydrophobic and lipophilic surface with 3D
printed artificial biomimetic eggbeater shaped structures, which enabled us to remove the oil from
oil-water mixtures in complex and various environments with high efficiency (refer to Figure 52).

Figure 52 Biomimetic superhydrophobic and superlipophilic structure. (a) The salvinia molesta
floating leaf densely covered with eggbeater shaped structure[185]; (b) a spherical water droplet
was gripped by the eggbeater shaped structure [185]; (c) SEM image of the natural multi-scale
eggbeater shaped structure with hydrophilic nanoscale tips [185]; (d) the water-air interface
generated by eggbeater structures due to the superhydrophobic property; (e) schematic diagram
of surface covered with large scale of bionic eggbeater structures and each eggbeater structure is
designed with four hairs; (f) the oil separated by bionic eggbeater structures from water and oil
mixture; (g) SEM images of the 3D printed eggbeater shaped structures with four hairs and there
are MWCNTs on the surface of tip of eggbeater structures; (h) schematic diagram of the water-
oil interface generated by bioinspired eggbeater shaped structures due to the superhydrophobic
and superlipophilic property.

104
6.3 Design of biomimetic superhydrophobic and superoleophilic multi-scale
structure
In nature, salvinia molesta leaves shows significant superhydrophobic properties with
hydrophilic pins (refer to Figure 52b). When it is immersed in water, the surface of salvinia molesta
leaves can achieve big retention layer of air, since the upper side of the floating leaves is densely
covered with complex multicellular hairs and each hair are filled with hydrophilic nanoscale tips
[95, 185]. As shown in the Figure 52 (a-d), four microscale hairs are converged on the top to form
an eggbeater-shaped structure, of which the total height is 2 mm. Inspired by such eggbeater
shaped structure, we designed a functional surface with full scale reproduced eggbeater shaped
structures. We put a droplet of oil-water mixture on the top surface of eggbeater shaped structure,
the oil quickly merged into the 3D printed eggbeater shaped structures, and water droplet was left
on the top surface of eggbeater shaped structure (refer to Figure 52f).
The bionic eggbeater shaped structure is composed of a cylindrical stalk and a head mimicking
the salvinia molesta leaves. The height of the stalk of an artificial eggbeater shaped structures is
700 μm, and diameters of bottom portion and top portion of stalk are 300 μm and 150 μm
respectively. The head of eggbeater shaped structures is obtained by intersecting different number
of hair, of which a diameter is 35 μm and the height is 250μm (refer to Figure 52 e). To further
study the wettability of bionic eggbeater shaped structure, we designed eggbeater shaped structures
with different numbers of hair (N) and gap distance (d) (refer to Figure 53a-i). We further generated
the mathematic model to predict the contact angle of water droplet on the surface of bionic
eggbeater shaped structure with different numbers of hair and gap distance.

105

Figure 53 The wettability of bionic eggbeater shaped structure with different designs. (a) N=2,
d=0.3 mm; (b) N=4, d=0.3mm; (c) N=6, d=0.3mm; (d) N=8, d=0.3mm; (e) N=2, d=0.4mm; (f)
N=4, d=0.4mm; (g)N=6, d=0.4mm; (h) N=8, d=0.4mm;( i) N=2, d=0.5mm; (j) N=4,
d=0.5mm;( k)N=6, d=0.5mm; (h) N=8, d=0.5mm; (m) Contact angle and test pictures of bionic
eggbeater structure with different numbers of N and gap distance d; and (n) a comparison of the
experimental and simulated contact angles with CB-CB and W-CB models for different patterns.

106
The contact angle of water in the Wenzel (W) model can be represented as [187]:
*
cos cos
Y
r   (24)
where r is the roughness parameter, and θ* is the apparent contact angle, and θ
Y
is the ideal
contact angle on a smooth surface without roughness.
From a theoretical point of view, air layer is generated under the hair of designed eggbeater
shaped structure, Wenzel model cannot describe the phenomenon of bionic eggbeater shaped
structures. The Cassie-Baxter (CB) model considers both solid-liquid and liquid-air interface and
the contact angle of droplet can be calculated by Eq.25 [188]:
cos(𝜃 𝐶𝐵
) = 𝑓 𝑆𝐿
𝑐𝑜𝑠 𝜃 1
+ 𝑓 𝐿 𝐴 𝑐𝑜𝑠 𝜃 2
(25)
where
CB
 is the apparent contact angle, 𝑓 𝑆𝐿
is the fractional area of liquid and solid, 𝑓 𝐿𝐴
is the
fractional area of liquid and air 𝑓 𝑆𝐿
+ 𝑓 𝐿𝐴
= 1, 𝜃 1
is a contact angle of liquid on solid surface, and
𝜃 2
is contact angle of liquid in the air.
For the eggbeater shaped structure, the water droplet only contact the top hair and the air is
trapped inside the stalk of eggbeater shaped structure. The contact angle 𝜃 2
of water in the air is
180 degree, so the contact angle of water on the surface of eggbeater shaped structure can be
recalculated by the below equation:
cos(𝜃 𝐶𝐵
𝐶𝐵
) = 𝑓 𝑆𝐿
′
cos(𝜃 ) − 𝑓 𝐿𝐴
′
(26)

where 𝑓 𝑆𝐿
′
and 𝑓 𝐿𝐴
′
represent the fraction of eggbeater shaped structure -water interface and water-
air interface respectively.

107
The contact angle of each design was calculated based on the Eq. 26 and the real contact angle
of artificial eggbeater shaped structures was also measured as shown in Figure 53m. The
comparison showed that the experimental results of contact angle correspond well with theoretical
results (Figure 53n). The material we use showed hydrophilic property for a flat surface. Without
the eggbeater shaped structure, the contact angle of water was only 65 degree. The wettability of
printed surface was transferred from hydrophilic property into superhydrophobic after adding
eggbeater shaped structures. The contact angle of water decreased with the increasing of N, and
this is because the contact area between the hair of bionic eggbeater shaped structure and the water
droplet. The contact angles of water droplet on the eggbeater shaped structure designed with 2
hairs and 4 hairs was 170 degree and 152 degree respectively. The gap distance of eggbeater
shaped structure was changed from 300 μm to 500 μm. The contact angle of water droplet
increased with the increment of gap distance. This is attributed to the increasing of contact area
between the hair of eggbeater shaped structure and the water droplet.

The elasticity of eggbeater shaped structure also had effect on the wettability of eggbeater
beater shaped structure. Comsol Multiphysics was used to conduct the simulation of deformation
of eggbeater shaped structure under the load of water droplet. As shown in Figure 54, the stress
and strain was located on the hair of eggbeater shaped structure. The stress and strain of eggbeater
shaped structure with 2 hairs was much larger than the one with 8 hairs under the same load. This
is because the stress was divided equally on each of the hair, and the stress of each hair was reduced
with the increment of the number of hair. With the deformation of the hair, the contact area of
solid material and water droplet increased, so that the contact angle of water droplet was reduced.
Furthermore, the stress and strain located on the hair of eggbeater shaped structure increased if the
load direction was applied from the top to the top left 45 degree (refer to Figure 54).

108

Figure 54 The simulation of relative stress of the superhydrophobic eggbeater shaped structure
designed with different number of hair. (a) N=2; (b) N=4;(c) N=6; and (d) N=8.
Moreover, since the oil occupies the space, where the air is trapped between each stalk of
eggbeater shaped structures. The hollow space between each stalk of eggbeater shaped structures
can be enlarged to absorb more volume of oil, which can significantly improve the oil/water
separation efficiency of our eggbeater shaped structure. Therefore, the surfaces covered with
different densities of eggbeater shaped structures were designed as shown in Figure 55a-e. The
different gap distance (d1 / d2) between each stalk of eggbeater shaped structure ranges from 300
μm to 500 μm, and the corresponding air/solid ratio can be modified from 0.48 to 7.9. The relation
of air layer volume and the height of the stalk of eggbeater shaped structure was studied and the
result is shown in Figure 55g. The valid volume of absorbed oil can achieve 14.6 mm
3
per unit
with eggbeater shaped structures, of which the height was a 4 mm. The oil/water separation

109
efficiency of eggbeater shape structure can be modulated with the height of the stalk of eggbeater
shaped structures.

Figure 55 The air/solid ration of eggbeater shaped structures with different height. (a) The
eggbeater structures with 400 um gap distance and the height of stalk; (b) the height of stalk is
800 𝜇 m; (c) the height of stalk is 1.6 mm; (d) the height of stalk is 2.4 mm; (e) the height of stalk
is 3.2 mm; (f) the air/solid ratio of bioinspired eggbeater structures with different designs of gap
when the height of stalk is 0.8 mm; and (g) air layer volume change with different height and gap
distance.
6.4 Material development
Our separation methodology was solely using the difference in capillary forces acting on the
two phases between water and oil. The material we used was hydrophilic and superoleophilic

110
composite. Decorated with eggbeater shaped structure, the surface showed the superhydrophobic
and super oleophilic property. The superhydrophobic and super oleophilic property of the printed
eggbeater structure was attributed to the higher surface tension of water than oil. Due to the
different capillary force between water, oil and the polymer based composite, after pulling the oil-
water mixture, the oil droplet quickly merged into the 3D-printed eggbeater shaped structure and
kept the filtered water on the top surface of eggbeater 3D-printed eggbeater shaped structure. To
further improve the hydrophobic and oleophilic property, multi-wall carbon nanotubes (MWCNTs)
was added to increase the surface roughness based on the Eq.24. The photocurable polymer resin
E-glass was purchased from EnvisionTEC (Dearborn, MI) and used as received. 3% (w/v)
multiwall carbon Nanotubes (MWCNTs) (length (1 - 5 μm) and outer diameter (5 - 15 nm)),
purchased from Bucky USA. Inc) was mixed with photocurable polymer E-glass. The mixture was
stirred for 2h by using a magnetic stirrer, and then the mixture was put inside the ultrasonic bath
for 30 min. 0.05% Oil red dye purchased from Sigma Aldrich was added inside the E-glass to
reduce the cure depth. All photocurable material were degassed in the vacuum before usages. The
superhydrophobic and oleophobic nanocoating spray was purchased from Ultra-Ever Dry. [186].
As shown in Figure 56, the photocurable polymer E-glass shows super oleophilic property
where the oil spread quickly along the surface of the polymer. For oil/water separation, the material
we use should perform superhydrophobic and super oleophilic. Then we added the multi-wall
carbon nanotubes (MWCNTs) inside the photocurable polymer. Using this composite solution, the
microscale pillar array and the bionic eggbeater shaped structures were printed by using CISA
respectively, and the SEM images of printed structures are shown in Figure 56d and e.

111

Figure 56 The hydrophilic and lipophilic property of material. (a) Contact angle of oil and water
and macro and SEM images of flat surface fabricated by using E-glass; (b)optical microscopy, the
SEM image, contact angle of oil and water of printed flat surface with E-glass/ MWCNTs; (c)
optical microscopy, the SEM image, contact angle of oil and water of printed flat surface with E-
glass/ MWCNTs and Nano spray; (d) optical microscopy, the SEM image, contact angle of oil and
water of microscale pillar array using E-glass; (e) optical microscopy, the SEM image, contact
angle of oil and water of eggbeater structure with E-glass/ MWCNTs; and (f) optical microscopy,
the SEM image, contact angle of oil and water of eggbeater structure E-glass/ MWCNTs and Nano
spray.
The contact angle of the oil and water on the printed surface were measured, and both water
and oil were merged inside the microscale pillar array. This is because the capillary force of
traditional pillar-shaped structure was not big enough to separate the oil from water and oil mixture.

112
However, the eggbeater shaped structures showed superhydrophobic and oleophobic property.
There was a big air layer under the head of eggbeater shaped structure, where the oil was easily to
flow inside the stalk of eggbeater shaped structures. After sputtering Nano-coating by using Ulta-
Ever Dry, the eggbeater shaped showed superhydrophobic and superoleophobic. The water droplet
slipped easily from the surface, and meantime the oil droplets kept sphere shape on the 3D printed
eggbeater shaped structure (refer to Figure 56c).

Figure 57 The surface roughness effect on hydrophobic and oleophobic property of designed
material. (a) contact angle of water on E-glass made flat surface; (b) contact angle of water on E-
glass/MWCNTs made flat surface; (c) microscopy image of E-glass based eggbeater shaped
structure; (d) microscopy image of E-glass / MWCNTs based eggbeater shaped structure; (e)
microscopy image of E-glass / MWCNTs based eggbeater shaped structure with Nano spray; (f)
SEM images of E-glass made flat surface; (g) SEM images of E-glass / MWCNTs made flat surface;
(h) SEM images of E-glass based eggbeater shaped structure; (i) SEM images of E-glass /
MWCNTs based eggbeater shaped structure; (j) SEM images of E-glass/MWCNTs based
eggbeater shaped structure after sputtering of superhydrophobic coating; (k) and (l) super

113
hydrophobic effect of E-glass / MWCNTs based eggbeater shaped structure after sputtering of
superhydrophobic coating; (m) oleophobic property of E-glass / MWCNTs based eggbeater
shaped structure; (n) oleophobic property of E-glass / MWCNTs based eggbeater shaped structure
after sputtering of superhydrophobic coating; and (o) Optical microscopy and the schematic
diagrams shows the mixed Wenzel and Cassie-Baxter state for the printed eggbeater shaped
structures, and the SEM images of printed eggbeater shaped structure with E-glass and E-glass /
MWCNTs composites, respectively.
The experiments were further conducted to study the influence of surface roughness on the
superhydrophobic and oleophobic property. The surface roughness of the flat surface was
increased by the addition of MWCNTs as shown Figure 57c and d. It was noticed that the surface
roughness of the eggbeater structure was also increased by the addition of MWCNTs, and further
enhanced the superhydrophobicity of eggbeater shaped structure. As shown in Figure 57, the
contact angle (θ
W
) of water on the flat surface printed by E-glass was 65

degree, and the contact
angle  of flat surface printed by the E-glass/MWCNTs was 82 degree. Based on the Eq.24, the
roughness parameter r can be calculated, which was 2.9. After the sputtering Nano-coating on
eggbeater shaped structure, it was shown that nanoscale structures on the surface, leading to the
further enhancement of roughness. To separate oil/water mixture and grip the water droplet,
hydrophilic material was necessary for the fabricate the hair of eggbeater shaped structure.
6.5 Fabrication of biomimetic superhydrophobic and superoleophilic multi-
scale structure by CISA
Continuous immersed surface accumulation (CISA) based 3D printing was used to fabricate
the bionic eggbeater shaped structure (refer to Figure 58a). To fabricate bionic eggbeater shaped

114
structure, the light guide tool was merged inside a tank that was filled with photocurable composite
(E-glass/MWCNTs) (refer to Figure 58b). Combining both dynamically controlled light beam and
the 4-axis movement of the light guide tool, in CISA microscale features continuously were
constructed on the surface of a macroscale object by selectively curing liquid resin into solid (refer
to Figure 58e).

Figure 58 The fabrication of eggbeater shaped structure by CISA process. (a) Schematic diagram
of the fabrication process in CISA, and SEM image of the 3D-printed eggbeater shaped structure
arrays; (b) the optical system in the CISA process; (d) the magnification of light guide tool; (c)
CAD model of eggbeater shaped structure; and (e) an illustration of adding microscale eggbeater
structure on the free form surface of a lotus flower, which changes the surface wettability from
hydrophilic to super-hydrophobic by using a straight light guide tool.
In CISA process, we firstly sliced the CAD model of eggbeater shaped structure in to a set of
2D images. To generate high-resolution 2D patterned light beam, each image was adjusted with

115
different gray scale value, reaching the uniformed light intensity. The resin was solidified on the
surface of the pre-existing object by the 2D patterned light beam, which was projected on the
surface of light guide tool. The exposure time was adjusted based on the exposure area of the 2D
patterned light beam. For example, it cost 35 s and 10 s respectively to fabricate the hair and stalk
of bionic eggbeater shaped structure. Since we studied the resin filling of the CISA process in
Section 4.3, the speed of light guide tool was set at the valid value, enabling the sufficient resin to
fill back to the curing area with the movement of the light guide tool. In addition, the head of
eggbeater shaped structures was composed of hollow structures, so that the curing property of the
E-glass/MWCNTs was studied to avoid the over-cure features in the vertical plane. The cure depth
of composite material was adjusted to only 25 μm.
6.6 Experiment results and analysis
In the Section 6.6.1, the oil/water separation of the 3D printed eggbeater shaped structure, of
which the hair number and gad distance was 4 and 0.4 mm respectively, was firstly demonstrated
at different conditions. The oil remove test of the 3D printed eggbeater shaped structure was
demonstrated in Section 6.6.2. Four types of oil were tested both in the oil/water separation and
oil remove. The 3D printed eggbeater shaped structure acted well on the droplet manipulation,
including non-loss transportation, droplet merging, droplet splitting, and all the demonstrations
will be shown in Section 6.6.3.
6.6.1 Oil/water separation test
To test oil/water separation, we firstly fabricated the eggbeater shaped structures with 4 hairs,
and a syringe was used to put an oil/water mixture droplet onto the top of the eggbeater structure.
A microscope based set-up was built to observe and record the whole process of oil water

116
separation. To facilitate the observation, 0.02 g/ml Oil Red Dye was added inside the oil. To verify
this assumption that the direction of eggbeater shaped structures has no effect on the oil/ water
separation efficiency, we tiled the 3D printed eggbeater shaped structure array with different
angles, including 0, 45, 90 and upside down 180 degrees. The oil quickly merged into the eggbeater
shaped structure within 6 s, and water formed a spheroid shape on the top surface of 3D printed
eggbeater shaped structure array (refer to Figure 59a-e). This is because the capillary force was
much bigger than the gravity of liquid.

117
Figure 59 The oil separation of 3D-printed eggbeater shaped structure at different conditions. (a)
Dust vacuum oil separation of the 3D-printed eggbeater shaped structure within 6s under
horizontal arrangement; (b) corn oil separation of the 3D-printed eggbeater shaped structure
within 6s under horizontal arrangement; (c) corn oil separation of the 3D-printed eggbeater
structures tilt upside down for 180 degrees; (d) tilt for 45 degrees;(e) tilt for 90 degrees; f)
comparison of separation rate 3D-printed eggbeater shaped structure for different types of oil;
and (g) changes of weight gain with different cycles [186].
Furthermore, four kinds of oil, including light and heavy crude oil, vacuum oil, and corn oil,
were tested and the absorption rates of eggbeater structures was shown in Figure 59e.The weight
gain of oil after several cycles was shown in Figure 59f. The oil cleaning efficiency was
quantitatively studied by thermogravimetric analysis (TGA). The separation efficiency was as high
as 90% and the average absorption rate of oil water separation was 12 g.g
-1
. s
-1
.
6.6.2 Oil remove test
. The oil absorption capacity of bio-inspired eggbeater shaped structures was investigated, and
the absorption of four kind of oil, which had different density and viscosity, was also studied. To
observe the oil absorption process, a glass tank containing water was placed on a table and oil
droplets were placed on the surface of water. The oil droplets formed a circle shape on the water
surface due to the low density and surface tension (refer to Figure 60). The same observation set-
up was used to record the oil absorption process by using eggbeater shaped structures. When an
oil droplet contacted the 3D-printed eggbeater shaped structures, it quickly filled inside the empty
space between each stalk, where was originally occupied by air layer. The whole cleaning process
was completed within 2s, suggesting a superior oil cleaning property of the eggbeater shaped

118
structures. By comparing the TGA of oil, water, permeate oil, and water, the results indicated that
a separation efficiency of 3D printed eggbeater structures was bigger than 99.9 %.

Figure 60 The absorption of different oils by the 3D-printed eggbeater shaped structure (N=4,
d=0.4mm). (a) The test is a light, paraffin base crude oil; (b) a heavy, asphalt base crude oil; (c)
dyed corn oil with oil red dye; and (d) dust vacuum oil respectively.
Furthermore, the oil occupied the space, where was the air layer between each stalk of eggbeater
shaped structures. The emptier space was inside eggbeater shaped structures; the more volume of
oil could be cleaned. Therefore, the efficiency of oil/water separation could be further significantly
improved by increasing the height of the stalk of eggbeater shaped structures and the gap between
each eggbeater shaped structure. To verify this assumption, we made the comparison of weight
gain of 3D-printed eggbeater shaped structures with different heights. Figure 61a shows the
pictures of 3D-printed eggbeater shaped structures, and the heights of 3D-printed eggbeater shaped
structures were 1.1mm, 1.8mm, 2.5mm and 3.2 mm, respectively. The results revealed that an

119
increment of height of 3D-printed eggbeater shaped structures increased the volume of air layer
which further improved oil cleaning capability of bionic eggbeater shaped structures (refer to
Figure 61c). The weight gain of oil removed by 3D-printed eggbeater shaped structures with the
height of 3.2 mm was twice of the one with the height 1.1 mm. Figure 61d-g showed the oil
absorption process for 3D-printed eggbeater shaped structures with the height of 3.2 mm. Overall,
a patterned array with 3D-printed eggbeater shaped showed promising properties in terms of oil
spill cleanup. As environmentally friendly materials, the smart 3D-printed eggbeater shaped will
have various prospective applications in the oil cleaning, wastewater treatment, and oil/water
separation under harsh conditions.

Figure 61 Oil absorption comparison of the 3D-printed eggbeater shaped structure with different
height. (a) Microscopy pictures of the 3D-printed eggbeater shaped structure (N=4, four hair)
with different heights; (b) H=1.1mm; (c) comparison of weight gain of oil absorption eggbeater
shaped structure with different heights, (d-f) diagrams show oil absorption process for H=3.2 mm.

120
6.6.3 Droplet manipulations
Because of the similarities in the geometric design, the 3D-printed eggbeater shaped structures
showed similar performance that it can grip the water droplet with hydrophilic microscale printed
hair. Different morphologies (numbers of eggbeater hairs, and gap between eggbeater shaped
structures) on the surfaces result in the tunable effect of the adhesion. Benefit from the controllable
adhesion, a lot of applications, such as droplet-based microreactors, non-loss water transport, 3D
cell culture, can be achieved with 3D-printed eggbeater shaped structures.

Figure 62. The droplet manipulation by 3D-printed eggbeater shaped structure. (a) The merge
process of several water droplets with different colors on the surface of 3D-printed eggbeater
shaped structure; (b) Split process of one dyed water droplet by 3D-printed eggbeater shaped
structure, both with N = 4; and (c) pictures show the non-loss water transport process from an
3D-printed eggbeater shaped structure (N=4) to another one with larger adhesive force (N=8).

121
Traditionally, liquid droplets are usually transported by a flowing carrier, which is filled with
the immiscible liquid [189], and the manipulation of micro-droplets highly relies on the auxiliary
systems [190]. In these system, operation devices are rather cumbersome and complex to achieve
the microscale droplet manipulation. In contrast, 3D-printed eggbeater structures, featuring special
high adhesion of liquid, can be used as an effective solution of manipulating droplets in a much
easier way (see in Figure 62). For instance, based on stable retention of micro-droplet, 3D printed
eggbeater structures can be applied as microreactors to observe the chemical reaction or cell
growth. As shown in Figure 62a, our 3D-printed superhydrophobic structures were used be the
reaction substrate for four types of water droplets. The droplet kept a sphere shape due to the
superhydrophobic property of the eggbeater surface. Based on micro droplet retention, eggbeater
shaped structure based microreactors have the advantages of high flexibility, significant material
saving, and low cost to distinguish different mole ratio of solvents. Such droplet-based micro
reactor has shown significant advances in protein crystallization, enzymatic kinetics, and other
biochemical reactions [190].
Besides, the water adhesive force of eggbeater structure was able to be adjusted by changing
the number of eggbeater arms (e.g. from 23 μN for N=2 to 55 μN for N=8). The water droplet can
be split by the different adhesion of 3D-printed eggbeater shaped structure, which is ascribed to
different wetting states and the changes of contact areas between the hair of 3D-printed eggbeater
shaped structure with a water droplet. For example, the dyed water droplet was split into two equal
parts due to the similar adhesive force (F1 ≈F2) by using eggbeater structures with N = 4, d=300
μm, and N = 4, d=400 μm (refer to Figure 62b). In comparison, the non-loss water droplet transport
can be achieved by using 3D-printed eggbeater shaped structure with significantly different
adhesive forces. As shown in Figure 62c, a water droplet can be completely transferred from 3D-

122
printed eggbeater shaped structure with N=4 to another 3D-printed eggbeater shaped structure with
N = 8. The sticky patterned surface was put upside down for the transport of water droplet, which
the adhesive forces of 3D-printed eggbeater shaped structures was F3 = 469 μN and F4 = 880 μN
respectively. Finally, the water droplet still maintained sphere shape when it was released on the
surface of 3D-printed eggbeater shaped structure with N=8. The difference of adhesive forces
enables 3D-printed eggbeater shaped structures to be a microscale robotic arm to transfer and split
small water droplets from a superhydrophobic surface to another one without loss or contamination,
which indicates various potential applications in biomedical engineering and microfluidic devices
[191-193].
6.7 Summary
In this chapter, we designed superhydrophobic and lipophilic surface with bionic multi-scale
eggbeater shaped structures inspired from salvinia molesta plant according to the real dimension
in nature. Then the lipophilic photocurable polymer with nanoscale multi-wall Carbon Nano-tubes
was developed to facilitate artificial eggbeater shaped structures macroscopically behaved
superhydrophobic and lipophilic. we used CISA to fabricate multi-scale eggbeater structures with
unique features, ranging from microscale to nanoscale. It opens intriguing perspectives for
designing artificial surfaces on the basis of 3D-printed eggbeater shaped structures to form a
superhydrophobic surface. The superhydrophobic property is associated with the hair number and
the gap distance between each stalk of 3D-printed eggbeater shaped structure. The wetting
characteristics of 3D printed superhydrophobic surface are governed by both chemical
composition and geometric structure. A new energy-efficient solution based on our 3D-printed
eggbeater shaped structures was demonstrated to separate different kinds of oil/water mixtures

123
with a high oil absorption rate. Furthermore, we quantified the oil absorption capacity of the
surface decorated with 3D-printed eggbeater shaped structures and investigated the influence of
the different design of 3D-printed eggbeater shaped structures on the efficiency of oil absorption.
The microscale water droplet also can be transported and separated by using 3D-printed eggbeater
shaped structures due to different attaching force. Therefore, as environmentally friendly materials,
the smart 3D-printed eggbeater shaped structures will have numerous prospective applications in
different areas [186].

124

Chapter 7 ISA based Hybrid Process Investigation for Biomimetic
Hierarchical Structure Fabrication
7.1 Overview
Nature provides inspirations of creative multi-functional materials and structures. However,
biological systems are usually composed of materials and structures in multi-scales with high
geometry complexity, which brings challenges to bioinspired design and fabrication process.
Additive manufacturing (AM) presents a potential solution due to its capability of creating three
dimensional (3D) objects with freeform surface and multi-materials. Vat photopolymerization is
an additive manufacturing process with good properties, such as high precision and fast fabrication
speed, and it has been widely used in various applications [37]. However, current vat
photopolymerization including both laser-based SL and mask-projection-based SL can only
achieve a single scale size with the ratio of fabrication extent to fabrication resolution at around
1000 [60-73]. However, multi-scale hierarchical structures in nature contain features ranging from
macroscale to nanoscale, where the ratio of the size span to the smallest geometry features could
be in the range of 10
8
- 10
9
[34, 86, 87]. The aim of Chapter 7 is to investigate the fabrication of
multi-scale bio-mimic textures by using hybrid process.
In this Chapter, we firstly investigated a multi-scale AM process that can fabricate bio-mimic
structures with multi-scale features ranging from nanoscale to macro- scales, and the ratio of the
size span to the smallest geometry features was more than 10
6
. In the proposed multi-scale printing
method, we applied ISA process to add microscale structures on the surface of macroscale and
mesoscale 3D structures fabricated by the MIP-SL process, and later the TPP process was used to

125
add nanoscale features on the surface of microscale structures printed by ISA process. Special
fixtures were developed in order to integrate the microscale ISA process with macro- and
mesoscale MIP-SL process and nanoscale TPP process. Two biomimetic applications were
demonstrated that using our proposed method has the capability to fabricate multi-scale objects.
Driven by the discoveries based on the multifunctional biological structures and recent material
advances, our multi-scale 3D printing process has the potential to enable the fabrication of smart
mechanical, fluidic and optical, structures that require large dimensional size span and small
geometric features [23, 34, 87]. We envision such fabrication capability would enable
multifunctional engineering systems for various applications in the future.
7.2 Background introduction
Many material systems from nature exhibit outstanding properties not found in artificial or
synthetic systems. The exceptional performances of natural material, such as the structural color
of butterfly [90], the superhydrophobic effect of salvinia paradox [95], and water collection of
desert beetles [169], benefits from hierarchical structures over a large range of scales from macro
scale to nanoscale [87]. Such multi-scale structures in nature provide inspiration for composite
material design, and show the promising multifunctional application in mechanical, optical,
thermal, and electrical fields [34]. Meanwhile, nature creatures and structures possess
comprehensive complexity in geometry, hierarchy, and material, setting challenges for mimicking
and reproducing the promising biological material system. Additive manufacturing (AM), as a
solution, can fabricate objects with complex design by depositing different material layer by layer
[1-3]. Due to this unique fabrication principle, a lot of progress has been made in the field of AM
of biomimetic structures and material [124].

126

Figure 63 An illustration of hierarchical structure in nature and the design & process planning of
multiscale hierarchical printing [89].
However, most of AM technologies can only fabricate 3D structure at one single scale, and
there is a big challenge by using AM process to reproduce multi-scale biological material. As we
discussed in Chapter 2, several printing approaches were recently developed to achieve the multi-
scale additive manufacturing. For example, multi-scale bio-mimic blood vessel was printed by
injecting biocompatible material into revisable gel through a microscale nozzle head [132-135].
To achieve high hydrodynamic performance, bio-mimic multi-scale shark skin was designed and
fabricated using microscale laser based printing process [194]. Advanced micro- and nanoscale
devices can be printed by electrohydrodynamic (EHD) printing [195]. However, time consumption

127
will be significantly high if this micro- and nanoscale leaning technique is used to fabricate large
scale structures [195]. Overall, most of the current AM fabrication procedures for printing multi-
scale hierarchical structures have experimental complexity and limited flexibility in structure
symmetry, especially when the desired multi-scale objects are developed with complex micro- and
nanoscale features [23]. Therefore, multi-scale AM hierarchical processes, which can cover a large
range of dimension from macroscale to nanoscale, is valuable to explore and develop for the
functional application in different fields of bionics.
Each type of AM processes is considered to have its most appropriate printing scope, in which
the optimized resolution is applied. It is desired to fabricate features at different scale using the 3D
printing process, whose scope matches the purposive fabrication. For example, based on the
natural architecture of gecko’s feet (refer to Figure 63), where macroscale, mesoscale, microscale
and nanoscale features are present, the fabrication process should be synchronized as the features
scale down in a hierarchical manner, where macroscale structures are built by macroscale process,
followed by micro and nanoscale features’ building through their scale-matching processes,
respectively. The stacking of these scale-exclusive fabrication methods demands the
understanding of the advantages of sub-process, and all processes should be integrated seamlessly.
Based on our previous work, MIP-SL is highly efficient in building objects with features that
are in macro- and microscale [9, 16, 196]. In the Chapter 3 and 4, ISA process was developed to
build microscale features on the surface of macroscale pre-existing object. Meanwhile, the TPP
process is an appropriate method to fabricate submicron features with high resolution and accuracy
[158,159]. In this chapter, a 3D hierarchical printing approach was proposed with an investigation
on integrating the MIP-SL, ISA and TPP processes for the fabrication of multi-scale structures
ranging from macroscale to nanoscale. To integrate multiple processes, process parameters were

128
further optimized for each sub-process, the key characteristics of the material was identified, and
alignment algorithm for the transition between procedures was investigated.
7.3 Multi-scale hybrid process development
7.3.1 Macroscale & mesoscale MIP-SL
MIP-SL at different scale have been investigated based on photopolymerization [65-67]. As
shown in Section 1.1.3, DMD is used to reflect the light from a light source into the surface of
polymer resin in MIP-SL (refer to Figure 64). The DMD is an electro-optical component that
comprises millions of micromirrors within a small area [164]. The angle of each mirror is
adjustable, and the illumination intensity of each pixel is tunable by adjusting the angle of the
corresponding micro mirror [164], In a macroscale or mesoscale MIP-SL setup, a computerized
system is used to process the CAD model and generates a set of mask images that constructs 3D
shape of printed object [4]. The projection light reflected by the DMD chip will be focused by a
group of optical lenses and scaled to a certain size, which defines the dimensional extent in the
fabrication [65-67]. A clear and sharp image is eventually projected at the resin surface for curing
material [51]. After one layer is solidified, a computerized Z stage will move up to refill liquid
resin, enabling the fabrication of next layer [53]. The MIP-SL process allows different sizes of
projection images to be used in the fabrication process, which provides flexibility on the
fabrication resolution at macroscale and mesoscale.
The projection areas of our macro- and mesoscale MIP-SL systems are designed to be 60 mm
× 45 mm and 15 mm × 11.25 mm, respectively. The DMD chips used in the developed MIP-SL
systems has 1024 × 768 micro-mirrors. Hence the pixel sizes in the macro- and mesoscale MIP-
SL systems are 58.6 µm and 14.6 µm, respectively. The smallest geometry features that can be

129
fabricated are around 50 – 100 µm. However, since the single pixel resolution cannot be easily
achieved in our MIP-SL setups, the macroscale and mesoscale MIP-SL is unable to fabricate
microscale structures in the range of 5 - 50 µm.
Because there are several problems in focusing quality of projection light, fabricating
microscale features using only several pixels is difficult in the MIP-SL process (refer to Figure
64). In our macro and mesoscale MIP-SL, 3D structures are fabricated by precisely controlling the
energy of each pixel in the projection image. The exposure curves of resins employed in MIP-SL
are quite nonlinear. Hence it is hard to achieve the fabrication of micro and mesoscale features by
using same exposure time in the MIP-SL processes [61, 62]. Besides, the curing resolution in the
vertical plane is also a limitation for the microscale feature fabrication by using the macro- and
mesoscale MIP-SL process, since the cure depth of material has to be reduced from 250 µm to 50
µm or even smaller for the microscale feature fabrication.
Meanwhile, the MIP-SL process is highly efficient in building objects with features that are in
macro- and mesoscale [38]. Because each layer of material can be cured by one-time exposure
using high resolution 2D light beam in MIP-SL process. Compared with other printing processes,
where the printing tool moves along all the fabrication area, MIP-SL showed fabrication efficiency
on a large-scale printing. Besides, the DMD chip contains millions of micro-mirrors, and the mirror
can generate complex and diverse 2D patterned light beam, enabling us to fabricate the large-scale
structures with fine details. However, the micro-features built by MIP-SL are less sharp than the
ones made by the ISA process due to the light accumulation effect. Hence, an ideal multi-scale
hierarchical printing method is to integrate both ISA and MIP-SL processes for the accurate
fabrication of a single object together with macro, and microscale features.

130
(a)
(b)
Figure 64 Schematic diagram and physical prototype of the mesoscale MIP-SL. (a)The design of
the macro & mesoscale MIP-SL; and (b) the physical prototype of MIP-SL process.
7.3.2 Microscale ISA
As shown in Chapter 3 and 4, the ISA process contains an optical imaging system, vision
alignment system, and multi-axis motion system (refer to Figure 65). The ISA process is unique
on the fabrication of 3D microscale structures. It is capable of fabricating complex geometries with
feature sizes that are around 10 µm. More importantly, since the ISA process shows flexibility in

131
building around insert, it can be used to add microscale surface structures on the surface of an
object that was fabricated by other manufacturing processes. By using different accumulation tools,
both inside and outside surfaces can be decorated with desired microscale structures (refer to
Figure 65b). Such fabrication capability by building-around-insert cannot be achieved by other
microscale fabrication processes. The above two key features enable ISA to be a promising
microscale solution to integrate with other processes for the fabrication of multi-scale structures.

Figure 65 The diagram of the ISA process. (a) The layout of the ISA technology; and (b) the
process planning to fabricate 3D microscale surface textures using the ISA technology.
Although the ISA process is versatile and flexible on fabricating microstructures, it has
limitations on building macro- and mesoscale structures. The largest dimension of the focused 2D
patterned beam is 2mm. It is time-consuming to use the ISA process to fabricate large scale
structures where the beam needs to scan the entire portion of each layer. What’s more, because the

132
projection image has to be focused at the place, where the microscale structures are added, the
curvature of the object is a critical parameter in the ISA process during the fabrication process.
Macro- and mesoscale fabrication increases the difficulty for the tool path planning, and the
collision of the light guide tool with the new cured features and pre-existing object shape may
occur due to special curvature of the object. Taking advantage of the ISA’s microscale fabrication
capability and the MIP-SL’s fast fabrication speed, it is beneficial to integrate the MIP-SL and
ISA processes to fabricate 3D objects with feature sizes that may vary from macro-, meso-, and
microscales.
7.3.3 Nanoscale TPP
Two-photon polymerization (TPP) is one type of microscale and nanoscale fabrication
technology [158, 159]. Because TPP is writing based approach, which the material is not
accumulated in the layer manner, TPP has capability to build geometries without topological
constraints [123]. Moreover, microscale and nanoscale structures are fabricated by the nonlinear
optical of light absorption of the photocurable polymer, so that TPP has the capability to fabricate
the submicron structures, which is not possible to build by other printing processes.
As shown in Figure 66a, the photosensitive polymer was crosslinked under the near infrared
emission, and the liquid material turned to be solid when it goes through the polymerization
process [123]. In TPP, high numerical aperture lenses and ultra-short and fast pulsed laser are used
to accomplish a multiphoton absorption [123]. Unlike the single photon absorption in
photopolymerization triggered by visible light or UV light, two or more photons are
simultaneously absorbed by the resin to activate the polymerization [123]. In TPP, the
photopolymerization only happens at the place where the light intensity reach the critical value,
and the value is determined by the photosensitivity of material [123]. Compared with MIP-SL, it

133
requires much higher energy to solidify the material. Thus, TPP can fabricate submicron features,
which is hard for the light at other wavelengths [123]. The volume of the focused laser beam
determines the fabrication resolution and efficiency of TPP [123].

Figure 66 Experimental setup and test result in TPP. (a) Set-up design; (b) mesoscale part built
by TPP; and (c) submicron model made by TPP [123].
TPP process is an excellent method for fabricating microstructures with submicron finesse
[123]. However, it is not ideal for creating macro- and mesoscale structures by using TPP. There
are several challenges in making large structures by TPP. Firstly, since the width of the focused
laser beam (voxel) is limited in micro level, it is time-consuming to fabricate macro- and mesoscale
structures by TPP, where the beam should scan the entire portion of the model. Furthermore,
because the laser beam should be focused at the same horizontal position, the flatness of the object
is a critical factor in the TPP during the fabrication. Therefore, macro- and mesoscale fabrication
increases the adhesion difficulty for the perfect flatness, and material cracks due to the poor
mechanical performance of material during the fabrication (refer to Figure 66b). Figure 66b
showed an example of the material crack during the fabrication, where a hybrid organic-inorganic
material was used to create mesoscale structure by TPP process, and it cost several hours to
fabricate the part with the whole dimension of 2 mm. However, TPP showed strength in the
microscale and nanoscale fabrication and the fabrication resolution is at submicron (refer to Figure

134
66c). To take advantages of high precision of TPP and expand it into future applications, it is
necessary to integrate the TPP with MIP-SL and ISA.
7.3.4 Integration of MIP-SL, ISA and TPP
In the integration of two processes, we first set up the prototype machine of both MIP-SL and
TPP processes, and then experiments were conducted to optimize the fabrication parameters for
each process respectively. A special fixture was investigated for the transition from MIP-SL
process to the TPP process. In order to detect and align each process, a vision assisted approach
was developed to discern the spatial relationship between the light beam and pre-printed objects.
7.3.4.1 Transition region optimization
During the transition between consecutive processes, e.g., fabricating subsequent features at
the same layer of the pre-built object, the attachment between the different scale of features at
transition region is a critical challenge. Particularly, for the overhanging feature with no solid
support from the structure built by the previous larger scale process, the newly cured feature only
attaches to the side wall of pre-built structure with limited adherence. Without any support
structures from the bottom layer of pre-built object, such newly cured feature maintains its shape
by the bending stress at the endpoint of each side, as shown in Figure 67. The bending force of
these features is minor due to limited contact area A and tensile stress of material 𝜎 𝜏 , so that the
deformation, which is caused by the material self-gravity load, results in the damage of microscale
structure when the new features are written on it. The deformation under self-gravity load can be
calculated based on the below equation:
𝑑 =
5𝜌𝑔 𝐿 5
384𝐸𝐼
(27)

135
where 𝜌 is the density, 𝑔 is the gravitational acceleration, L is the length of printed part, E is the
young’s modulus, and I is second moment of area.

Figure 67 An illustration of the transition process between multiple processes. (a) An illustration
of the transition structure design for the integration of multiple processes; (b) the glass slide mount
on the macroscale MIP-SL set-up; (c) the glass slide mount on the microscale ISA set-up; and (d)
the glass slide mount on nanoscale TPP set-up.
To solve the above problem, the area under the overhanging feature should firstly be filled
with a certain amount of supporting polymer that provides reinforcement underneath for the
microscale features to attach (refer to Figure 67a). Compared with original contact area A, the
deformation will be largely eliminated with straight supporting from the layer of material
underneath, and the micro and nanoscale features can be further built on the top surface of the
support. Overall, it is necessary for input CAD model to go through transition optimization process,
and the additional features have added to support the fabrication at each scale.

136
Besides, for adding micro and nanoscale structures on the top of the support, the material used
for building needs to be transparent to both visible and near-infrared photons. This is because TPP
was performed in this experiment using an up-right system like microscope, where transmission
light microscopy was employed to identify predetermined locations (refer to Figure 67b). The light
penetrable photocurable polymers, such as poly(ethylene glycol) diacrylate (PEGDA) and E-glass,
are therefore chosen to fabricate the support and the micro/nanoscale structures on top. The
transparency of material also helped the locating process of the fabrication region in the vision
assisted alignment system. Meanwhile, a glass fabrication platform was used as the building
substrate in our integration of MIP-SL, ISA, and TPP processes for its easy transition and superior
transparency. We cured several base layers of transparent material at the thickness of 500 m using
dental cure machine, to prevent thin layers of cured polymer from slipping off the smooth surface
of glass. Furthermore, we designed special glass substrate holder with magnet blocks for easy
transition of the microscope glass slide between MIP-SL, ISA, and TPP processes, and the
mounting mechanism of transparent glass substrate is shown in Figure 67b-d.
7.3.4.2 Vision assisted alignment
When multiple processes are being integrated, it is necessary to determine the relative location
of current curing light beam with respect to previous built geometry. To address this issue, an
optical vision system was designed and added to both microscale ISA and nanoscale TPP prototype
machine, allowing us to observe the location of the current curing light and pre-built object for the
purpose of calibration.
The vision system contains a beam splitter, a focusing convex lens, and a complementary
metal–oxide–semiconductor (CMOS) camera (refer to Figure 68). The light beam illuminating the

137
pre-built object travels back through the objective lens and beam-splitter, and is focused by a
convex lens. To observe the image of pre-built object and the focused light beam, a CMOS camera
is set up behind the focusing convex lens, such that the focused light can be captured by the CMOS
camera, and a clear image of fabrication region is obtained. The captured detection image is
displayed on a computer monitor with 50  magnifications. Using above designed vision assisted
system and the image obtained, the relative location of current curing light beam and the pre-built
object can be determined through following procedures.

Figure 68 An illustration of the design of vision system in ISA and TPP processes
Assume the center portion of the light beam (1920 x 1080) is used to build micro-features. The
movements in the XY directions are required for the alignment of the light beam and the pre-built
object. The X and Y stages move both the tank and the pre-built object in the related directions.
The target of the alignment of the two different processes is to get the relative position of the curing
light and pre-built object. For example, the positions of the curing light and pre-built object in

138
microscale ISA are showed in Figure 69. After moving X and Y stages to the home position, P0 ~
P3 are the four corners of the curing light beam, and M0 ~ M3 are the four corners of the processing
area of the pre-built object where the micro-features will be located. For each corner, there were
two related coordinates. One is the pixel position in the projection image, defined as CI (0, 1, 2, 3),
and the other is the actual position of the pre-built object on the platform, defined as CM. After
identifying the relative position between the M0 point and the P0 ~ P3 points, we learned the
relationship between M0 ~ M3 and P0 ~ P3. By adjusting the X and Y stage, the curing light beam
can be located at the relevant processing fabrication area.

Figure 69 An illustration of the alignment in ISA and TPP processes
The alignment process has the following steps:
Step 1: put a transparent checkboard on the top surface of the tank, which is used for marking
the positions of the curing light beam and pre-built object.
Step 2: project a full-size curing light beam. Then mark the four corners P0 ~ P3 and the relevant
position of the fabrication area in the microscale ISA system.

139
Step 3: mount the pre-built object and move the X and Y stages to make the point M0 closer to
the projection image_1’s corner P0. During the movement of the X and Y stages, record the steps
for each axis. The size of the pre-built object is big enough for us to make markers at the corners
M0~M4.
Step 4: when the two corners are close, try to capture the two corners both in the camera view.
Then move the X and Y stages by smaller steps and observe from the calibration visual system.
When they are merged together, we get the relative position between the two corners from the
captured image.
Step 5: use the same process for the other three corners of the pre-built object, M1~M3. We get
all the spatial relationship between the curing light beam and the place of the pre-built object on
which the feature with smaller scale will be built.
Our TPP system has a similar machine vision system and an XYZ stage to observe the pre-
fabricated micro-features related to the NIR laser. During the fabrication of submicron features on
the pre-built object by the ISA process, the calibration of the relative position is similar to the
aforementioned integration process of macroscale MIP-SL and microscale ISA systems. Based on
the above-developed integration process, the ISA and TPP setups are used to fabricate various
micro and nano scales objects.
7.3.4.3 Initial position identification
Besides, when fabricating the micro-features, the initial Z position is also important since the
deformation of the used PDMS film under pressure can be 100 m, which is significantly larger
than the layer thickness of 10 - 50 m used in the microscale ISA process. Thus, first few layers

140
of the feature cannot be fabricated at the beginning of the fabrication process, if there is no resin
filling in the projection area due to the deformation of PDMS. To avoid such situation, it is critical
to figure out the initial position of pre-existing object in the Z direction, when it is mounted on the
building platform of ISA system.
An imaging analysis method based on the machine vision system was applied in our study. As
the macroscale object moved down, the distance L between the PDMS and the macroscale object
decreased; hence, the volume of the resin between these two objects turned to be smaller. In the
visual system of the micro ISA setup, the color of resin shown in the captured image became lighter
with the decrease of distance L. When the outer surface of macroscale object touched the top
surface of PDMS, the resin between macroscale object and PDMS was squeezed out and the
surface of the macro-object appeared in the captured images. By dynamically performing the
image-analysis of the captured images, the relationship between the initial distance Li and the color
level R of the captured image was fitted. Based on the above method, the distance Li was calculated
during the movement of the platform. Finally, the platform was moved to the initial position until
the value of Li equals to the fabrication layer thickness of microscale ISA process.

Figure 70 An illustration of positioning the light guide tool. (a) Initial distance Li =0; (b) liquid
resin fills in when the inserted object moves up; (c) one feature is fabricated on the object surface;
and (d) multiple features can be built on the surface of the pre-existing object.

141
7.4 Case study I: macro & micro scale biomimetic cell culture
microenvironment fabrication
7.4.1 Design of biomimetic cell culture microenvironment
The nephron progenitor cells (NPCs) show poor expansion in vitro test, which brings the
difficulty in accessibility to have slowed basic study of renal development and diseases [197]. The
nephron progenitor cells require 3D culture environment to support the long-term expansion of
fetal NPCs. Inspired by natural micro texture, we designed multiple 3D cell culture environment
with different concave shaped scaffold as shown in Figure 71. In the concave-shaped scaffold, we
added a bar, rod and lobs shaped microstructures in the middle of anti-tetrahedron scaffold, which
can slightly to prevent all cells clustering in the center.

Figure 71 The CAD model of biomimetic microscale structures for long-term cell culture. (a) The
anti-tetrahedral-shaped microstructure; (b) rod-shaped microstructure; (c) bar-shaped
microstructure; and (d) lobe-shaped microstructure.

142
7.4.2 Material development
We designed and fabricated 3D cell culture microenvironment with anti-tetrahedron shaped
scaffold. The fabricated anti-tetrahedron shaped scaffold should be transparent so that the light can
easily transmit from the printed microstructure, and the cell growth status on the anti-tetrahedron
shaped scaffold can be observed by using an optical microscope. Besides, anti-tetrahedron shaped
scaffold should be non-degradable and biocompatible. Based on the above requirements, we
selected the poly (ethylene diacrylate) PEGDA as the material to fabricate the 3D cell culture
microenvironment. 60% (w/v) PEGDA based hydrogel solution is synthesize by the below
procedures. Firstly, we dissolved the 1 wt%photoinitiator photoinitiator (Irgacure 819, purchased
from BASF) into phosphate buffered saline (PBS). Then we added 60% PEGDA (Mw 700,
purchased from Sigma-Aldirich) into the above PBS solution, and used the magnetic stirrer to mix
the solution for 2h. Before the fabrication, the PEGDA solution was degassed by vacuum and kept
away from ambient light.
7.4.3 Fabrication of biomimetic cell culture microenvironment
In this Section, we presented the fabrication of biomimetic cell culture microenvironment by
using our proposed multi-scale 3D printing method, which integrated macro-& mesoscale MIP-
SL process and the microscale ISA process. The basic idea, hardware, and process development
as well as the detailed implementation were presented. The process planning of both MIP-SL and
ISA are developed, and the curing characteristics of materials in different processes were studied.
3D cell culture microenvironment with multi-scale were used as the test cases to demonstrate the
effectiveness of our proposed hybrid approach. By adding microscale structures inside macroscale
cell culture plate, the cells can be cultured in a much complex environment compared with the

143
original flat surface. Our newly developed multi-scale AM process could open broad prospects for
cell culture study with 3D biomimetic cell culture microenvironment.
7.4.3.1 Process planning
The MIP-SL process is highly efficient in building objects with macro- and mesoscale features,
and the ISA process is capable of fabricating microscale 3D structure on the surface of pre-existing
substrate. Hence an ideal multi-scale fabrication method is to integrate both MIP- SL and ISA
processes to fabricate 3D biomimetic cell culture microenvironment. In order to fabricate 3D
biomimetic cell culture microenvironment, we firstly built a mesoscale block inside macroscale
cell culture plate by using the MIP-SL process. Then, we printed the microscale cell culture
microscale scaffold inside the mesoscale block using light guide tool. (refer to Figure 72). After
that, the cell medium can be dropped into the 3D printed cell culture microenvironment.

Figure 72 The process planning diagram of the ISA based hybrid process for the fabrication of 3D
cell culture microenvironment
7.4.3.2 Fabrication results and analysis
Three-dimensional microscale biomimetic structures are necessary for long term cell and
embryoid body culture. For the traditional injection molding process, the only simple 2D mesh can
be fabricated. Such limited shape complexity restricts the understanding of cell growth in various

144
3D environments in medical research. To address the problem, we applied our multi-scale AM
process to fabricate the microscale anti-tetrahedron shaped scaffold inside a standard 24-walls cell
culture plate. Such a plate can be 3D printed by the macroscale MIP-SL process or pre-fabricated
by the injection molding. Several cell culture scaffolds with complex geometric shapes were built
to verify the printing capability of our proposed multi-scale AM process (refer to Figure 73 and
74).
The diameter of the original cell culture well was 15.6 mm, which consumed a significant
amount of cells and chemical solutions that may be expensive. To reduce the culture size, the
artificial cell culture environment contained the outside wall of scaffold, which prevent the leaking
of culture media including both cells and chemical solutions (refer to Figure 73a). The height of
such wall was 15 mm and the total XY size was 10 × 10 mm. We first used the mesoscale MIP-
SL process to build such wall using the biocompatible 60% PEGDA solution (refer to Figure 73d).
Inside the wall we designed 3D scaffold with anti-tetrahedron shaped scaffold (refer to the top
view of a cell in the culture plate). The digital model of the microscale anti-tetrahedron shaped
scaffold is shown in Figure 73b. The total height of the anti-tetrahedron shaped scaffold was 150
μm and the total XY size of the anti-tetrahedron shaped scaffold was 2.5 mm × 2 mm.
In the test, the light guide tool based on the optical acrylic rod was used to build the anti-
tetrahedron shaped scaffold inside the artificial culture environment printed by the MIP-SL process
(refer to Figure 73e). To achieve high resolution surface quality required by the microscale scaffold,
the slicing thickness of the anti-tetrahedron shaped scaffold was set at 3μm per layer. The setup
during the fabrication process and special fixture are shown in Figure 73f. The fabrication results
of the artificial cell culture environment with and without 3D anti-tetrahedron shaped scaffold are

145
shown in Figure 73h and j, respectively. The added 3D anti-tetrahedron shaped scaffolds instead
of a flat surface can provide significant advantages for the long-term cell growth study.

Figure 73 The design and fabrication of 3D cell culture microenvironment using the integrated
MIP-SL and ISA process. (a) Design of multi-scale artificial cell culture microenvironment; (b, e)
the design and fabrication of 3D anti-tetrahedron shaped scaffold; (c, d) the design and fabrication
of mesoscale scaffold wall; (f) the integration of ISA printing for the fabrication of 3D anti-
tetrahedron shaped scaffold; (g, i) the microscope image of without and with 3D printed scaffold;
and (h, j) the 3D cell culture microenvironment without and with microscale scaffold.

146
In order to better understand the cell culture growth on the micro-textured scaffold, we
designed scaffold with multiple geometric shapes. As shown in Figure 74, the concave scaffold of
anti-tetrahedron shape was built with one bar 30 µm in the center to inhibit the forming of
spheroids. To study the effect of dimension on the microscale concave scaffold, the printed
scaffold with textures in multiple dimensions were designed as shown in Figure 74b. Furthermore,
as shown in Figure 74c, microscale scaffold with different shapes can also be designed and
fabricated. The microscopy images of the 3D printed scaffolds with 3D concave textures are shown
in Figure 74c, f and i. The surface quality of the fabricated microscale scaffold is satisfactory,
which can be seen from the microscopy images. In addition, our multi-scale AM process enabled
cell culture study on the printed scaffold with different wall types.

Figure 74 The scaffolds with 3D microscale concave textures are printed using our integrated
MIP-SL and ISA processes. (a, d, g) Multiple designs of scaffold; (b, e, h) the fabrication results

147
of artificial cell culture scaffold by our proposed multi-scale AM process; (c, f, i) the microscope
image of the microscale scaffolds fabricated by the ISA process; and (j) the whole fabrication
process of artificial cell culture environments using our multi-scale AM process.
7.4.4 Case study I summary
The aim of this section is to investigate the fabrication of multi-scale 3D cell culture
microenvironment using the integrated process. The MIP-SL process was firstly introduced. Then
a multi-scale 3D printing method by integrating the ISA and MIP-SL processes was present, and
the process parameters to achieve high-resolution fabrication results were discussed. The
fabrication of biomimetic 3D cell culture microenvironment for long term cell culture was
demonstrated, and the experimental results verified the effectiveness and efficiency of this
proposed multi-scale 3D printing approach by integrating MIP-SL and ISA process.
7.5 Case study II: bioinspired optical filter with hierarchical structures ranging
from macroscale to nanoscale
7.5.1 Bioinspired design of hierarchical optical structure
In the natural world, living creatures shows dazzling color to attract other’s attention. For
example, the butterfly’s iridescent wings, fish’s glaring scales, and peacock’s gorgeous features.
These splendid colors are produced by the optical interaction of light with hierarchical biological
features [168]. In nature, living creatures have evolved to reflect and refract light in multiple ways
by their variety of nanoscale structures. Such structural color inspires the innovations of artificial
optical system [106]. In this section, we designed one kind of optical filter that contains the
nanoscale structures with the dimension near the wavelength of light.

148
To study the physical phenomenon of the optical filter with submicron structures, we
developed an optical filter with multi-scale hieratical structures. The schematic of optical
structures is shown in Figure 75. Firstly, we designed the macroscale holder to mount the optical
fiber so that the laser can go through the optical fiber. Then we designed a microscale optical filter
holder to support nanoscale structures. After that, we designed a nanoscale optical filter with
functional structures on the microscale holder. When the laser goes through the optical filter, the
light will be filtered by the nanoscale structures.

Figure 75 The design of bioinspired optical filter with hierarchical structures
7.5.2 Material development
In this optical application, the macrostructures were printed using composite material. The
composite material was prepared by mixing E-glass resin (purchased from EnvisionTEC Inc.) and
MWCNTs (purchased from Bucky USA. Inc). 14 mg MWCNTs (purchased from Bucky USA.
Inc) was added in 5mL polymer resin E-glass, and the new solution was stirred for 2h by using a

149
magnetic stirrer. To uniform distribute the MWCNTs, the mixture was put into ultrasonic bath for
30 mins. The microstructures were printed using transparent E-glass resin (purchased from
EnvisionTEC Inc.). All the materials were put in the vacuum to degas for 2h.
7.5.3 Fabrication of biomimetic optical filter by hybrid process
7.5.3.1 Process planning
The process planning of hierarchical printing was developed. We firstly printed macroscale
structure using MIP-SL, followed by the fabrication of microstructures on the top side of the pre-
printed macroscale structure by using ISA process. After that, we fabricated nanoscale structures
on the top of the micro features by using TPP, which is most advantageous at nanoscale fabrication.
Vision assisted alignment, which was discussed in Section 7.3.4, was applied to integrate MIP-SL,
ISA and TPP processes so that both microscale and nanoscale features was printed correctly on
the desired place.
7.5.3.2 Process parameter optimization
Photopolymerization based AM is an efficient 3D printing process that layers of photopolymer
resin are cross-linked upon exposure of 2D patterned light beam generated by the high-resolution
DMD chip. Each micro mirror corresponds to one pixel of the focused image, and the exposure
area of the 2D patterned light beam can be adjusted from several microns’ square to several
millimeters’ square [65-70]. The radiation of each pixel in the 2D patterned light beam implies a
Gaussian intensity profile, and the light intensity of the whole exposure area, consisted of multiple
Gaussian beams, is accumulated [58, 61]. The accumulated light beam photoinduced
polymerization reaction of photocurable polymer resin, when the absorbed energy surpasses the
reaction threshold. The uneven photopolymerization and unsatisfactory fabrication quality in the

150
photopolymerization based AM is usually caused by a non-uniform visible light intensity
distribution of the 2D light beam. The calibration of projection light is necessary to get accurate
fabrication result of photopolymerization based AM process, especially for microscale ISA
process [66].
A uniformed light intensity of projection light was generated by adjusting the gray scale of
each pixel in the 2D mask image. Meanwhile, the crosslinking of polymer is a nonlinear process
due to the multi-order of polymerization kinetics [39]. Therefore, the setting of exposure time was
optimized according to different light intensity based on the curing characteristics of photopolymer.
Besides, the cure depth of photopolymer was considered as the function of the concentrations of
photoinitiator and light absorber. The cure depth of material was investigated so that the newly
cured layer can be fully attached on the previous cured structure, without over-curing features
existing in the overhang structures along the fabrication direction.
A schema of photopolymerization based AM process at the mesoscale is displayed in Figure
76. The mesoscale optical fiber holder with gradually decreasing microscale inner hole features
was sliced into a series of 2D mask images. To fabricate this tapered shape hole feature, the mask
images were displayed in order on the DMD chip, and the 2D patterned light beam was projected
onto the photocurable resin with gradually increasing exposure area. Along with the movement of
Z stage, the mesoscale cylindrical holder was built layer by layer with series of mask images with
incremental size. The total height of cylindrical holder was 1.8 mm and the z stage moved up 60
μm / layer to maintain the same thickness of new refilling material. The radiation of each pixel Pi
(𝑥 𝑖 , 𝑦 𝑖 ) in the 2D patterned light beam implied a Gaussian intensity profile, and the overall light
intensity 𝐼 𝑥𝑦
of the entire exposure area, consisted of multiple Gaussian-conforming beams, was
accumulated. The whole light intensity can be represented by the below Equation:

151
𝐼 𝑥𝑦
= ∑ 𝐴 𝑖 𝑒𝑥𝑝 {−
1
2
[
(𝑥 −𝑥𝑖 )
2
𝜎 𝑥 2
+
(𝑦 −𝑦𝑖 )
2
𝜎 𝑥 2
]}
𝑛 𝑖 =1
(28)
where Ai represents the peak value of light intensity of pixel Pi, n is the number of pixels in the
projection area, 𝜎 𝑥 and 𝜎 𝑦 represents the shape of light distribution of pixel Pi .
The energy 𝐸 𝑥𝑦
that resin received from the projection with n pixels after exposure time T, can be
calculated by the below Equation:
𝐸 𝑥𝑦
= ∑
∬ 𝐴 𝑖 𝑇 ∗ 𝑒𝑥𝑝 {−
1
2
[
(𝑥 −𝑥𝑖 )
2
𝜎 𝑥 2
+
(𝑦 −𝑦𝑖 )
2
𝜎 𝑥 2
]}
𝑛 𝑖 =1
𝑑𝑥𝑑𝑦 (29)
When the absorbed energy 𝐸 𝑥𝑦
surpasses the critical threshold 𝐸 𝐶 of photo-induced
polymerization reaction, the corresponding area of photocurable polymer resin will be solidified.

Figure 76 The process parameter optimization in photopolymerization based AM process
We used the same exposure time as the first layer used, 𝑇 1
, to fabricate subsequent layers.

152
There were extra over-cured features existing in boundary of the final printed inner holes, resulting
in smaller inner holes than the desired shape. This phenomenon is because the illuminated areas
of following layers gradually turn to be larger compared with the one of first layer. Conversely,
the inner hole of first layer become larger in dimension if the optimal exposure time of final layer,
𝑇 35
, is used for all layers. We hypothesized that a less exposure time was required to fabricate the
layer whose corresponding mask image contained smaller inner black hole. Furthermore, when
decreasing the exposure time 𝑇 𝑛 in a linear fashion, there were still some extra over-cured features
at the boundary of the last few layers of inner holes. The inconsistent exposure setting resulted
from the non-linear nature of free radical polymerization of resin with energy accumulation of
neighborhood pixels. We hence applied a non-linear exposure time setting to maintain the printing
accuracy of MIP-SL, by adjusting the exposure time based on actual exposure area. Consequently,
the optimal 𝑇 1
and 𝑇 35
were set as 15 s and 8 s, respectively, and the time for interim layers were
obtained through interpolation in a nonlinear manner based on Eq.29. The desired tapered inner
hole was successfully built by MIP-SL. Several cross sections from the printing result of the hole
feature are shown in Figure 76. The nonlinear exposure setting can effectively eliminate
unexpected over-cured features previously existing due to energy accumulation, and significantly
enhanced the finishing quality of the part designed with microscale features.
7.5.3.3 Fabrication results
The hierarchical design of the multi-scale optical device with nanoscale filter was fabricated
to illustrate the resolution capability of the proposed hybrid process (refer to Figure 77a).
Specifically, the digital model of micro and nanoscale optical filter is shown in Figure 77b.
Microscale filter support (refer to Figure 77d) was printed with E-glass material on the top surface
of mesoscale optical fiber holder, by using the light beam generated based on the calibrated mask

153
image shown in Figure 77c. The mesoscale optical fiber holder was fabricated with the E-Glass
mixed with MWCNTs. The layer based macroscale MIP-SL and microscale ISA were used to
build this part, and the layer thicknesses of MIP-SL and ISA were set as 100 μm and 20 μm,
respectively. The nanoscale features are built using home-made LabView software capable of
importing 2D and 3D digital files and transforming them in polyline coordinates that are read by
the stacked XYZ stages used for TPP. The submicron optical filter was printed inside the hole of
microscale filter holder shown in Figure 71f.

Figure 77 The preliminary printing result of optical filter with hierarchical structures ranging
from macroscale to nanoscale. (a) The side view of design of optical filter with hierarchical
structures; (b)the microscale structure with nanoscale optical filter; (c) the picture of mask image
and 2D patterned light beam which is used to print the microscale optical filter support; (d) the
SEM image of microscale optical filter support printed by ISA process; (e)the mesoscale optical
fiber holder with microscale optical filter support; and (f) the nanoscale optical filter fabricated
on the surface of microscale optical filter support by TPP process.

154
7.5.4 Case study II summary
The aim of this case study is to demonstrate a fabrication capability of ISA based hybrid
process for the fabrication of hierarchical objects with macro-, micro- and nanoscale features.
Butterfly wing inspired optical filter was successfully fabricated by using the developed hybrid
process by integrating MIP-SL, ISA, TPP processes. Firstly, the mesoscale optical holder was
fabricated by using MIP-SL process, and the microscale filter support was fabricated by ISA
process. At the end, nanoscale feature was added on the top surface of microscale optical substrate
by TPP process. The fabrication setting of each process was studied and integration of each process
was also validated in this case study.
7.6 Summary
In this chapter, special hierarchical design and process planning towards integrating multiple
printing processes were first demonstrated. Then the integrated process of nanoscale TPP,
microscale ISA, and macroscale MIP-SL was introduced to address the multi-scale fabrication
challenge. MIP-SL, ISA, and TPP system were developed, and a machine vision assisted approach
was applied to the integration of different AM processes. Process parameters were optimized to
improve the fabrication quality of each process. Two testing cases were demonstrated to validate
the feasibility and efficiency of the proposed multi-scale hybrid printing approach. The results
showed the fabrication capability of the proposed multi-scale hierarchical printing process,
enabling the potential applications in different fields, such as interface technology, optics, biology,
and biomedical engineering.
155

Chapter 8 Conclusions and Recommendations
8.1 Answering the research questions
As stated in this thesis, this research mainly focused on the investigation and development of
novel printing approach called ISA and strategies for the fabrication of multi-scale biomimetic
structures. To achieve these goals, we proposed approaches to improve the printing capabilities in
accuracy by studying the material curing properties and optimizing the process parameters and to
expand the scale in photopolymerization based AM processes by integrating with macroscale MIP-
SL and nanoscale TPP. The research questions in this dissertation are list below:
Q1: How to fabricate high accurate microscale features by using photopolymerization based
microscale AM process?
Q2: How to reproduce multi-scale biomimetic design for functional applications?
For each question, there is three sub questions. The answer of each sub question is summarized as
follows:
Answer to Q 1.1: How to improve the surface quality of printed microscale features by using
ISA process?
The light guide tool is inserted in the resin tank, and the material was continuously accumulated
with the movement of light guide tool. The fabrication parameter and process planning of CISA
process were described in Chapter 4. Both theoretical and empirical quantitative validations were
presented using the examples in Section 4.4 and 4.7. In the demonstrated cases, the surface quality

156
of printed structures by using the CISA process was much smoother than the one printed by layer
based fabrication process.
Answer to Q 1.2: How to fabricate microscale structures with complex geometry by using ISA
process?
ISA process was developed to fabricate photocurable material (refer to Chapter 3 & 4). In the
ISA process, 2D patterned light beam, of which the resolution is 2.5 μm/pixel, was generated by a
DMD based optical design. To project the 2D patterned light beam on the fabrication area, light
guide tool was developed by using optical transparent material (acrylic rod and optical fiber). By
immersing the light guide tool in the resin tank, the material can be accumulated to form microscale
texture under the light exposure. The microscale texture built by ISA process can be as small as 8
μm. Besides, empirical qualitative validation was presented using examples (refer to Section 4.5),
and two case studies of bioinspired functional structures’ fabrication in Chapter 5 and 6. In all
cases, ISA showed the fabrication capability of microscale structures with complex 3D geometry.
Answer to Q 1.3: How should composite material be prepared for the microscale fabrication
of ISA to achieve nanoscale functionalities?
To fabricate MWCNTs based composite material, MWCNTs was mixed with photocurable
liquid resin, and when the photocurable liquid resin was solidified under the light exposure,
MWCNTs was embedded inside the cured material. Since the length of MWCNTs is 1-5 μm and
outer the diameter is 5-15 nm, the MWCNTs based composite material can be fabricated into the
microscale texture by using the ISA process. Meanwhile, MWCNTs generated nanoscale textures
on the surface of the printed object, and the roughness of the printed surface was increased by
adding MWCNTs, which can further modify the wettability of the printed surface. Based on the
empirical qualitative and quantitative validation in Section 5.3.3 and Section 6.3, we can make the

157
conclusion that MWCNTs based composite material can be printed by the ISA process at
microscale and achieve nanoscale functional texture at the same time.
Answer to Q 2.1: How to accurately build microscale textures on the surface of macroscale
substrate?
In this dissertation, vision assisted alignment algorithm was developed to precisely move the
light guide tool to the place where planned to add the microscale features. Besides, the initial
position identification was conducted to calculate the distance between light guide tool and the
surface of pre-existing macroscale object. Both studies in Section 7.3.4 and case study in Section
7.4 provides empirical qualitative and quantitative validation that the ISA process can integrate
with macroscale MIP-SL process to fabricate microscale features on the surface of macroscale
object.
Answer to Q 2.2: How to add nanoscale features on the surface of microscale substrate
fabricated by ISA process?
In this dissertation, vision assisted alignment algorithm of TPP process was developed to
precisely locate the printed laser beam to the place where will add the nanoscale features, and the
transition region optimization for the TPP and ISA process was studied. Both research in Section
7.3.4 and case study in Section 7.5 provides empirical and theoretical qualitative validation that
the ISA process can integrate with nanoscale TPP to fabricate nanoscale features on the surface of
microscale object.
Answer to Q 2.3: How to solve the transition problem between two different processes?
ISA process can be used to add microscale texture on the surface of the macroscale substrate.
In Section 7.4, multi-scale functional long term cell culture device was fabricated by integrating
macro and mesoscale MIP-SL with ISA. Similarly, multi-scale biomimetic optical structures with

158
the feature ranging from nanoscale to macroscale was fabricated by using a hybrid process, which
integrated macroscale MIP-SL, microscale ISA, and nanoscale TPP (refer to Section 7.5).
Optimization result in Section 7.3.4 and case study in Section 7.5 provides empirical and
theoretical qualitative validation that geometry optimization of multi-scale biomimetic design can
provide the support of printed features at different scale. ISA process is easy to integrate with other
manufacturing processes and the developed hybrid process can be used to fabricate multi-scale
biomimetic structures.
To answer the above hypotheses, we analyzed the problem and established appropriate
mathematical models, and the experiment was designed and conducted for theoretical analysis of
the problem. Based on the experiment data, a quantitative solution was further provided to solve
the problem. Four kinds of functional case study of the multi-scale biomimetic structures
fabrication based on ISA process were discussed in Chapter 5-7 in details.
8.2 Achievements and research contributions
During the past thirty years, the revolution of additive manufacturing (AM) has enabled
advances in producing complex products that are novel and sustainable. A paradigm shift in AM
from geometry-centered usage to function-focused applications is taking place recently, and
bioinspired manufacturing, among others, has become one of the major focuses [23, 34]. The
millions of years of evolution brings biological material system an extraordinary functional
performance, and the biological material system provides us the inspirations for the innovation in
different fields [34]. Integration of AM and biomimicry promotes possibilities in the fabrication
and processing of functional material [34]. My research offers a distinctive perspective on the

159
development of novel advanced manufacturing processes, with emphasis on manipulating and
mimicking the multi-scale structures in nature.
The dissertation topic is 3D printing of multi-scale biomimetic structures via Immersed Surface
Accumulation (ISA) printing. In this work, a novel Nano-composite printing process (ISA) was
developed to fabricate biomimetic multi-scale structures for specific functional requirements. It
utilized digital light projection, multi-axis motion control, and machine vision to achieve
functional material fabrication (Chapter 3 and 4). This newly developed process enabled to design
the morphology of artificial surfaces on the base of biomimetic structures to form functional
surfaces. For example, bionic spine array inspired by cactus was reproduced by ISA process to
achieve high water collection (Chapter 5), and multi-scale eggbeater shaped structures on the
Salvinia molesta leaf with intriguing “super-hydrophobic” property, was replicated by the ISA
process to provide various potential applications in multiple fields (Chapter 6). In addition, by
integrating ISA with other manufacturing technologies, a hybrid process was developed to enable
the design and manufacturing of multi-scale functional devices in a variety of areas (Chapter 7).
To demonstrate the multi-scale printing capability, an optical component designed with
bioinspired hierarchical structures, of which the size ranges from macroscale to nanoscale, was
fabricated by the proposed hybrid process (Section 7.5). In my dissertation, I conducted research
in the field of and the main contribution are listed below:
Contribution related to microscale fabrication (H1):
 Develop a novel microscale printing process, enabling to build microscale biomimetic
structures on the surface of the macroscale substrate with flexible light guide tool.
Different from the traditional microscale fabrication approaches, ISA printing showed

160
unique printing capability of building around insert due to the developed light guide tool.
Microscale features can be built on the surface of macroscale substrate with different
curvatures by using special shaped light guide tool (Section 3.4).
 Generate the energy distribution model of the 2D patterned light beam for the radical
polymerization in microscale fabrication. It enables the microscale fabrication because
of the uniformed light intensity. An algorithm, which is based on the light energy
distribution model, was proposed to increase the fabrication accuracy of ISA process
(Section 3.5).
 Experimentally model the curing performance of composite material with the exposure
of 2D patterned light beam and develop the appropriate photocurable material for
different functionalities. The overcuring of material in both horizontal and vertical
directions were eliminated based on the study of curing performance of material (Section
4.1). According to the functional requirements, the photocurable material was developed
and the curing performance was fitted by the proposed model (Section 5.3, Section 6.4,
Section 7.4.2 and Section 7.5.2).
 Experimentally study physical models of fluid flow in the printing process and optimized
the process parameters. The material filling and moving speed are two key factors in
ISA process. After study the physical model of fluid flow in the printing area, the valid
fabrication dimension can be calculated based on the moving speed of light guide tool
(Section 4.4). What’s more, the proposed material filling model is applicable in a wide
variety of situations where the material is photo cured by projection image in the layer
or continuous manner.

161
 Developed ISA based continuous microscale printing process to achieve smooth surface
quality. Most of AM process build 3D object in a layer manner, which generates serious
stair-effect. The surface quality of printed part with stair effect heavily influence its
functional performance. A continuous ISA printing process was developed to improve
the surface quality. The surface quality of structure printed by continuous ISA process
is much smoother than the layer based approach (Section 4.7).
Contribution related to multi-scale biomimetic design fabrication (H2):
 Develop algorithm to design transition region to support the structures fabricated by the
manufacturing process at different scale. To fabricate the biomimetic design with multi-
scale structures by using hybrid process, the biomimetic design has to be decomposed
into three sections: macroscale structures, microscale structures and nanoscale structures.
The shape of each portion requires to be optimized based on the developed algorithm by
adding transition region for the purpose of manufacturing (Section 7.3.4). With the
developed algorithm, the structures fabricated by the manufacturing process at different
scale can bonding together without the deformation. It provides a design method for
multi-scale fabrication, which can be widely used in hybrid process.
 Experimentally explore the integration of ISA with other photopolymerization based
printing process by using vision assisted system. The vision assisted system was
developed to integrate multiple photopolymerization based printing process (Section
7.3.4). This vision assisted method can be also used in the multi-scale hybrid process
development by integrating different manufacturing methods.

162
 Investigate the integrated multi-scale manufacturing process solution for the fabrication
of biomimetic functional material and structures, and formulate quantitative models to
characterize and optimize the manufacturing process parameters. To solve the challenge
of multi-scale fabrication, a solution was provided in this dissertation based on the
multiple photopolymerization based AM process (MIP-SL, ISA, TPP). The problem of
integration was formed and solved by the developed method (Section 7.3.4). The
manufacturing parameter for each process was characterized and optimized to achieve
the fabrication of multi-scale biomimetic design.
Meanwhile, in my Ph.D. study, I collaborated with researchers in Center of Craniofacial
Molecular Biology, School of Dentistry of USC, on the Alfred E. Mann Institute for Biomedical
device funded project “Innovative craniofacial and long bone regeneration with 3D printed
scaffold”, in which bio-materials was explored and corresponding AM tools was further developed
to design and fabricate biodegradable scaffolds that contribute to bone formation. This innovation
would greatly impact current healing procedures of bone critical defects, and has the potential to
reform and expedite bone regeneration process. Besides, hair inspired liquid sensor was developed
based on swelling behavior of hydrogel and conductive multi-wall carbon Nanotubes [68]. In this
work, multi-material stereolithography based 3D printing process was applied in to the design and
fabrication of multi-functional flexible liquid sensors [68]. The liquid sensor for micro-droplet,
made with hydrophilic hydrogel, shows remarkable high sensitivity and quick response. The liquid
sensing property of PEGDA based composite hydrogel is related to swelling deformation and
absorbency capability of hydrogel [68]. PEGDA hydrogel with and without MWCNTs show
opposite signal reactions at the same wetting condition. The micro droplet sensing of our 3D-
printed liquid sensor indicates various potential applications in biomedical engineering and

163
environmental science [68]. It opens intriguing perspectives for designing 3D-printed monolithic
sensor with special distribution of materials to achieve anisotropic detection. Considering the
efficient sensing mechanism, the 3D-printed liquid sensor for micro-droplet has potentially
implications in a variety of fields, such as disease detection, micro fluid monitor, micro-actuator
and robot.
8.3 Limitation and future work
In the past, AM processes are mainly used to build structural components from a wide variety
of plastics, metals, and rubber materials; the current trend of AM research is moving toward the
construction of functional synthetic devices based on interdisciplinary technologies and process
innovations [23, 34]. The research interest of this dissertation is to develop advanced
manufacturing processes with emphasis on building multi-scale biomimetic structures. The study
includes, but not limited to, material development, advanced manufacturing process development,
and relevant applications’ investigation in multidisciplinary areas.
In this dissertation, ISA was developed mainly focused on the fabrication of microscale
features, and it enables the fabrication of microscale features array with one time exposure. Even
ISA improved the fabrication efficiency compared with traditional laser based approach, it still
takes long time to repeat microscale patterned features with large scale area. There is a tradeoff
between the fabrication resolution and fabrication efficiency for the ISA process. How to solve
this challenge is the one problem of ISA process. Another problem is ISA based hybrid process.
Only multi-scale object with certain type of hierarchical structures can be fabricated by the
proposed hybrid process. It is hard to fabricate the object with nested multi-scale features. What’s
more, the material, which can be fabricated by ISA process, is developed based on photocurable

164
polymer. The mechanical strength of printed part is not strong enough for long time. It can be
improved with the help of material investigation of the integration of physical filed.
Some of the research topics based on ISA process that can further be pursued in the coming
years are listed as follows.
8.3.1 Novel additive manufacturing process development
Current AM processes mainly fabricate a part with single or few types of material in a layer-
based manner. Besides the advantages of AM processes, it also brings some concomitant issues,
typically in material properties, surface finish, fabrication speed, etc. In future research, new
concepts will be explored based on ISA process for the development of new AM processes. Some
potential directions are listed as follows.
8.3.1.1 Multi-directional layer-less 3D printing
A new material refilling method will be explored to further improve the ISA based continuous
printing process. With optimized material refilling, a solid part with large cross-section area can
be continuously fabricated with ultra-fast speed. Besides, ISA process will be extended to expedite
the fabrication of high viscous composite material [162, 186], e.g., nanotube ink, which usually
takes several hours by current AM methods. Furthermore, current 3D objects are fabricated in a
2.5-dimensional manner, where the material is deposited along with one direction [198, 199]. A
layer-less printing will be developed to directly build the 3D objects.
8.3.1.2 Physical field assisted multi-material printing
The possibility of printing multiple types of materials in a single piece is attracting a lot of
attention due to many strategic applications in different fields [200, 201]. However, it is still

165
difficult to fabricate structures designed with different types of materials by using most of current
3D printing processes. Physical field assisted printing, therefore, becomes necessary to fill the
vacancy of the increasing demand. Nacre inspired lightweight and high strength materials can be
jointly built by electrically assisted 3D printing method. As an extension of this study, physical
field assisted ISA printing process will be developed in my future research, aiming at achieving
multiple types of materials fabrication by using single process. Subsequent research, including
physical field design, process optimization, geometric modeling, as well as functional applications,
will be conducted.
8.3.2 Bioinspired functional material fabrication via ISA based nanocomposite printing
Biomimicry is driving a paradigm shift in biomimetic AM from natural material system to
functional applications [34]. However, replicating natural materials and structures by current AM
technologies remains a challenge. Many opportunities exist for developing new AM methods for
the fabrication of bioinspired functional devices.
8.3.2.1 3D-printing of bioinspired hydrodynamic functional surface
Nature provides inspirations on the design principles for the construction of multifunctional
hydrodynamic functional surface. One of case study in this dissertation is focused on Salvinia
Molesta inspired superhydrophobic structures, which demonstrated potential applications in micro
droplet manipulation and oil/water separation [186]. One future research interest of ISA lies in the
incorporation of Nano-composite printing process and digital design tools into the fundamental
study on development of functional liquid manipulation robot and eco-friendly cleaning products
with different types of bioinspired hydrodynamic surfaces, e.g., low/high adhesive
superhydrophobic surface, superamphiphobic surface, and thermal insulation surface [202].

166
8.3.2.2 Multi-scale printing of bioinspired hierarchical structures
Material systems from nature exhibit outstanding properties not found in artificial or synthetic
systems. The exceptional performances of natural material benefit from hierarchical structures
over a large range of scales from macroscale to nanoscale [34]. It is critical to understand the
biological hierarchical structures and replicate their material system for the functional application
sin engineered systems [34]. However, the manufacturing capability of current AM technologies
is intrinsically specified within a certain scope. It is desired to develop multi-scale advanced
manufacturing process to narrow this gap between scale of features in hierarchical structures. In
future, multi-scale hybrid fabrication processes based on ISA will be explored as a solution to
build bioinspired hierarchical structures for applications such as flexible electronics, body on a
chip, microfluidic device, and structural color filtering device [34].
8.3.3 3D printing medical devices for healthcare
Bio-printing shows promising advances in the development of biomedical devices, and diverse
types of AM processes are developed to construct functional biomedical devices. However,
effective AM process for functional biomedical devices is still scarce, and will be one application
area of ISA process in future.
8.3.3.1 3D printing functional scaffold for tissue regeneration
One application of multiscale ISA Nano-composite printing process can be the fabrication of
implantable scaffold with programmable drug release. In addition to providing a good growing
environment for stem cell, this implantable scaffold enables local and sustained release of
bioactive molecules to the target place for the acceleration of bone tissue regeneration. This kind
of implantable scaffold eliminates the need to inject drugs dosage at specific time intervals and

167
shows advantages over traditional scaffolds. The geometric design, photo kinetics, and drug
loading play major roles in drug release during the degradation of scaffolds. Part of future work of
ISA based scaffold fabrication can be focused on addressing the open issues in material selection,
geometric shape design, and drug release mechanism of biomedical scaffold for bone tissue
regeneration.
8.3.3.2 3D printing cell culture microenvironment
Tissues are made up of a highly complex 3D arrangement of cells. Conventional 2D cell culture
system fails to mimic complex tissues. But well-defined 3D cell culture microenvironment would
be able to reflect the behavior of tissue growth in vivo [163]. Construction of 3D cell culture
microenvironments by AM technologies provides the possibility to closely simulate the natural
tissue. In Section 6.6.3, the hanging drop based 3D cell culture microenvironment enabled by
superhydrophobic structures were developed to understand 3D aggregation of the cancer cell. Due
to the advantages of ISA process for freeform fabrication of biological materials, more innovative
designs of 3D cell culture microenvironments, such as structure-based platform and micro-fluid
based cell culture plate, will be explored in the upcoming years. This kind of 3D printed bio-
inspired 3D cell culture microenvironment has potential applications in a variety of fields, such as
therapeutic efficacy of drugs, gene function analysis, model for cell-cell interactions, 3D model in
nanomedicine research, and medical diagnosis [203].
168

Reference
[1] Mueller, B., 2012. Additive manufacturing technologies–Rapid prototyping to direct digital
manufacturing. Assembly Automation, 32(2).
[2] Wong, K.V. and Hernandez, A., 2012. A review of additive manufacturing. ISRN Mechanical
Engineering, 2012.
[3] Vaezi, M., Seitz, H. and Yang, S., 2013. A review on 3D micro-additive manufacturing
technologies. The International Journal of Advanced Manufacturing Technology, 67(5-8),
pp.1721-1754.
[4] Xu, K. and Chen, Y., 2015. Mask image planning for deformation control in projection-based
stereolithography process. Journal of Manufacturing Science and Engineering, 137(3), p.031014.
[5] Lipson, H. and Kurman, M., 2013. Fabricated: The new world of 3D printing. John Wiley &
Sons.
[6] Hull, C., 2012. On stereolithography. Virtual and Physical Prototyping, 7(3), pp.177-177.
[7] Duda, T. and Raghavan, L.V., 2016. 3D metal printing technology. IFAC-PapersOnLine,
49(29), pp.103-110.
[8] Wang, X., Jiang, M., Zhou, Z., Gou, J. and Hui, D., 2017. 3D printing of polymer matrix
composites: A review and prospective. Composites Part B: Engineering, 110, pp.442-458.
[9] Yang, Y., Li, X., Chu, M., Sun, H., Jin, J., Yu, K., Wang, Q., Zhou, Q. and Chen, Y., 2019.
Electrically assisted 3D printing of nacre-inspired structures with self-sensing capability. Science
advances, 5(4), p.eaau9490.
[10] Song, J.H., Edirisinghe, M.J. and Evans, J.R., 1999. Formulation and multilayer jet printing
of ceramic inks. Journal of the American Ceramic Society, 82(12), pp.3374-3380.

169
[11] Klein, J., Stern, M., Franchin, G., Kayser, M., Inamura, C., Dave, S., Weaver, J.C., Houk,
P., Colombo, P., Yang, M. and Oxman, N., 2015. Additive manufacturing of optically transparent
glass. 3D Printing and Additive Manufacturing, 2(3), pp.92-105.
[12] Zhang, Q., Zhang, F., Medarametla, S.P., Li, H., Zhou, C. and Lin, D., 2016. 3D printing
of graphene aerogels. Small, 12(13), pp.1702-1708.
[13] Wei, X., Li, D., Jiang, W., Gu, Z., Wang, X., Zhang, Z. and Sun, Z., 2015. 3D printable
graphene composite. Scientific reports, 5, p.11181.
[14] Mironov, V., Boland, T., Trusk, T., Forgacs, G. and Markwald, R.R., 2003. Organ printing:
computer-aided jet-based 3D tissue engineering. TRENDS in Biotechnology, 21(4), pp.157-161.
[15] Mironov, V., Visconti, R.P., Kasyanov, V., Forgacs, G., Drake, C.J. and Markwald, R.R.,
2009. Organ printing: tissue spheroids as building blocks. Biomaterials, 30(12), pp.2164-2174.
[16] Li, X., Xie, B., Jin, J., Chai, Y. and Chen, Y., 2018. 3D Printing Temporary Crown and
Bridge by Temperature Controlled Mask Image Projection Stereolithography. Procedia
Manufacturing, 26, pp.1023-1033.
[17] Cummins, G. and Desmulliez, M.P., 2012. Inkjet printing of conductive materials: a review.
Circuit world, 38(4), pp.193-213.
[18] Olakanmi, E.O., Cochrane, R.F. and Dalgarno, K.W., 2015. A review on selective laser
sintering/melting (SLS/SLM) of aluminium alloy powders: Processing, microstructure, and
properties. Progress in Materials Science, 74, pp.401-477.
[19] Meteyer, S., Xu, X., Perry, N. and Zhao, Y.F., 2014. Energy and material flow analysis of
binder-jetting additive manufacturing processes. Procedia CIRP, 15, pp.19-25.

170
[20] Mohamed, O.A., Masood, S.H. and Bhowmik, J.L., 2015. Optimization of fused deposition
modeling process parameters: a review of current research and future prospects. Advances in
Manufacturing, 3(1), pp.42-53.
[21] Wanji, C. and Zhen, W., 2008. A selective review on recent development of displacement-
based laminated plate theories. Recent patents on mechanical engineering, 1(1), pp.29-44.
[22] Lewis, J.A., Smay, J.E., Stuecker, J. and Cesarano, J., 2006. Direct ink writing of three ‐
dimensional ceramic structures. Journal of the American Ceramic Society, 89(12), pp.3599-3609.
[23] Leung, Y.S., Kwok, T.H., Li, X., Yang, Y., Wang, C.C. and Chen, Y., 2019. Challenges
and Status on Design and Computation for Emerging Additive Manufacturing
Technologies. Journal of Computing and Information Science in Engineering, 19(2), p.021013.
[24] Ngo, T.D., Kashani, A., Imbalzano, G., Nguyen, K.T. and Hui, D., 2018. Additive
manufacturing (3D printing): A review of materials, methods, applications and challenges.
Composites Part B: Engineering, 143, pp.172-196.
[25] Fu, K., Yao, Y., Dai, J. and Hu, L., 2017. Progress in 3D printing of carbon materials for
energy ‐related applications. Advanced Materials, 29(9), p.1603486.
[26] Kang, H.W., Lee, S.J., Ko, I.K., Kengla, C., Yoo, J.J. and Atala, A., 2016. A 3D bioprinting
system to produce human-scale tissue constructs with structural integrity. Nature biotechnology,
34(3), p.312.
[27] Giffi, C.A., Gangula, B. and Illinda, P., 2014. 3D opportunity for the automotive industry.
Additive manufacturing hits the road.
[28] Zhu, Z., Guo, S.Z., Hirdler, T., Eide, C., Fan, X., Tolar, J. and McAlpine, M.C., 2018. 3D
printed functional and biological materials on moving freeform surfaces. Advanced Materials,
30(23), p.1707495.

171
[29] https://www.geeky-gadgets.com/cocojet-chocolate-3d-printer-unveiled-by-3d-systems-at-
ces-2015-07-01-2015/
[30] https://www.machinedesign.com/3d-printing/metal-3d-printing-brings-new-design-
capabilities-aerospace-engines/gallery?slide=7
[31] Sun, K., Wei, T.S., Ahn, B.Y., Seo, J.Y., Dillon, S.J. and Lewis, J.A., 2013. 3D printing of
interdigitated Li ‐Ion microbattery architectures. Advanced materials, 25(33), pp.4539-4543.
[32] Khoshnevis, B., 2005. Automated Construction by Contour Crafting (Presentation). In IIE
Annual Conference. Proceedings (p. 1). Institute of Industrial and Systems Engineers (IISE).
[33] Tofail, Syed AM, Elias P. Koumoulos, Amit Bandyopadhyay, Susmita Bose, Lisa
O’Donoghue, and Costas Charitidis, 2018. Additive manufacturing: scientific and technological
challenges, market uptake and opportunities. Materials today 21(1), pp. 22-37.
[34] Yang, Y., Song, X., Li, X., Chen, Z., Zhou, C., Zhou, Q. and Chen, Y., 2018. Recent
progress in biomimetic additive manufacturing technology: from materials to functional
structures. Advanced Materials, 30(36), p.1706539.
[35] Melchels, F.P., Feijen, J. and Grijpma, D.W., 2010. A review on stereolithography and its
applications in biomedical engineering. Biomaterials, 31(24), pp.6121-6130.
[36] Li X, Mao H, Pan Y, Chen Y., 2019. Mask Video Projection-Based Stereolithography With
Continuous Resin Flow. ASME. J Journal of Manufacturing Science and Engineering,141(8),
pp.081007-081007-10. doi:10.1115/1.4043765.
[37] Gibson, I., Rosen, D. and Stucker, B., 2015. Vat photopolymerization processes. In
Additive Manufacturing Technologies (pp. 63-106). Springer, New York, NY.
[38] Jacobs, P.F., 1995. Stereolithography and other RP&M technologies: from rapid
prototyping to rapid tooling. Society of Manufacturing Engineers.

172
[39] Andrzejewska, E., 2001. Photopolymerization kinetics of multifunctional monomers.
Progress in polymer science, 26(4), pp.605-665.
[40] Bertsch, A. and Renaud, P., 2016. Microstereolithography. In Three-Dimensional
Microfabrication Using Two-photon Polymerization (pp. 20-44). William Andrew Publishing.
[41] Yang, Y., Chen, Z., Song, X., Zhu, B., Hsiai, T., Wu, P.I., Xiong, R., Shi, J., Chen, Y.,
Zhou, Q. and Shung, K.K., 2016. Three dimensional printing of high dielectric capacitor using
projection based stereolithography method. Nano Energy, 22, pp.414-421.
[42] Zheng, X., Lee, H., Weisgraber, T.H., Shusteff, M., DeOtte, J., Duoss, E.B., Kuntz, J.D.,
Biener, M.M., Ge, Q., Jackson, J.A. and Kucheyev, S.O., 2014. Ultralight, ultrastiff mechanical
metamaterials. Science, 344(6190), pp.1373-1377.
[43] Song, X., Chen, Y., Lee, T.W., Wu, S. and Cheng, L., 2015. Ceramic fabrication using
mask-image-projection-based stereolithography integrated with tape-casting. Journal of
Manufacturing Processes, 20, pp.456-464.
[44] Raman, R. and Bashir, R., 2015. Stereolithographic 3D bioprinting for biomedical
applications. In Essentials of 3D Biofabrication and Translation (pp. 89-121). Academic Press.
[45] Chen, X., Liu, W., Dong, B., Lee, J., Ware, H.O.T., Zhang, H.F. and Sun, C., 2018. High ‐
speed 3D printing of millimeter ‐size customized aspheric imaging lenses with sub 7 nm surface
roughness. Advanced Materials, 30(18), p.1705683.
[46] Chen, X., 2018. Scalable Micro and Nano Manufacturing of Photonic Devices (Doctoral
dissertation, NORTHWESTERN UNIVERSITY).
[47] Liu, Feng, Biqin Dong, and Xiaohan Liu. "Bio-Inspired Photonic Structures: Prototypes,
Fabrications and Devices." In Optical Devices in Communication and Computation. IntechOpen,
2012.

173
[48] Kuang, M., Wang, J., Bao, B., Li, F., Wang, L., Jiang, L. and Song, Y., 2014. Inkjet printing
patterned photonic crystal domes for wide viewing ‐angle displays by controlling the sliding three
phase contact line. Advanced Optical Materials, 2(1), pp.34-38.
[49] Bártolo, P.J. ed., 2011. Stereolithography: materials, processes and applications. Springer
Science & Business Media.
[50] Mao, H., Leung, Y.S., Li, Y., Hu, P., Wu, W. and Chen, Y., 2017. Multiscale
Stereolithography Using Shaped Beams. Journal of Micro and Nano-Manufacturing, 5(4),
p.040905.
[51] Xu, K. and Chen, Y., 2017. Photocuring temperature study for curl distortion control in
projection-based stereolithography. Journal of Manufacturing Science and Engineering, 139(2),
p.021002.
[52] Zhang, J., Yang, Y., Zhu, B., Li, X., Jin, J., Chen, Z., Chen, Y. and Zhou, Q., 2018.
Multifocal point beam forming by a single ultrasonic transducer with 3D printed
holograms. Applied Physics Letters, 113(24), p.243502.
[53] Pan, Y., Zhou, C. and Chen, Y., 2012. A fast mask projection stereolithography process
for fabricating digital models in minutes. Journal of Manufacturing Science and Engineering,
134(5), p.051011.
[54] Zhou, C., Chen, Y., Yang, Z. and Khoshnevis, B., 2013. Digital material fabrication using
mask-image-projection-based stereolithography. Rapid Prototyping Journal, 19(3), pp.153-165.
[55] Skoog, S.A., Goering, P.L. and Narayan, R.J., 2014. Stereolithography in tissue
engineering. Journal of Materials Science: Materials in Medicine, 25(3), pp.845-856.

174
[56] Pan, Y., Chen, Y. and Yu, Z., 2017. Fast mask image projection-based micro-
stereolithography process for complex geometry. Journal of Micro and Nano-Manufacturing, 5(1),
p.014501.
[57] Choi, J.W., Wicker, R., Lee, S.H., Choi, K.H., Ha, C.S. and Chung, I., 2009. Fabrication
of 3D biocompatible/biodegradable micro-scaffolds using dynamic mask projection
microstereolithography. Journal of Materials Processing Technology, 209(15-16), pp.5494-5503.
[58] Zhou, C. and Chen, Y., 2009, August. Calibrating large-area mask projection
stereolithography for its accuracy and resolution improvements. In Proceedings of Solid Freeform
Fabrication Symposium, Austin, Texas.
[59] Kang, H.W., Park, J.H. and Cho, D.W., 2012. A pixel based solidification model for
projection based stereolithography technology. Sensors and Actuators A: Physical, 178, pp.223-
229.
[60] Xu, K. and Chen, Y., 2015. Mask image planning for deformation control in projection-
based stereolithography process. Journal of Manufacturing Science and Engineering, 137(3),
p.031014.
[61] Zhou, C. and Chen, Y., 2012. Additive manufacturing based on optimized mask video
projection for improved accuracy and resolution. Journal of Manufacturing Processes, 14(2),
pp.107-118.
[62] Zhou, C., Chen, Y. and Waltz, R.A., 2009, January. Optimized mask image projection for
solid freeform fabrication. In ASME 2009 International Design Engineering Technical
Conferences and Computers and Information in Engineering Conference (pp. 543-557). American
Society of Mechanical Engineers.

175
[63] Chen, Y., Mao, H. and Li, X., University of Southern California (USC), 2016. Mask Video
Projection Based Stereolithography With Continuous Resin Flow. U.S. Patent Application
15/187,713.
[64] Liravi, F., Das, S. and Zhou, C., 2015. Separation force analysis and prediction based on
cohesive element model for constrained-surface Stereolithography processes. Computer-Aided
Design, 69, pp.134-142.
[65] Pan, Y. and Dagli, C., 2017. Dynamic resolution control in a laser projection-based
stereolithography system. Rapid Prototyping Journal, 23(1), pp.190-200.
[66] Choi, J.W., Ha, Y.M., Lee, S.H. and Choi, K.H., 2006. Design of microstereolithography
system based on dynamic image projection for fabrication of three-dimensional microstructures.
Journal of mechanical science and technology, 20(12), pp.2094-2104.
[67] Choi, J.W., 2006. Development of projection-based microstereolithography apparatus
adapted to large surface and microstructure fabrication for human body application (Doctoral
dissertation, Doctoral dissertation).
[68] Li, X., Yang, Y., Xie, B., Chu, M., Sun, H., Hao, S., Chen, Y. and Chen, Y., 2019. 3D
Printing of Flexible Liquid Sensor Based on Swelling Behavior of Hydrogel with Carbon
Nanotubes. Advanced Materials Technologies, 4(2), p.1800476.
[69] Emami, M.M., Barazandeh, F. and Yaghmaie, F., 2014. Scanning-projection based
stereolithography: Method and structure. Sensors and Actuators A: Physical, 218, pp.116-124.
[70] Sun, C., Fang, N., Wu, D.M. and Zhang, X., 2005. Projection micro-stereolithography
using digital micro-mirror dynamic mask. Sensors and Actuators A: Physical, 121(1), pp.113-120.
[71] Monneret, S., Loubere, V. and Corbel, S., 1999, March. Microstereolithography using a
dynamic mask generator and a noncoherent visible light source. In Design, Test, and

176
Microfabrication of MEMS and MOEMS (Vol. 3680, pp. 553-562). International Society for
Optics and Photonics.
[72] Devaux, F., Mosset, A., Lantz, E., Monneret, S. and Le Gall, H., 2001. Image upconversion
from the visible to the UV domain: application to dynamic UV microstereolithography. Applied
optics, 40(28), pp.4953-4957.
[73] Bertsch, A., Bernhard, P. and Renaud, P., 2001, October. Microstereolithography:
Concepts and applications. In ETFA 2001. 8th International Conference on Emerging
Technologies and Factory Automation. Proceedings (Cat. No. 01TH8597) (Vol. 2, pp. 289-298).
IEEE.
[74] Park, I.B., Ha, Y.M. and Lee, S.H., 2010. Cross-section segmentation for improving the
shape accuracy of microstructure array in projection microstereolithography. The International
Journal of Advanced Manufacturing Technology, 46(1-4), pp.151-161.
[75] Limaye, A.S. and Rosen, D.W., 2007. Process planning method for mask projection micro-
stereolithography. Rapid Prototyping Journal, 13(2), pp.76-84.
[76] Ha, Y.M., Choi, J.W. and Lee, S.H., 2008. Mass production of 3-D microstructures using
projection microstereolithography. Journal of mechanical science and technology, 22(3), p.514.
[77] Han, L.H., Mapili, G., Chen, S. and Roy, K., 2008. Projection microfabrication of three-
dimensional scaffolds for tissue engineering. Journal of Manufacturing Science and
Engineering, 130(2), p.021005.
[78] Xia, C., Lee, H. and Fang, N., 2010. Solvent-driven polymeric micro beam device. Journal
of Micromechanics and Microengineering, 20(8), p.085030.

177
[79] Song, X., Zhang, Z., Chen, Z. and Chen, Y., 2017. Porous structure fabrication using a
stereolithography-based sugar foaming method. Journal of Manufacturing Science and
Engineering, 139(3), p.031015.
[80] Chiu, S.H., Pong, S.H., Wu, D.C. and Lin, C.H., 2008. A study of photomask correction
method in area-forming rapid prototyping system. Rapid Prototyping Journal, 14(5), pp.285-292.
[81] Choi, J.W., Kim, H.C. and Wicker, R., 2011. Multi-material stereolithography. Journal of
Materials Processing Technology, 211(3), pp.318-328.
[82] Wegst, U.G., Bai, H., Saiz, E., Tomsia, A.P. and Ritchie, R.O., 2015. Bioinspired structural
materials. Nature materials, 14(1), p.23.
[83] Liu, K. and Jiang, L., 2011. Bio-inspired design of multiscale structures for function
integration. Nano Today, 6(2), pp.155-175.
[84] Bhushan, B., 2009. Biomimetics: lessons from nature–an overview.
[85] Ke, P., Jiao, X.N., Ge, X.H., Xiao, W.M. and Yu, B., 2014. From macro to micro: structural
biomimetic materials by electrospinning. RSC advances, 4(75), pp.39704-39724.
[86] Liu, K. and Jiang, L., 2011. Multifunctional integration: from biological to bio-inspired
materials. ACS nano, 5(9), pp.6786-6790.
[87] Fratzl, P. and Weinkamer, R., 2007. Nature’s hierarchical materials. Progress in materials
Science, 52(8), pp.1263-1334.
[88] Kizilkan, E., Strueben, J., Staubitz, A. and Gorb, S.N., 2017. Bioinspired photocontrollable
microstructured transport device. Sci. Robot, 2(2).
[89] Sitti, M. and Fearing, R.S., 2003. Synthetic gecko foot-hair micro/nano-structures as dry
adhesives. Journal of adhesion science and technology, 17(8), pp.1055-1073.

178
[90] Zheng, Y., Gao, X. and Jiang, L., 2007. Directional adhesion of superhydrophobic butterfly
wings. Soft Matter, 3(2), pp.178-182.
[91] Shao, Z. and Vollrath, F., 2002. Materials: Surprising strength of silkworm
silk. Nature, 418(6899), p.741.
[92] Zheng, Y., Bai, H., Huang, Z., Tian, X., Nie, F.Q., Zhao, Y., Zhai, J. and Jiang, L., 2010.
Directional water collection on wetted spider silk. Nature, 463(7281), p.640.
[93] Saranathan, V., Osuji, C.O., Mochrie, S.G., Noh, H., Narayanan, S., Sandy, A., Dufresne,
E.R. and Prum, R.O., 2010. Structure, function, and self-assembly of single network gyroid (I4132)
photonic crystals in butterfly wing scales. Proceedings of the National Academy of
Sciences, 107(26), pp.11676-11681.
[94] Whyburn, G.P., Li, Y. and Huang, Y., 2008. Protein and protein assembly based material
structures. Journal of Materials Chemistry, 18(32), pp.3755-3762.
[95] Barthlott, W., Schimmel, T., Wiersch, S., Koch, K., Brede, M., Barczewski, M., Walheim,
S., Weis, A., Kaltenmaier, A., Leder, A. and Bohn, H.F., 2010. The Salvinia paradox:
superhydrophobic surfaces with hydrophilic pins for air retention under water. Advanced
Materials, 22(21), pp.2325-2328.
[96] Sethi, S.K. and Manik, G., 2018. Recent progress in super hydrophobic/hydrophilic self-
cleaning surfaces for various industrial applications: a review. Polymer-Plastics Technology and
Engineering, 57(18), pp.1932-1952.
[97] Eliason, C.M. and Shawkey, M.D., 2011. Decreased hydrophobicity of iridescent feathers:
a potential cost of shiny plumage. Journal of Experimental Biology, 214(13), pp.2157-2163.

179
[98] Darmanin, T. and Guittard, F., 2014. Recent advances in the potential applications of
bioinspired superhydrophobic materials. Journal of Materials Chemistry A, 2(39), pp.16319-
16359.
[99] Song, C. and Zheng, Y., 2014. Wetting-controlled strategies: From theories to bio-
inspiration. Journal of colloid and interface science, 427, pp.2-14.
[100] Barber, A.H., Lu, D. and Pugno, N.M., 2015. Extreme strength observed in limpet
teeth. Journal of The Royal Society Interface, 12(105), p.20141326.
[101] Nudelman, F., Gotliv, B.A., Addadi, L. and Weiner, S., 2006. Mollusk shell formation:
mapping the distribution of organic matrix components underlying a single aragonitic tablet in
nacre. Journal of structural biology, 153(2), pp.176-187.
[102] Gu, G.X., Su, I., Sharma, S., Voros, J.L., Qin, Z. and Buehler, M.J., 2016. Three-
dimensional-printing of bio-inspired composites. Journal of biomechanical engineering, 138(2),
p.021006.
[103] Drotlef, D.M., Amjadi, M., Yunusa, M. and Sitti, M., 2017. Bioinspired composite
microfibers for skin adhesion and signal amplification of wearable sensors. Advanced
Materials, 29(28), p.1701353.
[104] Baltrusaitis, R.M., Cady, R., Cassiday, G.L., Cooperv, R., Elbert, J.W., Gerhardy, P.R., Ko,
S., Loh, E.C., Salamon, M., Steck, D. and Sokolsky, P., 1985. The Utah Fly's eye detector. Nuclear
Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors
and Associated Equipment, 240(2), pp.410-428.
[105] Porter, M.M., Adriaens, D., Hatton, R.L., Meyers, M.A. and McKittrick, J., 2015. Why the
seahorse tail is square. Science, 349(6243), p.aaa6683.

180
[106] Gao, W., Zhang, Y., Ramanujan, D., Ramani, K., Chen, Y., Williams, C.B., Wang, C.C.,
Shin, Y.C., Zhang, S. and Zavattieri, P.D., 2015. The status, challenges, and future of additive
manufacturing in engineering. Computer-Aided Design, 69, pp.65-89.
[107] Guo, N. and Leu, M.C., 2013. Additive manufacturing: technology, applications and
research needs. Frontiers of Mechanical Engineering, 8(3), pp.215-243.
[108] Espalin, D., Alberto Ramirez, J., Medina, F. and Wicker, R., 2014. Multi-material, multi-
technology FDM: exploring build process variations. Rapid Prototyping Journal, 20(3), pp.236-
244.
[109] Thiele, S., Arzenbacher, K., Gissibl, T., Giessen, H. and Herkommer, A.M., 2017. 3D-
printed eagle eye: Compound microlens system for foveated imaging. Science advances, 3(2),
p.e1602655.
[110] Huber, B., Engelhardt, S., Meyer, W., Krüger, H., Wenz, A., Schönhaar, V., Tovar, G.,
Kluger, P. and Borchers, K., 2016. Blood-vessel mimicking structures by stereolithographic
fabrication of small porous tubes using cytocompatible polyacrylate elastomers,
biofunctionalization and endothelialization. Journal of functional biomaterials, 7(2), p.11.
[111] Li, Y., Mao, H., Liu, H., Yao, Y., Wang, Y., Song, B., Chen, Y. and Wu, W., 2015.
Stereolithography with variable resolutions using optical filter with high-contrast gratings. Journal
of Vacuum Science & Technology B, Nanotechnology and Microelectronics: Materials,
Processing, Measurement, and Phenomena, 33(6), p.06F604.
[112] Li, X., Yang, Y., Chen, Y., 2017, Bio-inspired Microscale Texture Fabrication based on
Immersed Surface Accumulation Process. Proceedings of the World Congress on Micro and Nano
Manufacturing Conference WCMNM2017. PP:33-36.

181
[113] Libonati, F. and Buehler, M.J., 2017. Advanced structural materials by
bioinspiration. Advanced Engineering Materials, 19(5), p.1600787.
[114] Bhushan, B. and Her, E.K., 2010. Fabrication of superhydrophobic surfaces with high and
low adhesion inspired from rose petal. Langmuir, 26(11), pp.8207-8217.
[115] Chen, Y., Zhou, C. and Lao, J., 2011. A layerless additive manufacturing process based on
CNC accumulation. Rapid Prototyping Journal, 17(3), pp.218-227.
[116] Chen, Y. and Zhou, C., University of Southern California (USC), 2015. Computer
numerical control (CNC) additive manufacturing. U.S. Patent 9,221,216.
[117] Mao, H., Zhou, C. and Chen, Y., 2016. LISA: Linear immersed sweeping
accumulation. Journal of Manufacturing Processes, 24, pp.406-415.
[118] Zhao, X., Pan, Y., Zhou, C., Chen, Y. and Wang, C.C., 2013. An integrated CNC
accumulation system for automatic building-around-inserts. Journal of Manufacturing
Processes, 15(4), pp.432-443.
[119] Choi, J.W., Wicker, R.B., Cho, S.H., Ha, C.S. and Lee, S.H., 2009. Cure depth control for
complex 3D microstructure fabrication in dynamic mask projection microstereolithography. Rapid
Prototyping Journal, 15(1), pp.59-70.
[120] Gong, H., Beauchamp, M., Perry, S., Woolley, A.T. and Nordin, G.P., 2015. Optical
approach to resin formulation for 3D printed microfluidics. RSC advances, 5(129), 106621-
106632
[121] Lee, J.H., Prud'Homme, R.K. and Aksay, I.A., 2001. Cure depth in photopolymerization:
experiments and theory. Journal of Materials Research, 16(12), 3536-3544.

182
[122] Pedersen, K., Emblemsvag, J., Bailey, R., Allen, J.K. and Mistree, F., 2000, September.
Validating design methods and research: the validation square. In ASME Design Engineering
Technical Conferences (pp. 1-12).
[123] Baldacchini, T. ed., 2015. Three-dimensional microfabrication using two-photon
polymerization: fundamentals, technology, and applications. William Andrew.
[124] Studart, A.R., 2016. Additive manufacturing of biologically-inspired materials. Chemical
Society Reviews, 45(2), pp.359-376.
[125] Huang, S.H., Liu, P., Mokasdar, A. and Hou, L., 2013. Additive manufacturing and its
societal impact: a literature review. The International Journal of Advanced Manufacturing
Technology, 67(5-8), pp.1191-1203.
[126] Chua, C.K. and Leong, K.F., 2014. 3D Printing and Additive Manufacturing: Principles
and Applications (with Companion Media Pack) of Rapid Prototyping Fourth Edition. World
Scientific Publishing Company.
[127] Schubert, C., Van Langeveld, M.C. and Donoso, L.A., 2014. Innovations in 3D printing: a
3D overview from optics to organs. British Journal of Ophthalmology, 98(2), pp.159-161.
[128] Huang, J.Y. and Lai, Y.K., 2015. TiO2-Based Surfaces with Special Wettability–From
Nature to Biomimetic Application. In Wetting and Wettability. IntechOpen.
[129] Bhushan, B., 2011. Biomimetics inspired surfaces for drag reduction and
oleophobicity/philicity. Beilstein journal of nanotechnology, 2(1), pp.66-84.
[130] Choi, J.W., Yamashita, M., Sakakibara, J., Kaji, Y., Oshika, T. and Wicker, R.B., 2010.
Combined micro and macro additive manufacturing of a swirling flow coaxial phacoemulsifier
sleeve with internal micro-vanes. Biomedical microdevices, 12(5), pp.875-886.

183
[131] Kolesky, D.B., Truby, R.L., Gladman, A.S., Busbee, T.A., Homan, K.A. and Lewis, J.A.,
2014. 3D bioprinting of vascularized, heterogeneous cell ‐laden tissue constructs. Advanced
materials, 26(19), pp.3124-3130.
[132] Au, A.K., Huynh, W., Horowitz, L.F. and Folch, A., 2016. 3D ‐printed
microfluidics. Angewandte Chemie International Edition, 55(12), pp.3862-3881.
[133] Wu, W., DeConinck, A. and Lewis, J.A., 2011. Omnidirectional printing of 3D
microvascular networks. Advanced materials, 23(24), pp.H178-H183.
[134] Hinton, T.J., Jallerat, Q., Palchesko, R.N., Park, J.H., Grodzicki, M.S., Shue, H.J.,
Ramadan, M.H., Hudson, A.R. and Feinberg, A.W., 2015. Three-dimensional printing of complex
biological structures by freeform reversible embedding of suspended hydrogels. Science
advances, 1(9), p.e1500758.
[135] Bhattacharjee, T., Zehnder, S.M., Rowe, K.G., Jain, S., Nixon, R.M., Sawyer, W.G. and
Angelini, T.E., 2015. Writing in the granular gel medium. Science advances, 1(8), p.e1500655.
[136] Lee, V.K., Lanzi, A.M., Ngo, H., Yoo, S.S., Vincent, P.A. and Dai, G., 2014. Generation
of multi-scale vascular network system within 3D hydrogel using 3D bio-printing
technology. Cellular and molecular bioengineering, 7(3), pp.460-472.
[137] Lewis, J.A. and Gratson, G.M., 2004. Direct writing in three dimensions. Materials
today, 7(7-8), pp.32-39.
[138] Li, M.G., Tian, X.Y. and Chen, X.B., 2009. A brief review of dispensing-based rapid
prototyping techniques in tissue scaffold fabrication: role of modeling on scaffold properties
prediction. Biofabrication, 1(3), p.032001.
[139] Qin, Z., Compton, B.G., Lewis, J.A. and Buehler, M.J., 2015. Structural optimization of
3D-printed synthetic spider webs for high strength. Nature communications, 6, p.7038.

184
[140] Bai, L., Xie, Z., Wang, W., Yuan, C., Zhao, Y., Mu, Z., Zhong, Q. and Gu, Z., 2014. Bio-
inspired vapor-responsive colloidal photonic crystal patterns by inkjet printing. ACS nano, 8(11),
pp.11094-11100.
[141] Liu, W., Zhang, Y.S., Heinrich, M.A., De Ferrari, F., Jang, H.L., Bakht, S.M., Alvarez,
M.M., Yang, J., Li, Y.C., Trujillo ‐de Santiago, G. and Miri, A.K., 2017. Rapid continuous
multimaterial extrusion bioprinting. Advanced materials, 29(3), p.1604630.
[142] Emami, M.M., Barazandeh, F. and Yaghmaie, F., 2015. An analytical model for scanning-
projection based stereolithography. Journal of Materials Processing Technology, 219, pp.17-27.
[143] Ha, Y.M., Park, I.B., Kim, H.C. and Le, S.H., 2010. Three-dimensional microstructure
using partitioned cross-sections in projection microstereolithography. International Journal of
Precision Engineering and Manufacturing, 11(2), pp.335-340.
[144] Lee, M.P., Cooper, G.J., Hinkley, T., Gibson, G.M., Padgett, M.J. and Cronin, L., 2015.
Development of a 3D printer using scanning projection stereolithography. Scientific reports, 5,
p.9875.
[145] Calvert, P., 2001. Inkjet printing for materials and devices. Chemistry of materials, 13(10),
pp.3299-3305.
[146] Kim, J.Y., Pfeiffer, K., Voigt, A., Gruetzner, G. and Brugger, J., 2012. Directly fabricated
multi-scale microlens arrays on a hydrophobic flat surface by a simple ink-jet printing
technique. Journal of Materials Chemistry, 22(7), pp.3053-3058.
[147] Perez, K.B. and Williams, C.B., 2013, August. Combining additive manufacturing and
direct write for integrated electronics—a review. In 24th International Solid Freeform Fabrication
Symposium—An Additive Manufacturing Conference, SFF (pp. 962-979).

185
[148] Karunakaran, K.P., Suryakumar, S., Pushpa, V. and Akula, S., 2010. Low cost integration
of additive and subtractive processes for hybrid layered manufacturing. Robotics and Computer-
Integrated Manufacturing, 26(5), pp.490-499.
[149] Giannitelli, S.M., Mozetic, P., Trombetta, M. and Rainer, A., 2015. Combined additive
manufacturing approaches in tissue engineering. Acta biomaterialia, 24, pp.1-11.
[150] Gissibl, T., Thiele, S., Herkommer, A. and Giessen, H., 2016. Two-photon direct laser
writing of ultracompact multi-lens objectives. Nature Photonics, 10(8), p.554.
[151] Malinauskas, M., Lukoševičius, L., Mackevičiūtė, D., Balčiūnas, E., Rekštytė, S. and
Paipulas, D., 2014, May. Multiscale 3D manufacturing: combining thermal extrusion printing with
additive and subtractive direct laser writing. In Laser Sources and Applications II (Vol. 9135, p.
91350T). International Society for Optics and Photonics.
[152] Kim, G., Ahn, S., Yoon, H., Kim, Y. and Chun, W., 2009. A cryogenic direct-plotting
system for fabrication of 3D collagen scaffolds for tissue engineering. Journal of Materials
Chemistry, 19(46), pp.8817-8823.
[153] Puppi, D., Mota, C., Gazzarri, M., Dinucci, D., Gloria, A., Myrzabekova, M., Ambrosio,
L. and Chiellini, F., 2012. Additive manufacturing of wet-spun polymeric scaffolds for bone tissue
engineering. Biomedical microdevices, 14(6), pp.1115-1127.
[154] Farrugia, B.L., Brown, T.D., Upton, Z., Hutmacher, D.W., Dalton, P.D. and Dargaville,
T.R., 2013. Dermal fibroblast infiltration of poly (ε-caprolactone) scaffolds fabricated by melt
electrospinning in a direct writing mode. Biofabrication, 5(2), p.025001.
[155] Kim, M. and Kim, G., 2015. Physical and biological activities of newly designed, macro-
pore-structure-controlled 3D fibrous poly (ε-caprolactone)/hydroxyapatite composite
scaffolds. RSC Advances, 5(34), pp.26954-26964.

186
[156] Giannitelli, S.M., Mozetic, P., Trombetta, M. and Rainer, A., 2015. Combined additive
manufacturing approaches in tissue engineering. Acta biomaterialia, 24, pp.1-11.
[157] Kataria, A. and Rosen, D.W., 2001. Building around inserts: methods for fabricating
complex devices in stereolithography. Rapid Prototyping Journal, 7(5), pp.253-262.
[158] Maruo, S., Nakamura, O. and Kawata, S., 1997. Three-dimensional microfabrication with
two-photon-absorbed photopolymerization. Optics letters, 22(2), pp.132-134.
[159] Malinauskas, M., Purlys, V., Rutkauskas, M. and Gadonas, R., 2009, February. Two-
photon polymerization for fabrication of three-dimensional micro-and nanostructures over a large
area. In Micromachining and Microfabrication Process Technology XIV (Vol. 7204, p. 72040C).
International Society for Optics and Photonics.
[160] Bhushan, B., 2011. Biomimetics inspired surfaces for drag reduction and
oleophobicity/philicity. Beilstein journal of nanotechnology, 2(1), pp.66-84.
[161] Ke, P., Jiao, X.N., Ge, X.H., Xiao, W.M. and Yu, B., 2014. From macro to micro: structural
biomimetic materials by electrospinning. RSC advances, 4(75), pp.39704-39724.
[162] Li, X. and Chen, Y., 2017. Microscale feature fabrication using immersed surface
accumulation. Journal of Manufacturing Processes, 28, pp.531-540.
[163] Chen, Y. and Li, X., University of Southern California (USC), 2018. Surface projection
tool for multi-axis additive manufacturing. U.S. Patent Application 15/868,891.
[164] Lee, B., 2018. Introduction to±12 Degree Orthogonal Digital Micromirror Devices
(DMDs). Texas Instrument Incorporated, 3.
[165] Ikuta, K., Ogata, T., Tsubio, M. and Kojima, S., 1996, February. Development of mass
productive micro stereo lithography (Mass-IH Process). In Proceedings of Ninth International
Workshop on Micro Electromechanical Systems (pp. 301-306). IEEE.

187
[166] Dendukuri, D., Pregibon, D.C., Collins, J., Hatton, T.A. and Doyle, P.S., 2006.
Continuous-flow lithography for high-throughput microparticle synthesis. Nature materials, 5(5),
p.365.
[167] Richardson, S., 1972. Hele Shaw flows with a free boundary produced by the injection of
fluid into a narrow channel. Journal of Fluid Mechanics, 56(4), pp.609-618.
[168] Varadan, V.K., Jiang, X. and Varadan, V.V., 2001. Microstereolithography and other
fabrication techniques for 3D MEMS. John Wiley & Sons Inc.
[169] Parker, A.R. and Lawrence, C.R., 2001. Water capture by a desert
beetle. Nature, 414(6859), p.33.
[170] Dorrer, C. and Rü he, J., 2008. Mimicking the Stenocara Beetle Dewetting of Drops from
a Patterned Superhydrophobic Surface. Langmuir, 24(12), pp.6154-6158.
[171] Tian, X., Chen, Y., Zheng, Y., Bai, H. and Jiang, L., 2011. Controlling water capture of
bioinspired fibers with hump structures. Advanced Materials, 23(46), pp.5486-5491.
[172] Dong, H., Wang, N., Wang, L., Bai, H., Wu, J., Zheng, Y., Zhao, Y. and Jiang, L., 2012.
Bioinspired electrospun knotted microfibers for fog harvesting. ChemPhysChem, 13(5), pp.1153-
1156.
[173] Ju, J., Bai, H., Zheng, Y., Zhao, T., Fang, R. and Jiang, L., 2012. A multi-structural and
multi-functional integrated fog collection system in cactus. Nature communications, 3, p.1247.
[174] Ju, J., Xiao, K., Yao, X., Bai, H. and Jiang, L., 2013. Bioinspired conical copper wire with
gradient wettability for continuous and efficient fog collection. Advanced Materials, 25(41),
pp.5937-5942.
[175] Song, C. and Zheng, Y., 2014. Wetting-controlled strategies: From theories to bio-
inspiration. Journal of colloid and interface science, 427, pp.2-14.

188
[176] Cai, S.Q. and Bhunia, A., 2017. Dropwise Condensation on/in High Roughness
Structures. Journal of Heat Transfer, 139(4), p.041501.
[177] Ju, J., Xiao, K., Yao, X., Bai, H. and Jiang, L., 2013. Bioinspired conical copper wire with
gradient wettability for continuous and efficient fog collection. Advanced Materials, 25(41),
pp.5937-5942.
[178] Jung, G.Y., Li, Z., Wu, W., Chen, Y., Olynick, D.L., Wang, S.Y., Tong, W.M. and
Williams, R.S., 2005. Vapor-phase self-assembled monolayer for improved mold release in
nanoimprint lithography. Langmuir, 21(4), pp.1158-1161.
[179] Ju, J., Yao, X., Yang, S., Wang, L., Sun, R., He, Y. and Jiang, L., 2014. Cactus Stem
Inspired Cone ‐Arrayed Surfaces for Efficient Fog Collection. Advanced Functional
Materials, 24(44), pp.6933-6938.
[180] Xue, Z., Cao, Y., Liu, N., Feng, L. and Jiang, L., 2014. Special wettable materials for
oil/water separation. Journal of Materials Chemistry A, 2(8), pp.2445-2460.
[181] Zhang, J. and Seeger, S., 2011. Polyester materials with superwetting silicone
nanofilaments for oil/water separation and selective oil absorption. Advanced Functional
Materials, 21(24), pp.4699-4704.
[182] Xue, Z., Wang, S., Lin, L., Chen, L., Liu, M., Feng, L. and Jiang, L., 2011. A novel
superhydrophilic and underwater superoleophobic hydrogel ‐coated mesh for oil/water
separation. Advanced Materials, 23(37), pp.4270-4273.
[183] Kota, A.K., Kwon, G., Choi, W., Mabry, J.M. and Tuteja, A., 2012. Hygro-responsive
membranes for effective oil–water separation. Nature communications, 3, p.1025.

189
[184] Scott, T.C. and Wham, R.M., US Department of Energy, 1988. Surface area generation
and droplet size control in solvent extraction systems utilizing high intensity electric fields. U.S.
Patent 4,767,515.
[185] Zeiger, C., da Silva, I.C.R., Mail, M., Kavalenka, M.N., Barthlott, W. and Hölscher, H.,
2016. Microstructures of superhydrophobic plant leaves-inspiration for efficient oil spill cleanup
materials. Bioinspiration & biomimetics, 11(5), p.056003.
[186] Yang, Y., Li, X., Zheng, X., Chen, Z., Zhou, Q. and Chen, Y., 2018. 3D ‐Printed
Biomimetic Super ‐Hydrophobic Structure for Microdroplet Manipulation and Oil/Water
Separation. Advanced materials, 30(9), p.1704912.
[187] Wolansky, G. and Marmur, A., 1999. Apparent contact angles on rough surfaces: the
Wenzel equation revisited. Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 156(1-3), pp.381-388.
[188] Choi, W., Tuteja, A., Mabry, J.M., Cohen, R.E. and McKinley, G.H., 2009. A modified
Cassie–Baxter relationship to explain contact angle hysteresis and anisotropy on non-wetting
textured surfaces. Journal of colloid and interface science, 339(1), pp.208-216.
[189] Demello, A.J., 2006. Control and detection of chemical reactions in microfluidic
systems. Nature, 442(7101), pp.394-402.
[190] Su, B., Wang, S., Song, Y. and Jiang, L., 2012. A heatable and evaporation-free miniature
reactor upon superhydrophobic pedestals. Soft Matter, 8(3), pp.631-635.
[191] Jin, M., Feng, X., Feng, L., Sun, T., Zhai, J., Li, T. and Jiang, L., 2005. Superhydrophobic
aligned polystyrene nanotube films with high adhesive force. Advanced Materials, 17(16),
pp.1977-1981.

190
[192] Wong, W.S., Gutruf, P., Sriram, S., Bhaskaran, M., Wang, Z. and Tricoli, A., 2016. Strain
Engineering of Wave ‐like Nanofibers for Dynamically Switchable Adhesive/Repulsive
Surfaces. Advanced Functional Materials, 26(3), pp.399-407.
[193] Lai, Y., Pan, F., Xu, C., Fuchs, H. and Chi, L., 2013. In situ surface ‐modification ‐induced
superhydrophobic patterns with reversible wettability and adhesion. Advanced Materials, 25(12),
pp.1682-1686.
[194] Wen, L., Weaver, J.C. and Lauder, G.V., 2014. Biomimetic shark skin: design, fabrication
and hydrodynamic function. Journal of Experimental Biology, 217(10), pp.1656-1666.
[195] Porter, B.F., Mkhize, N. and Bhaskaran, H., 2017. Nanoparticle assembly enabled by EHD-
printed monolayers. Microsystems & Nanoengineering, 3, p.17054.
[196] Li, X., Baldacchin, T., Song, X. and Chen, Y., 2016. Multi-scale additive manufacturing:
an investigation on building objects with macro-, micro-and nanoscales features. In 11th
International Conference on Micro Manufacturing, Irvine, CA, Mar (pp. 29-31).
[197] Li, Z., Araoka, T., Wu, J., Liao, H.K., Li, M., Lazo, M., Zhou, B., Sui, Y., Wu, M.Z.,
Tamura, I. and Xia, Y., 2016. 3D culture supports long-term expansion of mouse and human
nephrogenic progenitors. Cell stem cell, 19(4), pp.516-529.
[198] Mao, H., Kwok, T.H., Chen, Y. and Wang, C.C., 2019. Adaptive slicing based on efficient
profile analysis. Computer-Aided Design, 107, pp.89-101.
[199] Li, Y., Mao, H., Hu, P., Hermes, M., Lim, H., Yoon, J., Luhar, M., Chen, Y. and Wu, W.,
2019. Bioinspired Functional Surfaces Enabled by Multiscale Stereolithography. Advanced
Materials Technologies, p.1800638.

191
[200] Leung, Y.S., Kwok, T.H., Mao, H. and Chen, Y., 2019. Digital Material Design Using
Tensor-Based Error Diffusion for Additive Manufacturing. Computer-Aided Design, 114, pp.224-
235.
[201] Leung, Y.S., Mao, H. and Chen, Y., 2018, August. Approximate Functionally Graded
Materials for Multi-Material Additive Manufacturing. In ASME 2018 International Design
Engineering Technical Conferences and Computers and Information in Engineering Conference
(pp. V01AT02A030-V01AT02A030). American Society of Mechanical Engineers.
[202] Chen, W., Fadeev, A.Y., Hsieh, M.C., Öner, D., Youngblood, J. and McCarthy, T.J., 1999.
Ultrahydrophobic and ultralyophobic surfaces: some comments and examples. Langmuir, 15(10),
pp.3395-3399.
[203] Haycock, J.W., 2011. 3D cell culture: a review of current approaches and techniques. In 3D
Cell Culture (pp. 1-15). Humana Press.

Abstract (if available)

Abstract Nature provides us a vast of promising functional material systems consisted of hierarchical structures, where complex geometry and dimensional discrepancy across sections are present. Such hierarchical structures possess functionalities that inspired various biomimetic applications but are overall difficult to be built with traditional manufacturing process primarily due to the limitation of manufacturing capability set by conventional machine tools. Nowadays, additive manufacturing (AM) methods are commonly deployed to tackle with three-dimensional objects with complex structure by gradually accumulating material layer by layer. However, the hierarchical structure measured from macroscale to nanoscale still raises a challenge to the current commercial AM process, whose manufacturing capability is intrinsically specified within a certain scope. It is desirable to develop a new AM process with the capability to fabricate biomimetic functional material, and scalability to accommodate hierarchical structures at multi-scale. The major thrust of this research, therefore, is to investigate and develop novel printing approaches and strategies for manipulating and mimicking the intrinsically multi-scale, multi-material, and multifunctional structures in nature. ❧ In this research work, a novel microscale AM processes called immersed surface accumulation (ISA) is first developed to fabricate functional cactus inspired multi-scale structure. In this process, a special optical system is developed with a surface-based light guide tool to deliver light beam, and the calibration of light intensity is further conducted to obtain 2 dimensional (2D) patterned light beam with uniform light distribution. In order to fabricate cactus inspired microstructure, a light guide tool is immersed inside a tank filled with liquid based composite material, and material is gradually accumulated on the flat surface with the light exposure. Moreover, the system design and nonlinear exposure settings were proposed and implemented to fabricate cactus inspired microstructure arrays. Water collection of artificial three dimensional (3D) printed microstructure array is studied and high-efficiency water collection device is further developed and tested. ❧ To investigate the high-efficient fabrication of multi-scale bio-mimic superhydrophobic and superoleophilic structures, continuous immersed surface accumulation printing (CISA) approach was developed. In order to achieve continuous accumulation of material in printing direction, the self-filling and curing performance of the material is investigated and analyzed. In addition, the material refilling of CISA process is discussed to get the good surface quality of bioinspired microstructures, and the mechanical performance of the printed part with different speed is investigated. The composite material composed of photocurable material and multi-walled carbon nanotubes (MWCNTs) is developed, and special bio-inspired superhydrophobic and superoleophilic microstructures are designed and fabricated by the proposed ISA process. Droplet manipulation and Oil/water separation/removing are demonstrated respectively to show the superhydrophobicity and superoleophilicity of 3D printed bio-inspired microstructures. ❧ Based on the above macroscale and microscale AM processes, the integrated photopolymerization based hybrid process is developed to investigate the fabrication of multi-scale biomimetic texture. The macro- and mesoscale mask image projection based stereolithography is introduced and the corresponding process parameters are studied and optimized. Then the integration of the microscale ISA process is discussed to achieve high-resolution microscale fabrication results. A multi-scale 3D printing method is presented by integrating the ISA and mask image projection based stereolithography (MIP-SL) processes. To achieve high-resolution fabrication, process planning, material curing performance, and the printing parameters are discussed. To enable cell culture study, multi-scale cell culture environment with multiple bio-inspired textures are fabricated with bio-compatible material based on developed multi-scale AM process. The experimental results verify the efficiency and accuracy of the developed multi-scale 3D printing method on the fabrication of multi-scale functional structures. ❧ To fabricate bio-inspired hierarchical structure with macro-, micro- and nano- scales features, a 3D hierarchical printing approach with an investigation on integrating multiple printing processes is proposed. Firstly, special hierarchical design and process planning towards integrating multiple printing processes are demonstrated. Then sub-processes including macroscale mask image projection based stereolithography, microscale ISA, and nanoscale two photon polymerization are developed, and the key process parameters for each sub-process, e.g., the curing characteristics of the material, are optimized to improve the fabrication quality of each process. To address the challenge of the integration of multiple processes, a vision assisted approach is introduced, and the alignment of each subprocess for the transition is introduced and developed. Throughout this research work, a systematic study for multi-scale AM development including the material development, biomimetic structure design, the process development in multiple manufacturing scales is presented in order to achieve certain functionalities inspired by nature.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Selective Separation Shaping (SSS): large scale cementitious fabrication potentials

PDF

Energy control and material deposition methods for fast fabrication with high surface quality in additive manufacturing using photo-polymerization

PDF

Hybrid vat photopolymerization: methods and systems

PDF

Hybrid vat photopolymerization processes for viscous photocurable and non-photocurable materials

PDF

Metallic part fabrication wiht selective inhibition sintering (SIS) based on microscopic mechanical inhibition

PDF

Slurry based stereolithography: a solid freeform fabrication method of ceramics and composites

PDF

Scalable polymerization additive manufacturing: principle and optimization

PDF

Fabrication of ultrasound transducer and 3D-prinitng ultrasonic device

PDF

Fabrication-aware machine learning for accuracy control in additive manufacturing

PDF

3D printing of polymeric parts using Selective Separation Shaping (SSS)

PDF

Deformable geometry design with controlled mechanical property based on 3D printing

PDF

Additive manufacturing of piezoelectric and composite for biomedical application

PDF

Analyses of strength of layered structures fabricated by Contour Crafting

$Metasurfaces in 3D applications: multiscale stereolithography and inverse design of diffractive optical elements for structured light$

PDF

Metasurfaces in 3D applications: multiscale stereolithography and inverse design of diffractive optical elements for structured light

PDF

Enhancing the surface quality and dimensional accuracy of SIS-metal parts with application to high temperature alloys

PDF

Selective separation shaping: an additive manufacturing method for metals and ceramics

PDF

Some scale-up methodologies for advanced manufacturing

PDF

Mechanics and additive manufacturing of bio-inspired polymers

PDF

Machine learning-driven deformation prediction and compensation for additive manufacturing

PDF

Biosensing and biomimetic electronics

Asset Metadata

Creator Li, Xiangjia (author)

Core Title Multi-scale biomimetic structure fabrication based on immersed surface accumulation

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Industrial and Systems Engineering

Publication Date 07/28/2019

Defense Date 06/05/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag additive manufacturing,biomimetic fabrication,immersed surface accumulation,multi-scale,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chen, Yong (committee chair), Lu, Chih-Yang (committee member), Zhou, Qifa (committee member)

Creator Email cindylinuaa@gmail.com,xiangjil@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-196728

Unique identifier UC11663551

Identifier etd-LiXiangjia-7640.pdf (filename),usctheses-c89-196728 (legacy record id)

Legacy Identifier etd-LiXiangjia-7640.pdf

Dmrecord 196728

Document Type Dissertation

Format application/pdf (imt)

Rights Li, Xiangjia

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA