University of Southern California Dissertations and Theses

High-resolution data acquisition with neural and dermal interfaces

(USC Thesis Other)

High-resolution data acquisition with neural and dermal interfaces

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content High-Resolution Data Acquisition with Neural and Dermal Interfaces
by
James Jung Yoo
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
(BIOMEDICAL ENGINEERING)
May 2024
Copyright 2024 James Jung Yoo

ii
To Stella

iii
ACKNOWLEDGEMENTS
Ellis, thank you for taking a chance on this aged industry runaway. I certainly thought going
back to graduate school after so long away would have made my acceptance a slim possibility,
but here I am. I deeply admire your patience and calmness when it comes to navigating research
projects and all their associated setbacks, deadlines, and pressure. You have been incredibly
gracious to me, helping me navigate parenthood amidst grad school.
While the lab has greatly shifted and moved (literally) during my time here, every member
has been integral to my development as a researcher. I have so many people to thank, but here
goes (in no particular order). Trevor, I think we were kindred spirits from the first time we met;
thanks for the laughs, the advice, and the heartfelt conversations. Eugene, your enthusiasm for
research and commitment to reaching out to younger students has been an inspiration to me.
Angelica, the lab has always felt like something between family and work, largely because of
your efforts to bring us together; thanks for the guidance and the fun. Law, I owe so much of my
initial fab knowledge to you; thanks for your wisdom. Alex, your persistence through grad
school while battling your own issues made me feel like I wasn’t alone; thanks for your honesty.
Chris, you’re good at asking the right questions, and there have been many instances where I’d
find new insights after talking to you. I often felt isolated as a parent in grad school, so when
Silas came along, I was doubly happy. Thanks for lending an ear. Ahuva, I admire your work in
lab, but the lesson I value most from you is resting. You showed me that setting aside time to
disconnect from work and spending it with family is possible for the PhD life. Jessica, you have
always been dependable, helping me out when I really needed it; thanks for your support. Kee,
you have been a fount of knowledge and an excellent sounding board for device troubleshooting.

iv
Back when I first applied to MEMS, I felt very unsure of myself, but your encouragement during
that time was something I didn’t know I needed. Sue, your work ethic and organization are
forces to be reckoned with; I wish I could be more like you. Brianna, talking with you has always
been refreshing and fun; thank you for being thoughtful of everyone in lab. Nikolas, I am always
impressed by your attention to detail whenever you present your work, and I expect great things
from you. Emmanuel, I’m glad I got to work alongside you, and even though the coding aspect
wasn’t in your wheelhouse, I think you did a great job. Thanks for your persistence and your 3D
printer. Ruitong, the lab has definitely gotten more lively since you joined as a PhD student. I
know organizing hangouts can be time-consuming, but I appreciate all your efforts. I admire
your dedication to both the work and the people in lab. Alberto, you exude kindness, and talking
to you has always come easy. Thanks for always checking in.
Much of my work would have stopped at the bench were it not for the efforts of students
from other labs. Aria Samiei, my ASIC partner, whenever we met, you were so full of warmth,
and your enthusiasm was infectious. You were there when we tested PUB successfully on a chip
for the first time, and I’m glad we got to share that moment. Huijing Xu, our lab owes so much
to the work you’ve done and continue to do; thank you for the advice on device and experiment
design. Wenxuan Jiang, thank you for being patient with me and for being so generous with your
time. Sahar, you are a relentless knowledge seeker, and that makes you such a great researcher to
work with. Leah is lucky to have you as her mother.
To my parents: thank you for valuing education. Despite our fraught relationships and
miscommunication, I know you are proud of me. To my sisters: 누나, I did this in large part
because you’ve encouraged my “nerdiness” since I was a wee child; thanks for making me

v
unashamed to pursue knowledge. 지영, you’re my role model when it comes to engaging the
culture, and I don’t know anyone who has as much passion for their research as you do.
Anastasia, Caspian, Maxwell: you bring me so much joy. If I had to pick, I would say
fatherhood is harder than doctoral studies, but I would also say that it brings a deeper satisfaction.
Seeing my devices finally work after years of toil was a great feeling. But watching you walk for
the first time, eat with your hands for the first time, or even use the potty by yourself—these
moments far outshine them all.
Stella, my guiding star, you have been the most pivotal person not just during graduate
school but all my life. Whenever I was wracked with uncertainty or doubt or shame, you were
there to hold my heart together. You have always loved me even when my own mind and body
did not. I literally would not have survived if not for you. I love you more than words can bear.
Finally, I thank God for this period of life—trials and all—and for the people who have
helped me along the way.

vi
TABLE OF CONTENTS
Acknowledgements........................................................................................................................iii
List of Tables ............................................................................................................................... viii
List of Figures................................................................................................................................ ix
Abstract......................................................................................................................................... xv
Chapter 1: Biological Applications for MEMS........................................................................... 1
1.1 Microelectromechanical Systems.................................................................... 1
1.2 MEMS in the Brain ......................................................................................... 2
1.3 MEMS in the Skin........................................................................................... 7
1.4 Objectives...................................................................................................... 10
References................................................................................................................ 11
Chapter 2: Interconnect Bonding Methods for Parylene Devices............................................. 21
2.1 Background ................................................................................................... 21
2.2 Challenges to ASIC Integration .................................................................... 23
2.3 Existing Interconnect Methods...................................................................... 25
2.4 Methods Investigated .................................................................................... 31
2.5 Device Alignment.......................................................................................... 38
2.6 Test Device Design and Fabrication ............................................................. 39
2.7 Interconnect Testing Methods....................................................................... 44
2.8 Testing Results.............................................................................................. 46
2.9 Discussion & Conclusion .............................................................................. 51
References................................................................................................................ 56
Chapter 3: ASIC Integration of Parylene Devices..................................................................... 62
3.1 Chip Integration with Flexible Polymers ...................................................... 62
3.2 Intan RHD2164 Layout................................................................................. 63
3.3 Adapting to Commercial Chips..................................................................... 65
3.4 Working with the Intan RHD2164 BGA....................................................... 68
3.5 Bonding to a Custom ASIC........................................................................... 72
3.6 Bonding to an Active BGA ........................................................................... 79
3.7 Obstacles Encountered with Active ASIC Bonding...................................... 86
3.8 Bonding to an Active Bare Die ..................................................................... 87
3.9 In Vitro Comparison of ASIC-Integrated and Passive pMEA...................... 95
3.10 Conclusion..................................................................................................... 99
References.............................................................................................................. 100
Chapter 4: Microneedles for Continuous Health Monitoring.................................................. 102
4.1 The Market for Continuous Health Monitoring .......................................... 102
4.2 Biological Sensing Methods........................................................................ 103

vii
4.3 Electrochemical Aptamer-Based Sensing ................................................... 105
4.4 Challenges in Addressing Interstitial Fluid ................................................. 109
4.5 Microneedle Design .................................................................................... 111
4.6 Microneedle Fabrication.............................................................................. 113
4.7 Microneedle Performance ........................................................................... 114
4.8 First Potentiostat Prototype Design............................................................. 117
4.9 Second Potentiostat Prototype Design ........................................................ 121
4.10 Potentiostat Prototype PCB Verification..................................................... 124
4.11 Conclusion................................................................................................... 128
References.............................................................................................................. 130
Chapter 5: Conclusion ............................................................................................................. 140
Bibliography ............................................................................................................................... 142
Appendices.................................................................................................................................. 167
Appendix A: Bonding Ribbon Cable Fabrication ............................................... 167
Appendix B: Polymer MEA Fabrication............................................................. 168
Appendix C: Other PUB Bonding Applications ................................................. 170
Appendix D: Silicon Microneedle Fabrication.................................................... 176
Appendix E: EAB Sensor Functionalization Process ......................................... 177
Appendix F: Gold Wire Testing in Ferrocyanide ............................................... 178
Appendix G: EAB Sensor Testing in Vancomycin ............................................. 179

viii
LIST OF TABLES
Table 2-1: Comparison of existing bonding methods................................................................... 30
Table 2-2. Comparison of Interconnect Strategies for Bonding Flexible Polymer to Rigid
Structures. Reprinted from [53]. ....................................................................................... 54
Table 2-3. Comparison of Interconnect Strategies for Bonding Flexible Polymer to Rigid
Structures. Reprinted from [53]. ....................................................................................... 55
Table 4-1: Etch rates for oxide using 48% HF ........................................................................... 113
Table A-1: DRIE deposition and etch parameters for each loop ................................................ 167
Table A-2: DRIE deposition and etch parameters for each loop ................................................ 169
Table A-3: Titration amounts for target concentration ............................................................... 179

ix
LIST OF FIGURES
Figure 1-1: The first silicon MEMS microprobe, containing 2 recording sites. Reprinted,
with permission, from [5]. .................................................................................................. 2
Figure 1-2: Common medical device materials are harder than human tissue. Adapted
with permission from Springer Nature [64]........................................................................ 4
Figure 1-3: (left) A single unit of pary-xylxylene and (right) a single unit of Parylene C,
wherein a hydrogen atom has been replaced with chlorine. ............................................... 5
Figure 1-4: Trends of recording sites per probe increase year over year........................................ 6
Figure 1-5: a) One variation of a drug delivery device in Gerstel and Place’s patent. b)
the first reported microneedles used in human skin. © 1998 IEEE. Reprinted,
with permission, from [78] ................................................................................................. 8
Figure 1-6: Various kinds of microneedles targeting the dermis. Reprinted from [91],
Copyright 2012, with permission from Elsevier................................................................. 8
Figure 2-1: a) The inside of a microprocessor showing wire bonds along the perimeter of
the chip, b) a close-up highlighting wire bonds and bond pads, and c) a scanning
electron micrograph (SEM) showing a typical wire wedge bond.
Motorola68040die.jpg by Gregg M. Erickson is licensed under CC BY 3.0. .................. 21
Figure 2-2: For this neural implant, the probe array connects to a printed circuit board by
way of a zero-insertion force (ZIF) connector, which is the interconnect........................ 22
Figure 2-3: The number of recording sites on a neural probe is rapidly increasing, and
many modern devices utilize chip integration to handle the increased packaging
burden. .............................................................................................................................. 23
Figure 2-4: Comparing critical temperatures of MEMS materials to temperature ranges
of standard interconnect techniques shows that many existing methods are
unsuitable for polymers..................................................................................................... 24
Figure 2-5: In ACF bonding, the film is placed between two substrates with raised
contact surfaces; once pressure and heat are applied, MPSes are compressed by
the contacts while the adhesive flows into the channels between the contacts. ............... 27
Figure 2-6: Technical drawings of two popular spring connectors: a) Bal Seal’s Sygnus
(Courtesy of Bal Seal Engineering); b) Molex’s ZIF ....................................................... 29
Figure 2-7: A summary of the four interconnect methods investigated. a) conductive
epoxy connected the metal on the device to the metal on the chip by filling the
well, b) wedge bonding both electrically and mechanically connected the device
metal to the chip metal, c) ACF was placed over the chip and then compressed to
form an electrical and mechanical connection, d) PUB bonds were gold wire
bumps on the chip that were subjected to ultrasonic energy. Reprinted from [53]. ......... 32
Figure 2-8: Ball bonding with gold wire on top of platinum contact pads on Parylene
lifts the metal off the polymer where bonds are attempted. Scale bar is 500 µm............. 33
Figure 2-9: Much like Microflex bonding, ball bonding can also rivet Parylene structures
down to a substrate, but the finest pitch capable on this machine with 25 µm wire
was greater than 100 µm. Scale bar is 500 µm. ................................................................ 33
Figure 2-10: Pull tests indicate that curing at a lower temperature and longer duration
than the manufacturer's recommendation produces comparable bond strength. .............. 34
Figure 2-11: Optical micrograph of waffle tool face. Scale bar is 250 µm. ................................. 35

x
Figure 2-12: The major steps of PUB bonding include 1) placing a "bump" of 25 µm
gold wire, 2) aligning the polymer device, 3) switching to a waffle tool, and 4)
applying ultrasonic energy to fuse the thin-film metal to the gold bump......................... 36
Figure 2-13: Cryogenic cross-sectioning was not sufficient to produce a visible edge on
the Parylene device, and the width of the razor removed some of the bond pad
underneath......................................................................................................................... 37
Figure 2-14: FIB can form a clear cross-section in the area of interest........................................ 37
Figure 2-15: FIB-SEM imaging of the PUB bond reveals the intimate contact between
the gold bump and the thin-film metal of the Parylene device. Scale bar is 1 µm.
Reprinted from [53]. ......................................................................................................... 38
Figure 2-16: The alignment jig holds the cable holder and chip holder separately until
they are brought together during alignment, after which the two holders are
secured together. Reprinted from [53]. ............................................................................. 39
Figure 2-17: The surrogate ribbon cable design contained eight alignments holes along
the perimeter, 28 contact pads, and 28 bond pads in the center. Additionally, a)
design A bond pads had varying lengths and no insulating layer while b) design
B retained the insulated layer and had through-holes in the bond pads. Scale bars
are 2 mm. Reprinted from [53]. ........................................................................................ 40
Figure 2-18: The surrogate chip design contained 28 "bond pads" that were connected
together to a common contact pad. Scale bar is 2 mm. Reprinted from [53]. .................. 41
Figure 2-19: The pull test structure featured an anchor hole for the test setup and 12 sets
of contact pads and bond pads; these bond pads contain the differentiating
features of the surrogate cable designs in addition to an experimental set of pads.
Pitch was 100 µm. Scale bar is 2 mm. Reprinted from [53]............................................. 42
Figure 2-20: Parylene devices were fabricated as follows: a) deposition of Parylene onto
silicon carrier wafer; b) photoresist lithography, e-beam metal deposition, and
metal liftoff; c) deposition of Parylene insulating layer; d) deep reactive ion etch
(DRIE) to form through-holes; and e) DRIE to remove insulation in the bonding
area for design A and to form through-hole ledges for design B. Reprinted from
[53].................................................................................................................................... 43
Figure 2-21: A completed Parylene-gold-Parylene surrogate cable and glass-gold
surrogate chip sit next to a U.S. dime. .............................................................................. 44
Figure 2-22: Bonded devices using the four interconnect methods. Scale bar is 0.5 mm.
Reprinted from [53]. ......................................................................................................... 45
Figure 2-23: The environmental chamber heated and cooled between 0 and 60 °C for the
duration of the thermal cycling test. The stepped ramping was necessary to
approximate the ramp rate recommended by the JEDEC standard. ................................. 46
Figure 2-24: Yields for all four interconnect methods are comparable at 100 µm pitch,
and at coarser pitch PUB bonding achieves 100%. Reprinted from [53]. ........................ 47
Figure 2-25: The average resistances of each interconnect method are comparable except
at 100 µm pitch, where PUB is clearly lower than the others. Reprinted from [53].
........................................................................................................................................... 47
Figure 2-26: In the long-term storage test, the average resistance of the bonds remained
steady throughout all 1000 hours for all four methods. Reprinted from [53]................... 49
Figure 2-27: During long-term storage, only 1 of the 46 wire bonds failed................................. 49

xi
Figure 2-28: Throughout the 36 cycles of the thermal cycling test, the average resistance
of all bonds remained stable, and no failures occurred. Reprinted from [53]. ................. 50
Figure 2-29: Pull testing obviates the mechanical strength of adhesive bonds, but
ultrasonic bonds can be reinforced with underfill to produce a similar yield
strength. Reprinted from [53]. .......................................................................................... 51
Figure 2-30: Representative failure modes show that both epoxy (a) and ACF (d) are
stronger than the Parylene itself. Likewise, the ultrasonic bonds themselves are
stronger than the Parylene, which tore at the bond pads for both wedge (b-c) and
PUB bonds (e). Reprinted from [53]................................................................................. 51
Figure 3-1: Simplified diagram for the Intan RHD2164 with analog inputs and digital
communication lines highlighted. Reprinted with permission from [10]. ........................ 64
Figure 3-2: Bond pad layout of Intan RHD2164 bare die. The top row of pads (red) are
the 64 analog inputs with 1 reference electrode on each side, spaced at a pitch of
101.6 µm. On the bottom are power, communication, and auxiliary lines.
Reprinted with permission from [10]................................................................................ 65
Figure 3-3: The underside of the Intan RHD2164 BGA chip features low-temperature
solder bumps. Scale bar is 1 mm. ..................................................................................... 67
Figure 3-4: After our desoldering process, most of the solder is removed; any remaining
solder is thin enough that the bonder can pierce the solder. Scale bar is 1 mm. .............. 67
Figure 3-5: The BGA probe array features contact pads for debugging on top, etched
bonding area in the center, matching bond pads in the center, and contact pads on
the bottom for ZIF connection. Scale bar is 20 mm. ........................................................ 69
Figure 3-6: Prior to curing, strips of ACF are placed over the bond pads of the BGA
probe array. Scale bar is 1 mm.......................................................................................... 70
Figure 3-7: After curing, the ACF does not appear to connect the chip to the device.
Even the adhesive does not spread properly. Scale bar is 1 mm. ..................................... 70
Figure 3-8: For the PUB process, a) gold wire bonds can be made directly to the metal
underneath the solder and b-c) the BGA probe array can be bonded to those
bumps. Scale bars for a-b) are 1 mm. Scale bar for c) is 0.5 mm. Reprinted from
[13].................................................................................................................................... 71
Figure 3-9: Diagram of the custom neural stimulation chip......................................................... 72
Figure 3-10: The custom ASIC cable also features an etched bonding area in the center
and contact pads on either side for ZIF connection. Reprinted from [13]........................ 73
Figure 3-11: Gold wire bonds were successfully placed directly on top of the bond pads
of the custom ASIC; trailing segments of wire can be removed with tweezers or
prevented entirely with wire bonder settings. Scale bar is 250 µm. ................................. 74
Figure 3-12: Gold wire bonds were successfully pre-treated with the waffle tool. Scale
bar is 250 µm. ................................................................................................................... 74
Figure 3-13: The custom ASIC cable was aligned to the ASIC, PUB bonded, and
underfilled successfully. Scale bar is 500 µm. Reprinted from [13]. ............................... 75
Figure 3-14: Sixteen wires on either side of the bonded ASIC were necessary for
connection to power and programming; they were soldered directly to the flat
flexible cable. Scale bar is 25 mm. ................................................................................... 75
Figure 3-15: The ASIC produced the programmed square wave successfully; likewise,
the microelectrode detected the wave through the PBS, giving us the transient
response shown here. Reprinted from [13]. ...................................................................... 77

xii
Figure 3-16: A custom acrylic surgery jig built using a CO2 laser cutter holds two
microelectrodes at a prescribed distance (250 µm) and a reference wire. Scale bar
is 1 cm............................................................................................................................... 78
Figure 3-17: Recordings of spontaneous electrical activity in the hippocampus ......................... 78
Figure 3-18: Recordings of the electrical activity in the rat brain include stimulation
artifacts every 10 seconds; upon closer inspection, these waveforms cannot be
distinguished from neural activity. ................................................................................... 79
Figure 3-19: The Intan BGA cable features two sets of contact pads for ZIF connection
and etched bonding area. The four holes in the center allow for easier
underfilling. Scale bar is 2 mm......................................................................................... 80
Figure 3-20: The Intan BGA cable was successfully PUB bonded to the BGA chip.
Scale bar is 1 mm.............................................................................................................. 81
Figure 3-21: SMT components were attached to the cable using conductive epoxy.................... 81
Figure 3-22: The bonded BGA cable was placed into a custom PCB. Scale bar is 10 mm. ........ 82
Figure 3-23: A function generator outputting a sine wave connected to one channel (A063) while others reported some crosstalk........................................................................ 83
Figure 3-24: The bonded BGA cable recorded the triangle wave output by the function
generator successfully. Crosstalk appeared between the recording channel and
the adjacent channel. Reprinted from [19] © 2024 IEEE. ................................................ 84
Figure 3-25: The reference electrode trace (highlighted in red) on the BGA sits next to
power and digital signal lines (black traces). Scale bar is 2 mm. ..................................... 85
Figure 3-26: The PCB for the BGA cable was two layers. Many of the top (red) and
bottom (blue) traces overlap, another avenue for crosstalk. ............................................. 85
Figure 3-27: First-generation bare die polymer MEA design with bypass traces for
independent interconnect verification on the sides and a ZIF connection on the
bottom. .............................................................................................................................. 86
Figure 3-28: Bubbling appeared on two of the positive voltage power lines. The left
trace has a disconnect........................................................................................................ 87
Figure 3-29: Polymer MEA design with black traces showing digital lines and red
showing analog lines. Scale bars are 2 mm and 500 µm (inset)....................................... 88
Figure 3-30: pMEA design, with red indicating base layer and blue indicating top layer
of Parylene ........................................................................................................................ 89
Figure 3-31: Closeup of bondpad region showing varying pitch to account for Parylene
shrinkage during fabrication. Top portions are spaced 101.6 µm apart while
bottom portions are 104.1 µm apart.................................................................................. 89
Figure 3-32: a) Image of custom PCB along with Omnetics and Molex connectors; b)
PCB with pMEA seated in ZIF with side PCB portions removed. Scale bars are
10 mm. .............................................................................................................................. 90
Figure 3-33: (left) Image of chip holder on top of wedge bonder workstage; (right)
image of the pMEA-holding PCB mated to the chip holder............................................. 91
Figure 3-34: Image of cover for PCB with a bar across the chip area to keep Parylene
flat against the chip. Scale bar is 2 mm. ........................................................................... 92
Figure 3-35: Image of ASIC-integrated pMEA lying next to a US penny. Reprinted from
[19] © 2024 IEEE. ............................................................................................................ 93
Figure 3-36: Micrograph of PUB bonds—66 along the top, 23 along the bottom—with
inset showing a closeup of the bonds. Here, the bonds appear circular because a

xiii
ball bonder was used to form the initial gold bumps. Scale bars are 1 mm and 50
µm (inset). Reprinted from [19] © 2024 IEEE................................................................. 93
Figure 3-37: Images of four stacked ASIC-integrated pMEAs with Mylar spacers..................... 94
Figure 3-38: Overlay of image focused on the top- and bottom-most of 4 stacked
pMEAs; lateral deviation appears on probe 1 on the left. Scale bar is 0.5 mm................ 95
Figure 3-39: Setup schematic for initial benchtop testing using a BGA-integrated cable
as a “headstage” ................................................................................................................ 96
Figure 3-40: Initial digital recordings from ASIC-integrated pMEA with probe tips in
PBS with a 1 Hz sine wave input; average of 11 periods. ................................................ 96
Figure 3-41: Recordings from ASIC-integrated pMEA and from a wire connected to a
BGA-integrated cable immersed in PBS with a 1 Hz sine wave input............................. 97
Figure 3-42: Setup schematic of in vitro test where digital signals were recorded via
Intan USB board and analog signals were recorded via Plexon products.
Reprinted from [19] © 2024 IEEE.................................................................................... 98
Figure 3-43: Recordings from the digital and analog sides of the ASIC-integrated pMEA,
indicating that performance is comparable. ...................................................................... 99
Figure 4-1: While many biosensing methods are generalizable, continuous, or specific,
only EAB sensing characterizes all three. FSCV image used with permission of
Royal Society of Chemistry, from [22]; permission conveyed through Copyright
Clearance Center, Inc. CGM image reproduced with permission from [23]. All
other images public domain............................................................................................ 104
Figure 4-2: Standard EAB sensors are gold wires coated in aptamers that undergo
conformational change in the presence of the target molecule....................................... 107
Figure 4-3: In SWV, current is measured at the peak and trough of applied potential E0,
then at E0 + ΔE, etc. Results are graphed as potential vs. net current (difference
between forward and reverse current)............................................................................. 108
Figure 4-4: A) Square wave voltammetry can detect the changes in electron transfer rate
due to the conformational change of the aptamer. B) At some frequencies, target
binding induces a “signal-on” response and C) at other frequencies, target
binding induces a “signal-off” response. Used with permission of the Royal
Society of Chemistry, from [58]; permission conveyed through Copyright
Clearance Center, Inc...................................................................................................... 109
Figure 4-5: A wearable platform for measuring ISF would be possible using EAB
sensors............................................................................................................................. 110
Figure 4-6: The design of the wafer has microneedle arrays in eight combinations of
length, spacing, and number as well as rectangular spacers. .......................................... 112
Figure 4-7: The microneedle 3D array concept consists of stacked 2D arrays and spacers
bound together with a pin. .............................................................................................. 112
Figure 4-8: Image of oxide-coated (left) and Au-coated (right) Si microneedle. Scale bar
is 0.5 mm......................................................................................................................... 114
Figure 4-9: SWV at 100 Hz reveals that the Si array performs comparably to Au for
detecting vancomycin. .................................................................................................... 115
Figure 4-10: SWV at 100 Hz shows that the Si shard can detect varying concentrations
of vancomycin................................................................................................................. 116
Figure 4-11: The signal gain response from the Si shard shows that 100 Hz is an
excellent signal-on frequency for detecting vancomycin. .............................................. 116

xiv
Figure 4-12: Block diagram for proposed potentiostat............................................................... 117
Figure 4-13: Two paths for developing a potentiostat were combining MCU and BLE
(left) or MCU and AFE (right)........................................................................................ 118
Figure 4-14: Block diagram of the AD5940............................................................................... 118
Figure 4-15: Block diagram of the ADuCM355. The left portion is identical to that of
the AD5940..................................................................................................................... 119
Figure 4-16: Breadboard setup connecting as STM32WB55 to an AD5941; insets show
both of the QFN-48 (quad flat no-lead) packaged chips, measuring 7 mm by 7
mm. ................................................................................................................................. 120
Figure 4-17: Image of breadboard setup with ADuCM355 on a breakout PCB next to the
AduCM355 evaluation board.......................................................................................... 121
Figure 4-18: Simplified block diagram for second potentiostat prototype; instead of BLE
as in Figure 4-13 (right), communication is through USB-C. ........................................ 122
Figure 4-19: Complete circuit schematic for miniature potentiostat .......................................... 122
Figure 4-20: For gold wires immersed in ferrocyanide, a peak response appears for all
three potentiostats: Gamry benchtop, evaluation board, and breadboard....................... 123
Figure 4-21: SWV on EAB wires in PBS shows that each potentiostat is sensitive to
frequency. Moreover, the breadboard response is identical to that of the
evaluation board.............................................................................................................. 124
Figure 4-22: Potentiostat layout with major components highlighted; see Figure 4-18 for
a block diagram............................................................................................................... 125
Figure 4-23: Photograph of fabricated and assembled potentiostat prototype ........................... 126
Figure 4-24: SWV on gold wires in ferrocyanide shows that each potentiostat is
sensitive to frequency. .................................................................................................... 127
Figure 4-25: Both the benchtop potentiostat and the miniaturized potentiostat PCB can
detect the increasing concentration of vancomycin in solution...................................... 128
Figure A-1: The earliest test of PUB bonding a Parylene-Pt-Parylene device to a PCB
showed excellent mechanical adhesion. Scale bar is 0.5 mm......................................... 171
Figure A-2: The PIE foundry uses PUB bonding to attach Parylene C microelectrode
arrays directly to PCBs for compact packaging.............................................................. 172
Figure A-3: Parylene C devices were directly PUB bonded to Pt wire, though soldering
the Pt wire to bond pads was necessary for gold ball bonding. ...................................... 173
Figure A-3: Shorted contact pads were PUB bonded to gold. The remaining contact pads
were for practice. Scale bar is 0.5 mm............................................................................ 174
Figure A-4: Optical micrographs of a) a bird’s eye view and b) an oblique view of two
Parylene-Pt-Parylene devices that have been PUB bonded together. The PCB
shown is not bonded to the devices. Scale bars are 0.5 mm. .......................................... 175

xv
ABSTRACT
By bringing sensing and actuation to the microscale, microelectromechanical systems
(MEMS) has profoundly shaped the modern world. Particularly in the biomedical sphere, MEMS
sensors can operate in close proximity to the targets of choice, be they organs, cells, or biofluids.
Still, there remain engineering challenges to the practical realization of these technologies into
useable products. For neural interfaces, bulky electrical connections hamper higher resolution.
For wearables such as microneedles, benchtop laboratory equipment must become portable. By
focusing on electronics and packaging, future generations of MEMS devices can continue to
improve lives.
In this work, chapter 1 discusses the role of MEMS in biological sensing and the challenges
faced therein. Chapter 2 introduces bonding methods by which rigid electronic chips can be
integrated with flexible polymer devices. Next, chapter 3 details the application of one such
bonding method—polymer ultrasonic on bump (PUB)—to integrate a neural recording chip with
a 64-channel Parylene C multielectrode array. Finally, chapter 4 captures the development of two
key components of a wearable microneedle system: the microneedle array itself and the compact
electronics necessary to take measurements.

1
CHAPTER 1
BIOLOGICAL APPLICATIONS FOR MEMS
1.1 MICROELECTROMECHANICAL SYSTEMS
Microelectromechanical systems (MEMS) are microscale devices that have been fabricated
using semiconductor technology and designed to interact with the physical world. From
accelerometers in automobiles to microphones in phones to micromirrors in optical data
transmission, MEMS technology has become integral to modern life. Thanks to their miniature
size, MEMS sensors can access spaces too small for conventional sensors, accomplishing the
same task at a fraction of the footprint. The same is true for biological and medical applications,
whose devices are called “bioMEMS.”
Many MEMS fabrication techniques find their roots in integrated circuit (IC) fabrication,
where silicon is the predominant material of choice due to its excellent semiconductor and
mechanical properties [1]. While it is far beyond the scope of this thesis to detail the numerous
fabrication techniques used to make MEMS devices, a brief overview can be found in Judy [2],
and an in-depth primer can be found in Madou [3]. In general, MEMS fabrication is
characterized by batch-scale additive and subtractive micromachining processes, the dimensions
of which are often controlled by photolithography. At these small scales, MEMS devices offer
unprecedented access to biological systems that were previously impossible or impractical to
pursue. In this thesis, we examine two areas where MEMS has had a large impact—areas where
we hope to contribute as well.

2
Figure 1-1: The first silicon MEMS microprobe, containing 2 recording sites. Reprinted,
with permission, from [5].
1.2 MEMS IN THE BRAIN
Until 1970, neural recording devices for the brain were metal-filled glass pipettes or
microwires—typically tungsten or platinum-coated metal [4]. Wise et al. were the first to apply
MEMS fabrication techniques to produce both single- and double-electrode microprobes for
recording from a cat cortex [5](Figure 1-1). The probe measured roughly 50 µm long, 100 µm
wide, and 55 µm thick. By taking advantage of planar construction in silicon bulk
micromachining, they were able to control the spacing between the double electrodes to 50 µm.
Iterations of this device would later be known as the “Michigan probe.” In 1991, Campbell et al.
reported a silicon needle array for intracortical recording [6]. This device featured 100 individual

3
electrodes that measured 80 µm in diameter and 1500 µm in length; they were fashioned from a
novel dicing and chemical etching strategy. This device later become known as the “Utah
electrode array.”
The Utah array saw incredible success as a brain-computer interface [7,8], but its out-ofplane design meant that each needle could house just one electrode, thereby limiting resolution to
100 channels. In contrast, the planar design of the Michigan probe was only limited by overall
intended probe size and minimum trace width and spacing—i.e., lithography resolution, a shared
goal in integrated circuit (IC) fabrication. Thus, while silicon manufacturing techniques
continued to improve and more features could be resolved in finer dimensions, researchers
produced neural probes with ever increasing number of recording sites [9–16] or number of
shanks themselves [6,14,17–19]. The highest-density silicon probe, as of this writing, is reported
in a preprint as having 4416 electrodes on a single shank [20].
The knowledge gained from both silicon and metal implants has been invaluable to brain
research [21–32] and prostheses [33], but one drawback to these rigid devices is the loss of
electrode signal during chronic implantation [34–40]. Upon initial injury to tissue, probes
inserted into the brain elicit an acute inflammatory response [41–46]. While this inflammatory
reaction may decrease over time, a dense sheath of reactive astrocytes and microglia continues to
surround the device [47,48]. One purported cause of this chronic damage is relative motion
between the implant and surrounding tissue [48–51], exacerbated by the wide gap in material
hardness between implant and tissue. After all, the Young’s modulus of silicon is approximately
150 GPa whereas that of tissue ranges from 10s to 100s of kPa (Figure 1-2). Another factor that
exacerbates failure of rigid implants is corrosion [52,53], and so a protective coating, typically a
polymer, must shield these materials everywhere but the recording site.

4
Figure 1-2: Common medical device materials are harder than human tissue. Adapted
with permission from Springer Nature [64].
One way to mitigate failures associated with hardness is to replace the substrate with
flexible polymers. Today, polymers such as polyimide, polydimethylsiloxane (PDMS), SU-8,
liquid crystal polymer (LCP) [54], and Parylene [55–57] comprise the bulk of these flexible
implants. Indeed, several researchers have demonstrated the long-term recording capability of
flexible polymer probes: 4 months [58], 5 months [59], 5.5 months [60], 300 days [61], and even
678 days [62]. By contrast, Biran et al. showed that silicon devices exhibited 40% neuronal loss
at 4 weeks post-implantation [48]. In a later study, Seymour et al. showed that Parylene C
devices experienced just 12-17% neuronal loss at the same 4-week time point [63].
Parylene is a thermoplastic polymer that is commonly coated on electronic devices to protect
against moisture. It is formed from the polymerization of para-xylylene, which consists of a
benzene ring and 2 CH2 bridges (Figure 1-3). When one or more hydrogen atoms in the ring are
replaced with another functional group, the resulting polymer exhibits different thermal,
mechanical, optical properties. The most common of these variants is Parylene C, in which a
hydrogen atom is replaced with chlorine (Figure 1-3). Other variations include Parylene N, D,
and HT, to name a few. Parylene C was the first to be qualified as biocompatible according to the
ISO 10993 and USP class VI standards, though Parylenes N and HT have also passed that bar.

5
Depositing Parylene C is quicker than Parylene N and does not require substrate cooling as is the
case with Parylene HT. Of the four, Parylene C also exhibits the lowest gas permeability.
Parylene C is typically deposited by vaporizing and splitting Parylene C dimer and then
driving the resulting monomers with a temperature gradient into a low-pressure deposition
chamber, where the resulting polymer forms a conformal, pinhole-free coating. Hence, it has
been widely used as insulation for electronic technology, providing excellent water resistance
and chemical inertness. Additionally, Parylene C can be micromachined using classic MEMS
techniques such as lithographic patterning and dry etching. It can be further molded using a
thermoforming process [65] as well as annealed for improved water resistance [66].
Figure 1-3: (left) A single unit of pary-xylxylene and (right) a single unit of Parylene C,
wherein a hydrogen atom has been replaced with chlorine.
Parylene C’s amenability to micromachining, flexibility, and biocompatibility make it an
excellent substrate for bioMEMS devices. Early on, Parylene C proved to be an effective
insulator for microelectrodes in long-term implantation [67], but later, Parylene C form the basis
of entire devices rather than just a coating [68–73]. As with silicon devices, successive polymer
devices had more and more recording sites as micromachining technology advanced.
State-of-the-art flexible polymer brain implants can hold several dozens of recording sites
on a single probe, but silicon devices can hold nearly 1000 sites on a probe [15,16](Figure 1-4).
The disparity lies not only in the advanced architecture possible with silicon—namely, vias and

6
small feature size, but also in that material’s semiconductor properties that allow adjacent
placement of integrated circuitry. While polymer microfabrication is still an active area of
research, compact polymer-based circuitry has yet to be realized. Instead, to achieve high-density
neural interfaces, simple polymer-metal-polymer probes can be arrayed and stacked to form a
three-dimensional array [58]. However, without the aid of on-board digitization and multiplexing,
packaging poses a challenge to making such devices. Furthermore, bulkier packaging introduces
greater risk of subjects’ tampering, which can lead to device failure [37].
Figure 1-4: Trends of recording sites per probe increase year over year.
To reduce the burden of packaging and pursue higher density of recording sites, more
compact interconnects are necessary. Chapter 2 of this thesis introduces basic concepts of
common bonding methods and the challenges that impede the integration of rigid chips with
flexible devices. This chapter also provides a solution to these challenges in the form of a new
bonding method called polymer-ultrasonic-bump (PUB) bonding. Chapter 3 discusses the
deployment of these studies for the assimilation of application-specific integrated circuit (ASIC)
chips with Parylene devices.

7
1.3 MEMS IN THE SKIN
Microneedles may have been first conceived in 1976 when a patent for a drug delivery
device was registered by Gerstel and Place [74](Figure 1-5). They described a device with a
“plurality of projections” through which a drug could be administered once these projections
pierced the stratum corneum, the outermost layer of the skin. Reports of microneedles, however,
did not appear in the literature until MEMS techniques were more mature [75]. While the first
reported microneedle was a glass micropipette with a 1 µm tip used for genetic engineering of
nematodes in 1994 [76], it was shortly followed by a silicon microneedle array fabricated from a
well-known MEMS process—anisotropic KOH etching [77]. The first to apply microneedles for
human use were Henry et al. who, in 1998, used reactive ion etching to produce a bed of silicon
spikes that were 80 µm in diameter and 150 µm long [78]. Coincidentally, these microneedles
closely resembled the Utah electrode array in form, but at a tenth of the length [6]. Still, these
dimensions were enough to pierce the stratum corneum—the topmost part of the epidermis,
thereby increasing the permeability of human skin. The next year, inspired by Wise’s silicon
microprobe and its reappearance alongside integrated circuitry (IC) [79], Lin and Pisano
produced a planar silicon microneedle with a single channel in the style of a hypodermic needle
[80]. Later that same year Brazzle et al. reported an array of similar planar hollow microneedles
[81], and McAllister et al. revealed out-of-plane hollow and silicon microneedles [82,83].
Though another planar, hypodermic-style microneedle appeared in the literature in 2000 [84], the
vast majority of microneedles thereafter were out-of-plane arrays [85–88].

8
Figure 1-5: a) One variation of a drug delivery device in Gerstel and Place’s patent. b)
the first reported microneedles used in human skin. © 1998 IEEE. Reprinted, with
permission, from [78]
The promise of painless penetration motivated many to pursue microneedles as vehicles for
transdermal drug delivery [89,90]. Researchers explored new geometries—both solid and hollow,
materials, fabrication methods, and deployment strategies to replace the hypodermic needle
[91,92](Figure 1-6). Meanwhile, others sought to use the microneedle platform not as a way to
inject drugs but as a way to extract information about biological processes.
Figure 1-6: Various kinds of microneedles targeting the dermis. Reprinted from [91],
Copyright 2012, with permission from Elsevier.
Notably, in 2004, Mukerjee et al. attempted to draw out interstitial fluid (ISF) from the skin
by capillary action [87]. This device was an array of hollow, out-of-plane silicon microneedles
that were anodically bonded to a borosilicate glass cap. The needle lumens and channels were

9
deep reactive ion etched (DRIE) on one side of the silicon wafer, and the tips were diced then
reactive ion etched (RIE) on the other side to form sharp points. Though it was successful in
drawing out the ISF, researchers reported clogging issues depending on center alignment
between the lumen and the needle tip. Indeed, later researchers who similarly attempted to
extract fluid in vivo were keenly aware of the possibility of occlusion [93–95].
The next stage in the development of biosensing microneedles was placing a sensing
element on the needles themselves. Starting in 2011, the Wang group was prolific in producing
3D printed hollow microneedles filled with carbon paste electrodes that were functionalized with
an electroactive enzyme. With this approach, they were able to detect lactate [96,97], glutamate
[98], glucose [97–99], and more recently, organophosphates [100], alcohol [101], and the antiParkinson’s drug levodopa [102]. Other biosensing microneedles constructed of unique materials
were also reported: silicon [103], SU-8 [104–107], stainless steel [108–110], and a variety of
polymers [111–114]. In 2013 and 2014, ArKal Medical reported a silicon microneedle array
capable of extracting ISF into an adjacent glucose-sensing chamber [115,116]. Impressively,
they showed good agreement between fingerstick blood glucose and their 34 devices, which
were worn for 72 hours. The majority of these microneedles, too, relied on enzymatic sensing.
Within last few years, however, microneedles that incorporate electrochemical aptamer-based
(EAB) sensing have been reported [117–120], and while EAB sensing is a promising new
method of biomarker detection, it has yet to be incorporated into a portable, perhaps wearable,
system. Chapter 4 of this thesis presents a pathway for doing so.

10
1.4 OBJECTIVES
The projects presented in this dissertation represent two different ways MEMS devices can
address biological sensing. In the first, flexible polymer-based MEMS arrays can obtain chronic
neural recording at the cost of high spatial resolution, which silicon-based devices can
accomplish using monolithic semiconductor fabrication of ICs. Expanding channel counts on
polymer arrays will require advanced packaging to integrate application-specific ICs to perform
signal processing on the array. In this way, electrical connection to such a device would be more
compact and thus more manageable. To that end, we present a strategy for integrating an ASIC
with a Parylene C multielectrode array.
The second way that MEMS can access biological information is microneedles that can
painlessly interact with interstitial fluid in the dermis. Combined with electrochemical aptamerbased (EAB) sensing, these sensors would be capable of detecting nearly any analyte with
excellent temporal resolution. Testing these devices, however, requires compact, lightweight
electronics that can perform electrochemical measurement. Hence, we present the design and
fabrication of both silicon microneedle arrays and miniaturized potentiostat for future in vivo
testing.
In short, the overarching goal of this work is to develop biological sensors with enhanced
spatial and temporal resolution by incorporating compact electronics and packaging.

11
REFERENCES
[1] Petersen K E 1982 Silicon as a mechanical material Proc. IEEE 70 420–57
[2] Judy J W 2001 Microelectromechanical systems (MEMS): fabrication, design and
applications Smart Mater. Struct. 10 1115–34
[3] Madou M J 2018 Fundamentals of microfabrication: the science of miniaturization (CRC
press)
[4] Gesteland R, Howland B, Lettvin J and Pitts W 1959 Comments on Microelectrodes
Proc. IRE 47 1856–62
[5] Wise K D, Angell J B and Starr A 1970 An Integrated-Circuit Approach to Extracellular
Microelectrodes IEEE Trans. Biomed. Eng. BME-17 238–47
[6] Campbell P K, Jones K E, Huber R J, Horch K W and Normann R A 1991 A siliconbased, three-dimensional neural interface: manufacturing processes for an intracortical
electrode array IEEE Trans. Biomed. Eng. 38 758–68
[7] Hochberg L R, Serruya M D, Friehs G M, Mukand J A, Saleh M, Caplan A H, Branner
A, Chen D, Penn R D and Donoghue J P 2006 Neuronal ensemble control of prosthetic
devices by a human with tetraplegia Nature 442 164–71
[8] Hochberg L R, Bacher D, Jarosiewicz B, Masse N Y, Simeral J D, Vogel J, Haddadin S,
Liu J, Cash S S, van der Smagt P and Donoghue J P 2012 Reach and grasp by people
with tetraplegia using a neurally controlled robotic arm Nature 485 372–5
[9] Kuperstein M and Whittington D A 1981 A Practical 24 Channel Microelectrode for
Neural Recording in Vivo IEEE Trans. Biomed. Eng. BME-28 288–93
[10] BeMent S L, Wise K D, Anderson D J, Najafi K and Drake K L 1986 Solid-State
Electrodes for Multichannel Multiplexed Intracortical Neuronal Recording IEEE Trans.
Biomed. Eng. BME-33 230–41
[11] Blanche T J, Spacek M A, Hetke J F and Swindale N V 2005 Polytrodes: High-Density
Silicon Electrode Arrays for Large-Scale Multiunit Recording J. Neurophysiol. 93 2987–
3000
[12] Seidl K, Herwik S, Torfs T, Neves H P, Paul O and Ruther P 2011 CMOS-Based HighDensity Silicon Microprobe Arrays for Electronic Depth Control in Intracortical Neural
Recording J. Microelectromechanical Syst. 20 1439–48
[13] Scholvin J, Kinney J P, Bernstein J G, Moore-Kochlacs C, Kopell N, Fonstad C G and
Boyden E S 2016 Close-Packed Silicon Microelectrodes for Scalable Spatially
Oversampled Neural Recording IEEE Trans. Biomed. Eng. 63 120–30

12
[14] Rios G, Lubenov E V, Chi D, Roukes M L and Siapas A G 2016 Nanofabricated Neural
Probes for Dense 3-D Recordings of Brain Activity Nano Lett. 16 6857–62
[15] Jun J J, Steinmetz N A, Siegle J H, Denman D J, Bauza M, Barbarits B, Lee A K,
Anastassiou C A, Andrei A and Aydın Ç 2017 Fully integrated silicon probes for highdensity recording of neural activity Nature 551 232
[16] Raducanu B C, Yazicioglu R F, Lopez C M, Ballini M, Putzeys J, Wang S, Andrei A,
Rochus V, Welkenhuysen M, Helleputte N van, Musa S, Puers R, Kloosterman F, Hoof C
van, Fiáth R, Ulbert I and Mitra S 2017 Time Multiplexed Active Neural Probe with
1356 Parallel Recording Sites Sensors 17 2388
[17] Ylinen A, Bragin A, Nadasdy Z, Jando G, Szabo I, Sik A and Buzsaki G 1995 Sharp
wave-associated high-frequency oscillation (200 Hz) in the intact hippocampus: network
and intracellular mechanisms J. Neurosci. 15 30–46
[18] Csicsvari J, Henze D A, Jamieson B, Harris K D, Sirota A, Barthó P, Wise K D and
Buzsáki G 2003 Massively Parallel Recording of Unit and Local Field Potentials With
Silicon-Based Electrodes J. Neurophysiol. 90 1314–23
[19] Berényi A, Somogyvári Z, Nagy A J, Roux L, Long J D, Fujisawa S, Stark E, Leonardo
A, Harris T D and Buzsáki G 2014 Large-scale, high-density (up to 512 channels)
recording of local circuits in behaving animals J. Neurophysiol. 111 1132–49
[20] Trautmann E M, Hesse J K, Stine G M, Xia R, Zhu S, O’Shea D J, Karsh B, Colonell J,
Lanfranchi F F, Vyas S, Zimnik A, Steinmann N A, Wagenaar D A, Andrei A, Lopez C
M, O’Callaghan J, Putzeys J, Raducanu B C, Welkenhuysen M, Churchland M, Moore T,
Shadlen M, Shenoy K, Tsao D, Dutta B and Harris T 2023 Large-scale high-density
brain-wide neural recording in nonhuman primates 2023.02.01.526664
[21] Anderson L A, Christianson G B and Linden J F 2009 Mouse auditory cortex differs
from visual and somatosensory cortices in the laminar distribution of cytochrome oxidase
and acetylcholinesterase Brain Res. 1252 130–42
[22] Gage G J, Stoetzner C R, Wiltschko A B and Berke J D 2010 Selective Activation of
Striatal Fast-Spiking Interneurons during Choice Execution Neuron 67 466–79
[23] Winslow B D, Christensen M B, Yang W-K, Solzbacher F and Tresco P A 2010 A
comparison of the tissue response to chronically implanted Parylene-C-coated and
uncoated planar silicon microelectrode arrays in rat cortex Biomaterials 31 9163–72
[24] Belluscio M A, Mizuseki K, Schmidt R, Kempter R and Buzsaki G 2012 CrossFrequency Phase-Phase Coupling between Theta and Gamma Oscillations in the
Hippocampus J. Neurosci. 32 423–35
[25] Mizuseki K, Royer S, Diba K and Buzsáki G 2012 Activity dynamics and behavioral
correlates of CA3 and CA1 hippocampal pyramidal neurons Hippocampus 22 1659–80

13
[26] Royer S, Zemelman B V, Losonczy A, Kim J, Chance F, Magee J C and Buzsáki G 2012
Control of timing, rate and bursts of hippocampal place cells by dendritic and somatic
inhibition Nat. Neurosci. 15 769–75
[27] Patel J, Fujisawa S, Berényi A, Royer S and Buzsáki G 2012 Traveling Theta Waves
along the Entire Septotemporal Axis of the Hippocampus Neuron 75 410–7
[28] Hampson R E, Gerhardt G A, Marmarelis V, Song D, Opris I, Santos L, Berger T W and
Deadwyler S A 2012 Facilitation and restoration of cognitive function in primate
prefrontal cortex by a neuroprosthesis that utilizes minicolumn-specific neural firing J.
Neural Eng. 9 056012
[29] Haider B, Häusser M and Carandini M 2013 Inhibition dominates sensory responses in
the awake cortex Nature 493 97–100
[30] Sales-Carbonell C, Rueda-Orozco P E, Soria-Gomez E, Buzsaki G, Marsicano G and
Robbe D 2013 Striatal GABAergic and cortical glutamatergic neurons mediate
contrasting effects of cannabinoids on cortical network synchrony Proc. Natl. Acad. Sci.
110 719–24
[31] Lee J H, Whittington M A and Kopell N J 2013 Top-Down Beta Rhythms Support
Selective Attention via Interlaminar Interaction: A Model ed S Coombes PLoS Comput.
Biol. 9 e1003164
[32] Smith M A, Jia X, Zandvakili A and Kohn A 2013 Laminar dependence of neuronal
correlations in visual cortex J. Neurophysiol. 109 940–7
[33] Chestek C A, Gilja V, Nuyujukian P, Foster J D, Fan J M, Kaufman M T, Churchland M
M, Rivera-Alvidrez Z, Cunningham J P, Ryu S I and Shenoy K V 2011 Long-term
stability of neural prosthetic control signals from silicon cortical arrays in rhesus
macaque motor cortex J. Neural Eng. 8 045005
[34] Rousche P J and Normann R A 1998 Chronic recording capability of the Utah
Intracortical Electrode Array in cat sensory cortex J. Neurosci. Methods 82 1–15
[35] Freire M A M, Morya E, Faber J, Santos J R, Guimaraes J S, Lemos N A M, Sameshima
K, Pereira A, Ribeiro S and Nicolelis M A L 2011 Comprehensive Analysis of Tissue
Preservation and Recording Quality from Chronic Multielectrode Implants ed S D
Ginsberg PLoS ONE 6 e27554
[36] Prasad A, Xue Q-S, Sankar V, Nishida T, Shaw G, Streit W J and Sanchez J C 2012
Comprehensive characterization and failure modes of tungsten microwire arrays in
chronic neural implants J. Neural Eng. 9 056015
[37] Barrese J C, Rao N, Paroo K, Triebwasser C, Vargas-Irwin C, Franquemont L and
Donoghue J P 2013 Failure mode analysis of silicon-based intracortical microelectrode
arrays in non-human primates J. Neural Eng. 10 066014

14
[38] Prasad A, Xue Q-S, Dieme R, Sankar V, Mayrand R C, Nishida T, Streit W J and
Sanchez J C 2014 Abiotic-biotic characterization of Pt/Ir microelectrode arrays in chronic
implants Front. Neuroengineering 7
[39] Kozai T D Y, Catt K, Li X, Gugel Z V, Olafsson V T, Vazquez A L and Cui X T 2015
Mechanical failure modes of chronically implanted planar silicon-based neural probes for
laminar recording Biomaterials 37 25–39
[40] Debnath S, Prins N W, Pohlmeyer E, Mylavarapu R, Geng S, Sanchez J C and Prasad A
2018 Long-term stability of neural signals from microwire arrays implanted in common
marmoset motor cortex and striatum Biomed. Phys. Eng. Express 4 055025
[41] Agnew W F, Yuen T G H, McCreery D B and Bullara L A 1986 Histopathologic
evaluation of prolonged intracortical electrical stimulation Exp. Neurol. 92 162–85
[42] Edell D J, Toi V V, McNeil V M and Clark L D 1992 Factors influencing the
biocompatibility of insertable silicon microshafts in cerebral cortex IEEE Trans. Biomed.
Eng. 39 635–43
[43] McCreery D B, Yuen T G H, Agnew W F and Bullara L A 1997 A characterization of
the effects on neuronal excitability due to prolonged microstimulation with chronically
implanted microelectrodes IEEE Trans. Biomed. Eng. 44 931–9
[44] Schmidt S, Horch K and Normann R 1993 Biocompatibility of silicon-based electrode
arrays implanted in feline cortical tissue J. Biomed. Mater. Res. 27 1393–9
[45] Schmidt E M, Bak M J and McIntosh J S 1976 Long-term chronic recording from
cortical neurons Exp. Neurol. 52 496–506
[46] Schultz R L and Willey T J 1976 The ultrastructure of the sheath around chronically
implanted electrodes in brain J. Neurocytol. 5 621–42
[47] Szarowski D H, Andersen M D, Retterer S, Spence A J, Isaacson M, Craighead H G,
Turner J N and Shain W 2003 Brain responses to micro-machined silicon devices Brain
Res. 983 23–35
[48] Biran R, Martin D C and Tresco P A 2005 Neuronal cell loss accompanies the brain
tissue response to chronically implanted silicon microelectrode arrays Exp. Neurol. 195
115–26
[49] Subbaroyan J, Martin D C and Kipke D R 2005 A finite-element model of the
mechanical effects of implantable microelectrodes in the cerebral cortex J. Neural Eng. 2
103–13
[50] Lee H, Bellamkonda R V, Sun W and Levenston M E 2005 Biomechanical analysis of
silicon microelectrode-induced strain in the brain J. Neural Eng. 2 81–9

15
[51] Biran R, Martin D C and Tresco P A 2007 The brain tissue response to implanted silicon
microelectrode arrays is increased when the device is tethered to the skull J. Biomed.
Mater. Res. A 82A 169–78
[52] Patrick E, Orazem M E, Sanchez J C and Nishida T 2011 Corrosion of tungsten
microelectrodes used in neural recording applications J. Neurosci. Methods 198 158–71
[53] Vanhoestenberghe A and Donaldson N 2013 Corrosion of silicon integrated circuits and
lifetime predictions in implantable electronic devices J. Neural Eng. 10 031002
[54] Jeong J, Hyun Bae S, Seo J-M, Chung H and June Kim S 2016 Long-term evaluation of
a liquid crystal polymer (LCP)-based retinal prosthesis J. Neural Eng. 13 025004
[55] Stieglitz T and Schuettler M 2013 Material–tissue interfaces in implantable systems
Implantable Sensor Systems for Medical Applications (Elsevier) pp 39–67
[56] Kim B J, Kuo J T W, Hara S A, Lee C D, Yu L, Gutierrez C A, Hoang T Q, Pikov V and
Meng E 2013 3D Parylene sheath neural probe for chronic recordings J. Neural Eng. 10
045002
[57] Cobo A M, Larson C E, Scholten K, Miranda J A, Elyahoodayan S, Song D, Pikov V
and Meng E 2019 Parylene-Based Cuff Electrode With Integrated Microfluidics for
Peripheral Nerve Recording, Stimulation, and Drug Delivery J. Microelectromechanical
Syst. 28 36–49
[58] Chung J E, Joo H R, Fan J L, Liu D F, Barnett A H, Chen S, Geaghan-Breiner C,
Karlsson M P, Karlsson M and Lee K Y 2019 High-density, long-lasting, and multiregion electrophysiological recordings using polymer electrode arrays Neuron 101 21-31.
e5
[59] Böhler C, Vomero M, Soula M, Vöröslakos M, Porto Cruz M, Liljemalm R, Buzsaki G,
Stieglitz T and Asplund M 2023 Multilayer Arrays for Neurotechnology Applications
(MANTA): Chronically Stable Thin‐Film Intracortical Implants Adv. Sci. 10 2207576
[60] Luan L, Wei X, Zhao Z, Siegel J J, Potnis O, Tuppen C A, Lin S, Kazmi S, Fowler R A,
Holloway S, Dunn A K, Chitwood R A and Xie C 2017 Ultraflexible nanoelectronic
probes form reliable, glial scar–free neural integration Sci. Adv. 3 e1601966
[61] Zhao Z, Zhu H, Li X, Sun L, He F, Chung J E, Liu D F, Frank L, Luan L and Xie C 2023
Ultraflexible electrode arrays for months-long high-density electrophysiological mapping
of thousands of neurons in rodents Nat. Biomed. Eng. 7 520–32
[62] Sohal H S, Jackson A, Jackson R, Clowry G J, Vassilevski K, O’Neill A and Baker S N
2014 The sinusoidal probe: a new approach to improve electrode longevity Front.
Neuroengineering 7
[63] Seymour J P and Kipke D R 2007 Neural probe design for reduced tissue encapsulation
in CNS Biomaterials 28 3594–607

16
[64] Lacour S P, Courtine G and Guck J 2016 Materials and technologies for soft implantable
neuroprostheses Nat. Rev. Mater. 1 1–14
[65] Kim B J and Meng E 2016 Micromachining of Parylene C for bioMEMS Polym. Adv.
Technol. 27 564–76
[66] Ortigoza-Diaz J, Scholten K and Meng E 2018 Characterization and Modification of
Adhesion in Dry and Wet Environments in Thin-Film Parylene Systems J.
Microelectromechanical Syst. 27 874–85
[67] Schmidt E M, Mcintosh J S and Bak M J 1988 Long-term implants of Parylene-C coated
microelectrodes Med. Biol. Eng. Comput. 26 96–101
[68] Suzuki T, Mabuchi K and Takeuchi S 2003 A 3D flexible parylene probe array for
multichannel neural recording First International IEEE EMBS Conference on Neural
Engineering, 2003. Conference Proceedings. pp 154–6
[69] Takeuchi S, Ziegler D, Yoshida Y, Mabuchi K and Suzuki T 2005 Parylene flexible
neural probes integrated with microfluidic channels Lab. Chip 5 519
[70] Seymour J P and Kipke D R 2006 Fabrication of Polymer Neural Probes with Subcellular Features for Reduced Tissue Encapsulation 2006 International Conference of the
IEEE Engineering in Medicine and Biology Society 2006 International Conference of the
IEEE Engineering in Medicine and Biology Society pp 4606–9
[71] Kato Y, Nishino M, Saito I, Suzuki T and Mabuchi K 2006 Flexible Intracortical Neural
Probe with Biodegradable Polymer for Delivering Bioactive Components 2006
International Conference on Microtechnologies in Medicine and Biology 2006
International Conference on Microtechnologies in Medicine and Biology pp 143–6
[72] Kim Y, Lee H J, Kim D, Kim Y K, Lee S H, Yoon E-S and Cho I-J 2013 A new MEMS
neural probe integrated with embedded microfluidic channel for drug delivery and
electrode array for recording neural signal 2013 Transducers & Eurosensors XXVII: The
17th International Conference on Solid-State Sensors, Actuators and Microsystems
(TRANSDUCERS & EUROSENSORS XXVII) 2013 Transducers & Eurosensors XXVII:
The 17th International Conference on Solid-State Sensors, Actuators and Microsystems
(TRANSDUCERS & EUROSENSORS XXVII) pp 876–9
[73] Ziegler D, Suzuki T and Takeuchi S 2006 Fabrication of Flexible Neural Probes With
Built-In Microfluidic Channels by Thermal Bonding of Parylene J.
Microelectromechanical Syst. 15 1477–82
[74] Gerstel M S and Place V A 1976 Drug delivery device
[75] Laermer F and Schilp A 1996 Method of anisotropically etching silicon
[76] Hashmi S, Hashmi G and Gaugler R 1995 Genetic Transformation of an
Entomopathogenic Nematode by Microinjection J. Invertebr. Pathol. 66 293–6

17
[77] Trimmer W, Ling P, Chee-Kok Chin, Orton P, Gaugler R, Hashmi S, Hashmi G, Brunett
B and Reed M 1995 Injection of DNA into plant and animal tissues with
micromechanical piercing structures Proceedings IEEE Micro Electro Mechanical
Systems. 1995 IEEE Micro Electro Mechanical Systems. 1995 (Amsterdam, Netherlands:
IEEE) p 111
[78] Henry S, McAllister D V, Allen M G and Prausnitz M R 1998 Microfabricated
Microneedles: A Novel Approach to Transdermal Drug Delivery J. Pharm. Sci. 87 922–5
[79] Najafi K, Wise K D and Mochizuki T 1985 A high-yield IC-compatible multichannel
recording array IEEE Trans. Electron Devices 32 1206–11
[80] Lin L and Pisano A P 1999 Silicon-processed microneedles J. Microelectromechanical
Syst. 8 78–84
[81] Brazzle J, Papautsky I and Frazier A B 1999 Micromachined needle arrays for drug
delivery or fluid extraction IEEE Eng. Med. Biol. Mag. 18 53–8
[82] McAllister D V 1999 Three-dimensional hollow microneedle and microtube arrays Proc.
Transducers’ 99 pp 1098–101
[83] McAllister D V, Allen M G and Prausnitz M R 2000 Microfabricated Microneedles for
Gene and Drug Delivery Annu. Rev. Biomed. Eng. 2 289–313
[84] Smart W H and Subramanian K 2000 The Use of Silicon Microfabrication Technology
in Painless Blood Glucose Monitoring Diabetes Technol. Ther. 2 549–59
[85] Stoeber B and Liepmann D 2000 Fluid injection through out-of-plane microneedles 1st
Annual International IEEE-EMBS Special Topic Conference on Microtechnologies in
Medicine and Biology. Proceedings (Cat. No.00EX451) 1st Annual International IEEEEMBS Special Topic Conference on Microtechnologies in Medicine and Biology.
Proceedings (Cat. No.00EX451) pp 224–8
[86] Griss P, Enoksson P, Tolvanen-Laakso H K, Merilainen P, Ollmar S and Stemme G
2001 Micromachined electrodes for biopotential measurements J.
Microelectromechanical Syst. 10 10–6
[87] Mukerjee E V, Collins S D, Isseroff R R and Smith R L 2004 Microneedle array for
transdermal biological fluid extraction and in situ analysis Sens. Actuators Phys. 114
267–75
[88] Kim K, Park D S, Lu H M, Che W, Kim K, Lee J-B and Ahn C H 2004 A tapered
hollow metallic microneedle array using backside exposure of SU-8 J. Micromechanics
Microengineering 14 597
[89] Gill H S, Denson D D, Burris B A and Prausnitz M R 2008 Effect of microneedle design
on pain in human subjects Clin. J. Pain 24 585–94

18
[90] Kaushik S, Hord A H, Denson D D, McAllister D V, Smitra S, Allen M G and Prausnitz
M R 2001 Lack of Pain Associated with Microfabricated Microneedles Anesth. Analg. 92
502
[91] Kim Y-C, Park J-H and Prausnitz M R 2012 Microneedles for drug and vaccine delivery
Adv. Drug Deliv. Rev. 64 1547–68
[92] Prausnitz M R 2004 Microneedles for transdermal drug delivery Adv. Drug Deliv. Rev.
56 581–7
[93] Tsuchiya K, Nakanishi N, Uetsuji Y and Nakamachi E 2005 Development of Blood
Extraction System for Health Monitoring System Biomed. Microdevices 7 347–53
[94] Wang P M, Cornwell M and Prausnitz M R 2005 Minimally Invasive Extraction of
Dermal Interstitial Fluid for Glucose Monitoring Using Microneedles Diabetes Technol.
Ther. 7 131–41
[95] Corrie S R, Fernando G J P, Crichton M L, Brunck M E G, Anderson C D and Kendall
M A F 2010 Surface-modified microprojection arrays for intradermal biomarker capture,
with low non-specific protein binding Lab. Chip 10 2655
[96] Windmiller J R, Zhou N, Chuang M-C, Valdés-Ramírez G, Santhosh P, Miller P R,
Narayan R and Wang J 2011 Microneedle array-based carbon paste amperometric
sensors and biosensors Analyst 136 1846–51
[97] Miller P R, Skoog S A, Edwards T L, Lopez D M, Wheeler D R, Arango D C, Xiao X,
Brozik S M, Wang J, Polsky R and Narayan R J 2012 Multiplexed microneedle-based
biosensor array for characterization of metabolic acidosis Talanta 88 739–42
[98] Windmiller J R, Valdés-Ramírez G, Zhou N, Zhou M, Miller P R, Jin C, Brozik S M,
Polsky R, Katz E, Narayan R and Wang J 2011 Bicomponent Microneedle Array
Biosensor for Minimally-Invasive Glutamate Monitoring Electroanalysis 23 2302–9
[99] Valdés-Ramírez G, Li Y-C, Kim J, Jia W, Bandodkar A J, Nuñez-Flores R, Miller P R,
Wu S-Y, Narayan R, Windmiller J R, Polsky R and Wang J 2014 Microneedle-based
self-powered glucose sensor Electrochem. Commun. 47 58–62
[100] Mishra R K, Mohan A M V, Soto F, Chrostowski R and Wang J 2017 A microneedle
biosensor for minimally-invasive transdermal detection of nerve agents Analyst 142 918–
24
[101] Mohan A M V, Windmiller J R, Mishra R K and Wang J 2017 Continuous minimallyinvasive alcohol monitoring using microneedle sensor arrays Biosens. Bioelectron. 91
574–9
[102] Goud K Y, Moonla C, Mishra R K, Yu C, Narayan R, Litvan I and Wang J 2019
Wearable Electrochemical Microneedle Sensor for Continuous Monitoring of Levodopa:
Toward Parkinson Management ACS Sens. 4 2196–204

19
[103] Yoon Y, Lee G S, Yoo K and Lee J-B 2013 Fabrication of a Microneedle/CNT
Hierarchical Micro/Nano Surface Electrochemical Sensor and Its In-Vitro Glucose
Sensing Characterization Sensors 13 16672–81
[104] Moniz A R B, Michelakis K, Trzebinski J, Sharma S, Johnston D G, Oliver N and Cass
A 2012 Minimally Invasive Enzyme Microprobes: An Alternative Approach for
Continuous Glucose Monitoring J. Diabetes Sci. Technol. 6 479–80
[105] Trzebinski J, Sharma S, Moniz A R-B, Michelakis K, Zhang Y and G. Cass A E 2012
Microfluidic device to investigate factors affecting performance in biosensors designed
for transdermal applications Lab. Chip 12 348–52
[106] Sharma S, Huang Z, Rogers M, Boutelle M and Cass A E G 2016 Evaluation of a
minimally invasive glucose biosensor for continuous tissue monitoring Anal. Bioanal.
Chem. 408 8427–35
[107] Samavat S, Lloyd J, O’Dea L, Zhang W, Preedy E, Luzio S and Teng K S 2018 Uniform
sensing layer of immiscible enzyme-mediator compounds developed via a spray aerosol
mixing technique towards low cost minimally invasive microneedle continuous glucose
monitoring devices Biosens. Bioelectron. 118 224–30
[108] Invernale M A, Tang B C, York R L, Le L, Hou D Y and Anderson D G 2014
Microneedle Electrodes Toward an Amperometric Glucose-Sensing Smart Patch Adv.
Healthc. Mater. 3 338–42
[109] Lee S J, Yoon H S, Xuan X, Park J Y, Paik S-J and Allen M G 2016 A patch type nonenzymatic biosensor based on 3D SUS micro-needle electrode array for minimally
invasive continuous glucose monitoring Sens. Actuators B Chem. 222 1144–51
[110] Chinnadayyala S R, Park I and Cho S 2018 Nonenzymatic determination of glucose at
near neutral pH values based on the use of nafion and platinum black coated microneedle
electrode array Microchim. Acta 185 250
[111] Dardano P, Caliò A, Di Palma V, Bevilacqua M F, Di Matteo A and De Stefano L 2015
A Photolithographic Approach to Polymeric Microneedles Array Fabrication Materials 8
8661–73
[112] Caliò A, Dardano P, Di Palma V, Bevilacqua M F, Di Matteo A, Iuele H and De Stefano
L 2016 Polymeric microneedles based enzymatic electrodes for electrochemical
biosensing of glucose and lactic acid Sens. Actuators B Chem. 236 343–9
[113] Bollella P, Sharma S, Cass A E G and Antiochia R 2019 Microneedle-based biosensor
for minimally-invasive lactate detection Biosens. Bioelectron. 123 152–9
[114] Kim K B, Lee W-C, Cho C-H, Park D-S, Cho S J and Shim Y-B 2019 Continuous
glucose monitoring using a microneedle array sensor coupled with a wireless signal
transmitter Sens. Actuators B Chem. 281 14–21

20
[115] Chua B, Desai S P, Tierney M J, Tamada J A and Jina A N 2013 Effect of microneedles
shape on skin penetration and minimally invasive continuous glucose monitoring in vivo
Sens. Actuators Phys. 203 373–81
[116] Jina A, Tierney M J, Tamada J A, McGill S, Desai S, Chua B, Chang A and Christiansen
M 2014 Design, Development, and Evaluation of a Novel Microneedle Array-based
Continuous Glucose Monitor J. Diabetes Sci. Technol. 8 483–7
[117] Wu Y, Tehrani F, Teymourian H, Mack J, Shaver A, Reynoso M, Kavner J, Huang N,
Furmidge A, Duvvuri A, Nie Y, Laffel L M, Doyle F J, Patti M-E, Dassau E, Wang J and
Arroyo-Currás N 2022 Microneedle Aptamer-Based Sensors for Continuous, Real-Time
Therapeutic Drug Monitoring Anal. Chem. 94 8335–45
[118] Lin S, Cheng X, Zhu J, Wang B, Jelinek D, Zhao Y, Wu T-Y, Horrillo A, Tan J, Yeung J,
Yan W, Forman S, Coller H A, Milla C and Emaminejad S 2022 Wearable microneedlebased electrochemical aptamer biosensing for precision dosing of drugs with narrow
therapeutic windows Sci. Adv. 8 eabq4539
[119] Downs A M, Bolotsky A, Weaver B M, Bennett H, Wolff N, Polsky R and Miller P R
2023 Microneedle electrochemical aptamer-based sensing: Real-time small molecule
measurements using sensor-embedded, commercially-available stainless steel
microneedles Biosens. Bioelectron. 236 115408
[120] Friedel M, Werbovetz B, Drexelius A, Watkins Z, Bali A, Plaxco K W and Heikenfeld J
2023 Continuous molecular monitoring of human dermal interstitial fluid with
microneedle-enabled electrochemical aptamer sensors Lab. Chip 23 3289–99

21
CHAPTER 2
INTERCONNECT BONDING METHODS FOR PARYLENE DEVICES
2.1 BACKGROUND
Electronics packaging is the strategy by which an electronic device is protected from and
connected to the outside world [1]. For instance, a microprocessor may be wire bonded to a
larger resin-based shell with outer contact bumps in a grid array and then sealed with epoxy to
protect from dust and moisture (Figure 2-1). In a similar manner, connections from electronic
medical devices to a transmitter or computer must be protected from hazards biological,
mechanical, electrical, and thermal. These connections between two discrete components of a
device can be referred to as the interconnect. In the previous microprocessor example, the wire
bonds are the interconnects.
Figure 2-1: a) The inside of a microprocessor showing wire bonds along the perimeter of
the chip, b) a close-up highlighting wire bonds and bond pads, and c) a scanning
electron micrograph (SEM) showing a typical wire wedge bond. Motorola68040die.jpg
by Gregg M. Erickson is licensed under CC BY 3.0.

22
In a neural device, any and all connections from the sensing portion of the device to the
computer are interconnects. These could be printed circuit boards (PCBs), solder, wires,
proprietary connectors, or any number of other strategies (Figure 2-2). For our purposes, we are
concerned mainly with the interconnect from the active sensing portion of the device to the next
component.
Figure 2-2: For this neural implant, the probe array connects to a printed circuit board
by way of a zero-insertion force (ZIF) connector, which is the interconnect.
For any of these strategies, the trend for more signals has led to wider or longer connectors to
account for an increasing number of interconnects [2](Figure 2-3). As researchers and companies
alike pursue higher densities of signals on a single device, the size of the device becomes
impractically bulky. Thus, packaging poses a potential threat to progress. Reducing the size and
spacing of interconnects—aka pitch—is one solution. A more challenging proposition is
integrating circuits (IC) to digitize and then multiplex signals so that fewer connections need to
be made. For devices made of silicon, ICs can be built adjacent to the sensing portion of the
device [3], but for flexible polymer devices, such ICs must be bonded in a separate process. One

23
application-specific integrated circuit (ASIC) that is designed for neural signal processing is
Intan’s RHD series. These chips offer multiplexing and digitization capabilities, and they can be
found in a number of neural recording systems [4], [5], [6], [7], [8], [9], [10], [11], [12], [13],
[14], [15], [16].
Figure 2-3: The number of recording sites on a neural probe is rapidly increasing, and
many modern devices utilize chip integration to handle the increased packaging burden.
2.2 CHALLENGES TO ASIC INTEGRATION
The greatest challenges to integrating ASICs to polymer devices are fine pitch, thermal budget,
and precise alignment. Here, we define “fine pitch” as center-to-center spacing between bond
pads of 100 µm. Below this pitch, specialized or automatic equipment is required [17]. Intan’s
chips with 100 µm pitch, for example, can be wire bonded to a protective ball-grid array (BGA)
package using a semiautomatic bonder [18]. Because making pitch finer necessarily entails
making interconnects smaller, the proposed interconnect must be capable of making electrical
connections even at reduced size and pitch. This may prove problematic especially for
heterogeneous interconnects such as conductive epoxy. Additionally, if an interconnect is liquid

24
before curing, its viscosity must be such that it is not only deployable but also incapable of
flowing and shorting adjacent bond pads.
Figure 2-4: Comparing critical temperatures of MEMS materials to temperature ranges
of standard interconnect techniques shows that many existing methods are unsuitable for
polymers
The vast majority of interconnects are designed for rigid substrates such as silicon chips or
PCBs, materials that can withstand much higher temperatures than polymers (Figure 2-4). For
instance, silicon’s melting temperature is 1400 °C, but Parylene C begins to oxidize around
120 °C and melt at 290 °C. Hence, any bonding method for integrated ASICs with polymers
must keep high-temperature steps away from the polymer device.
Finally, alignment between polymer device and chip must be precise enough to prevent
shorting between neighboring bond pads due to overflow of material. For example, industrial
equipment that can accomplish this are flip chip bonders and pick and place machines.

25
2.3 EXISTING INTERCONNECT METHODS
A number of interconnect methods exist for MEMS devices, each bearing distinct situational
advantages. As with fabrication techniques, many packaging strategies find their roots in silicon
MEMS devices; still others have been developed specifically for flexible polymer devices. Here
follows a summary of interconnects in both academic and industrial MEMS applications.
Soldering is the formation of a joint between two bonding surfaces by melting a tin-based
alloy, flowing via capillary action, and solidifying. While hobbyists and professionals may solder
by hand using solder filament, batch soldering is possible by deploying multiple solder bumps on
a device and melting them simultaneously. Aligning an array of solder bumps blindly onto
another substrate may require a flip-chip bonder. In any case, solder often melts from 180 °C to
250 °C, but specially formulated low-temperature solder can melt below 150 °C [19]. Typical
surface mount technology (SMT) chips have pitch greater than or equal to 0.3 mm as per JEDEC
standard JEP95. Though pitches finer than 50 µm are possible, such methods require copper
bump or pillar formation [20], [21], [22], [23] and planarization [24]. Because of tin’s toxicity to
tissue, soldering is rarely used in medical devices unless it can be hermetically sealed off [25].
Wire bonding employs ultrasonic and/or thermal energy to fuse fine metal wire to metal
bond pads underneath, and there exist three types: thermocompression, ultrasonic, and
thermosonic [17]. In ultrasonic bonding, the bonding tool tip contacts the wire and then transmits
the ultrasonic energy, whereupon the wire fuses to the bond pad and takes on the shape of the
tool tip, which may or may not include a groove. The resulting bond resembles a wedge, so this
type of bonding is often called “wedge bonding.” In both thermocompression and thermosonic
bonding, an electric charge forms a ball on the end of the wire. Then the ball contacts the bond
pad before compression and sonication occur on top of a heated stage (> 125 °C). Both these

26
types are called “ball bonding.” Meyer et al. modified the ball bonding process—called
“MicroFlex”—to form spherical rivets between a flexible device and a rigid substrate (100-200
µm pitch) [26]. While wedge bonding can benefit from a heated work stage, it can also be
performed at room temperature with the right combination of materials. In the earliest reports of
MicroFlex, wedge bonding was explored, but has since been entirely ball-based [26], [27], [28],
[29]. The challenge in adapting wire bonding for fine pitch applications comes from human
operation and material size limits. While semiautomatic bonders can accommodate wire sizes
down to 18 µm, an incredibly steady hand and entirely static-free environment are essential, and
so such small wire sizes are relegated to automatic bonders, which are capable of 40 µm pitch [1],
[17].
Conductive epoxy is a mixture of resin and metal flakes. Because it is an isotropic electrical
conductor, bonding capability (i.e., pitch) depends on placement and size of epoxy drops, size
and distribution of metal flakes within, and viscosity of the resin. Common applicators include
epoxy dispensers and screen printers, the latter of which can be approximated manually with an
appropriate screen and “squeegee” applicator. Chang et al. used SU-8 walls atop a Parylene
structure to serve as a screen while they manually spread silver epoxy across to achieve via
interconnects (140 µm pitch) [30]. They also reported that shorts between bond pads due to
overflowing epoxy could be remedied with laser ablation.
Anisotropic conductive film (ACF) is a subset of anisotropic conductive adhesives (ACA)
where the adhesive is semi-solid, resembling double-sided tape (Figure 2-5). Metal-covered
polymer spheres (MPS) randomly populate the adhesive such that pressure in one direction
causes these MPS to deform into each other and form conductive paths. Bonding capability of
ACFs depends on pressure applied, cure temperature, size and distribution density of the MPS,

27
and the total area of the bonding surfaces. In practice, typical ACFs require that the bonded
substrates also contain recesses for excess adhesive to escape as well as elevated cure
temperature—at least 150 °C. Ledochowitsch et al. used an ACF bonder to connect a Parylene C
electrocorticography array to a PCB (200 µm pitch) [5]. The pitch capability of ACF depends
largely on the effective connection area—that is, the overlapping region between the top and
bottom bond pads. For this reason, manufacturers will typically rate ACFs not for pitch but for
area. Pursuing finer pitch for a given ACF may entail lengthening bond pads in the perpendicular
direction, but because such a strategy that may not be compatible with standard chip layouts,
newer ACFs rely on novel conductive material [31] layers for improved insulation .
Figure 2-5: In ACF bonding, the film is placed between two substrates with raised
contact surfaces; once pressure and heat are applied, MPSes are compressed by the
contacts while the adhesive flows into the channels between the contacts.
Aerosol jet printing (AJP) is an emerging technology capable of producing 10 µm resolution
traces with silver nanoparticle ink [32]. In AJP, the ink—a suspension of nanoparticles—is
atomized ultrasonically or pneumatically and then carried by a sheath gas to deposition. The
metal in the ink, usually silver, must be sintered at high temperature (> 150 °C) to form
conductive paths [33]. AJP is becoming more common in academic spaces, but it is not widely
used in industrial applications [34], due in part to the complexity of process optimization [34],
[35].

28
Printing is an attractive option for forming interconnects for its prototyping capability, use
with non-planar surfaces, and maskless process. Conductive inks are dispensed with
piezoelectric inkjet or aerosol jet nozzles to form desired patterns and then sintered to drive off
organic compounds and to fuse the metal particles together [36]. Silver is the most common
metal for its low cost compared to gold and its stability in air compared to metals that oxidize,
such as copper. To create interconnects, the electronic components can be placed side by side or
stacked on top of each other while the ink fills vias and then forms continuous conductive traces
[37], [38]. The sintering step is usually a simple application of heat with temperatures typically
above 150 °C [33], [36], but some research has been dedicated to finding alternative ways to
attain conductivity such as exposure to HCl vapors [39] or chemical reduction using ascorbic
acid [40]. Wang et al. achieved 5 µm interconnects by using a Nd:YVO4 laser to transfer silver
nanopaste to the bond pads, and then curing in an oven at 150-250 °C [41]. They used the same
technique to connect LEDs embedded in polyimide with 65 µm wide interconnects.
Spring-like connectors offer the capability of detaching and reattaching easily, but at the cost
of increased bulk (Figure 2-6). Springs often lie perpendicular to the surface to which they
connect, or in other cases, axially aligned springs can accomplish the same task [42]. Zeroinsertion force connectors (ZIF) are mechanical hinges with one or two rows of metal digits upon
which the substrate is clamped. ZIFs have been reported successfully for Parylene C devices [43],
but even the lowest pitch connectors as of this writing (currently 200 µm) cannot accommodate
ASICs. The BION microstimulator contains a mm-scale gold-coating spring designed to connect
the inner electronics and the outer stimulating electrode [44]. In an approach similar to spring
contacts, Aarts et al. were able to manufacture hanging electroplated gold contacts at 70 µm
pitch for mating out-of-plane silicon needles to a silicon base [45].

29
Figure 2-6: Technical drawings of two popular spring connectors: a) Bal Seal’s Sygnus
(Courtesy of Bal Seal Engineering); b) Molex’s ZIF
In most medical devices, Bal Seal connectors are a popular choice. These systems feature
cylindrical leads with ring-shaped contact pads, and the leads connect to the electronics by way
of ring-shaped conductive springs. This design makes the electrodes replaceable in case of
failure or revision. Examples include deep brain stimulators (DBS) [46] and cochlear implants
[47]. Bal Seal connectors, while robust, feature low connector density (> 476 mm3
/ch), thus
limiting scalability for high-density devices [42]. Another commercial spring connector is an
elastomeric connector, a strip of striated insulating and conducting layers, designed for
interconnecting two rigid substrates [48], [49]. While spacing between the conducting layers can
be as low as 50 um, the recommended pitch is 250 um or greater.
Another strategy for building interconnects between a flexible substrate and rigid ASIC is to
build on top of the chip itself. Chip integrated interconnect (CI2
) is a method invented by Rodger
et al. wherein chips are embedded in holes etched into the carrier wafer (200 µm pitch) [50].
With the face of the chip level with the surface of the carrier wafer, Parylene C and interconnect
lines are deposited directly on top. In practice, small gaps or height offsets can prevent achieving
high yield. FlexTrate is a fan-out wafer-level packaging process invented by the S. Iyer group
[51] that combines pick-and-place, capillary self-alignment, and PDMS molding. Silicon dielets

30
are placed onto a carrier with droplets of temporary adhesive securing their position, and PDMS
is poured on top and cured. This method is capable of very fine pitch (40 µm), but requires
specialized equipment.
Table 2-1: Comparison of existing bonding methods
Method Fine Pitch
Capability (µm)
Bond Temperature
(°C)
Necessary equipment
Manual soldering 300 150-400 Iron or heat gun
Solder bumps 50 150-400 Flip chip bonder
Manual wire bonding 100 23-150 Manual or semiautomatic wire
bonder
Automatic wire bonding 40 23-150 Automatic wire bonder
Conductive epoxy 140 140 Stencil printer or epoxy
dispenser
Anisotropic conductive
film
100 100-250 ACF bonder
Printing <10 150-300 Specialized ink, nozzle, and/or
laser
Spring connectors 200 Room
CLI
2 200 Metal deposition
temperature
Reactive ion etch, e-beam
evaporator
Flextrate 40 Metal deposition
temperature
Pick and place, electroplating,
evaporator
Flex2Chip 50 Room

31
Recently, a fine-pitch bonding method named Flex2Chip was reported as capable of 50 µm
pitch interconnects between a polyimide neural recording device and a commercial silicon-based
microelectrode array (CMOS-MEA, MaxWell Biosystems, Zurich, Switzerland)[52]. In this
method, springlike bond pad microstructures are etched into the 1 µm base layer, and after
aligning the device over the silicon chip in isopropanol (IPA), the evaporating IPA causes the
two devices to self-assemble and attach due to Van der Waals forces. The pixels of the CMOSMEA were attached using Flex2Chip and encapsulated with silicone elastomer, but the
remainder of the chip was wire bonded to a PCB. These methods are summarized in Table 2-1.
2.4 METHODS INVESTIGATED
Taking into consideration compatibility with Parylene C, track record, and cost, we
investigated three of the aforementioned interconnect technologies and along the way invented a
fourth: conductive epoxy, wedge bonding, ACF, and polymer ultrasonic bump (PUB) bonding.
Some modifications were necessary for use with Parylene C devices (Figure 2-7).
For conductive epoxy, we opted to forego a separate screen and instead to use the Parylene
C structure as the screen and device combined. After a drop of epoxy was applied to one side of
the wells, a 2 mm thick polyurethane shim spread the epoxy across the rest of the wells. This
maneuver was repeated until all wells were filled.

32
Figure 2-7: A summary of the four interconnect methods investigated. a) conductive
epoxy connected the metal on the device to the metal on the chip by filling the well, b)
wedge bonding both electrically and mechanically connected the device metal to the chip
metal, c) ACF was placed over the chip and then compressed to form an electrical and
mechanical connection, d) PUB bonds were gold wire bumps on the chip that were
subjected to ultrasonic energy. Reprinted from [53].
Both ball and wedge bonding were investigated for possible candidates, but early testing of
ball bonding showed that bond would not adhere to the thin-film gold on the Parylene and
instead rip the gold where the bonds were placed (Figure 2-8). Additionally, substrate
temperatures were too high for Parylene. In contrast, wedge bonding with gold wire can be
accomplished at room temperature. Given the same size of wire, ball bonding cannot produce as
fine a pitch as wedge bonding due to the size of the ball itself. We found that bonds would short
if ball bonds were placed in a row (Figure 2-9).

33
Figure 2-8: Ball bonding with gold wire on top of platinum contact pads on Parylene lifts
the metal off the polymer where bonds are attempted. Scale bar is 500 µm.
Figure 2-9: Much like Microflex bonding, ball bonding can also rivet Parylene structures
down to a substrate, but the finest pitch capable on this machine with 25 µm wire was
greater than 100 µm. Scale bar is 500 µm.
For ACF (CP13341-18AA, Dexerials), a few modifications were conceived to adapt this
method for Parylene C. In place of a dedicated ACF bonder, a pressure jig was fabricated. It
consisted of 2 aluminum plates with 4 alignment pins, secured together with 2 bolts. The top

34
plate also included a bar to apply pressure only on the ACF. Pressure was calculated as force per
area, and force was calculated from the following torque equation:
=
where T is torque, K is the coefficient of friction, D is the major bolt diameter, and F is force.
Next, because the minimum recommended cure temperature (150 °C) would cause Parylene to
oxidize, we investigated lower cure temperatures, finding that even at 50 °C with extended cure
time, comparable yield strength could be achieved (Figure 2-10).
Figure 2-10: Pull tests indicate that curing at a lower temperature and longer duration
than the manufacturer's recommendation produces comparable bond strength.
In subsequent ACF tests, the cure schedule was set to 100 °C for 75 minutes. Also, initial ACF
tests showed that disconnects due to moisture was a common failure mode, so a 30 minute
defrost step became necessary.

35
Polymer ultrasonic bump (PUB) bonding was initially conceived as a repair technique for
ACF bonding, postulating that unbonded metal-coated polymer spheres (MPS) could be forced
together using the wire bonder’s waffle tool (7045W-TI-10050-3/4-M TDF=046 W2=008
FX=01654 FR=BR=0015 (ref master # 490-00043-MA), Small Precision Tools, CA). This style
of tool is common in tape-automated bonding (TAB) where gold ribbon is fused to bond pads
(Figure 2-11). While ACF repair attempts were unsuccessful, replacing the ACF with bulk gold
wire was a next logical step. Upon visual confirmation of successful mechanical bonding,
attempts at bonding the thin-film gold on the Parylene directly to bond pads were made but were
ultimately unsuccessful. Early trials of PUB bonding had unpredictable yield until the 25 µm
gold wire was pretreated with low ultrasonic energy and force from the waffle tool—or
“bumped”—before aligning the Parylene device. In summary, the steps of PUB bonding are
bond, bump, align, and bond (Figure 2-12).
Figure 2-11: Optical micrograph of waffle tool face. Scale bar is 250 µm.

36
Figure 2-12: The major steps of PUB bonding include 1) placing a "bump" of 25 µm gold
wire, 2) aligning the polymer device, 3) switching to a waffle tool, and 4) applying
ultrasonic energy to fuse the thin-film metal to the gold bump.
Given PUB bonding’s infancy, cross-sectional imaging of the bonds was of keen interest.
However, a few challenges were presented due to the heterogeneity of the materials. Microtomes,
while useful for soft tissue and polymers, would not be able to cut glass or silicon cleanly or cut
precisely enough just the Parylene away from the chip. Cryogenic cross-sectioning by dipping
devices in liquid nitrogen and cutting with a razor resulted in deformed Parylene and thin-film
metal (Figure 2-13). Instead, a focused ion beam (FIB) scanning electron microscope (SEM)
could both cut micron-scale trenches with gallium ions and image the result (Figure 2-14). These
studies revealed that the bond pads on the Parylene C were in intimate contact with the gold
bumps (Figure 2-15).

37
Figure 2-13: Cryogenic cross-sectioning was not sufficient to produce a visible edge on
the Parylene device, and the width of the razor removed some of the bond pad
underneath.
Figure 2-14: FIB can form a clear cross-section in the area of interest.

38
Figure 2-15: FIB-SEM imaging of the PUB bond reveals the intimate contact between
the gold bump and the thin-film metal of the Parylene device. Scale bar is 1 µm.
Reprinted from [53].
On test devices, bonds made with wire and PUB were fragile, so an underfill was necessary
to support the regions surrounding the connections. By introducing a low-viscosity epoxy
(EpoTek 301) in the space between the Parylene C and chip, the epoxy would spread via
capillary attraction over the entire area. This epoxy can be cured at low temperature (65 °C for 1
hour or room temperature for 24 hours).
2.5 DEVICE ALIGNMENT
Precise alignment between bonded substrates is critical in fine pitch applications, so an
alignment jig was crafted. A custom substrate holder and device holder were fashioned out of
polyether ether ketone (PEEK), a thermoplastic whose coefficient of thermal expansion (CTE)
closely matches that of Parylene C (3.42×10-5 m/m/°C and 3.5 ×10-5 m/m/°C, respectively).
These holders were secured to translational microstages in 3 dimensions and a rotational stage
(Figure 2-16) [53]. A Parylene device could be secured to its holder by alignment pins and then

39
kept in place with a cover and screws. A chip could be secured to its holder by means of a
temporary adhesive. The easiest method was sugar-saturated water cured at 100 °C for 15
minutes.
Typical processing occurred as follows:
1. Attachment of chip to substrate holder and device to device holder
2. Pre-treatment of chip—placement of ACF or gold bumps for PUB
3. Alignment of device and chip under microscope
4. Securing the device holder and substrate holder together with screws
5. Removal of subassembly from stages
6. Application of conductive epoxy, ultrasonic energy for wedge bonding or PUB
bonding, or curing of ACF
Figure 2-16: The alignment jig holds the cable holder and chip holder separately until
they are brought together during alignment, after which the two holders are secured
together. Reprinted from [53].
2.6 TEST DEVICE DESIGN AND FABRICATION
Instead of evaluating bonding on a polymer ribbon cable and an active ASIC, surrogate cables
and chips were designed and fabricated. Both test devices featured 28 matching bond pads in
four combinations of pitch and width (µm): 100 × 70, 200 × 70, 200 × 140, 400 × 210. The finest
pitch setting mimicked the bond pad layout of the Intan RHD2164 chip while the rest were
multiples of this combination.

40
Figure 2-17: The surrogate ribbon cable design contained eight alignments holes along
the perimeter, 28 contact pads, and 28 bond pads in the center. Additionally, a) design A
bond pads had varying lengths and no insulating layer while b) design B retained the
insulated layer and had through-holes in the bond pads. Scale bars are 2 mm. Reprinted
from [53].
Additionally, since some interconnect processes require face-up bond pads while others
require face-down, two ribbon cable designs were developed: design A with open areas for ACF
and PUB (Figure 2-17a) and design B with through-holes for epoxy and wire bonding (Figure
2-17b). Design A’s bond pads varied in length to adjust overall connection area since ACF

41
bonding depends on both pitch and area. Eight holes were added to the periphery of the device
for alignment with matching pins in a custom jig.
The test chip contained bond pads matching those of the test cable (Figure 2-18), which
were shorted together into a single contact pad for continuity measurements.
Figure 2-18: The surrogate chip design contained 28 "bond pads" that were connected
together to a common contact pad. Scale bar is 2 mm. Reprinted from [53].
To test mechanical bond strength, a second polymer test device was designed (Figure 2-19),
featuring a hole for securing to the pull test setup and three rows of bond pads intended for 1)
epoxy and wire bonding, 2) ACF and PUB bonding, and 3) experimental use. The pitch and
width of these bond pads were 100 × 70 µm2
. The experimental pads revealed that a 25 µm
diameter circular face is too small of a bonding tool for PUB bonding and could puncture the
Parylene C.

42
Figure 2-19: The pull test structure featured an anchor hole for the test setup and 12 sets
of contact pads and bond pads; these bond pads contain the differentiating features of the
surrogate cable designs in addition to an experimental set of pads. Pitch was 100 µm.
Scale bar is 2 mm. Reprinted from [53].
Fabrication of polymer test cables followed previously reported process flows for Parylene
C devices [54] (Figure 2-20). First, 4" silicon carrier wafers were baked for dehydration (110 °C,
15 min, hot plate), which improves adhesion of Parylene C. A 10 µm layer of Parylene C was
deposited using a PDS 2010 (Specialty Coating Systems, Indianapolis, IN). To produce a liftoff
mask, 2 µm of image-reversal photoresist (AZ5214, Integrated Micro Materials, Argyle, TX)
was spun onto the Parylene-coated wafer, which was then prebaked (90 °C, 70 s, hot plate). The
metal mask was lithographically patterned onto the wafers using a contact aligner (Model 200,
OAI, Milpitas, CA). Then the wafer was soft baked (110 °C, 40 s, hot plate), flood exposed, and
developed (4:1 deionized water:AZ340, Integrated Micro Materials).

43
Figure 2-20: Parylene devices were fabricated as follows: a) deposition of Parylene onto
silicon carrier wafer; b) photoresist lithography, e-beam metal deposition, and metal
liftoff; c) deposition of Parylene insulating layer; d) deep reactive ion etch (DRIE) to
form through-holes; and e) DRIE to remove insulation in the bonding area for design A
and to form through-hole ledges for design B. Reprinted from [53].
Next, in an e-beam evaporator (BJD-1800, Temescal, Livermore, CA), 20 nm of titanium
and 250 nm of gold were deposited. Liftoff of unwanted metal and PR was performed in a bath
of 60 °C acetone. The wafers were then cleaned in isopropanol (IPA) and deionized water. After
an O2 plasma descum, a second layer of 5 µm Parylene C was deposited. To make an etch mask,
15 µm of photoresist was spun onto the wafers (AZ4620, Integrated Micro Materials). Next,
partial etching of the center through-holes in design B and the device outlines was accomplished
in a deep reactive ion etcher (DRIE, Plasmalab 100, Oxford Instruments, Bristol, UK) using an
oxygen-based switched chemistry recipe [55]. After cleaning the wafers in acetone, IPA, and
water, another 15 µm etch mask was formed using the same AZ4620 PR and lithography steps.
Once again, the wafers were etched in the DRIE, this time forming the bonding pad openings for
design A cables, the ledges of the through-holes for design B cables, and cable outlines and

44
alignment holes. Wafers were cleaned in acetone, IPA, and water. Individual test cables were
released by gently peeling them from the carrier wafer while immersed in water (Figure 2-21).
Fabrication of test chips began with 4” borosilicate glass wafers. As with the cable wafers, 2
µm of AZ5214 photoresist was spun onto the glass wafers. The metal mask was lithographically
patterned on the contact aligner, and Ti/Au was deposited in an e-beam evaporator. Liftoff was
performed in acetone, and wafers were cleaned in IP and water. Lastly, glass wafers were cut
into individual test chips measuring 7.5 by 4.5 mm on a dicing saw (DAD-2H/6M, Disco, Tokyo,
Japan).
Figure 2-21: A completed Parylene-gold-Parylene surrogate cable and glass-gold
surrogate chip sit next to a U.S. dime.
2.7 INTERCONNECT TESTING METHODS
All four interconnect methods demonstrated successful bonding at every combination of
pitch and width (Figure 2-22). Electrical performance was evaluated by testing for continuity and

45
resistance between the contact pads of the flexible test cable and the contact pad on the test chip
using a 2-point probe. Yield percentage was calculated at the number of successful bonds
divided by attempted bonds for each device. Both yield and resistance results were categorized
according to pitch and width.
Figure 2-22: Bonded devices using the four interconnect methods. Scale bar is 0.5 mm.
Reprinted from [53].
Reliability was evaluated using two thermal tests. The first set of bonded devices (n = 2
devices per method) was subjected to a thermal cycling test, following the Joint Electron Device
Engineering Council (JEDEC) standard JESD22-A100D. This test involved repeated thermal
cycling with 10-minute soaks at 60 °C and 0 °C, conducted 36 times in an environmental
chamber (EC0A, Sun Electronic Systems, Inc., Titusville, FL) (Figure 2-23). Resistance
measurements were taken every 10 cycles to monitor thermomechanical fatigue. The second set
of devices (n = 2 devices per method) underwent a long-term storage test, based on JEDEC’s

46
JESD22-A104D, where samples were kept at 60 °C for 1000 hours to simulate accelerated
lifetime conditions. Resistance measurements were recorded at doubling intervals of 24 hours
(e.g., 24 hours, 48, 96, etc.). The maximum temperature for both tests was 60 °C to prevent the
devices from exceeding the glass transition temperature of Parylene C. This precaution ensured
that the tests remained focused on interconnect performance rather than a Parylene material
evaluation.
Figure 2-23: The environmental chamber heated and cooled between 0 and 60 °C for the
duration of the thermal cycling test. The stepped ramping was necessary to approximate
the ramp rate recommended by the JEDEC standard.
Finally, mechanical strength was evaluated using a 90° T-peel test, adapted from MIL-STD883. Devices were pulled at a 90° angle using a motorized stage moving at 0.2 mm/s until failure
occurred. The force applied to the entire test structure was measured using a load cell and
recorded through a data acquisition system.
2.8 TESTING RESULTS
All four interconnect methods produced successful bonds at all combinations of pitch and
width (Figure 2-24). At 100 µm pitch, all methods produced comparable yield, and at coarser
pitch and larger pad width—200 µm × 140 µm and 400 µm × 210 µm), PUB bonding produced

47
100% yield on all samples. Comparison of average resistance shows that PUB bonds were
consistently lower in resistance than other interconnects (Figure 2-25). Additionally, PUB
bonding maintained the lowest average resistance over the course of thermal reliability testing as
well.
Figure 2-24: Yields for all four interconnect methods are comparable at 100 µm pitch,
and at coarser pitch PUB bonding achieves 100%. Reprinted from [53].
Figure 2-25: The average resistances of each interconnect method are comparable
except at 100 µm pitch, where PUB is clearly lower than the others. Reprinted from [53].

48
In the long-term storage test, average resistance for all methods remained steady rather than
rapidly increasing, as would be the case for failing bonds (Figure 2-26). For epoxy, ACF, and
PUB, all bonds remained connected throughout all 1000 hours, but 1 of the 46 wire bonds did
disconnect (Figure 2-27). Mean time to failure (MTTF) k, can be estimated using the Arrhenius
equation for reaction rates (Equation 1), where A is a pre-exponential factor, EA is the activation
energy of the substrate—here Parylene C, assumed to be 1.07 eV [56], kB is Boltmann’s constant,
and T is absolute temperature.
= exp (−

)
(1)
To estimate the MTTF of the same reaction at different temperature, Equation (1) can be
rearranged as follows:
2 = 1 exp (

(
1
1
−
1
2
))
(2)
With the collected data, we can estimate the room-temperature lifetime of wire bonds to be 11
months, and those of epoxy, ACF, and PUB to be over 18 years.

49
Figure 2-26: In the long-term storage test, the average resistance of the bonds remained
steady throughout all 1000 hours for all four methods. Reprinted from [53].
Figure 2-27: During long-term storage, only 1 of the 46 wire bonds failed.
In thermal cycling, all bonds again maintained their average resistance throughout all cycles
(Figure 2-28). Here, none of the bonds on all four interconnects failed by disconnecting.

50
Figure 2-28: Throughout the 36 cycles of the thermal cycling test, the average resistance
of all bonds remained stable, and no failures occurred. Reprinted from [53].
Pull testing showed that the peak force required to destroy samples was highest for epoxy,
followed by ACF, wire, and PUB, respectively (Figure 2-29). The primary failure mode for
epoxy and ACF samples was tearing of the Parylene, either at the device’s anchor point to the
test setup or at the region where adhesive was no longer contacting the Parylene (Figure 2-30a,d).
The failure mode for wire bonding also appeared to be tearing of the Parylene since the bonds
themselves stayed attached to the chip while leaving large holes in the Parylene structure (Figure
2-30b,c). Similarly, the gold bumps from PUB bonding remained on the chip (Figure 2-30e).
Upon closer inspection, portions of Ti/Au can be seen on some of the bumps, indicating that the
adhesive strength between the gold bump and thin-film metal may be comparable to the adhesive
strength between the thin-film metal and the Parylene. Given that epoxy and ACF employed
much larger surface area for adhesion whereas wire and PUB bonds were only 12 bonds, the
disparity between the adhesive-based interconnects and ultrasonic interconnects is not surprising.
For wire and PUB bonding, underfill epoxy can be employed to bolster mechanical strength to
the same level of adhesives (Figure 2-29).

51
Figure 2-29: Pull testing obviates the mechanical strength of adhesive bonds, but
ultrasonic bonds can be reinforced with underfill to produce a similar yield strength.
Reprinted from [53].
Figure 2-30: Representative failure modes show that both epoxy (a) and ACF (d) are
stronger than the Parylene itself. Likewise, the ultrasonic bonds themselves are stronger
than the Parylene, which tore at the bond pads for both wedge (b-c) and PUB bonds (e).
Reprinted from [53].
2.9 DISCUSSION & CONCLUSION
Next-generation, high-density neural probes will require advanced packaging to address
greater signal counts. Just as silicon-based devices can monolithically incorporate integrated
circuits to process these signals, polymer devices can benefit from ASIC functionality as well.

52
To that end, we selected three interconnect methods, modified them to work with Parylene C,
and tested them alongside a novel method called PUB bonding, which we developed especially
for bonding ASICs to Parylene C.
All four methods were able to bond Parylene C devices to gold-coated glass chips at four
combinations of bond pad width and pitch. At 100 µm, all methods produced similar yields (>
80%), but only PUB delivered 100% yield at coarser pitch. Finer pitch PUB bonding may be
possible with adjusted bond pad dimensions and smaller gold wire, but shrinking these would
push the limits of manual wire bonding. Optimization of PUB bond parameters, tool shape, tool
size, and wire size alongside automation could effectively drive pitch down to ≤ 50 µm.
Interestingly, PUB bonds maintained the lowest average resistance in initial testing and
throughout reliability testing. In terms of contact area, PUB bonds engage the most surface area
between device and interconnect. Epoxy and wire adjoin just a 10 µm wide square ring around a
600 µm2
square hole, and though ACF covers a wide surface area, conductive paths are formed
by interspersed 4 µm spheres. By contrast, PUB bonds are 200 µm long and 50-70 µm wide,
with extra “folds” created by the waffle pattern, effectively creating more surface area.
Furthermore, we ascertained the robustness of each bond method by subjecting bonded
devices to long-term storage, thermal cycling, and pull tests. A single wire bond disconnected
during the long-term storage test while the rest sustained steady resistance throughout. Likewise,
all four methods survived thermal cycling while maintaining average resistance. Pull testing
revealed that epoxy and ACF bonded samples were much stronger mechanically than wire or
PUB bonded ones, but underfill epoxy can mitigate this weakness and produce comparably
strong bonds.

53
Compared to other work in incorporating rigid substrates with flexible polymer devices
(Table 2-2, Table 2-3), PUB can achieve comparable or finer pitch without costly equipment or
extensive handling of the integrated circuit component while maintaining low temperature
process for the polymer. It is worth noting that nearly all the materials studied here have been
used in medical devices with the exception of ACF, whose long-term effects may render it
unsuitable for such applications [57]. The metals—gold wire in wedge or PUB bonding and
Ti/Au thin film on Parylene devices—pose no risk of leaching, and the epoxies (Epoxy
Technology, H20E and 301) can be purchased in medical grade variants.

54
Table 2-2. Comparison of Interconnect Strategies for Bonding Flexible Polymer to Rigid
Structures. Reprinted from [53].
Interconnect Strategy ZIF Solder Solder paste Epoxy Wire bond Wire bond ACF
Pitch (µm) 500 [43] 250 [58] 140 [59] 140 [60] 98 [61] 200 [62] 100 [63]
Number of connections 8 1368 8 256 4 21 4
Chip area (mm2
) N/A 10 × 10 0.27 × 0.22
0.6 × 0.3
1.05 × 0.95
1.0 × 0.5
N/A 2.5 × 2.5
Peak Temperature (° C) 230 249 * 80 * 140 140 [61] 100
Conductive material Au-plated
phosphor
bronze
Sn/Pb
solder
Cu Ag Au Au Au, Ni
Flexible substrate PxC PI PI PxC PI PxC PxC
Rigid substrate Second
ZIF connector
Silicon Surfacemounted parts
IC chip Neural
stimulator
chip
Ceramic
connector
CMOS
rectifier chip
Required equipment Flip chip
bonder
Pick
and place
Dispenser
or screen
printer
Ball
bonder
Ball
bonder
Flip chip
bonder
Abbreviations: PI = polyimide, PxC = Parylene C, PDMS = polydimethysiloxane
* Manufacturer’s recommended minimum temperature for curing

55
Table 2-3. Comparison of Interconnect Strategies for Bonding Flexible Polymer to Rigid
Structures. Reprinted from [53].
Interconnect Strategy ACF FlexTrate™ CL-I
2 Epoxy, wire,
ACF, PUB (this
work)
PUB (this
work)
PUB (this
work)
Pitch (µm) 200 [5] 40 [51] 200 [50] 100, 200, 400 500 100
Number of connections 256 12 40 28 89 32
Chip area (mm2
) N/A 1.6 × 0.8
0.62 × 0.32
3.5 × 1.85
1.26 × 0.76
0.48 × 0.48
2.5 × 2.6 7.5 × 4.5 7.0 × 9.0 1.75 × 1.2
Peak Temperature (° C) 200 23
100 (ACF)
23 (polymer)
300 (chip)
23
150 (chip)
Conductive material Au, Ni Evaporated
Ti/Au
Evaporated
Ti/Au
Ti/Au Al/Cu Al/Cu
Flexible substrate PxC PxC, PDMS PxC PxC PxC PxC
Rigid substrate Printed
circuit board
Silicon dielets ASIC Glass BGA Bare
ASIC
Required equipment ACF
bonder
Pick and place,
electroplating,
evaporator
Reactive ion etch,
e-beam
evaporator
Wedge bonder
(wire, PUB),
custom pressure
jig (ACF)
Wedge
bonder
Wedge
bonder
Abbreviations: CL-I
2 = chip-level integrated interconnect, PxC = Parylene C, PDMS = polydimethysiloxane

56
REFERENCES
[1] D. Lu and C. Wong, Materials for advanced packaging, vol. 181. Springer, 2009.
[2] J. Koch, M. Schuettler, C. Pasluosta, and T. Stieglitz, “Electrical connectors for neural
implants: design, state of the art and future challenges of an underestimated component,”
J. Neural Eng., vol. 16, no. 6, p. 061002, Oct. 2019, doi: 10.1088/1741-2552/ab36df.
[3] J. J. Jun et al., “Fully integrated silicon probes for high-density recording of neural
activity,” Nature, vol. 551, no. 7679, p. 232, 2017.
[4] J. E. Chung et al., “High-density, long-lasting, and multi-region electrophysiological
recordings using polymer electrode arrays,” Neuron, vol. 101, no. 1, pp. 21-31. e5, 2019.
[5] P. Ledochowitsch, R. J. Félus, R. R. Gibboni, A. Miyakawa, S. Bao, and M. M. Maharbiz,
“Fabrication and testing of a large area, high density, parylene MEMS µECoG array,” in
2011 IEEE 24th International Conference on Micro Electro Mechanical Systems, Jan.
2011, pp. 1031–1034. doi: 10.1109/MEMSYS.2011.5734604.
[6] C. Xie, J. Liu, T.-M. Fu, X. Dai, W. Zhou, and C. M. Lieber, “Three-dimensional
macroporous nanoelectronic networks as minimally invasive brain probes,” Nat. Mater.,
vol. 14, no. 12, pp. 1286–1292, Dec. 2015, doi: 10.1038/nmat4427.
[7] T.-M. Fu, G. Hong, T. Zhou, T. G. Schuhmann, R. D. Viveros, and C. M. Lieber, “Stable
long-term chronic brain mapping at the single-neuron level,” Nat. Methods, vol. 13, no.
10, pp. 875–882, Oct. 2016, doi: 10.1038/nmeth.3969.
[8] A. S. Caravaca et al., “A novel flexible cuff-like microelectrode for dual purpose, acute
and chronic electrical interfacing with the mouse cervical vagus nerve,” J. Neural Eng.,
vol. 14, no. 6, p. 066005, Dec. 2017, doi: 10.1088/1741-2552/aa7a42.
[9] F. Deku, Y. Cohen, A. Joshi-Imre, A. Kanneganti, T. J. Gardner, and S. F. Cogan,
“Amorphous silicon carbide ultramicroelectrode arrays for neural stimulation and
recording,” J. Neural Eng., vol. 15, no. 1, p. 016007, Feb. 2018, doi: 10.1088/1741-
2552/aa8f8b.
[10] W. F. Gillis et al., “Carbon fiber on polyimide ultra-microelectrodes,” J. Neural Eng., vol.
15, no. 1, p. 016010, Feb. 2018, doi: 10.1088/1741-2552/aa8c88.
[11] M. A. Kanchwala, G. A. McCallum, and D. M. Durand, “A Miniature Wireless Neural
Recording System for Chronic Implantation in Freely Moving Animals,” in 2018 IEEE
Biomedical Circuits and Systems Conference (BioCAS), Oct. 2018, pp. 1–4. doi:
10.1109/BIOCAS.2018.8584701.
[12] V. Woods et al., “Long-term recording reliability of liquid crystal polymer µ ECoG
arrays,” J. Neural Eng., vol. 15, no. 6, p. 066024, Dec. 2018, doi: 10.1088/1741-
2552/aae39d.

57
[13] X. Wei et al., “Nanofabricated Ultraflexible Electrode Arrays for High‐Density
Intracortical Recording,” Adv. Sci., vol. 5, no. 6, p. 1700625, Jun. 2018, doi:
10.1002/advs.201700625.
[14] X. Yang et al., “Bioinspired neuron-like electronics,” Nat. Mater., vol. 18, no. 5, pp.
510–517, May 2019, doi: 10.1038/s41563-019-0292-9.
[15] B. Fan et al., “Flexible, diamond-based microelectrodes fabricated using the diamond
growth side for neural sensing,” Microsyst. Nanoeng., vol. 6, no. 1, p. 42, Dec. 2020, doi:
10.1038/s41378-020-0155-1.
[16] J. D. Falcone et al., “A novel microwire interface for small diameter peripheral nerves in
a chronic, awake murine model,” J. Neural Eng., vol. 17, no. 4, p. 046003, Jul. 2020, doi:
10.1088/1741-2552/ab9b6d.
[17] G. Harman, Wire bonding in microelectronics. McGraw-Hill Education, 2010.
[18] R. Harrison, “private communication,” Aug. 17, 2017.
[19] G. Humpston and D. M. Jacobson, Principles of soldering. ASM international, 2004.
[20] Y. Orii et al., “Ultrafine-pitch C2 flip chip interconnections with solder-capped Cu pillar
bumps,” in 2009 59th Electronic Components and Technology Conference, May 2009, pp.
948–953. doi: 10.1109/ECTC.2009.5074127.
[21] R. Dohle, F. Schüssler, T. Friedrich, J. Goßler, T. Oppert, and J. Franke, “Adapted
assembly processes for flip-chip technology with solder bumps of 50 µm or 40 µm
diameter,” in 3rd Electronics System Integration Technology Conference ESTC, IEEE,
2010, pp. 1–8.
[22] N. Islam, M.-C. Hsieh, K. KeonTaek, and V. Pandey, “Fine pitch Cu pillar assembly
challenges for advanced flip chip package,” in Proceedings of the International WaferLevel Packaging Conference, 2017.
[23] Shinko Electric Industries Co., Ltd., “Copper Pillar Bumping,” IC Assembly. Accessed:
Oct. 29, 2021. [Online]. Available:
https://www.shinko.co.jp/english/product/package/assembly/cu-pillar.php
[24] F. Inoue, J. Derakhshandeh, M. Lofrano, and E. Beyne, “Fine-pitch bonding technology
with surface-planarized solder micro-bump/polymer hybrid for 3D integration,” Jpn. J.
Appl. Phys., vol. 60, no. 2, p. 026502, Jan. 2021, doi: 10.35848/1347-4065/abd69c.
[25] M. Schuettler, J. S. Ordonez, T. Silva Santisteban, A. Schatz, J. Wilde, and T. Stieglitz,
“Fabrication and test of a hermetic miniature implant package with 360 electrical
feedthroughs,” presented at the 2010 Annual International Conference of the IEEE
Engineering in Medicine and Biology, Aug. 2010, pp. 1585–1588. doi:
10.1109/IEMBS.2010.5626677.

58
[26] J.-U. Meyer, “Retina implant—a bioMEMS challenge,” Sens. Actuators Phys., vol. 97–
98, pp. 1–9, Apr. 2002, doi: 10.1016/S0924-4247(01)00807-X.
[27] J.-U. Meyer, T. Stieglitz, O. Scholz, W. Haberer, and H. Beutel, “High density
interconnects and flexible hybrid assemblies for active biomedical implants,” IEEE Trans.
Adv. Packag., vol. 24, no. 3, pp. 366–374, Aug. 2001, doi: 10.1109/6040.938305.
[28] J. F. Hetke, J. C. Williams, D. S. Pellinen, R. J. Vetter, and D. R. Kipke, “3-D silicon
probe array with hybrid polymer interconnect for chronic cortical recording,” in First
International IEEE EMBS Conference on Neural Engineering, 2003. Conference
Proceedings., Mar. 2003, pp. 181–184. doi: 10.1109/CNE.2003.1196787.
[29] S. Kisban et al., “Microprobe Array with Low Impedance Electrodes and Highly Flexible
Polyimide Cables for Acute Neural Recording,” in 2007 29th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society, Aug. 2007, pp.
175–178. doi: 10.1109/IEMBS.2007.4352251.
[30] J. H. Chang, R. Huang, and Y. Tai, “High density 256-channel chip integration with
flexible parylene pocket,” in 2011 16th International Solid-State Sensors, Actuators and
Microsystems Conference, Beijing, China, Jun. 2011, pp. 378–381. doi:
10.1109/TRANSDUCERS.2011.5969478.
[31] J. Tao, A. Mathewson, and K. M. Razeeb, “Study of fine pitch micro-interconnections
formed by low temperature bonded copper nanowires based anisotropic conductive film,”
in 2014 IEEE 64th Electronic Components and Technology Conference (ECTC), Orlando,
FL: IEEE, May 2014, pp. 1064–1070. doi: 10.1109/ECTC.2014.6897420.
[32] F. Cai, Y. Chang, K. Wang, W. T. Khan, S. Pavlidis, and J. Papapolymerou, “High
resolution aerosol jet printing of D- band printed transmission lines on flexible LCP
substrate,” in 2014 IEEE MTT-S International Microwave Symposium (IMS2014), Jun.
2014, pp. 1–3. doi: 10.1109/MWSYM.2014.6848597.
[33] W. Verheecke, M. Van Dyck, F. Vogeler, A. Voet, and H. Valkenaers, “Optimizing
aerosol jet printing of silver interconnects on polyimide film for embedded electronics
applications,” in Eighth International DAAAM Baltic Conference Industrial Engineering,
Tallinn, Estonia, 2012, pp. 19–21.
[34] E. B. Secor, “Principles of aerosol jet printing,” Flex. Print. Electron., vol. 3, no. 3, p.
035002, Jul. 2018, doi: 10.1088/2058-8585/aace28.
[35] A. Mahajan, C. D. Frisbie, and L. F. Francis, “Optimization of Aerosol Jet Printing for
High-Resolution, High-Aspect Ratio Silver Lines,” ACS Appl. Mater. Interfaces, vol. 5,
no. 11, pp. 4856–4864, Jun. 2013, doi: 10.1021/am400606y.
[36] J. Perelaer et al., “Printed electronics: the challenges involved in printing devices,
interconnects, and contacts based on inorganic materials,” J. Mater. Chem., vol. 20, no.
39, p. 8446, 2010, doi: 10.1039/c0jm00264j.

59
[37] T. Seifert, M. Baum, F. Roscher, M. Wiemer, and T. Gessner, “Aerosol Jet Printing of
Nano Particle Based Electrical Chip Interconnects,” Mater. Today Proc., vol. 2, no. 8, pp.
4262–4271, Jan. 2015, doi: 10.1016/j.matpr.2015.09.012.
[38] N. J. Wilkinson, M. A. A. Smith, R. W. Kay, and R. A. Harris, “A review of aerosol jet
printing—a non-traditional hybrid process for micro-manufacturing,” Int. J. Adv. Manuf.
Technol., vol. 105, no. 11, pp. 4599–4619, Dec. 2019, doi: 10.1007/s00170-019-03438-2.
[39] M. Layani, M. Gruchko, O. Milo, I. Balberg, D. Azulay, and S. Magdassi, “Transparent
Conductive Coatings by Printing Coffee Ring Arrays Obtained at Room Temperature,”
ACS Nano, vol. 3, no. 11, pp. 3537–3542, Nov. 2009, doi: 10.1021/nn901239z.
[40] S. M. Bidoki, D. M. Lewis, M. Clark, A. Vakorov, P. A. Millner, and D. McGorman,
“Ink-jet fabrication of electronic components,” J. Micromech. Microeng., vol. 17, no. 5,
pp. 967–974, Apr. 2007, doi: 10.1088/0960-1317/17/5/017.
[41] J. Wang, R. C. Y. Auyeung, H. Kim, N. A. Charipar, and A. Piqué, “Three-Dimensional
Printing of Interconnects by Laser Direct-Write of Silver Nanopastes,” Adv. Mater., vol.
22, no. 40, pp. 4462–4466, 2010, doi: 10.1002/adma.201001729.
[42] P. Rustogi and J. W. Judy, “Electrical Isolation Performance of Microgasket Technology
for Implant Packaging,” in 2020 IEEE 70th Electronic Components and Technology
Conference (ECTC), Jun. 2020, pp. 1601–1607. doi: 10.1109/ECTC32862.2020.00251.
[43] C. A. Gutierrez, C. Lee, B. Kim, and E. Meng, “Epoxy-less packaging methods for
electrical contact to parylene-based flat flexible cables,” in 2011 16th International SolidState Sensors, Actuators and Microsystems Conference, Beijing, China, Jun. 2011, pp.
2299–2302. doi: 10.1109/TRANSDUCERS.2011.5969538.
[44] G. E. Loeb, R. A. Peck, W. H. Moore, and K. Hood, “BIONTM system for distributed
neural prosthetic interfaces,” Med. Eng. Phys., vol. 23, no. 1, pp. 9–18, Jan. 2001, doi:
10.1016/S1350-4533(01)00011-X.
[45] A. A. A. Aarts, H. P. Neves, R. P. Puers, and C. V. Hoof, “An interconnect for out-ofplane assembled biomedical probe arrays,” J. Micromechanics Microengineering, vol. 18,
no. 6, p. 064004, May 2008, doi: 10.1088/0960-1317/18/6/064004.
[46] R. E. Fischell, D. R. Fischell, and A. R. Upton, “System for treatment of neurological
disorders,” 2000
[47] I. Hochmair et al., “MED-EL Cochlear Implants: State of the Art and a Glimpse Into the
Future,” Trends Amplif., vol. 10, no. 4, pp. 201–219, Dec. 2006, doi:
10.1177/1084713806296720.
[48] I. Szabó, K. Máthé, A. Tóth, I. Hernádi, and A. Czurkó, “The application of elastomeric
connector for multi-channel electrophysiological recordings,” J. Neurosci. Methods, vol.
114, no. 1, pp. 73–79, Feb. 2002, doi: 10.1016/S0165-0270(01)00515-5.

60
[49] O. Frey et al., “Biosensor microprobes with integrated microfluidic channels for bidirectional neurochemical interaction,” J. Neural Eng., vol. 8, no. 6, p. 066001, Oct. 2011,
doi: 10.1088/1741-2560/8/6/066001.
[50] D. C. Rodger, J. D. Weiland, M. S. Humayun, and Y.-C. Tai, “Scalable high lead-count
parylene package for retinal prostheses,” Sens. Actuators B Chem., vol. 117, no. 1, pp.
107–114, Sep. 2006, doi: 10.1016/j.snb.2005.11.010.
[51] G. Ezhilarasu, A. Hanna, R. Irwin, A. Alam, and S. S. Iyer, “A Flexible, Heterogeneously
Integrated Wireless Powered System for Bio-Implantable Applications using Fan-Out
Wafer-Level Packaging,” in 2018 IEEE International Electron Devices Meeting (IEDM),
San Francisco, CA, Dec. 2018, p. 29.7.1-29.7.4. doi: 10.1109/IEDM.2018.8614705.
[52] E. T. Zhao et al., “A CMOS-based highly scalable flexible neural electrode interface,”
Sci. Adv., 2023.
[53] J. J. Yoo and E. Meng, “Bonding Methods for Chip Integration with Parylene Devices,” J.
Micromechanics Microengineering, Feb. 2021, doi: 10.1088/1361-6439/abe246.
[54] B. J. Kim and E. Meng, “Micromachining of Parylene C for bioMEMS,” Polym. Adv.
Technol., vol. 27, no. 5, pp. 564–576, 2016, doi: 10.1002/pat.3729.
[55] E. Meng, P.-Y. Li, and Y.-C. Tai, “Plasma removal of Parylene C,” J. Micromechanics
Microengineering, vol. 18, no. 4, p. 045004, Feb. 2008, doi: 10.1088/0960-
1317/18/4/045004.
[56] W. Li, D. C. Rodger, E. Meng, J. D. Weiland, M. S. Humayun, and Y. Tai, “Wafer-Level
Parylene Packaging With Integrated RF Electronics for Wireless Retinal Prostheses,” J.
Microelectromechanical Syst., vol. 19, no. 4, pp. 735–742, Aug. 2010, doi:
10.1109/JMEMS.2010.2049985.
[57] C. A. Kuliasha and J. W. Judy, “In Vitro Reactive-Accelerated-Aging Assessment of
Anisotropic Conductive Adhesive and Back-End Packaging for Electronic Neural
Interfaces,” in 2019 41st Annual International Conference of the IEEE Engineering in
Medicine and Biology Society (EMBC), Jul. 2019, pp. 3766–3769. doi:
10.1109/EMBC.2019.8856692.
[58] C. Banda, R. W. Johnson, T. Zhang, Z. Hou, and H. K. Charles, “Flip Chip Assembly of
Thinned Silicon Die on Flex Substrates,” IEEE Trans. Electron. Packag. Manuf., vol. 31,
no. 1, pp. 1–8, Jan. 2008, doi: 10.1109/TEPM.2007.914217.
[59] G. Shin et al., “Flexible Near-Field Wireless Optoelectronics as Subdermal Implants for
Broad Applications in Optogenetics,” Neuron, vol. 93, no. 3, pp. 509–521, Feb. 2017, doi:
10.1016/j.neuron.2016.12.031.
[60] J. H. Chang, R. Huang, and Y. Tai, “High-density IC chip integration with parylene
pocket,” in 2011 6th IEEE International Conference on Nano/Micro Engineered and

61
Molecular Systems, Kaohsiung, Taiwan, Feb. 2011, pp. 1067–1070. doi:
10.1109/NEMS.2011.6017541.
[61] T. Stieglitz, H. Beutel, and J.-U. Meyer, “‘Microflex’—A New Assembling Technique
for Interconnects,” J. Intell. Mater. Syst. Struct., vol. 11, no. 6, pp. 417–425, Jun. 2000,
doi: 10.1106/R7BV-511B-21RJ-R2FA.
[62] M. Mueller, N. de la Oliva, J. del Valle, I. Delgado-Martínez, X. Navarro, and T. Stieglitz,
“Rapid prototyping of flexible intrafascicular electrode arrays by picosecond laser
structuring,” J. Neural Eng., vol. 14, no. 6, p. 066016, Nov. 2017, doi: 10.1088/1741-
2552/aa7eea.
[63] K. Okabe, H. P. Jeewan, S. Yamagiwa, T. Kawano, M. Ishida, and I. Akita, “Co-Design
Method and Wafer-Level Packaging Technique of Thin-Film Flexible Antenna and
Silicon CMOS Rectifier Chips for Wireless-Powered Neural Interface Systems,” Sensors,
vol. 15, no. 12, pp. 31821–31832, Dec. 2015, doi: 10.3390/s151229885.

62
CHAPTER 3
ASIC INTEGRATION OF PARYLENE DEVICES
3.1 CHIP INTEGRATION WITH FLEXIBLE POLYMERS
Efforts to integrate flexible neural interfaces with bare integrated circuitry (IC) chips are
rarely reported in detail. For instance, in Neuralink’s white paper debut, the authors mention a
custom flip-chip bonding process, but are scarce on the details and images [1], and in Science
Corporation’s preprint describing an optogenetic retinal prosthetic, there is hardly any mention
of the electronics packaging [2], though it is likely copper bump-based flip-chip bonding.
In the academic community, researchers working on integrating flexible devices with
application-specific integrated circuit (ASIC) chips have been forthright with their methods.
Recently, Park et al. combined flip-chip bonding and anisotropic conductive film (ACF) to
integrate a polyimide probe array with a 256-channel chip at 75 µm pitch with an impressive
99.7% yield [3]. They did, however, rely on wire bonding to connect their power and
communication lines. Last year, Odenthal et al. described not just their flip-chip bonding
protocol, but a methodology for assessing bonding yield with hierarchical polyimide test
structures, achieving 97.2% yield [4].
Again last year, Zhao et al. reported on using van der Waals forces to adhere thin-film metal
on an electrocorticography (ECoG) array to an off-the-shelf ASIC CMOS-MEA (MaxWell
Biosystems, Zurich, Switzerland), calling this method “Flex2Chip” [5]. Here they added
isopropanol between the chip and the flexible device, letting the solvent evaporate and letting
stiction take over. Remarkably, the chip featured 26,400 pixel sensors while the ECoG device

63
had either 720 or 2200 channels with bond pads set at 50 µm pitch; they claim no alignment was
required. Critically, the polyimide was 1 µm thin, with spring-like structures built around each
bond pad; there was no mention of how this method would fare with thicker substrates. Lastly,
the ASIC was electroplated with gold and its peripheral pads wire bonded to a PCB, indicating
that while Flex2Chip is an elegant solution, it is not yet viable for a fully integrated device.
Amongst Parylene devices, Forssell and Fedder embedded a multiplexing amplifier chip into
a probe on a glass substrate [6], and Pan et al. placed a stimulation chip in Parylene using a
combination of flip-chip bonding, wire bonding, and anisotropic conductive paste (ACP) [7].
While both of these groups used custom chips with bond pad pitch well above 100 µm, they
successfully tested their devices while fully immersed in phosphate-buffered saline. As
mentioned in Chapter 2, Chang and Rodger from Tai group both reported their respective
packaging methods [8], [9], but their methods are prone to reliability issues.
While polyimide device engineers have made massive strides in ASIC integration, their
solutions are not adaptable for Parylene due to high temperature or alignment requirements. Of
the Parylene devices listed, the methods are not capable of ≤ 100 µm pitch, so we continue to
study and apply other methods.
3.2 INTAN RHD2164 LAYOUT
Our intended target ASIC for integration is manufactured by Intan Technologies in Los
Angeles, CA. The RHD2164 is their 64-channel neural recording chip, which is comprised of
two 32-channel circuits combined into a double data rate multiplexer (Figure 3-1). Each 32-
channel circuit has 32 amplifiers for its analog inputs, an analog multiplexer, and an analog-todigital converter. Communication with the chip occurs over serial peripheral interface (SPI),

64
which are 4 pairs of data and timing lines. The chip is powered by three pairs of voltage and
ground. The remaining necessary lines are for ADC reference and electrostatic discharge (ESD)
protection. In summary, the RHD2164 can receive analog input from 64 electrodes and 2
reference electrodes on one side of the chip, and the other side contains 16 critical power and
communication lines (Figure 3-2).
Figure 3-1: Simplified diagram for the Intan RHD2164 with analog inputs and digital
communication lines highlighted. Reprinted with permission from [10].
Communication with the RHD2164 requires a field-programmable gate array (FPGA). Intan
offered their own USB evaluation board which contains such hardware. The USB evaluation
board can connect up to 4 chips via SPI communication and 12-pin Omnetics connectors. We
fashioned our own SPI cables by twisting together each pair of SPI lines (e.g., CS+ and CS-,
D
igit
al
co
m
mu
nic
ati
on
A
nal
og
inp
ut

65
CLK+ and CLK-, etc.) to prevent crosstalk to other SPI lines. One end of these wires terminated
in the 12-pin Omnetics connector (A79620-001) while the other was customized for our own
connectors.
Figure 3-2: Bond pad layout of Intan RHD2164 bare die. The top row of pads (red) are
the 64 analog inputs with 1 reference electrode on each side, spaced at a pitch of 101.6
µm. On the bottom are power, communication, and auxiliary lines. Reprinted with
permission from [10].
3.3 ADAPTING TO COMMERCIAL CHIPS
Bonding polymer devices to foundry-fabricated chips as opposed to surrogates requires
careful consideration of the materials in question. Though gold is an excellent metal for
ultrasonic bonding, it is expensive, so aluminum and copper are much more common in
consumer electronics both as wire and as bond pad. Bonding with dissimilar metals can result in
intermetallic alloys (e.g. AuAl2 or “purple plague” and Au5Al2 or “white plague”), which can
jeopardize the reliability of the bond [11]. Growth of these intermetallics can be minimized with
proper control over pad cleanliness, lower bonding temperatures, and shorter exposure to high
heat.

66
Because Intan’s bare die features aluminum bond pads, aluminum-covered silicon wafers
were used as test pads to find working parameters for gold-aluminum wedge bonds. The most
critical parameter change was increasing temperature from room temperature to upwards of
150 °C, aiding in keeping the surface clean and making the bonding materials softer. This
elevated temperature would only be required for bumping the chip, and not for the rest of the
PUB bond process.
The solder bumps of the ball-grid array (BGA) packages presented another challenge.
Without modification of the chip, ACF would be impractical because the spherical shape of the
solder bumps would merely push away and not compress any MPSes in the ACF (Figure 3-3).
For PUB, bonding gold to a solder sphere is also untenable. Thus, a desoldering process was
developed.
1. Set hotplate to 300 °C
2. Place two strips of desoldering wick on aluminum foil on workspace
3. Douse wick in flux
4. Place BGA with bumps facing down onto strips of wick
5. Move foil to hotplate and press on BGA with tweezer for 5 seconds
6. Repeat steps 2-5 with perpendicular rows of bumps
7. Swab clean with IPA
8. Repeat until all solder is removed
Once completed, this process revealed the flat underlying metal in the BGA, upon which
ACF or PUB bonding could be attempted (Figure 3-4).

67
Figure 3-3: The underside of the Intan RHD2164 BGA chip features low-temperature
solder bumps. Scale bar is 1 mm.
Figure 3-4: After our desoldering process, most of the solder is removed; any remaining
solder is thin enough that the bonder can pierce the solder. Scale bar is 1 mm.
Two other material properties for PUB bonding were investigated: Pt instead of Au
metallization and 2 µm thick Parylene instead of 10 µm. Pt-based devices could be bonded to

68
chips with no change in parameters. For thinner Parylene, a backing sheet of 10 µm Parylene
was placed between the device and bonding tool; again, no change in the bonder parameters was
required.
3.4 WORKING WITH THE INTAN RHD2164 BGA
Five non-functional BGA chips were provided by Intan Technologies. Internally, two of the
bond pads were connected, and once connected to a cable, they could be interrogated with a
continuity check to ascertain interconnect success. For bonding to these chips, a probe array was
designed (Figure 3-5). In the center were 100 bond pads measuring 300 µm in diameter and 500
µm in pitch—60 for electrode inputs, 23 outputs, and the remainder grounded together. The area
surrounding the bonds pads was etched away to make room for ACF. Above the bond pads were
60 contact pads for debugging, and at the bottom of the device were 23 contact pads designed to
fit into a zero-insertion force (ZIF) connector. Finally, at the top of the device was an array of 8
probe shanks each featuring six recording sites with 40 µm diameter openings. The alignment
jig’s chip holder was modified to accommodate the size of the BGA.

69
Figure 3-5: The BGA probe array features contact pads for debugging on top, etched
bonding area in the center, matching bond pads in the center, and contact pads on the
bottom for ZIF connection. Scale bar is 20 mm.
Initial attempts with ACF were carried out in a similar fashion to the surrogate glass chips.
After 30 minutes of warming up the ACF (CP13341-18AA, Dexerials) outside the freezer, four
strips of ACF were placed onto the Parylene device covering all the bond pads (Figure 3-6).
Once aligned and cured, the device was tested for continuity, but no electrical connections could
be ascertained [12]. This process was repeated on a second device with ACF, but again, no
connections were apparent (Figure 3-7).

70
Figure 3-6: Prior to curing, strips of ACF are placed over the bond pads of the BGA
probe array. Scale bar is 1 mm.
Figure 3-7: After curing, the ACF does not appear to connect the chip to the device. Even
the adhesive does not spread properly. Scale bar is 1 mm.
In similar fashion, a Parylene device was PUB bonded to a BGA. First, gold bumps were
bonded onto the BGA using increased temperature (Figure 3-8a). A continuity check revealed
that the bonds were successful. On a second device, bonds were successful as well.

71
Figure 3-8: For the PUB process, a) gold wire bonds can be made directly to the metal
underneath the solder and b-c) the BGA probe array can be bonded to those bumps.
Scale bars for a-b) are 1 mm. Scale bar for c) is 0.5 mm. Reprinted from [13].
Results from this initial testing suggest that ACF would be incompatible with BGA chips.
One likely cause is the levelness required to produce good ACF bonding. With a 1-dimensional
array of bond pads, levelness is difficult to achieve with a manual pressure jig, and this problem
is exacerbated with a 2-dimensional array. Standard ACF bonders are designed for a reel of tape
over a single row of bond pads, so bonding in a perpendicular direction would require a custom
solution. Given that PUB bonding on BGA was successful, we did not pursue further ACF
optimization.

72
3.5 BONDING TO A CUSTOM ASIC
Functional bare die ASIC chips were acquired from the Hashemi lab at USC [14]. These
chips were designed to output a neural stimulation waveform on a single channel. Once
connected to a computer, a custom MATLAB script dictates the shape and timing of the
waveform, which the chip will continue to output as long as it is powered. The bare die measures
1.2 by 1.4 mm and features 32 bond pads along the perimeter (Figure 3-9). Measuring 75 by 120
µm2
in size and 100 µm in pitch, these bond pads are nearly identical to those of the Intan
RHD2164 bare die. Another Parylene ribbon cable was designed to mate with this chip (Figure
3-10). It featured 4 alignment holes, 32 matching bond pads, and two sets of 16 contact pads for
mating to ZIF connectors.
Figure 3-9: Diagram of the custom neural stimulation chip

73
Figure 3-10: The custom ASIC cable also features an etched bonding area in the center
and contact pads on either side for ZIF connection. Reprinted from [13].
After PUB bonding the custom ASIC to the Parylene ribbon cable (Figure 3-11, Figure 3-12,
Figure 3-13), continuity checks on the power and ground lines ascertained whether interconnects
were successful. On the first device, all power and ground pads were indeed connected together.
Next, the ends of the cable were manually cut off, and the device was fitted between two ZIF
connectors (Figure 3-14). These connected then fanned out into individual wires. These wires
were hooked up to a custom breadboard designed for programming and debugging and then to a
computer where the custom script ran.

74
Figure 3-11: Gold wire bonds were successfully placed directly on top of the bond pads
of the custom ASIC; trailing segments of wire can be removed with tweezers or prevented
entirely with wire bonder settings. Scale bar is 250 µm.
Figure 3-12: Gold wire bonds were successfully pre-treated with the waffle tool. Scale
bar is 250 µm.

75
Figure 3-13: The custom ASIC cable was aligned to the ASIC, PUB bonded, and
underfilled successfully. Scale bar is 500 µm. Reprinted from [13].
Figure 3-14: Sixteen wires on either side of the bonded ASIC were necessary for
connection to power and programming; they were soldered directly to the flat flexible
cable. Scale bar is 25 mm.
The first bonded device was non-functional, showing unusually high current at the power
lines, possibly indicating damage from electrostatic discharge (ESD) that could have occurred

76
anywhere between fabrication, shipping, handling, PUB bonding, and ZIF packaging. On a
second bonded device, again all power and ground lines were connected, but a short was
discovered between two non-adjacent pads, rendering the chip unusable. On a third device,
shorts were discovered between two pairs of adjacent bond pads. The shorts were caused by
unnecessarily long wire left behind after the bumping process. This device was reworked by
using the edge of a 27 GA needle to separate the wire that was shorting each pair of bond pads.
Afterwards, the bonded chip was fully functional.
The first functional test on the bonded ASIC was observing the output from the stimulation
line. We connected the output line of the chip to a resistor while an oscilloscope measured the
voltage across the resistor. After the MATLAB script was programmed to send a biphasic pulse,
we verified that the same shape appeared on the oscilloscope.
Next, the resistor was replaced with microelectrodes and 1× phosphate buffered saline (PBS)
to mimic conditions for a live experiment. One stainless steel microelectrode was connected to
the output line of the bonded ASIC for stimulation, and a second was connected to the
oscilloscope for recording. The resulting output resembled the response of a series RC circuit as
expected from a recording electrode with double capacitance immersed in a conductive solution
(Figure 3-15).

77
Figure 3-15: The ASIC produced the programmed square wave successfully; likewise, the
microelectrode detected the wave through the PBS, giving us the transient response
shown here. Reprinted from [13].
The next experiment for this bonded ASIC was directing stimulation pulses in a live animal
subject. To prepare, a series of prototype microelectrode holders were fabricated on a laser cutter
(Figure 3-16). The holder was designed to maintain constant spacing between a pair of
microelectrodes as well as attach to a stereotactic stage. During surgery, the recording electrode
was connected to the suite’s recording equipment, including a digital-analog converter and an
amplifier that could express hippocampal activity as distinct popping noises [15]. After the adult
rat was anesthetized, the portion of the skull above the hippocampus was exposed, and the
electrodes were lowered until the recording electrode was in the CA1 region and characteristic
hippocampal activity could be observed (Figure 3-17). Then, a series of stimulation pulse trains
were sent through the stimulation electrode. The stimulation current was varied from 10, 50, to
100 µA while wavelength, inter-phase duration, number of pulses, and overall duration was kept
constant. At currents of 50 and 100 µA, unidentifiable electrical activity was recorded (Figure
3-18). Upon filtering for 60 Hz background noise, the source of the activity could not be
ascertained. Likely, noise from the electronics and power supply overwhelmed any neural

78
activity. At the least, the recording electrode was able to pick up stimulus artifacts from the
stimulating electrode.
Figure 3-16: A custom acrylic surgery jig built using a CO2 laser cutter holds two
microelectrodes at a prescribed distance (250 µm) and a reference wire. Scale bar is 1
cm.
Figure 3-17: Recordings of spontaneous electrical activity in the hippocampus

79
Figure 3-18: Recordings of the electrical activity in the rat brain include stimulation
artifacts every 10 seconds; upon closer inspection, these waveforms cannot be
distinguished from neural activity.
3.6 BONDING TO AN ACTIVE BGA
The next stage towards a fully integrated ASIC was to bond a BGA chip to a Parylene cable
and simulate function of the chip itself. The reasoning for this stage was two-fold: the coarse
pitch of the BGA chip would make bonding easier than the bare die and communication with the
chip could be tested and debugged. For the latter, Intan offers a USB-connected evaluation board
containing a field-programmable gate array (FPGA) capable of reading and writing to the chip’s
registers. Intan has made their software open-source for a processor or FPGA with sufficient
clock speed [16]. While Intan offers serial peripheral interface (SPI) cables designed for use with
their headstages, a custom cable was necessary to splice the individual signal lines to the PCBmounted Parylene device.
We purchased functional RHD2164 BGA chips from Intan to bond with a newly designed
Parylene cable (Figure 3-19). It featured the same bond pads as the probe array for the BGA, but
the top of the cable was 71 contact pads for a ZIF connector. Over the central portion, four port
holes were etched to make underfilling easier. Fabrication of these cables followed the same

80
steps as previously reported. Next, the Intan BGA was stripped of solder bumps and then PUB
bonded to the new cable (Figure 3-20). As specified in the datasheet for the Intan RHD2164, two
100 nF surface-mount technology (SMT) capacitors were placed between ground and VDD lines,
and a 10 nF capacitor between ground and ADC reference (Figure 3-21). The cable was then
placed in two ZIFs that were reflow soldered to a custom PCB (solder paste, hot plate, 300 °C,
approximately 20 seconds or until solder formed).
Figure 3-19: The Intan BGA cable features two sets of contact pads for ZIF connection
and etched bonding area. The four holes in the center allow for easier underfilling. Scale
bar is 2 mm.

81
Figure 3-20: The Intan BGA cable was successfully PUB bonded to the BGA chip. Scale
bar is 1 mm.
Figure 3-21: SMT components were attached to the cable using conductive epoxy.

82
Figure 3-22: The bonded BGA cable was placed into a custom PCB. Scale bar is 10 mm.
We tested the bonded BGA by connecting it to a function generator and the USB evaluation
board, which was connected to computer running Intan’s RHX software. The bonded BGA and
the USB board were placed in a Faraday cage for shielding from ambient electrical noise. On
initial startup, nearly all electrode lines were connected (Figure 3-22). The function generator
output a sine wave to a single channel while the software scanned for signals (Figure 3-23). Each
channel was checked in this manner one at a time, resulting in 59 good connections out of 64.
Meanwhile, the remaining digital lines were all connected, giving a total 84/89 connections.
Finally, the function generator was set to output a triangle wave, and the signals of the intended
recording channel and adjacent channels were recorded (Figure 3-24). The results show that the
wave could be faithfully represented, but crosstalk appeared in neighboring channels.

83
Figure 3-23: A function generator outputting a sine wave connected to one channel (A063) while others reported some crosstalk.

84
Figure 3-24: The bonded BGA cable recorded the triangle wave output by the function
generator successfully. Crosstalk appeared between the recording channel and the
adjacent channel. Reprinted from [19] © 2024 IEEE.
This crosstalk is likely the result of capacitive coupling between analog and digital lines in
two directions: laterally on the Parylene device and vertically on the PCB. For example, ref_elec
on the chip acts as a reference for the electrodes and is meant to connect to a ground line that is
usually attached to a point on the animal skull such as a bone screw (Figure 3-25). When placed
near the bottom portion of the cable where all the lines are digital, the analog reference line sees
crosstalk from the high-frequency signaling of neighboring digital lines. Similarly, on two-layer
PCBs, analog lines that run across digital lines can see crosstalk (Figure 3-26). Solutions to these
are careful design of the Parylene cable and PCB wiring.

85
Figure 3-25: The reference electrode trace (highlighted in red) on the BGA sits next to
power and digital signal lines (black traces). Scale bar is 2 mm.
Figure 3-26: The PCB for the BGA cable was two layers. Many of the top (red) and
bottom (blue) traces overlap, another avenue for crosstalk.

86
3.7 OBSTACLES ENCOUNTERED WITH ACTIVE ASIC BONDING
A pMEA was designed to bond with an Intan RHD2164 (Figure 3-27). Its 8 probes held 8
electrodes, and all 64 channels fed into the bond pads and then out to contact pads for redundant
ZIF connection, meant as a method of diagnosing failures. Like the previous Parylene C devices
for BGA, the output from the ASIC fed to ZIF contact pads.
Figure 3-27: First-generation bare die polymer MEA design with bypass traces for
independent interconnect verification on the sides and a ZIF connection on the bottom.
After PUB bonding these first-generation bare die pMEAs to the bare dies themselves, no
connection could be established with the Intan evaluation board. Instead, bubbling of the
Parylene and broken traces were observed on two of the positive voltage traces (Figure 3-28).
Dies were recovered by boiling the entire bonded device in 100 °C isopropanol and then peeling
off the Parylene C device while a vacuum chuck held the chip down.

87
Figure 3-28: Bubbling appeared on two of the positive voltage power lines. The left trace
has a disconnect.
The cause of failure was likely resistive heating due to 1) narrow trace width, 2) lack of
bypass capacitors, and 3) high current. In normal operation, the RHD2164 can consume
anywhere from 2 to 16 mA depending on sampling rate, but without bypass capacitors to shunt
any stray voltage back to ground, the 250 nm thin traces cannot handle the large current. Notably,
the broken trace on the right of Figure 3-28 is 35 µm wide while its neighbor is 45 µm wide.
These considerations necessitated a re-design of the pMEA.
3.8 BONDING TO AN ACTIVE BARE DIE
A second-generation bare die polymer multielectrode array (pMEA) was designed to address
some of the weaknesses of earlier designs and to accommodate for future testing. The new
design included bypass traces to interrogate a subset of electrodes with both digital signaling via
ASIC and analog signaling via traditional neuroscience techniques (Figure 3-29, Figure 3-30).
This design retained the alignment holes, interdigitated electrodes, and strain-relief slots from
previous designs, but underfill ports were removed because the slots could serve this function

88
(Figure 3-30). Because Parylene C can shrink up to 2.5% after thermal annealing [17], bond pads
were designed to accommodate between 101.6 µm (chip pitch) and 104.1 µm pitch (Figure 3-31).
Additionally, the trace widths on voltage and ground lines were increased from a 35 µm
minimum to an 82 µm minimum.
Figure 3-29: Polymer MEA design with black traces showing digital lines and red
showing analog lines. Scale bars are 2 mm and 500 µm (inset).

89
Figure 3-30: pMEA design, with red indicating base layer and blue indicating top layer
of Parylene
Figure 3-31: Closeup of bondpad region showing varying pitch to account for Parylene
shrinkage during fabrication. Top portions are spaced 101.6 µm apart while bottom
portions are 104.1 µm apart.
Polymer multielectrode arrays (pMEAs) were fabricated by the Polymer Implantable
Electrode (PIE) Foundry [18] . They consisted of 10 µm of Parylene C, Ti/Pt/Au/Pt, and another
layer of Parylene C (see Appendix for fabrication details). Electrodes on the pMEAs were

90
inspected using an LCR meter (Agilent E4980A). A droplet of 1× phosphate-buffered saline
(PBS) was placed on top of the probes and a platinum wire connected to the one terminal. The
other terminal contacted each bond pad or ZIF contact pad while the impedance magnitude was
reported. Electrodes with impedances ≤ 1 MΩ were considered good.
Custom PCBs were designed for dual function: electrical connection and mechanical
alignment (Figure 3-32). Holes for alignment, clearance holes for screws, and a large window
were included in the design. Alignment holes were undersized to account for the manufacturer’s
tolerance and later opened to the desired diameter using a drill press. Then dowel pins were
mated with an interference fit to the holes. PCBs included one 61-position ZIF for the Parylene
device, one 10-pin Molex Pico-Clasp connector for SPI communication, and one 36-pin
Omnetics Nano Strip connector for analog signals. Critically, this design included 4-layers to
shield signal lines using a ground plane as layer 2. Lastly, the PCB featured a series of
breakaway slots on both sides so that the clearance holes and associated area could be removed
after bonding the pMEA.
Figure 3-32: a) Image of custom PCB along with Omnetics and Molex connectors; b)
PCB with pMEA seated in ZIF with side PCB portions removed. Scale bars are 10 mm.

91
Alignment consisted of placing the pMEA on the alignment pins and securing with a cover
(Figure 3-33). Initial designs of the cover optimized small size and included a large window for
viewing the device. Fabricated from 1/8” acrylic on the CO2 laser cutter, the cover tended to
break from tapping screw holes. Additionally, at least one corner of the Parylene over the chip
tended to flex enough to break bonds. Hence, the cover was redesigned to be more robust and to
include a 1 mm bar across the window to press the Parylene down against the chip during
bonding (Figure 3-34). The cover was secured to the PCB using #2-64 screws, and this
subassembly was secured to the substrate holder using #4-40 screws with oversized clearance
holes for ease of alignment. The substrate holder included a hole to act as a pass-through for
vacuum from the bonder workstage to secure the chip. Once the chip was placed on the substrate
holder and adhered with vacuum, the pMEA was placed on top, and alignment was performed
under the wire bonder’s microscope.1,2
Figure 3-33: (left) Image of chip holder on top of wedge bonder workstage; (right) image
of the pMEA-holding PCB mated to the chip holder
1 Matching opaque thin-film metal pads can be challenging, but using the wire bonder’s tool tip gently to impress the
Parylene over the gold bumps helped to visualize alignment more clearly.
2 The pMEAs were consistently aligned against the chip bond pads in a region whose pitch was 102.1 µm in the
CAD layout, suggesting that Parylene shrinkage may have been (102.1-101.6)/102.1 = 0.5%.

92
Figure 3-34: Image of cover for PCB with a bar across the chip area to keep Parylene
flat against the chip. Scale bar is 2 mm.
During alignment, the pMEA was connected via Molex Pico-Clasp to 12-pin Omnetics
A7620-001, which interfaces with the Intan USB evaluation board. The workstage was set to
60 °C for bonding. After PUB bonding all bond pads, the USB board was turned on, and the
RHX software opened to ascertain if all critical lines are connected (i.e., SPI and power). On the
RHX screen, 64 channels were visible, indicating chip functionality. At this point, underfill
epoxy was placed over the bond pads, where it seeped into the strain-relief slits and flowed
between the pMEA and chip via capillary action. The epoxy was cured by leaving it on the
workstage set to 60 °C for 1.5 hours, resulting in a complete ASIC-integrated pMEA (Figure
3-35, Figure 3-36).

93
Figure 3-35: Image of ASIC-integrated pMEA lying next to a US penny. Reprinted from
[19] © 2024 IEEE.
Figure 3-36: Micrograph of PUB bonds—66 along the top, 23 along the bottom—with
inset showing a closeup of the bonds. Here, the bonds appear circular because a ball
bonder was used to form the initial gold bumps. Scale bars are 1 mm and 50 µm (inset).
Reprinted from [19] © 2024 IEEE.
Initial impedance tests were performed using the RHX software’s on-board electrode
impedance test. A droplet of 1× PBS was placed on top of the electrodes, and the test was

94
performed at 1 kHz. Results indicated that 63 of 64 electrodes and both reference electrodes were
functioning.
Two additional ASICs were PUB bonded to pMEAs, and on both of these, the on-board
impedance test showed that 64 of 64 electrodes were connected. Along with one non-functional
ASIC-integrated pMEA, these four chip-integrated pMEAs were stacked onto a PCB to assess
the feasibility of three-dimensional pMEAs. Five Mylar spacers of 0.1 mm thickness, cut using a
femtosecond laser (Optec, Multihead WS), were placed between each array (Figure 3-37).
Stacking all components required approximately 3 minutes. To measure the lateral misalignment,
images were taken with an industrial microscope focused on the top device and on the bottom
device (Figure 3-38). Images were then overlaid, and distance was measured on imaging
software. The largest lateral deviation was determined to be 23 µm.
Figure 3-37: Images of four stacked ASIC-integrated pMEAs with Mylar spacers

95
Figure 3-38: Overlay of image focused on the top- and bottom-most of 4 stacked pMEAs;
lateral deviation appears on probe 1 on the left. Scale bar is 0.5 mm.
3.9 IN VITRO COMPARISON OF ASIC-INTEGRATED AND PASSIVE PMEA
The probe tips of integrated pMEA #1 were submerged in 1×PBS along with one Pt wire
and one stainless steel wire (ground). The Pt wire was connected to a multifunctional DAQ
system (National Instruments USB-6366) that generated a 2 mVpp 1 Hz sine wave. The digital
output of the pMEA was connected to the Intan USB board, and the analog output was connected
to the BGA-integrated cable acting as a headstage (Figure 3-39), which was then connected to
the USB board. At first, only the digital signals were recorded on the Intan RHX software
(Figure 3-40). Then the analog signals were recorded via the BGA-integrated cable. Recordings

96
showed that both the digital and analog outputs matched the frequency of the generated sine
wave, though the signals through BGA were noisy (Figure 3-41).
Figure 3-39: Setup schematic for initial benchtop testing using a BGA-integrated cable
as a “headstage”
Figure 3-40: Initial digital recordings from ASIC-integrated pMEA with probe tips in
PBS with a 1 Hz sine wave input; average of 11 periods.

97
Figure 3-41: Recordings from ASIC-integrated pMEA and from a wire connected to a
BGA-integrated cable immersed in PBS with a 1 Hz sine wave input.
Next, the analog output integrated pMEA #1 was connected to a Plexon Omniplex running
NeuroExplorer software (Figure 3-42). The probe tips were submerged in 1×PBS with Pt wire
and stainless steel wire. A function generator (Keysight 33500B) was connected to the wires
while outputting a 5 mVpp 1 kHz square wave. In order to reduce crosstalk, digital signals were
recorded with the Omnetics connector removed, and analog signals were recorded with the
Molex connector removed.

98
Figure 3-42: Setup schematic of in vitro test where digital signals were recorded via
Intan USB board and analog signals were recorded via Plexon products. Reprinted from
[19] © 2024 IEEE.
Both digital and analog signals were successfully recorded on their respective configurations.
All recorded signals resembled a square wave modified by a resistive-capacitive time constant as
typical for recording in solution (Figure 3-43). While individual signals varied in DC offset or
amplitude, the mean signals for both digital and analog overlap. Given that the Plexon headstage
incorporates the same Intan RHD2164, this result is not surprising, but it does confirm that ASIC
integration gives comparable signal quality while greatly reducing packaging burden. While each
analog signal required a 1:1 connection, the ASIC-integrated pMEA required 10 wires to the
PCB: 8 SPI lines, 1 voltage, and 1 ground.

99
Figure 3-43: Recordings from the digital and analog sides of the ASIC-integrated pMEA,
indicating that performance is comparable.
3.10 CONCLUSION
We have successfully demonstrated PUB bonding for integration of ASICs with Parylene
ribbon cables and MEAs with bare die and packaged chips. Using the lessons learned from
bonding to custom bare die and BGA chips, device and PCB designs were improved, and
intermittent mid-bond testing was developed. In total, three bare die Intan RHD2164 chips were
successfully integrated with pMEAs. Testing on the first of these showed that 63 of 64
connections were active, and in vitro recordings showed that an ASIC-integrated pMEA
performs comparably to a passive pMEA on a traditional recording setup. Thus, PUB bonding
presents a path to fabricating high-density, 3D pMEAs for large-scale neural recording. On these

100
devices, the backside silicon of the bare die is exposed, but with proper encapsulation, they could
be ready for in vivo studies. Once the ASIC-integrated pMEA is mated to the PCB, the PCB is
robust enough for handling and for attaching to stereotaxic equipment. For large-scale recording
with a 3D stack of ASIC-integrated pMEAs, flex PCBs and proper fixturing of cables would be
necessary for simultaneous recording.
REFERENCES
[1] E. Musk and Neuralink, “An Integrated Brain-Machine Interface Platform With
Thousands of Channels,” J. Med. Internet Res., vol. 21, no. 10, p. e16194, Oct. 2019, doi:
10.2196/16194.
[2] E. B. Knudsen et al., “A thin-film optogenetic visual prosthesis.” Neuroscience, Feb. 03,
2023. doi: 10.1101/2023.01.31.526482.
[3] S.-Y. Park et al., “A Miniaturized 256-Channel Neural Recording Interface With AreaEfficient Hybrid Integration of Flexible Probes and CMOS Integrated Circuits,” IEEE
Trans. Biomed. Eng., vol. 69, no. 1, pp. 334–346, Jan. 2022, doi:
10.1109/TBME.2021.3093542.
[4] M. C. Odenthal, V. Claar, O. Paul, and P. Ruther, “Hierarchical Bonding Yield Test
Structure for Flexible High Channel-Count Neural Probes Interfacing ASIC Chips,” in
2023 IEEE 36th International Conference on Micro Electro Mechanical Systems
(MEMS), Jan. 2023, pp. 409–412. doi: 10.1109/MEMS49605.2023.10052500.
[5] E. T. Zhao et al., “A CMOS-based highly scalable flexible neural electrode interface,”
Sci. Adv., 2023.
[6] M. Forssell and G. K. Fedder, “Parylene neural probe with embedded CMOS
multiplexing amplifier,” in 2018 40th Annual International Conference of the IEEE
Engineering in Medicine and Biology Society (EMBC), Jul. 2018, pp. 3374–3377. doi:
10.1109/EMBC.2018.8512938.
[7] K. Pan et al., “A flexible retinal device with CMOS smart electrodes fabricated on
parylene C thin-film and bioceramic substrate,” Jpn. J. Appl. Phys., vol. 62, no. SC, p.
SC1022, Jan. 2023, doi: 10.35848/1347-4065/acaca5.
[8] D. C. Rodger, J. D. Weiland, M. S. Humayun, and Y.-C. Tai, “Scalable high lead-count
parylene package for retinal prostheses,” Sens. Actuators B Chem., vol. 117, no. 1, pp.
107–114, Sep. 2006, doi: 10.1016/j.snb.2005.11.010.

101
[9] J. H. Chang, R. Huang, and Y. Tai, “High density 256-channel chip integration with
flexible parylene pocket,” in 2011 16th International Solid-State Sensors, Actuators and
Microsystems Conference, Beijing, China, Jun. 2011, pp. 378–381. doi:
10.1109/TRANSDUCERS.2011.5969478.
[10] Intan Technologies, LLC, “RHD2164 Digital Electrophysiology Interface Chip.”
RHD2164 datasheet, Dec. 01, 2017. Accessed: Nov. 18, 2020. [Online]. Available:
https://intantech.com/files/Intan_RHD2164_datasheet.pdf
[11] D. Lu and C. Wong, Materials for advanced packaging, vol. 181. Springer, 2009.
[12] R. Harrison, “private communication,” Aug. 07, 2018.
[13] J. J. Yoo and E. Meng, “Bonding Methods for Chip Integration with Parylene Devices,” J.
Micromechanics Microengineering, Feb. 2021, doi: 10.1088/1361-6439/abe246.
[14] A. Samiei and H. Hashemi, “Energy efficient neural stimulator with dynamic supply
modulation,” Electron. Lett., vol. 57, no. 4, pp. 173–174, 2021, doi: 10.1049/ell2.12024.
[15] H. Xu, A. W. Hirschberg, K. Scholten, T. W. Berger, D. Song, and E. Meng, “Acute in
vivo testing of a conformal polymer microelectrode array for multi-region hippocampal
recordings,” J. Neural Eng., vol. 15, no. 1, p. 016017, 2018.
[16] “Intan-RHX.” Intan Technologies, Jun. 25, 2021. Accessed: Aug. 04, 2021. [Online].
Available: https://github.com/Intan-Technologies/Intan-RHX
[17] B. J. Kim and E. Meng, “Micromachining of Parylene C for bioMEMS,” Polym. Adv.
Technol., vol. 27, no. 5, pp. 564–576, 2016, doi: 10.1002/pat.3729.
[18] “PIE Foundry.” Accessed: Sep. 09, 2023. [Online]. Available: https://piefoundry.usc.edu/
[19] J. J. Yoo and E. Meng, “ASIC Integration via Polymer Ultrasonic Bump Bonding to A
64-Channel Penetrating Parylene Multielectrode Array,” in 2024 IEEE 37th International
Conference on Micro Electro Mechanical Systems (MEMS), Austin, TX, USA: IEEE, Jan.
2024, pp. 392–395. doi: 10.1109/MEMS58180.2024.10439599.

102
CHAPTER 4
MICRONEEDLES FOR CONTINUOUS HEALTH MONITORING
4.1 THE MARKET FOR CONTINUOUS HEALTH MONITORING
Within consumer electronics, there is a growing demand for nonstop health monitoring
through the adoption of wearable technology. Globally, the wearables market is valued at $63.2
billion and is expected to grow 14.8% annually [1]. Smartwatches and smart bands offer not only
constant connection to one’s smartphone, but also steady surveillance of health data. In these
devices, a handful of sensors record biological information such as heart rate, oxygen saturation,
temperature, daily steps, and sleep quality. With access to longitudinal data, a signal like heart
rate can be processed and analyzed to detect irregular health conditions like atrial fibrillation [2].
In the regulated medical device space, however, only continuous glucose monitoring (CGM)
is capable of such long-term data collection. In 2021, an estimated 536.6 million adults had
diabetes, and this number is expected to increase to 783.2 million by 2045 [3]. We can only
expect CGM to become more important in treating diabetes [4].
The growth of the wearables market and the success of CGMs point to a need for portable
biosensors. The challenge, however, is finding the right combination of sensing technology and
portable electronics. Methods for detecting biological analytes including drugs and proteins
abound in the literature, but only electrochemical aptamer-based (EAB) sensing can achieve
short time resolution—in vivo—on a target-customizable platform [5], [6].

103
4.2 BIOLOGICAL SENSING METHODS
A common approach to biological molecular sensing, and the one employed by CGMs, is
enzymatic sensing. In these sensors, an enzyme capable of catalyzing a specific target is
immobilized on the sensing surface. Once this enzymatic reaction takes place, a mediator species
accepts the electrons and carries them to the electrode surface. By maintaining a potential across
this mediator, any change in current can be interpreted as an enzymatic reaction and therefore a
change in target concentration [7]. In the case of CGMs, glucose oxidase (GOx) is the enzyme of
choice, and the electron acceptor is usually oxygen or an artificial compound such as ferrocene
or organic salts [8]. This strategy has been successful for a number of targets with associated
enzymes: glutamate [9], [10], lactate [11], [12], acetylcholine [13], ethanol [14], [15], and many
others [16], [17], [18], [19], [20], [21]. When faced with in vivo conditions, these sensors must
contend with interference from biological molecules whether through adsorption or enzyme
inactivation and loss of enzyme activity due to protein degradation [7]. While researchers are
actively working to solve these and related issues, the fundamental limitation of enzyme-based
sensing is the limited availability of enzymes themselves. Enzymatic sensors can only detect
species for which catalytic enzymes exist and thus are not generalizable to other targets.

104
Figure 4-1: While many biosensing methods are generalizable, continuous, or specific,
only EAB sensing characterizes all three. FSCV image used with permission of Royal
Society of Chemistry, from [22]; permission conveyed through Copyright Clearance
Center, Inc. CGM image reproduced with permission from [23]. All other images public
domain.
Another approach to molecular sensing is targeting electroactive compounds directly. For
instance, some neurotransmitters can be detected by electrochemically oxidizing them [24], so
their concentrations can be measured using amperometry and fast-scan cyclic voltammetry [25],
[26], [27]. Amongst neurotransmitters, catecholamines (dopamine, norepinephrine, epinephrine)
and tryptophan-derived compounds (serotonin) are prime targets because their oxidation
potentials lie within the safe potential limits of common electrodes. Other compounds that can be
detected in this fashion include hydrogen peroxide [28], oxygen [29], [30], and ascorbate [10].
Direct electrochemical interrogation, however, is non-specific and relies on electrode surface
area. Thus, each of these molecules can interfere with the measurement of another, confounding

105
in vivo studies. Finally, this method only works for electroactive species, making it, too, not
generalizable to other targets.
Optical sensing methods offer excellent spatial resolution for detecting analytes. There
are a variety of optical biosensors, and most of these have been characterized only in the
laboratory or with in vitro studies. Optical nanosensors, on the other hand, have been successful
in detecting sodium [31], histamine [32], and hydrogen peroxide [33] in vivo. However, these
nanoscale sensors must be injected into subjects and imaged using highly sensitive and bulky
light detection systems. Targets for optical sensing are limited by available dyes or nanoparticles
that can react with sufficient luminescence to be captured behind tissue.
Piezoelectric sensors are a class of devices that measure the resonant frequency of
mechanical structures to assess changes in mass. In order to detect molecules, a specialized
coating must be applied, such an antibody, DNA, or polymers [34]. They have been used to
measure antigens and viruses [35]. However, these types of sensors cannot operate in vivo due to
the dynamic environment.
Lastly, benchtop laboratory methods such as chromatography, mass spectrometry, or
enzyme linked immunosorbent assays (ELISA) are gold standards for molecular analysis, but so
long as they require reagents, these processes cannot be truly “continuous.”
4.3 ELECTROCHEMICAL APTAMER-BASED SENSING
In contrast to established biosensing methods, EAB sensing is a broadly applicable
sensing modality that offers real-time measurement and has been demonstrated in vivo. Its basic
principle is the transduction of aptamer folding into electrical signals (Figure 2-1). So long as the
transducer is gold, which can be deposited in myriad ways, EAB sensing can be implemented.

106
Aptamers are single strands of DNA that “fold” upon binding with its target. EAB
sensing takes advantage of this conformational change by transducing it into a signal. The first
EAB sensor consisted of a thrombin-binding aptamer, a methylene blue label, and a gold
electrode [36]—inspired by an earlier reported electronic DNA sensor [37]. At the core, the basic
fabrication of these sensors has remained the same. Construction of EAB sensors begins with
aptamer selection using a process known as Systematic evolution of ligands by exponential
enrichment (SELEX) [38], [39]. Next the aptamer is covalently bonded to a thiol on one end and
a redox reporter, such as methylene blue, on the other. The aptamer strands then form a selfassembled monolayer (SAM) atop a gold surface, creating thiol-gold covalent bonds. Another
SAM compound such as 6-mercaptohexanol may be introduced at this stage to pack the aptamer
strands closer together. In this configuration, the electron transfer rate between the redox reporter
and the surface of the electrode can be measured. When the aptamer is in the presence of its
target, the binding-induced conformational change moves the redox reporter closer to the gold
surface, thereby inducing a change in the electron transfer rate, which again can be measured.
Thus, the concentration of a target analyte can be assessed in real-time with electrochemical
methods such as electrode impedance spectroscopy [40], alternating current voltammetry [41],
chronoamperometry [42], intermittent pulse amperometry [43], and square wave voltammetry
(SWV) [44].

107
Figure 4-2: Standard EAB sensors are gold wires coated in aptamers that undergo
conformational change in the presence of the target molecule.
In theory, EAB sensing can be accomplished with any molecule since aptamers can be
engineered to bind to specific targets. As such, EAB sensors have been fabricated and reported
for a wide variety of substances, ranging from antibiotics such as tobramycin and vancomycin
[5], [6], [42], [46], [47], [48], [49] to chemotherapeutics [6], [46], [50] to amino acids [51], [52],
[53]. Investigators have pursued detection of ATP [47], adenosine [54], insulin [55], and even
cocaine in the brain [56]. Furthermore, due to this bespoke engineering of aptamers, nonspecific
binding does not disturb the electrochemical signal except in cases where structurally similar
molecules exist in concentrations above the limit of detection [6]. This should not be a problem
for detecting exogenous targets.
Because EAB measurement only depends on the binding-induced structural shift of the
aptamer, adsorption of non-target species does not affect the signal. This translates to resilience
in vivo. Where many sensing modes may fail due to the presence of proteins and cells, EAB
sensors can successfully measure analytes. In fact, many of the aforementioned investigations
have been carried out in whole blood, in the jugular vein of adult rats [5], [6], [42], [46], [48],
[49], [50], [52]. Still, EAB sensors must contend with drift over time, but kinetic differential
measurement (KDM) can correct such variation [57], [58].

108
While a number of electrochemical methods can be deployed to measure an EAB sensor’s
response to its target analyte, our lab has employed square-wave voltammetry (SWV) to take
advantage of KDM. In SWV, the applied potential resembles a square wave imposed on a
staircase function (Figure 4-3). At a given potential E0, an additional pulse is applied, and the
forward current If is measured. Then, the polarity of the applied potential switches, and the
reverse current Ir is measured. This is repeated at a higher potential E0 + ΔE. While all these are
important parameters, it is frequency that has the most effect on EAB sensing response. At
higher frequencies, peak current response increases with increasing target concentration—“signal
on”—whereas at lower frequencies, the peak current response decreases with increasing target
concentration—“signal off” (Figure 4-4). By tracking two such frequencies in concert, the
differential measurement can account for signal drift.
Figure 4-3: In SWV, current is measured at the peak and trough of applied potential E0,
then at E0 + ΔE, etc. Results are graphed as potential vs. net current (difference between
forward and reverse current).

109
Figure 4-4: A) Square wave voltammetry can detect the changes in electron transfer rate
due to the conformational change of the aptamer. B) At some frequencies, target binding
induces a “signal-on” response and C) at other frequencies, target binding induces a
“signal-off” response. Used with permission of the Royal Society of Chemistry, from [58];
permission conveyed through Copyright Clearance Center, Inc.
4.4 CHALLENGES IN ADDRESSING INTERSTITIAL FLUID
Existing in vivo demonstrations of EAB sensing have been with whole blood, but a less
invasive method should be the next goal. Interstitial fluid (ISF) is a rich biofluid that makes up
70% the dermis [59], but it has been historically difficult to access. ISF contains not only
identical analytes as blood plasma in nearly the same concentrations [60], [61], [62], [63], but
other biomarkers not found in plasma as well [64]. Studies suggest that ISF is a promising
alternative to blood for monitoring of drugs such as vancomycin [65], [66].
Current methods of interacting with ISF include biopsy [67], iontophoresis [68], [69],
sonophoresis (ultrasonic extraction) [70], suction blisters [71], and laser ablation [72], but these
methods require bulky hospital or laboratory equipment that would make continuous monitoring

110
impossible. Two newer, portable methods are microneedles and subcutaneous sensors. Using
microneedles to extract ISF avoids the pain of implantation, but even under the best conditions,
tens of minutes and vacuum pressure are required [73]. Subcutaneous sensors have been a
commercially viable method of measuring ISF and have been deployed in CGMs such as
Medtronic’s MiniMed [74].
Instead, we propose measurement of ISF in situ rather than extraction using EAB
microneedles. A sufficiently long microneedle coated in functionalized aptamers can interact
with the ISF without the painstakingly long fluid withdrawal step. Additionally, the small form
factor of microneedles ensures a less painful insertion as compared to existing subcutaneous
sensors [75]. In the future, this approach could translate to the development of wearable
microneedles (Figure 4-5).
Figure 4-5: A wearable platform for measuring ISF would be possible using EAB sensors.
One last hurdle before realizing a portable EAB microneedle system is the measurement
electronics. All EAB demonstrations to date, including microneedles, have been performed with
a laboratory potentiostat [76], [77], [78], [79]. Efforts to miniaturize potentiostats have been
reported since 1987, when researchers scaled down to the chip scale [80], [81]. In subsequent
reports, miniaturization targeted the printed circuit board (PCB) and its components [82], [83],

111
[84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96] rather than the integrated
circuitry. Motivations for these projects included reduced cost for broader access to
electrochemistry education and reduced size for portability and field work. In most of these
works, the authors produced their own custom transimpedance amplifier circuit, the heart of a
potentiostat, but with the availability of commercial analog front end (AFE) chips, simpler
designs were possible [94], [96]. Rarely did these reports include integration with a biosensor.
One notable exception is the case of Tehrani et al. who combined their miniature potentiostat
with lactate- and glucose-sensing microneedles [97]. As of this writing, no one has combined a
portable potentiostat with EAB sensors.
4.5 MICRONEEDLE DESIGN
For preliminary studies, we chose a basic planar microneedle design. This way, the
fabrication could be simplified to a single lithographic step and two etch steps. The length of the
microneedles was critical because they need to reach into the dermis of rats and possibly humans.
Luckily, the epidermis of rats is 32 µm while that of humans is 47 µm [98], so at a minimum,
microneedles should be at least 50 µm long. Since longer microneedles can elicit more pain than
shorter ones [99], the length should not exceed 1000 µm.
The surface area of the gold wire used in previous experiments is much greater than that of a
single microneedle, so multiple microneedles will be necessary to approximate the same
coverage. To that end, the wafer design included both solitary and arrayed needles on a single
handle (Figure 4-6). These handles can be stacked to create a 3D grid of needles ranging in
length, spacing, and number of needles. Each piece also has a square hole so they can be aligned

112
and then locked together with a pin. Acrylic pieces cut using a laser cutter can act as a holder
while epoxy secures all the pieces (Figure 4-7).
Figure 4-6: The design of the wafer has microneedle arrays in eight combinations of
length, spacing, and number as well as rectangular spacers.
Figure 4-7: The microneedle 3D array concept consists of stacked 2D arrays and spacers
bound together with a pin.
microneedle
array
spacer
pin
casing

113
4.6 MICRONEEDLE FABRICATION
The microneedle fabrication plan consisted of conventional silicon processing steps: spin
coating photoresist atop oxide-coated wafers, photolithography, hydrofluoric acid (HF) etching
of oxide, deep reactive ion etching (Bosch process) of Si, pressure-enhance chemical vapor
deposition (PECVD) of SiO2 for insulation, and finally e-beam evaporation of titanium and gold.
Before starting fabrication in earnest, we set out to optimize the photolithography.
Preliminary work with HF etching consisted of patterning AZ4620 atop oxide-coated wafers
(1000 nm), scribing and splitting the wafer into quarters, and using an LDPE dropper to place HF
over exposed areas of the wafer shard. Because SiO2 is hydrophilic while Si is hydrophobic, the
HF visibly “runs away” from the exposed silicon when etching is complete. Table 4-1 captures
the estimated oxide etch rates using 48% HF, which are similar to previously reported etch rates
[100].
Table 4-1: Etch rates for oxide using 48% HF
Parts DI water to 1 part 48% HF Etch rate (nm/min)
0 750
1 300
2 100
A significant challenge in lithography of oxide is the adhesion between the sacrificial layer
and oxide. If the two layers do not maintain contact throughout the etching process, HF can make
its way between them and cause the lateral etch rate of the oxide to become unacceptably large.
To improve adhesion, we added an HMDS vapor deposition step between dehydration and spin

114
coating. While this step made a marked improvement to etch profiles, one more change—
substituting buffered HF (BHF) in place of HF—resulted in a negligible lateral etch rate.
The remainder of the fabrication was straightforward. The wafer was dipped in HF to
remove native oxide formed during storage. Then it was placed in the DRIE using a standard
Bosch process to remove approximately 200 µm of Si. At this point, it was attached to a carrier
wafer using drops of AZ5214 as an adhesive. Then the wafer was further etched until 200 µm
were removed. The wafer was dislodged from the carrier by soaking in acetone. A quarter of the
wafer was coated in 250-500 nm of oxide in the PECVD and then 10 nm Ti for adhesion and 200
nm Au (Figure 4-8).
Figure 4-8: Image of oxide-coated (left) and Au-coated (right) Si microneedle. Scale bar
is 0.5 mm.
4.7 MICRONEEDLE PERFORMANCE
Gold-coated silicon microneedles were functionalized with vancomycin-detecting aptamer
(see Appendices) alongside gold wire and gold-coated silicon shards. Silicon samples were
covered with Kapton tape that had an opening with the same size as the gold wire surface area.
Sensing was performed using square wave voltammetry (SWV) on all samples in 1× PBS at

115
varying frequencies and concentrations of vancomycin against a standard Ag/AgCl reference and
Pt wire counter on a benchtop potentiostat (Gamry).
All three samples measured increasing vancomycin concentration at 100 Hz. Unfortunately,
during testing, the silicon array suffered damage to one of the microneedles, thereby exposing
the silicon as well as reducing the overall sensing area. Still, the damaged array was able to sense
vancomycin on a scale similar to that of gold wire (Figure 4-9).
Figure 4-9: SWV at 100 Hz reveals that the Si array performs comparably to Au for
detecting vancomycin.
The silicon shard, on the other hand, proved to be a capable substrate for EAB sensing. Its
peak current for detecting vancomycin far exceeds that of the gold wire. However, rather than
proving silicon’s extraordinary performance, this may be an indication of the Kapton tape’s
inability to seal off the surface of the shard, thereby producing a larger sensing area. Still, when
considering the silicon shard’s performance on its own, SWV at 100 Hz shows a signal-on
response, with peak current rising as vancomycin concentration rises (Figure 4-10). Furthermore,

116
the signal gain between 10 and 100 Hz is wide and correlates with vancomycin concentration
(Figure 4-11), indicating that silicon is a good candidate for EAB sensing.
Figure 4-10: SWV at 100 Hz shows that the Si shard can detect varying concentrations of
vancomycin
Figure 4-11: The signal gain response from the Si shard shows that 100 Hz is an
excellent signal-on frequency for detecting vancomycin.

117
4.8 FIRST POTENTIOSTAT PROTOTYPE DESIGN
During early attempts to modify an existing electrochemistry electronics board, we
encountered operating system incompatibility, Bluetooth dysfunction, and incorrect aptamer
sequencing. Though we were able to test two gold wires with this board, the results were not
consistent with data from the benchtop potentiostat, indicating that the schematic needed
modification. However, the Bluetooth module ceased to function, and the developers of the
board could not be reached. We decided to build a miniature potentiostat based on other work
where SWV was accomplished [94], [101], [102], [103]. These strategies usually entailed the
combination of an analog front end (AFE) capable of electrochemistry and a microcontroller unit
(MCU), sometimes paired with Bluetooth Low-Energy (BLE) wireless communication (Figure
4-12). We chose the AD5940/59413
from Analog Devices because it is capable of SWV and
because the other commonly used AFE, Texas Instruments LMP91000, was not available due to
supply chain issues.
Figure 4-12: Block diagram for proposed potentiostat.
3 The AD5940 has 8 general-purpose input/output (GPIO) pins while the AD5941 has 3. The remaining architecture
is identical.

118
In the interest of producing as small a board as possible, we opted for multifunctional chips
that could combine some of the major functions, namely processing, sensing, and
communication. Thus, two options presented themselves: MCU and BLE in one chip
(STM32WB55 series) or MCU and AFE in one chip (ADuCM355)(Figure 4-13).

Figure 4-13: Two paths for developing a potentiostat were combining MCU and BLE
(left) or MCU and AFE (right).
The ADuCM355 is a system-on-chip (SOC) that integrates an ARM Cortex-M3 processor with
the AD5940 (Figure 4-14 and Figure 4-15).
Figure 4-14: Block diagram of the AD5940

119
Figure 4-15: Block diagram of the ADuCM355. The left portion is identical to that of the
AD5940.
Conveniently, Analog Devices, Inc. (ADI) manufactures both the AD5940 and the ADuCM355
and offers example code, including SWV, on GitHub [104]. We adapted this code to program an
STM32WB55 evaluation board to communicate with an AD5940 evaluation board using serialperipheral interface (SPI). This code randomly generated data, wrote it to the AFE, and then read
back the data through the MCU over 10000 cycles. Once SPI communication was successfully
established on evaluation boards, we replicated this setup using breadboard components and
bare-metal chips that were reflow soldered onto breakout boards (Figure 4-16). Here, too, the
code ran successfully, indicating successful SPI connection.

120
Figure 4-16: Breadboard setup connecting as STM32WB55 to an AD5941; insets show
both of the QFN-48 (quad flat no-lead) packaged chips, measuring 7 mm by 7 mm.
When attempting to adapt the SWV code provided by ADI, we were unable to obtain data
resembling the output from a benchtop potentiostat (Gamry). Several rounds of debugging were
unable to pinpoint the source of discrepancy, so we pivoted to the MCU/AFE combination chip
ADuCM355 using code from a working SWV potentiostat available on GitHub [102]. One
advantage of having the AFE combined with the MCU is the response time between the two
units. In the earlier code, the MCU commanded the AD5940/5941 through SPI to program the
sequencer to then program the digital-to-analog converter (DAC), the ADuCM355 requires
neither SPI nor the sequencer and can instead directly communicate with the DAC.

121
4.9 SECOND POTENTIOSTAT PROTOTYPE DESIGN
The adapted SWV code for the ADuCM355 evaluation board produced data that
resembled that of the Gamry benchtop potentiostat. To replicate the results on breadboard, a
breakout PCB was designed in-house and then fabricated externally (JLCPCB, Hong
Kong)(Figure 4-17). Much like the evaluation board, the mounted ADuCM355 was connected
via Universal Asynchronous Receiver / Transmitter (UART) protocol and serial wire debug
(SWD) protocol for programming (Figure 4-18 and Figure 4-19).
Figure 4-17: Image of breadboard setup with ADuCM355 on a breakout PCB next to the
AduCM355 evaluation board
For this iteration, we opted to exclude BLE communication and focused solely on
producing a miniature potentiostat that was compatible with our sensors. At this stage,
communication to the MCU would be wired through USB-C, which could later be replaced with
a BLE module.

122
Figure 4-18: Simplified block diagram for second potentiostat prototype; instead of BLE
as in Figure 4-13 (right), communication is through USB-C.
Figure 4-19: Complete circuit schematic for miniature potentiostat
To compare these three potentiostats—benchtop, evaluation board, and breadboard, SWV
was executed on a well-known electrochemical reaction: ferrocyanide and its accompanying
redox reaction (Equation 3). For the standard three-electrode setup, Au wire acted as the working
electrode, Pt wire as the counter, and Ag/AgCl as the reference.

123
[()6]
3− +
− ⇌ [()6]
4− (3)
Results indicated that all three potentiostats were capable of measuring the redox reaction at the
same voltage (Figure 4-20). The magnitude of the current response was not a concern at this time
because 1) these values can be scaled in the code and 2) for a single potentiostat, the difference
between current responses at varying analyte concentrations is more important for device
resolution.
Figure 4-20: For gold wires immersed in ferrocyanide, a peak response appears for all
three potentiostats: Gamry benchtop, evaluation board, and breadboard.
Next, the three-electrode setup was repeated with Au wire as working electrode in a solution of
1× PBS + 2 mM MgCl2, where the salt is necessary for later vancomycin introduction.

124
Indeed, all three potentiostats measured peak responses that correlate with frequency). Although
the voltage on the AduCM355 is shifted compared to the benchtop, this is something that can be
altered in the code. More importantly, the breadboard performance was identical to that of the
evaluation board, indicating that the schematic was properly designed (Figure 4-21).
Figure 4-21: SWV on EAB wires in PBS shows that each potentiostat is sensitive to
frequency. Moreover, the breadboard response is identical to that of the evaluation board.
4.10 POTENTIOSTAT PROTOTYPE PCB VERIFICATION
Next, the potentiostat schematic was realized in PCB design software while optimizing
overall size. Where possible, surface-mount technology (SMT) components were selected. A set
of generic header pin holes was included for SWD programming and for connection to two sets

125
of three-electrode setups (working, reference, counter). The PCB’s overall size was 0.68 × 0.95
in2
(Figure 4-22).
Figure 4-22: Potentiostat layout with major components highlighted; see Figure 4-18 for
a block diagram

126
Figure 4-23: Photograph of fabricated and assembled potentiostat prototype
Upon complete fabrication and assembly of the potentiostat PCB (JLCPCB, Hong Kong)(Figure
4-23), the prototype PCB was compared to the benchtop Gamry and ADuCM355 evaluation
board using the ferrocyanide reaction with Au wires (Figure 4-24). Interrogating the reaction at
both 10 and 100 Hz showed that all three potentiostats showed a signal change with frequency.
While the current peaks from the PCB prototype did not scale exactly with the benchtop’s, these
values can be adjusted within the code, a task better left for aptamer-based testing.

127
Figure 4-24: SWV on gold wires in ferrocyanide shows that each potentiostat is sensitive
to frequency.
Finally, the prototype PCB was compared to the benchtop Gamry using EAB sensors with
vancomycin. Both potentiostats measured vancomycin in solution with concentrations from 0 to
1000 µM at the signal-on frequency of 100 Hz and at the signal-off frequency of 10 Hz. The
normalized difference of the peak current at these frequencies was graphed (Figure 4-25). While
there was an abnormality in the PCB data at the 316 µM concentration, these results otherwise
demonstrate that the miniature potentiostat we built can detect a large signal gain as well as the
benchtop potentiostat.

128
Figure 4-25: Both the benchtop potentiostat and the miniaturized potentiostat PCB can
detect the increasing concentration of vancomycin in solution.
4.11 CONCLUSION
EAB sensing fulfils the need for continuous, real-time measurement of biologically relevant
molecules without the use of detecting agents that must be replaced. In combination with
microneedles and miniaturized electronics, EAB sensing stands to become a portable technology
as well.
To that end, we have made the first steps towards an EAB sensing microneedle array using
silicon. With MEMS techniques, silicon wafers can be fashioned into oxide-insulated, goldcoated planar microneedles. After aptamer functionalization, these perform comparably to
standard gold wires.

129
Building upon a commercial analog front end (AFE) chip, the ADuCM355, we have
developed a thumb-sized potentiostat capable of implementing square-wave voltammetry (SWV),
whose performance is on par with that of benchtop equipment as verified in experiments with
gold wire in ferrocyanide and with EAB sensors in vancomycin.

130
REFERENCES
[1] “Global Wearable Technology Market 2023-2027,” Technavio, Aug. 2023.
[2] D. D. McManus et al., “A novel application for the detection of an irregular pulse using
an iPhone 4S in patients with atrial fibrillation,” Heart Rhythm, vol. 10, no. 3, pp. 315–
319, Mar. 2013, doi: 10.1016/j.hrthm.2012.12.001.
[3] H. Sun et al., “IDF Diabetes Atlas: Global, regional and country-level diabetes
prevalence estimates for 2021 and projections for 2045,” Diabetes Res. Clin. Pract., vol.
183, p. 109119, Jan. 2022, doi: 10.1016/j.diabres.2021.109119.
[4] D. C. Klonoff, D. Ahn, and A. Drincic, “Continuous glucose monitoring: A review of the
technology and clinical use,” Diabetes Res. Clin. Pract., vol. 133, pp. 178–192, Nov.
2017, doi: 10.1016/j.diabres.2017.08.005.
[5] P. Dauphin-Ducharme et al., “Electrochemical Aptamer-Based Sensors for Improved
Therapeutic Drug Monitoring and High-Precision, Feedback-Controlled Drug Delivery,”
ACS Sens., vol. 4, no. 10, pp. 2832–2837, Oct. 2019, doi: 10.1021/acssensors.9b01616.
[6] N. Arroyo-Currás, J. Somerson, P. A. Vieira, K. L. Ploense, T. E. Kippin, and K. W.
Plaxco, “Real-time measurement of small molecules directly in awake, ambulatory
animals,” Proc. Natl. Acad. Sci., vol. 114, no. 4, pp. 645–650, Jan. 2017, doi:
10.1073/pnas.1613458114.
[7] G. Rocchitta et al., “Enzyme Biosensors for Biomedical Applications: Strategies for
Safeguarding Analytical Performances in Biological Fluids,” Sensors, vol. 16, no. 6, p.
780, May 2016, doi: 10.3390/s16060780.
[8] J. Wang, “Electrochemical Glucose Biosensors,” Chem. Rev., vol. 108, no. 2, pp. 814–
825, Feb. 2008, doi: 10.1021/cr068123a.
[9] U. Wollenberger, F. W. Scheller, A. Böhmer, M. Passarge, and H.-G. Müller, “A specific
enzyme electrode for l-glutamate-development and application,” Biosensors, vol. 4, no. 6,
pp. 381–391, Jan. 1989, doi: 10.1016/0265-928X(89)80004-5.
[10] N. R. Ferreira, A. Ledo, J. Laranjinha, G. A. Gerhardt, and R. M. Barbosa, “Simultaneous
measurements of ascorbate and glutamate in vivo in the rat brain using carbon fiber
nanocomposite sensors and microbiosensor arrays,” Bioelectrochemistry, vol. 121, pp.
142–150, Jun. 2018, doi: 10.1016/j.bioelechem.2018.01.009.
[11] F. Mizutani, T. Yamanaka, Y. Tanabe, and K. Tsuda, “An enzyme electrode forl-lactate
with a chemically-amplified response,” Anal. Chim. Acta, vol. 177, pp. 153–166, Jan.
1985, doi: 10.1016/S0003-2670(00)82947-5.
[12] A. Wolf et al., “Evaluation of Continuous Lactate Monitoring Systems within a
Heparinized In Vivo Porcine Model Intravenously and Subcutaneously,” Biosensors, vol.
8, no. 4, Art. no. 4, Dec. 2018, doi: 10.3390/bios8040122.

131
[13] K. M. Mitchell, “Acetylcholine and Choline Amperometric Enzyme Sensors
Characterized in Vitro and in Vivo,” Anal. Chem., vol. 76, no. 4, pp. 1098–1106, Feb.
2004, doi: 10.1021/ac034757v.
[14] M. Gamella et al., “A novel non-invasive electrochemical biosensing device for in situ
determination of the alcohol content in blood by monitoring ethanol in sweat,” Anal.
Chim. Acta, vol. 806, pp. 1–7, Jan. 2014, doi: 10.1016/j.aca.2013.09.020.
[15] G. Rocchitta et al., “Development and characterization of an implantable biosensor for
telemetric monitoring of ethanol in the brain of freely moving rats,” Anal. Chem., vol. 84,
no. 16, pp. 7072–7079, Aug. 2012, doi: 10.1021/ac301253h.
[16] M. Liu et al., “An amperometric biosensor based on ascorbate oxidase immobilized in
poly(3,4-ethylenedioxythiophene)/multi-walled carbon nanotubes composite films for the
determination of L-ascorbic acid,” Anal. Sci. Int. J. Jpn. Soc. Anal. Chem., vol. 27, no. 5,
p. 477, 2011, doi: 10.2116/analsci.27.477.
[17] V. Aggarwal, J. Malik, A. Prashant, P. K. Jaiwal, and C. S. Pundir, “Amperometric
determination of serum total cholesterol with nanoparticles of cholesterol esterase and
cholesterol oxidase,” Anal. Biochem., vol. 500, pp. 6–11, May 2016, doi:
10.1016/j.ab.2016.01.019.
[18] E. Casero et al., “Laccase biosensors based on different enzyme immobilization strategies
for phenolic compounds determination,” Talanta, vol. 115, pp. 401–408, Oct. 2013, doi:
10.1016/j.talanta.2013.05.045.
[19] A. L. Simonian, E. I. Rainina, J. Wild, and P. F. Fitzpatrick, “A Biosensor for LTryptophan Determination Based on Recombinant Pseudomonas savastanoi Tryptophan2-Monooxygenase,” Anal. Lett., vol. 28, no. 10, pp. 1751–1761, Jul. 1995, doi:
10.1080/00032719508000353.
[20] J. Niu and J. Y. Lee, “Renewable-surface graphite–ceramic enzyme sensors for the
determination of hypoxanthine in fish meat,” Anal. Commun., vol. 36, no. 3, pp. 81–83,
1999, doi: 10.1039/A900896I.
[21] S. V. Dzyadevych, V. N. Arkhypova, A. P. Soldatkin, A. V. El’skaya, C. Martelet, and N.
Jaffrezic-Renault, “Amperometric enzyme biosensors: Past, present and future,” IRBM,
vol. 29, no. 2, pp. 171–180, Apr. 2008, doi: 10.1016/j.rbmret.2007.11.007.
[22] P. Puthongkham and B. J. Venton, “Recent advances in fast-scan cyclic voltammetry,”
Analyst, vol. 145, no. 4, pp. 1087–1102, Feb. 2020, doi: 10.1039/C9AN01925A.
[23] B. H. McAdams and A. A. Rizvi, “An Overview of Insulin Pumps and Glucose Sensors
for the Generalist,” J. Clin. Med., vol. 5, no. 1, Art. no. 1, Jan. 2016, doi:
10.3390/jcm5010005.

132
[24] D. L. Robinson, A. Hermans, A. T. Seipel, and R. M. Wightman, “Monitoring Rapid
Chemical Communication in the Brain,” Chem. Rev., vol. 108, no. 7, pp. 2554–2584, Jul.
2008, doi: 10.1021/cr068081q.
[25] E. S. Bucher and R. M. Wightman, “Electrochemical Analysis of Neurotransmitters,”
Annu. Rev. Anal. Chem., vol. 8, no. 1, pp. 239–261, Jul. 2015, doi: 10.1146/annurevanchem-071114-040426.
[26] J. G. Roberts and L. A. Sombers, “Fast-Scan Cyclic Voltammetry: Chemical Sensing in
the Brain and Beyond,” Anal. Chem., vol. 90, no. 1, pp. 490–504, Jan. 2018, doi:
10.1021/acs.analchem.7b04732.
[27] J. J. Clark et al., “Chronic microsensors for longitudinal, subsecond dopamine detection
in behaving animals,” Nat. Methods, vol. 7, no. 2, Art. no. 2, Feb. 2010, doi:
10.1038/nmeth.1412.
[28] N. V. Kulagina and A. C. Michael, “Monitoring Hydrogen Peroxide in the Extracellular
Space of the Brain with Amperometric Microsensors,” Anal. Chem., vol. 75, no. 18, pp.
4875–4881, Sep. 2003, doi: 10.1021/ac034573g.
[29] J. B. Zimmerman and R. Mark. Wightman, “Simultaneous electrochemical measurements
of oxygen and dopamine in vivo,” Anal. Chem., vol. 63, no. 1, pp. 24–28, Jan. 1991, doi:
10.1021/ac00001a005.
[30] J. K. Thompson, M. R. Peterson, and R. D. Freeman, “Single-Neuron Activity and Tissue
Oxygenation in the Cerebral Cortex,” Science, vol. 299, no. 5609, pp. 1070–1072, Feb.
2003, doi: 10.1126/science.1079220.
[31] J. M. Dubach, E. Lim, N. Zhang, K. P. Francis, and H. Clark, “In vivo sodium
concentration continuously monitored with fluorescent sensors,” Integr. Biol., vol. 3, no.
2, pp. 142–148, Feb. 2011, doi: 10.1039/c0ib00020e.
[32] K. J. Cash and H. A. Clark, “In Vivo Histamine Optical Nanosensors,” Sensors, vol. 12,
no. 9, Art. no. 9, Sep. 2012, doi: 10.3390/s120911922.
[33] D. Lee et al., “In vivo imaging of hydrogen peroxide with chemiluminescent
nanoparticles,” Nat. Mater., vol. 6, no. 10, Art. no. 10, Oct. 2007, doi: 10.1038/nmat1983.
[34] M. Pohanka, “Overview of Piezoelectric Biosensors, Immunosensors and DNA Sensors
and Their Applications,” Materials, vol. 11, no. 3, Art. no. 3, Mar. 2018, doi:
10.3390/ma11030448.
[35] F. Narita et al., “A Review of Piezoelectric and Magnetostrictive Biosensor Materials for
Detection of COVID-19 and Other Viruses,” Adv. Mater., vol. 33, no. 1, p. 2005448,
2021, doi: 10.1002/adma.202005448.

133
[36] Y. Xiao, A. A. Lubin, A. J. Heeger, and K. W. Plaxco, “Label-Free Electronic Detection
of Thrombin in Blood Serum by Using an Aptamer-Based Sensor,” Angew. Chem., vol.
117, no. 34, pp. 5592–5595, 2005, doi: 10.1002/ange.200500989.
[37] C. Fan, K. W. Plaxco, and A. J. Heeger, “Electrochemical interrogation of
conformational changes as a reagentless method for the sequence-specific detection of
DNA,” Proc. Natl. Acad. Sci., vol. 100, no. 16, pp. 9134–9137, Aug. 2003, doi:
10.1073/pnas.1633515100.
[38] C. Tuerk and L. Gold, “Systematic Evolution of Ligands by Exponential Enrichment:
RNA Ligands to Bacteriophage T4 DNA Polymerase,” Science, vol. 249, no. 4968, p.
505, Aug. 1990.
[39] A. D. Ellington and J. W. Szostak, “In vitro selection of RNA molecules that bind
specific ligands,” Nature, vol. 346, no. 6287, Art. no. 6287, Aug. 1990, doi:
10.1038/346818a0.
[40] A. M. Downs, J. Gerson, K. L. Ploense, K. W. Plaxco, and P. Dauphin-Ducharme,
“Subsecond-Resolved Molecular Measurements Using Electrochemical Phase
Interrogation of Aptamer-Based Sensors,” Anal. Chem., vol. 92, no. 20, pp. 14063–14068,
Oct. 2020, doi: 10.1021/acs.analchem.0c03109.
[41] R. J. White, N. Phares, A. A. Lubin, Y. Xiao, and K. W. Plaxco, “Optimization of
Electrochemical Aptamer-Based Sensors via Optimization of Probe Packing Density and
Surface Chemistry,” Langmuir, vol. 24, no. 18, pp. 10513–10518, Sep. 2008, doi:
10.1021/la800801v.
[42] N. Arroyo-Currás, P. Dauphin-Ducharme, G. Ortega, K. L. Ploense, T. E. Kippin, and K.
W. Plaxco, “Subsecond-Resolved Molecular Measurements in the Living Body Using
Chronoamperometrically Interrogated Aptamer-Based Sensors,” ACS Sens., vol. 3, no. 2,
pp. 360–366, Feb. 2018, doi: 10.1021/acssensors.7b00787.
[43] M. Santos-Cancel, R. A. Lazenby, and R. J. White, “Rapid Two-Millisecond
Interrogation of Electrochemical, Aptamer-Based Sensor Response Using Intermittent
Pulse Amperometry,” ACS Sens., vol. 3, no. 6, pp. 1203–1209, Jun. 2018, doi:
10.1021/acssensors.8b00278.
[44] R. J. White and K. W. Plaxco, “Exploiting Binding-Induced Changes in Probe Flexibility
for the Optimization of Electrochemical Biosensors,” Anal. Chem., vol. 82, no. 1, pp. 73–
76, Jan. 2010, doi: 10.1021/ac902595f.
[45] N. Arroyo-Currás, J. Somerson, P. A. Vieira, K. L. Ploense, T. E. Kippin, and K. W.
Plaxco, “Real-time measurement of small molecules directly in awake, ambulatory
animals,” Proc. Natl. Acad. Sci., vol. 114, no. 4, pp. 645–650, Jan. 2017, doi:
10.1073/pnas.1613458114.

134
[46] H. Li et al., “A Biomimetic Phosphatidylcholine-Terminated Monolayer Greatly
Improves the In Vivo Performance of Electrochemical Aptamer-Based Sensors,” Angew.
Chem. Int. Ed., vol. 56, no. 26, pp. 7492–7495, 2017, doi: 10.1002/anie.201700748.
[47] H. Li et al., “High frequency, calibration-free molecular measurements in situ in the
living body,” Chem. Sci., vol. 10, no. 47, pp. 10843–10848, Dec. 2019, doi:
10.1039/C9SC04434E.
[48] N. Arroyo-Currás et al., “High-Precision Control of Plasma Drug Levels Using
Feedback-Controlled Dosing,” ACS Pharmacol. Transl. Sci., vol. 1, no. 2, pp. 110–118,
Nov. 2018, doi: 10.1021/acsptsci.8b00033.
[49] P. A. Vieira et al., “Ultra-High-Precision, in-vivo Pharmacokinetic Measurements
Highlight the Need for and a Route Toward More Highly Personalized Medicine,” Front.
Mol. Biosci., vol. 6, 2019, Accessed: Feb. 08, 2024. [Online]. Available:
https://www.frontiersin.org/articles/10.3389/fmolb.2019.00069
[50] A. Idili et al., “Seconds-resolved pharmacokinetic measurements of the chemotherapeutic
irinotecan in situ in the living body,” Chem. Sci., vol. 10, no. 35, pp. 8164–8170, Sep.
2019, doi: 10.1039/C9SC01495K.
[51] A. Idili, C. Parolo, G. Ortega, and K. W. Plaxco, “Calibration-Free Measurement of
Phenylalanine Levels in the Blood Using an Electrochemical Aptamer-Based Sensor
Suitable for Point-of-Care Applications,” ACS Sens., vol. 4, no. 12, pp. 3227–3233, Dec.
2019, doi: 10.1021/acssensors.9b01703.
[52] A. Idili, J. Gerson, T. Kippin, and K. W. Plaxco, “Seconds-Resolved, In Situ
Measurements of Plasma Phenylalanine Disposition Kinetics in Living Rats,” Anal.
Chem., vol. 93, no. 8, pp. 4023–4032, Mar. 2021, doi: 10.1021/acs.analchem.0c05024.
[53] A. Idili, J. Gerson, C. Parolo, T. Kippin, and K. W. Plaxco, “An electrochemical aptamerbased sensor for the rapid and convenient measurement of L-tryptophan,” Anal. Bioanal.
Chem., vol. 411, no. 19, pp. 4629–4635, Jul. 2019, doi: 10.1007/s00216-019-01645-0.
[54] D. Zhang et al., “Electrochemical aptamer-based microsensor for real-time monitoring of
adenosine in vivo,” Anal. Chim. Acta, vol. 1076, pp. 55–63, Oct. 2019, doi:
10.1016/j.aca.2019.05.035.
[55] Y. Wu, B. Midinov, and R. J. White, “Electrochemical Aptamer-Based Sensor for RealTime Monitoring of Insulin,” ACS Sens., vol. 4, no. 2, pp. 498–503, Feb. 2019, doi:
10.1021/acssensors.8b01573.
[56] I. M. Taylor et al., “Aptamer-functionalized neural recording electrodes for the direct
measurement of cocaine in vivo,” J. Mater. Chem. B, vol. 5, no. 13, pp. 2445–2458, Mar.
2017, doi: 10.1039/C7TB00095B.

135
[57] B. S. Ferguson et al., “Real-Time, Aptamer-Based Tracking of Circulating Therapeutic
Agents in Living Animals,” Sci. Transl. Med., vol. 5, no. 213, Nov. 2013, doi:
10.1126/scitranslmed.3007095.
[58] E. Verrinder, K. K. Leung, M. Kaan Erdal, L. Sepunaru, and K. W. Plaxco, “Comparison
of voltammetric methods used in the interrogation of electrochemical aptamer-based
sensors,” Sens. Diagn., 2023, doi: 10.1039/D3SD00083D.
[59] K. Aukland and G. Nicolaysen, “Interstitial fluid volume: local regulatory mechanisms.,”
Physiol. Rev., vol. 61, no. 3, pp. 556–643, Jul. 1981, doi: 10.1152/physrev.1981.61.3.556.
[60] J. Kool et al., “Suction blister fluid as potential body fluid for biomarker proteins,”
PROTEOMICS, vol. 7, no. 20, pp. 3638–3650, 2007, doi: 10.1002/pmic.200600938.
[61] A. C. Müller et al., “A Comparative Proteomic Study of Human Skin Suction Blister
Fluid from Healthy Individuals Using Immunodepletion and iTRAQ Labeling,” J.
Proteome Res., vol. 11, no. 7, pp. 3715–3727, Jul. 2012, doi: 10.1021/pr3002035.
[62] B. Q. Tran et al., “Proteomic Characterization of Dermal Interstitial Fluid Extracted
Using a Novel Microneedle-Assisted Technique,” J. Proteome Res., vol. 17, no. 1, pp.
479–485, Jan. 2018, doi: 10.1021/acs.jproteome.7b00642.
[63] J. Heikenfeld et al., “Accessing analytes in biofluids for peripheral biochemical
monitoring,” Nat. Biotechnol., vol. 37, no. 4, pp. 407–419, Apr. 2019, doi:
10.1038/s41587-019-0040-3.
[64] P. R. Miller et al., “Extraction and biomolecular analysis of dermal interstitial fluid
collected with hollow microneedles,” Commun. Biol., vol. 1, no. 1, pp. 1–11, Oct. 2018,
doi: 10.1038/s42003-018-0170-z.
[65] U. O. Häfeli et al., “Comparison of vancomycin concentrations in blood and interstitial
fluid: a possible model for less invasive therapeutic drug monitoring,” Clin. Chem. Lab.
Med. CCLM, vol. 49, no. 12, pp. 2123–2125, Dec. 2011, doi: 10.1515/CCLM.2011.727.
[66] T. K. L. Kiang, V. Schmitt, M. H. H. Ensom, B. Chua, and U. O. Häfeli, “Therapeutic
Drug Monitoring in Interstitial Fluid: A Feasibility Study Using a Comprehensive Panel
of Drugs,” J. Pharm. Sci., vol. 101, no. 12, pp. 4642–4652, Dec. 2012, doi:
10.1002/jps.23309.
[67] H. Haslene-Hox et al., “A New Method for Isolation of Interstitial Fluid from Human
Solid Tumors Applied to Proteomic Analysis of Ovarian Carcinoma Tissue,” PLOS ONE,
vol. 6, no. 4, p. e19217, Apr. 2011, doi: 10.1371/journal.pone.0019217.
[68] G. Rao, P. Glikfeld, and R. H. Guy, “Reverse Iontophoresis: Development of a
Noninvasive Approach for Glucose Monitoring,” Pharm. Res., vol. 10, no. 12, pp. 1751–
1755, Dec. 1993, doi: 10.1023/A:1018926215306.

136
[69] S. Y. Rhee et al., “Clinical Experience of an Iontophoresis Based Glucose Measuring
System,” J. Korean Med. Sci., vol. 22, no. 1, pp. 70–73, Feb. 2007, doi:
10.3346/jkms.2007.22.1.70.
[70] S. Mitragotri, M. Coleman, J. Kost, and R. Langer, “Analysis of ultrasonically extracted
interstitial fluid as a predictor of blood glucose levels,” J. Appl. Physiol., vol. 89, no. 3,
pp. 961–966, Sep. 2000, doi: 10.1152/jappl.2000.89.3.961.
[71] U. Kiistala, “Suction Blister Device for Separation of Viable Epidermis from Dermis*,” J.
Invest. Dermatol., vol. 50, no. 2, pp. 129–137, Feb. 1968, doi: 10.1038/jid.1968.15.
[72] S. Gebhart et al., “Glucose Sensing in Transdermal Body Fluid Collected Under
Continuous Vacuum Pressure Via Micropores in the Stratum Corneum,” Diabetes
Technol. Ther., vol. 5, no. 2, pp. 159–166, Apr. 2003, doi:
10.1089/152091503321827812.
[73] P. P. Samant and M. R. Prausnitz, “Mechanisms of sampling interstitial fluid from skin
using a microneedle patch,” Proc. Natl. Acad. Sci., vol. 115, no. 18, pp. 4583–4588, May
2018, doi: 10.1073/pnas.1716772115.
[74] J. J. Mastrototaro, “The MiniMed Continuous Glucose Monitoring System,” Diabetes
Technol. Ther., vol. 2, no. supplement 1, pp. 13–18, Dec. 2000, doi:
10.1089/15209150050214078.
[75] S. Kaushik et al., “Lack of Pain Associated with Microfabricated Microneedles,” Anesth.
Analg., vol. 92, no. 2, p. 502, Feb. 2001, doi: 10.1213/00000539-200102000-00041.
[76] Y. Wu et al., “Microneedle Aptamer-Based Sensors for Continuous, Real-Time
Therapeutic Drug Monitoring,” Anal. Chem., vol. 94, no. 23, pp. 8335–8345, Jun. 2022,
doi: 10.1021/acs.analchem.2c00829.
[77] S. Lin et al., “Wearable microneedle-based electrochemical aptamer biosensing for
precision dosing of drugs with narrow therapeutic windows,” Sci. Adv., vol. 8, no. 38, p.
eabq4539, Sep. 2022, doi: 10.1126/sciadv.abq4539.
[78] A. M. Downs et al., “Microneedle electrochemical aptamer-based sensing: Real-time
small molecule measurements using sensor-embedded, commercially-available stainless
steel microneedles,” Biosens. Bioelectron., vol. 236, p. 115408, Sep. 2023, doi:
10.1016/j.bios.2023.115408.
[79] M. Friedel et al., “Continuous molecular monitoring of human dermal interstitial fluid
with microneedle-enabled electrochemical aptamer sensors,” Lab. Chip, vol. 23, no. 14,
pp. 3289–3299, Jul. 2023, doi: 10.1039/D3LC00210A.
[80] R. F. B. Turner, D. J. Harrison, and H. P. Baltes, “A CMOS potentiostat for
amperometric chemical sensors,” IEEE J. Solid-State Circuits, vol. 22, no. 3, pp. 473–
478, Jun. 1987, doi: 10.1109/JSSC.1987.1052753.

137
[81] R. G. Kakerow, H. Kappert, E. Spiegel, and Y. Manoli, “Low-power Single-chip CMOS
Potentiostat,” in Proceedings of the International Solid-State Sensors and Actuators
Conference - TRANSDUCERS ’95, Stockholm, Sweden: IEEE, 1995, pp. 142–145. doi:
10.1109/SENSOR.1995.717115.
[82] Huaqing Li et al., “Multi-channel electrochemical detection system based on labVIEW,”
in International Conference on Information Acquisition, 2004. Proceedings., Hefei,
China: IEEE, 2004, pp. 224–227. doi: 10.1109/ICIA.2004.1373356.
[83] C.-Y. Huang et al., “A Portable Potentiostat with Molecularly Imprinted Polymeric
Electrode for Dopamine Sensing,” in 2009 IEEE Circuits and Systems International
Conference on Testing and Diagnosis, Chengdu , China: IEEE, Apr. 2009, pp. 1–4. doi:
10.1109/CAS-ICTD.2009.4960767.
[84] M. D. Steinberg and C. R. Lowe, “A micropower amperometric potentiostat,” Sens.
Actuators B Chem., vol. 97, no. 2, pp. 284–289, Feb. 2004, doi:
10.1016/j.snb.2003.09.002.
[85] A. A. Rowe et al., “CheapStat: An Open-Source, ‘Do-It-Yourself’ Potentiostat for
Analytical and Educational Applications,” PLOS ONE, vol. 6, no. 9, p. e23783, Sep.
2011, doi: 10.1371/journal.pone.0023783.
[86] E. S. Friedman, M. A. Rosenbaum, A. W. Lee, David. A. Lipson, B. R. Land, and L. T.
Angenent, “A cost-effective and field-ready potentiostat that poises subsurface electrodes
to monitor bacterial respiration,” Biosens. Bioelectron., vol. 32, no. 1, pp. 309–313, Feb.
2012, doi: 10.1016/j.bios.2011.12.013.
[87] M. D. M. Dryden and A. R. Wheeler, “DStat: A Versatile, Open-Source Potentiostat for
Electroanalysis and Integration,” PLOS ONE, vol. 10, no. 10, p. e0140349, Oct. 2015, doi:
10.1371/journal.pone.0140349.
[88] T. Dobbelaere, P. M. Vereecken, and C. Detavernier, “A USB-controlled
potentiostat/galvanostat for thin-film battery characterization,” HardwareX, vol. 2, pp.
34–49, Oct. 2017, doi: 10.1016/j.ohx.2017.08.001.
[89] A. Ainla et al., “Open-Source Potentiostat for Wireless Electrochemical Detection with
Smartphones,” Anal. Chem., vol. 90, no. 10, pp. 6240–6246, May 2018, doi:
10.1021/acs.analchem.8b00850.
[90] Y. C. Li et al., “An Easily Fabricated Low-Cost Potentiostat Coupled with User-Friendly
Software for Introducing Students to Electrochemical Reactions and Electroanalytical
Techniques,” J. Chem. Educ., vol. 95, no. 9, pp. 1658–1661, Sep. 2018, doi:
10.1021/acs.jchemed.8b00340.
[91] S. Adams, E. H. Doeven, K. Quayle, and A. Kouzani, “MiniStat: Development and
evaluation of a mini-potentiostat for electrochemical measurements,” IEEE Access, pp.
1–1, 2019, doi: 10.1109/ACCESS.2019.2902575.

138
[92] D. M. Jenkins, B. E. Lee, S. Jun, J. Reyes-De-Corcuera, and E. S. McLamore, “ABE-Stat,
a Fully Open-Source and Versatile Wireless Potentiostat Project Including
Electrochemical Impedance Spectroscopy,” J. Electrochem. Soc., vol. 166, no. 9, p.
B3056, Mar. 2019, doi: 10.1149/2.0061909jes.
[93] M. W. Glasscott, M. D. Verber, J. R. Hall, A. D. Pendergast, C. J. McKinney, and J. E.
Dick, “SweepStat: A Build-It-Yourself, Two-Electrode Potentiostat for Macroelectrode
and Ultramicroelectrode Studies,” J. Chem. Educ., vol. 97, no. 1, pp. 265–270, Jan. 2020,
doi: 10.1021/acs.jchemed.9b00893.
[94] O. S. Hoilett, J. F. Walker, B. M. Balash, N. J. Jaras, S. Boppana, and J. C. Linnes,
“KickStat: A Coin-Sized Potentiostat for High-Resolution Electrochemical Analysis,”
Sensors, vol. 20, no. 8, p. 2407, Apr. 2020, doi: 10.3390/s20082407.
[95] P. Irving, R. Cecil, and M. Z. Yates, “MYSTAT: A compact potentiostat/galvanostat for
general electrochemistry measurements,” HardwareX, vol. 9, p. e00163, Apr. 2021, doi:
10.1016/j.ohx.2020.e00163.
[96] E. W. Brown et al., “ACEstat: A DIY Guide to Unlocking the Potential of Integrated
Circuit Potentiostats for Open-Source Electrochemical Analysis,” Anal. Chem., vol. 94,
no. 12, pp. 4906–4912, Mar. 2022, doi: 10.1021/acs.analchem.1c04226.
[97] F. Tehrani et al., “An integrated wearable microneedle array for the continuous
monitoring of multiple biomarkers in interstitial fluid,” Nat. Biomed. Eng., vol. 6, no. 11,
Art. no. 11, Nov. 2022, doi: 10.1038/s41551-022-00887-1.
[98] E. C. Jung and H. I. Maibach, “Animal models for percutaneous absorption,” J. Appl.
Toxicol., vol. 35, no. 1, pp. 1–10, 2015, doi: 10.1002/jat.3004.
[99] H. S. Gill, D. D. Denson, B. A. Burris, and M. R. Prausnitz, “Effect of microneedle
design on pain in human subjects,” Clin. J. Pain, vol. 24, no. 7, pp. 585–594, Sep. 2008,
doi: 10.1097/AJP.0b013e31816778f9.
[100] K. R. Williams, K. Gupta, and M. Wasilik, “Etch rates for micromachining processingpart II,” J. Microelectromechanical Syst., vol. 12, no. 6, pp. 761–778, Dec. 2003, doi:
10.1109/JMEMS.2003.820936.
[101] F. Tehrani et al., “An integrated wearable microneedle array for the continuous
monitoring of multiple biomarkers in interstitial fluid,” Nat. Biomed. Eng., May 2022, doi:
10.1038/s41551-022-00887-1.
[102] E. W. Brown et al., “ACEstat: A DIY Guide to Unlocking the Potential of Integrated
Circuit Potentiostats for Open-Source Electrochemical Analysis,” Anal. Chem., vol. 94,
no. 12, pp. 4906–4912, Mar. 2022, doi: 10.1021/acs.analchem.1c04226.
[103] R. Ahmad et al., “KAUSTat: A Wireless, Wearable, Open-Source Potentiostat for
Electrochemical Measurements,” in 2019 IEEE SENSORS, Montreal, QC, Canada: IEEE,
Oct. 2019, pp. 1–4. doi: 10.1109/SENSORS43011.2019.8956815.

139
[104] “analogdevicesinc/ad5940-examples: AD594x related application examples and block
level examples.” Accessed: Dec. 01, 2023. [Online]. Available:
https://github.com/analogdevicesinc/ad5940-examples

140
CHAPTER 5
CONCLUSION
With the aid of microelectromechanical systems (MEMS) devices, researchers have
unprecedented access to dynamic information about biological systems. In the brain, MEMS
silicon microprobes have allowed us to insert ever increasing numbers of recording electrodes by
incorporating integrated circuitry (IC). Flexible polymer probes have given us long-term
recording capability at the cost of fewer channels, limited by electronic packaging. We sought to
remedy this hindrance by joining application-specific IC (ASIC) chips with Parylene C
multielectrode arrays (MEA). In other parts of the body, namely the skin, MEMS microneedles
have enabled new ways of interacting with the rich interstitial fluid (ISF) under the dermis.
Whereas most existing biosensing methods are constrained by reagents, by limited options for
measurable reactions, or by confounding from nonspecific targets, electrochemical aptamerbased (EAB) sensing offers truly continuous, generalizable, and specific monitoring of target
molecules. We sought to bring together both microneedles and miniaturized electronics to
produce a portable platform.
To bond a rigid substrate to a flexible one requires careful consideration of thermal budget
and desired bond pad pitch. In the case of Parylene C, due to the dearth of available methods, we
modified conductive epoxy, ultrasonic wire, and anisotropic conductive film (ACF) bonding to
suit our needs. Additionally, we invented a new method called “polymer ultrasonic on bump”
(PUB) bonding. Though all four of these methods performed well on test structures, PUB
bonding was deemed most capable of interconnecting Parylene C devices with bare silicon dies.
Indeed, PUB bonding gave us communication with a fully functioning 32-pad neural stimulating

141
chip through a Parylene ribbon cable and with a 64-channel ball-grid array (BGA) neural
recording chip, also through a Parylene ribbon cable. Next we demonstrated that PUB bonding
could successfully integrate a 64-channel Parylene MEA with a 64-channel neural recording bare
die, and by comparing both the output of that chip with the output from a standard recording
setup, we showed that ASIC integrated arrays perform identically, but with the added benefit of
requiring just 10 connections rather than 64. PUB bonding continues to be used as the printed
circuit board (PCB) interconnect strategy for hundreds of Parylene C devices across thousands of
bond pads that have been distributed to research institutions nationwide by the Polymer
Implantable Electrode (PIE) Foundry.
A wearable microneedle array consists of both the microneedles themselves and the portable
electronics to measure biological data. To that end, we developed silicon microneedles using
standard MEMS techniques for wet and dry etching and then demonstrated that gold-coated
silicon is a viable substrate for EAB sensing. Next, we designed and fabricated miniature
electronics using off-the-shelf components, and we wrote custom code so that square wave
voltammetry (SWV) could be carried out. In comparisons with benchtop laboratory equipment,
our miniature potentiostat produced similar results on gold sensors in ferrocyanide solution and
on EAB sensors in vancomycin solution. The remaining step is to combine microneedles with
this thumb-sized device to create a truly portable EAB microneedle array.
We hope that the tools introduced in this thesis will help future researchers and engineers
advance the use of MEMS not just for the brain and the skin but for all arenas of biological
research and healthcare.

142
BIBLIOGRAPHY
A. A. A. Aarts, H. P. Neves, R. P. Puers, and C. V. Hoof, “An interconnect for out-of-plane
assembled biomedical probe arrays,” J. Micromech. Microeng., vol. 18, no. 6, p. 064004,
May 2008, doi: 10.1088/0960-1317/18/6/064004.
S. Adams, E. H. Doeven, K. Quayle, and A. Kouzani, “MiniStat: Development and evaluation of
a mini-potentiostat for electrochemical measurements,” IEEE Access, pp. 1–1, 2019, doi:
10.1109/ACCESS.2019.2902575.
V. Aggarwal, J. Malik, A. Prashant, P. K. Jaiwal, and C. S. Pundir, “Amperometric
determination of serum total cholesterol with nanoparticles of cholesterol esterase and
cholesterol oxidase,” Anal Biochem, vol. 500, pp. 6–11, May 2016, doi:
10.1016/j.ab.2016.01.019.
W. F. Agnew, T. G. H. Yuen, D. B. McCreery, and L. A. Bullara, “Histopathologic evaluation of
prolonged intracortical electrical stimulation,” Experimental Neurology, vol. 92, no. 1, pp.
162–185, Apr. 1986, doi: 10.1016/0014-4886(86)90132-9.
R. Ahmad et al., “KAUSTat: A Wireless, Wearable, Open-Source Potentiostat for
Electrochemical Measurements,” in 2019 IEEE SENSORS, Montreal, QC, Canada: IEEE,
Oct. 2019, pp. 1–4. doi: 10.1109/SENSORS43011.2019.8956815.
A. Ainla et al., “Open-Source Potentiostat for Wireless Electrochemical Detection with
Smartphones,” Anal. Chem., vol. 90, no. 10, pp. 6240–6246, May 2018, doi:
10.1021/acs.analchem.8b00850.
L. A. Anderson, G. B. Christianson, and J. F. Linden, “Mouse auditory cortex differs from visual
and somatosensory cortices in the laminar distribution of cytochrome oxidase and
acetylcholinesterase,” Brain Research, vol. 1252, pp. 130–142, Feb. 2009, doi:
10.1016/j.brainres.2008.11.037.
N. Arroyo-Currás et al., “High-Precision Control of Plasma Drug Levels Using FeedbackControlled Dosing,” ACS Pharmacol. Transl. Sci., vol. 1, no. 2, pp. 110–118, Nov. 2018,
doi: 10.1021/acsptsci.8b00033.
N. Arroyo-Currás, P. Dauphin-Ducharme, G. Ortega, K. L. Ploense, T. E. Kippin, and K. W.
Plaxco, “Subsecond-Resolved Molecular Measurements in the Living Body Using
Chronoamperometrically Interrogated Aptamer-Based Sensors,” ACS Sens., vol. 3, no. 2,
pp. 360–366, Feb. 2018, doi: 10.1021/acssensors.7b00787.
N. Arroyo-Currás, J. Somerson, P. A. Vieira, K. L. Ploense, T. E. Kippin, and K. W. Plaxco,
“Real-time measurement of small molecules directly in awake, ambulatory animals,”
Proceedings of the National Academy of Sciences, vol. 114, no. 4, pp. 645–650, Jan.
2017, doi: 10.1073/pnas.1613458114.

143
K. Aukland and G. Nicolaysen, “Interstitial fluid volume: local regulatory mechanisms.,”
Physiological Reviews, vol. 61, no. 3, pp. 556–643, Jul. 1981, doi:
10.1152/physrev.1981.61.3.556.
C. Banda, R. W. Johnson, T. Zhang, Z. Hou, and H. K. Charles, “Flip Chip Assembly of Thinned
Silicon Die on Flex Substrates,” IEEE Transactions on Electronics Packaging
Manufacturing, vol. 31, no. 1, pp. 1–8, Jan. 2008, doi: 10.1109/TEPM.2007.914217.
J. C. Barrese et al., “Failure mode analysis of silicon-based intracortical microelectrode arrays in
non-human primates,” J. Neural Eng., vol. 10, no. 6, p. 066014, Dec. 2013, doi:
10.1088/1741-2560/10/6/066014.
M. A. Belluscio, K. Mizuseki, R. Schmidt, R. Kempter, and G. Buzsaki, “Cross-Frequency
Phase-Phase Coupling between Theta and Gamma Oscillations in the Hippocampus,”
Journal of Neuroscience, vol. 32, no. 2, pp. 423–435, Jan. 2012, doi:
10.1523/JNEUROSCI.4122-11.2012.
S. L. BeMent, K. D. Wise, D. J. Anderson, K. Najafi, and K. L. Drake, “Solid-State Electrodes
for Multichannel Multiplexed Intracortical Neuronal Recording,” IEEE Transactions on
Biomedical Engineering, vol. BME-33, no. 2, pp. 230–241, Feb. 1986, doi:
10.1109/TBME.1986.325895.
A. Berényi et al., “Large-scale, high-density (up to 512 channels) recording of local circuits in
behaving animals,” Journal of Neurophysiology, vol. 111, no. 5, pp. 1132–1149, Mar.
2014, doi: 10.1152/jn.00785.2013.
S. M. Bidoki, D. M. Lewis, M. Clark, A. Vakorov, P. A. Millner, and D. McGorman, “Ink-jet
fabrication of electronic components,” J. Micromech. Microeng., vol. 17, no. 5, pp. 967–
974, Apr. 2007, doi: 10.1088/0960-1317/17/5/017.
R. Biran, D. C. Martin, and P. A. Tresco, “Neuronal cell loss accompanies the brain tissue
response to chronically implanted silicon microelectrode arrays,” Experimental
Neurology, vol. 195, no. 1, pp. 115–126, Sep. 2005, doi:
10.1016/j.expneurol.2005.04.020.
R. Biran, D. C. Martin, and P. A. Tresco, “The brain tissue response to implanted silicon
microelectrode arrays is increased when the device is tethered to the skull,” Journal of
Biomedical Materials Research Part A, vol. 82A, no. 1, pp. 169–178, 2007, doi:
10.1002/jbm.a.31138.
T. J. Blanche, M. A. Spacek, J. F. Hetke, and N. V. Swindale, “Polytrodes: High-Density Silicon
Electrode Arrays for Large-Scale Multiunit Recording,” Journal of Neurophysiology, vol.
93, no. 5, pp. 2987–3000, May 2005, doi: 10.1152/jn.01023.2004.
C. Böhler et al., “Multilayer Arrays for Neurotechnology Applications (MANTA): Chronically
Stable Thin‐Film Intracortical Implants,” Adv Sci (Weinh), vol. 10, no. 14, p. 2207576,
Mar. 2023, doi: 10.1002/advs.202207576.

144
P. Bollella, S. Sharma, A. E. G. Cass, and R. Antiochia, “Microneedle-based biosensor for
minimally-invasive lactate detection,” Biosensors and Bioelectronics, vol. 123, pp. 152–
159, Jan. 2019, doi: 10.1016/j.bios.2018.08.010.
J. Brazzle, I. Papautsky, and A. B. Frazier, “Micromachined needle arrays for drug delivery or
fluid extraction,” IEEE Eng. Med. Biol. Mag., vol. 18, no. 6, pp. 53–58, Dec. 1999, doi:
10.1109/51.805145.
E. W. Brown et al., “ACEstat: A DIY Guide to Unlocking the Potential of Integrated Circuit
Potentiostats for Open-Source Electrochemical Analysis,” Anal. Chem., vol. 94, no. 12,
pp. 4906–4912, Mar. 2022, doi: 10.1021/acs.analchem.1c04226.
E. W. Brown et al., “ACEstat: A DIY Guide to Unlocking the Potential of Integrated Circuit
Potentiostats for Open-Source Electrochemical Analysis,” Anal. Chem., vol. 94, no. 12,
pp. 4906–4912, Mar. 2022, doi: 10.1021/acs.analchem.1c04226.
E. S. Bucher and R. M. Wightman, “Electrochemical Analysis of Neurotransmitters,” Annual
Rev. Anal. Chem., vol. 8, no. 1, pp. 239–261, Jul. 2015, doi: 10.1146/annurev-anchem071114-040426.
F. Cai, Y. Chang, K. Wang, W. T. Khan, S. Pavlidis, and J. Papapolymerou, “High resolution
aerosol jet printing of D- band printed transmission lines on flexible LCP substrate,” in
2014 IEEE MTT-S International Microwave Symposium (IMS2014), Jun. 2014, pp. 1–3.
doi: 10.1109/MWSYM.2014.6848597.
A. Caliò et al., “Polymeric microneedles based enzymatic electrodes for electrochemical
biosensing of glucose and lactic acid,” Sensors and Actuators B: Chemical, vol. 236, pp.
343–349, Nov. 2016, doi: 10.1016/j.snb.2016.05.156.
P. K. Campbell, K. E. Jones, R. J. Huber, K. W. Horch, and R. A. Normann, “A silicon-based,
three-dimensional neural interface: manufacturing processes for an intracortical electrode
array,” IEEE Transactions on Biomedical Engineering, vol. 38, no. 8, pp. 758–768, Aug.
1991, doi: 10.1109/10.83588.
A. S. Caravaca et al., “A novel flexible cuff-like microelectrode for dual purpose, acute and
chronic electrical interfacing with the mouse cervical vagus nerve,” J. Neural Eng., vol.
14, no. 6, p. 066005, Dec. 2017, doi: 10.1088/1741-2552/aa7a42.
E. Casero et al., “Laccase biosensors based on different enzyme immobilization strategies for
phenolic compounds determination,” Talanta, vol. 115, pp. 401–408, Oct. 2013, doi:
10.1016/j.talanta.2013.05.045.
K. J. Cash and H. A. Clark, “In Vivo Histamine Optical Nanosensors,” Sensors, vol. 12, no. 9,
Art. no. 9, Sep. 2012, doi: 10.3390/s120911922.
J. H. Chang, R. Huang, and Y. Tai, “High density 256-channel chip integration with flexible
parylene pocket,” in 2011 16th International Solid-State Sensors, Actuators and

145
Microsystems Conference, Beijing, China, Jun. 2011, pp. 378–381. doi:
10.1109/TRANSDUCERS.2011.5969478.
J. H. Chang, R. Huang, and Y. Tai, “High-density IC chip integration with parylene pocket,” in
2011 6th IEEE International Conference on Nano/Micro Engineered and Molecular
Systems, Kaohsiung, Taiwan, Feb. 2011, pp. 1067–1070. doi:
10.1109/NEMS.2011.6017541.
C. A. Chestek et al., “Long-term stability of neural prosthetic control signals from silicon cortical
arrays in rhesus macaque motor cortex,” J. Neural Eng., vol. 8, no. 4, p. 045005, Aug.
2011, doi: 10.1088/1741-2560/8/4/045005.
S. R. Chinnadayyala, I. Park, and S. Cho, “Nonenzymatic determination of glucose at near
neutral pH values based on the use of nafion and platinum black coated microneedle
electrode array,” Microchim Acta, vol. 185, no. 5, p. 250, Apr. 2018, doi:
10.1007/s00604-018-2770-1.
B. Chua, S. P. Desai, M. J. Tierney, J. A. Tamada, and A. N. Jina, “Effect of microneedles shape
on skin penetration and minimally invasive continuous glucose monitoring in vivo,”
Sensors and Actuators A: Physical, vol. 203, pp. 373–381, Dec. 2013, doi:
10.1016/j.sna.2013.09.026.
J. E. Chung et al., “High-density, long-lasting, and multi-region electrophysiological recordings
using polymer electrode arrays,” Neuron, vol. 101, no. 1, pp. 21-31. e5, 2019.
J. J. Clark et al., “Chronic microsensors for longitudinal, subsecond dopamine detection in
behaving animals,” Nat Methods, vol. 7, no. 2, Art. no. 2, Feb. 2010, doi:
10.1038/nmeth.1412.
A. M. Cobo et al., “Parylene-Based Cuff Electrode With Integrated Microfluidics for Peripheral
Nerve Recording, Stimulation, and Drug Delivery,” Journal of Microelectromechanical
Systems, vol. 28, no. 1, pp. 36–49, Feb. 2019, doi: 10.1109/JMEMS.2018.2881908.
S. R. Corrie, G. J. P. Fernando, M. L. Crichton, M. E. G. Brunck, C. D. Anderson, and M. A. F.
Kendall, “Surface-modified microprojection arrays for intradermal biomarker capture,
with low non-specific protein binding,” Lab Chip, vol. 10, no. 20, p. 2655, 2010, doi:
10.1039/c0lc00068j.
J. Csicsvari et al., “Massively Parallel Recording of Unit and Local Field Potentials With
Silicon-Based Electrodes,” Journal of Neurophysiology, vol. 90, no. 2, pp. 1314–1323,
Aug. 2003, doi: 10.1152/jn.00116.2003.
P. Dardano, A. Caliò, V. Di Palma, M. F. Bevilacqua, A. Di Matteo, and L. De Stefano, “A
Photolithographic Approach to Polymeric Microneedles Array Fabrication,” Materials,
vol. 8, no. 12, Art. no. 12, Dec. 2015, doi: 10.3390/ma8125484.

146
P. Dauphin-Ducharme et al., “Electrochemical Aptamer-Based Sensors for Improved
Therapeutic Drug Monitoring and High-Precision, Feedback-Controlled Drug Delivery,”
ACS Sens., vol. 4, no. 10, pp. 2832–2837, Oct. 2019, doi: 10.1021/acssensors.9b01616.
S. Debnath et al., “Long-term stability of neural signals from microwire arrays implanted in
common marmoset motor cortex and striatum,” Biomed. Phys. Eng. Express, vol. 4, no. 5,
p. 055025, Aug. 2018, doi: 10.1088/2057-1976/aada67.
F. Deku, Y. Cohen, A. Joshi-Imre, A. Kanneganti, T. J. Gardner, and S. F. Cogan, “Amorphous
silicon carbide ultramicroelectrode arrays for neural stimulation and recording,” J. Neural
Eng., vol. 15, no. 1, p. 016007, Feb. 2018, doi: 10.1088/1741-2552/aa8f8b.
T. Dobbelaere, P. M. Vereecken, and C. Detavernier, “A USB-controlled
potentiostat/galvanostat for thin-film battery characterization,” HardwareX, vol. 2, pp.
34–49, Oct. 2017, doi: 10.1016/j.ohx.2017.08.001.
R. Dohle, F. Schüssler, T. Friedrich, J. Goßler, T. Oppert, and J. Franke, “Adapted assembly
processes for flip-chip technology with solder bumps of 50 µm or 40 µm diameter,” in
3rd Electronics System Integration Technology Conference ESTC, IEEE, 2010, pp. 1–8.
A. M. Downs et al., “Microneedle electrochemical aptamer-based sensing: Real-time small
molecule measurements using sensor-embedded, commercially-available stainless steel
microneedles,” Biosensors and Bioelectronics, vol. 236, p. 115408, Sep. 2023, doi:
10.1016/j.bios.2023.115408.
A. M. Downs, J. Gerson, K. L. Ploense, K. W. Plaxco, and P. Dauphin-Ducharme, “SubsecondResolved Molecular Measurements Using Electrochemical Phase Interrogation of
Aptamer-Based Sensors,” Anal. Chem., vol. 92, no. 20, pp. 14063–14068, Oct. 2020, doi:
10.1021/acs.analchem.0c03109.
M. D. M. Dryden and A. R. Wheeler, “DStat: A Versatile, Open-Source Potentiostat for
Electroanalysis and Integration,” PLOS ONE, vol. 10, no. 10, p. e0140349, Oct. 2015,
doi: 10.1371/journal.pone.0140349.
J. M. Dubach, E. Lim, N. Zhang, K. P. Francis, and H. Clark, “In vivo sodium concentration
continuously monitored with fluorescent sensors,” Integrative Biology, vol. 3, no. 2, pp.
142–148, Feb. 2011, doi: 10.1039/c0ib00020e.
S. V. Dzyadevych, V. N. Arkhypova, A. P. Soldatkin, A. V. El’skaya, C. Martelet, and N.
Jaffrezic-Renault, “Amperometric enzyme biosensors: Past, present and future,” IRBM,
vol. 29, no. 2, pp. 171–180, Apr. 2008, doi: 10.1016/j.rbmret.2007.11.007.
D. J. Edell, V. V. Toi, V. M. McNeil, and L. D. Clark, “Factors influencing the biocompatibility
of insertable silicon microshafts in cerebral cortex,” IEEE Trans. Biomed. Eng., vol. 39,
no. 6, pp. 635–643, Jun. 1992, doi: 10.1109/10.141202.
A. D. Ellington and J. W. Szostak, “In vitro selection of RNA molecules that bind specific
ligands,” Nature, vol. 346, no. 6287, Art. no. 6287, Aug. 1990, doi: 10.1038/346818a0.

147
G. Ezhilarasu, A. Hanna, R. Irwin, A. Alam, and S. S. Iyer, “A Flexible, Heterogeneously
Integrated Wireless Powered System for Bio-Implantable Applications using Fan-Out
Wafer-Level Packaging,” in 2018 IEEE International Electron Devices Meeting (IEDM),
San Francisco, CA, Dec. 2018, p. 29.7.1-29.7.4. doi: 10.1109/IEDM.2018.8614705.
J. D. Falcone et al., “A novel microwire interface for small diameter peripheral nerves in a
chronic, awake murine model,” J. Neural Eng., vol. 17, no. 4, p. 046003, Jul. 2020, doi:
10.1088/1741-2552/ab9b6d.
B. Fan et al., “Flexible, diamond-based microelectrodes fabricated using the diamond growth
side for neural sensing,” Microsyst Nanoeng, vol. 6, no. 1, p. 42, Dec. 2020, doi:
10.1038/s41378-020-0155-1.
C. Fan, K. W. Plaxco, and A. J. Heeger, “Electrochemical interrogation of conformational
changes as a reagentless method for the sequence-specific detection of DNA,”
Proceedings of the National Academy of Sciences, vol. 100, no. 16, pp. 9134–9137, Aug.
2003, doi: 10.1073/pnas.1633515100.
B. S. Ferguson et al., “Real-Time, Aptamer-Based Tracking of Circulating Therapeutic Agents in
Living Animals,” Sci. Transl. Med., vol. 5, no. 213, Nov. 2013, doi:
10.1126/scitranslmed.3007095.
N. R. Ferreira, A. Ledo, J. Laranjinha, G. A. Gerhardt, and R. M. Barbosa, “Simultaneous
measurements of ascorbate and glutamate in vivo in the rat brain using carbon fiber
nanocomposite sensors and microbiosensor arrays,” Bioelectrochemistry, vol. 121, pp.
142–150, Jun. 2018, doi: 10.1016/j.bioelechem.2018.01.009.
R. E. Fischell, D. R. Fischell, and A. R. Upton, “System for treatment of neurological disorders,”
2000
M. Forssell and G. K. Fedder, “Parylene neural probe with embedded CMOS multiplexing
amplifier,” in 2018 40th Annual International Conference of the IEEE Engineering in
Medicine and Biology Society (EMBC), Jul. 2018, pp. 3374–3377. doi:
10.1109/EMBC.2018.8512938.
M. A. M. Freire et al., “Comprehensive Analysis of Tissue Preservation and Recording Quality
from Chronic Multielectrode Implants,” PLoS ONE, vol. 6, no. 11, p. e27554, Nov. 2011,
doi: 10.1371/journal.pone.0027554.
O. Frey et al., “Biosensor microprobes with integrated microfluidic channels for bi-directional
neurochemical interaction,” J. Neural Eng., vol. 8, no. 6, p. 066001, Oct. 2011, doi:
10.1088/1741-2560/8/6/066001.
M. Friedel et al., “Continuous molecular monitoring of human dermal interstitial fluid with
microneedle-enabled electrochemical aptamer sensors,” Lab Chip, vol. 23, no. 14, pp.
3289–3299, Jul. 2023, doi: 10.1039/D3LC00210A.

148
E. S. Friedman, M. A. Rosenbaum, A. W. Lee, David. A. Lipson, B. R. Land, and L. T.
Angenent, “A cost-effective and field-ready potentiostat that poises subsurface electrodes
to monitor bacterial respiration,” Biosensors and Bioelectronics, vol. 32, no. 1, pp. 309–
313, Feb. 2012, doi: 10.1016/j.bios.2011.12.013.
T.-M. Fu, G. Hong, T. Zhou, T. G. Schuhmann, R. D. Viveros, and C. M. Lieber, “Stable longterm chronic brain mapping at the single-neuron level,” Nat Methods, vol. 13, no. 10, pp.
875–882, Oct. 2016, doi: 10.1038/nmeth.3969.
G. J. Gage, C. R. Stoetzner, A. B. Wiltschko, and J. D. Berke, “Selective Activation of Striatal
Fast-Spiking Interneurons during Choice Execution,” Neuron, vol. 67, no. 3, pp. 466–479,
Aug. 2010, doi: 10.1016/j.neuron.2010.06.034.
M. Gamella et al., “A novel non-invasive electrochemical biosensing device for in situ
determination of the alcohol content in blood by monitoring ethanol in sweat,” Anal
Chim Acta, vol. 806, pp. 1–7, Jan. 2014, doi: 10.1016/j.aca.2013.09.020.
S. Gebhart et al., “Glucose Sensing in Transdermal Body Fluid Collected Under Continuous
Vacuum Pressure Via Micropores in the Stratum Corneum,” Diabetes Technology &
Therapeutics, vol. 5, no. 2, pp. 159–166, Apr. 2003, doi: 10.1089/152091503321827812.
M. S. Gerstel and V. A. Place, “Drug delivery device,” US3964482A, Jun. 22, 1976 Accessed:
Aug. 26, 2021. [Online]. Available: https://patents.google.com/patent/US3964482A/en
R. Gesteland, B. Howland, J. Lettvin, and W. Pitts, “Comments on Microelectrodes,” Proc. IRE,
vol. 47, no. 11, pp. 1856–1862, Nov. 1959, doi: 10.1109/JRPROC.1959.287156.
H. S. Gill, D. D. Denson, B. A. Burris, and M. R. Prausnitz, “Effect of microneedle design on
pain in human subjects,” Clin J Pain, vol. 24, no. 7, pp. 585–594, Sep. 2008, doi:
10.1097/AJP.0b013e31816778f9.
W. F. Gillis et al., “Carbon fiber on polyimide ultra-microelectrodes,” J. Neural Eng., vol. 15, no.
1, p. 016010, Feb. 2018, doi: 10.1088/1741-2552/aa8c88.
M. W. Glasscott, M. D. Verber, J. R. Hall, A. D. Pendergast, C. J. McKinney, and J. E. Dick,
“SweepStat: A Build-It-Yourself, Two-Electrode Potentiostat for Macroelectrode and
Ultramicroelectrode Studies,” J. Chem. Educ., vol. 97, no. 1, pp. 265–270, Jan. 2020, doi:
10.1021/acs.jchemed.9b00893.
K. Y. Goud et al., “Wearable Electrochemical Microneedle Sensor for Continuous Monitoring of
Levodopa: Toward Parkinson Management,” ACS Sens., vol. 4, no. 8, pp. 2196–2204,
Aug. 2019, doi: 10.1021/acssensors.9b01127.
P. Griss, P. Enoksson, H. K. Tolvanen-Laakso, P. Merilainen, S. Ollmar, and G. Stemme,
“Micromachined electrodes for biopotential measurements,” Journal of
Microelectromechanical Systems, vol. 10, no. 1, pp. 10–16, Mar. 2001, doi:
10.1109/84.911086.

149
C. A. Gutierrez, C. Lee, B. Kim, and E. Meng, “Epoxy-less packaging methods for electrical
contact to parylene-based flat flexible cables,” in 2011 16th International Solid-State
Sensors, Actuators and Microsystems Conference, Beijing, China, Jun. 2011, pp. 2299–
2302. doi: 10.1109/TRANSDUCERS.2011.5969538.
U. O. Häfeli et al., “Comparison of vancomycin concentrations in blood and interstitial fluid: a
possible model for less invasive therapeutic drug monitoring,” Clinical Chemistry and
Laboratory Medicine (CCLM), vol. 49, no. 12, pp. 2123–2125, Dec. 2011, doi:
10.1515/CCLM.2011.727.
B. Haider, M. Häusser, and M. Carandini, “Inhibition dominates sensory responses in the awake
cortex,” Nature, vol. 493, no. 7430, pp. 97–100, Jan. 2013, doi: 10.1038/nature11665.
R. E. Hampson et al., “Facilitation and restoration of cognitive function in primate prefrontal
cortex by a neuroprosthesis that utilizes minicolumn-specific neural firing,” J. Neural
Eng., vol. 9, no. 5, p. 056012, Oct. 2012, doi: 10.1088/1741-2560/9/5/056012.
G. Harman, Wire bonding in microelectronics. McGraw-Hill Education, 2010.
R. Harrison, “private communication,” Aug. 17, 2017.
R. Harrison, “private communication,” Aug. 07, 2018.
S. Hashmi, G. Hashmi, and R. Gaugler, “Genetic Transformation of an Entomopathogenic
Nematode by Microinjection,” Journal of Invertebrate Pathology, vol. 66, no. 3, pp. 293–
296, Nov. 1995, doi: 10.1006/jipa.1995.1103.
H. Haslene-Hox et al., “A New Method for Isolation of Interstitial Fluid from Human Solid
Tumors Applied to Proteomic Analysis of Ovarian Carcinoma Tissue,” PLOS ONE, vol.
6, no. 4, p. e19217, Apr. 2011, doi: 10.1371/journal.pone.0019217.
J. Heikenfeld et al., “Accessing analytes in biofluids for peripheral biochemical monitoring,” Nat
Biotechnol, vol. 37, no. 4, pp. 407–419, Apr. 2019, doi: 10.1038/s41587-019-0040-3.
S. Henry, D. V. McAllister, M. G. Allen, and M. R. Prausnitz, “Microfabricated Microneedles: A
Novel Approach to Transdermal Drug Delivery,” Journal of Pharmaceutical Sciences, vol.
87, no. 8, pp. 922–925, Aug. 1998, doi: 10.1021/js980042+.
J. F. Hetke, J. C. Williams, D. S. Pellinen, R. J. Vetter, and D. R. Kipke, “3-D silicon probe array
with hybrid polymer interconnect for chronic cortical recording,” in First International
IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings., Mar.
2003, pp. 181–184. doi: 10.1109/CNE.2003.1196787.
L. R. Hochberg et al., “Neuronal ensemble control of prosthetic devices by a human with
tetraplegia,” Nature, vol. 442, no. 7099, Art. no. 7099, Jul. 2006, doi:
10.1038/nature04970.

150
L. R. Hochberg et al., “Reach and grasp by people with tetraplegia using a neurally controlled
robotic arm,” Nature, vol. 485, no. 7398, pp. 372–375, May 2012, doi:
10.1038/nature11076.
I. Hochmair et al., “MED-EL Cochlear Implants: State of the Art and a Glimpse Into the Future,”
Trends in Amplification, vol. 10, no. 4, pp. 201–219, Dec. 2006, doi:
10.1177/1084713806296720.
O. S. Hoilett, J. F. Walker, B. M. Balash, N. J. Jaras, S. Boppana, and J. C. Linnes, “KickStat: A
Coin-Sized Potentiostat for High-Resolution Electrochemical Analysis,” Sensors, vol. 20,
no. 8, p. 2407, Apr. 2020, doi: 10.3390/s20082407.
C.-Y. Huang et al., “A Portable Potentiostat with Molecularly Imprinted Polymeric Electrode for
Dopamine Sensing,” in 2009 IEEE Circuits and Systems International Conference on
Testing and Diagnosis, Chengdu , China: IEEE, Apr. 2009, pp. 1–4. doi: 10.1109/CASICTD.2009.4960767.
Huaqing Li et al., “Multi-channel electrochemical detection system based on labVIEW,” in
International Conference on Information Acquisition, 2004. Proceedings., Hefei, China:
IEEE, 2004, pp. 224–227. doi: 10.1109/ICIA.2004.1373356.
G. Humpston and D. M. Jacobson, Principles of soldering. ASM international, 2004.
A. Idili et al., “Seconds-resolved pharmacokinetic measurements of the chemotherapeutic
irinotecan in situ in the living body,” Chem. Sci., vol. 10, no. 35, pp. 8164–8170, Sep.
2019, doi: 10.1039/C9SC01495K.
A. Idili, J. Gerson, T. Kippin, and K. W. Plaxco, “Seconds-Resolved, In Situ Measurements of
Plasma Phenylalanine Disposition Kinetics in Living Rats,” Anal. Chem., vol. 93, no. 8,
pp. 4023–4032, Mar. 2021, doi: 10.1021/acs.analchem.0c05024.
A. Idili, J. Gerson, C. Parolo, T. Kippin, and K. W. Plaxco, “An electrochemical aptamer-based
sensor for the rapid and convenient measurement of L-tryptophan,” Anal Bioanal Chem,
vol. 411, no. 19, pp. 4629–4635, Jul. 2019, doi: 10.1007/s00216-019-01645-0.
A. Idili, C. Parolo, G. Ortega, and K. W. Plaxco, “Calibration-Free Measurement of
Phenylalanine Levels in the Blood Using an Electrochemical Aptamer-Based Sensor
Suitable for Point-of-Care Applications,” ACS Sens., vol. 4, no. 12, pp. 3227–3233, Dec.
2019, doi: 10.1021/acssensors.9b01703.
F. Inoue, J. Derakhshandeh, M. Lofrano, and E. Beyne, “Fine-pitch bonding technology with
surface-planarized solder micro-bump/polymer hybrid for 3D integration,” Jpn. J. Appl.
Phys., vol. 60, no. 2, p. 026502, Jan. 2021, doi: 10.35848/1347-4065/abd69c.
Intan Technologies, LLC, “RHD2164 Digital Electrophysiology Interface Chip.” RHD2164
datasheet, Dec. 01, 2017. Accessed: Nov. 18, 2020. [Online]. Available:
https://intantech.com/files/Intan_RHD2164_datasheet.pdf

151
M. A. Invernale, B. C. Tang, R. L. York, L. Le, D. Y. Hou, and D. G. Anderson, “Microneedle
Electrodes Toward an Amperometric Glucose-Sensing Smart Patch,” Advanced
Healthcare Materials, vol. 3, no. 3, pp. 338–342, 2014, doi: 10.1002/adhm.201300142.
P. Irving, R. Cecil, and M. Z. Yates, “MYSTAT: A compact potentiostat/galvanostat for general
electrochemistry measurements,” HardwareX, vol. 9, p. e00163, Apr. 2021, doi:
10.1016/j.ohx.2020.e00163.
N. Islam, M.-C. Hsieh, K. KeonTaek, and V. Pandey, “Fine pitch Cu pillar assembly challenges
for advanced flip chip package,” in Proceedings of the International Wafer-Level
Packaging Conference, 2017.
D. M. Jenkins, B. E. Lee, S. Jun, J. Reyes-De-Corcuera, and E. S. McLamore, “ABE-Stat, a
Fully Open-Source and Versatile Wireless Potentiostat Project Including Electrochemical
Impedance Spectroscopy,” J. Electrochem. Soc., vol. 166, no. 9, p. B3056, Mar. 2019,
doi: 10.1149/2.0061909jes.
J. Jeong, S. Hyun Bae, J.-M. Seo, H. Chung, and S. June Kim, “Long-term evaluation of a liquid
crystal polymer (LCP)-based retinal prosthesis,” J. Neural Eng., vol. 13, no. 2, p. 025004,
Apr. 2016, doi: 10.1088/1741-2560/13/2/025004.
A. Jina et al., “Design, Development, and Evaluation of a Novel Microneedle Array-based
Continuous Glucose Monitor,” J Diabetes Sci Technol, vol. 8, no. 3, pp. 483–487, May
2014, doi: 10.1177/1932296814526191.
J. W. Judy, “Microelectromechanical systems (MEMS): fabrication, design and applications,”
Smart Mater. Struct., vol. 10, no. 6, pp. 1115–1134, Dec. 2001, doi: 10.1088/0964-
1726/10/6/301.
J. J. Jun et al., “Fully integrated silicon probes for high-density recording of neural activity,”
Nature, vol. 551, no. 7679, p. 232, 2017.
E. C. Jung and H. I. Maibach, “Animal models for percutaneous absorption,” Journal of Applied
Toxicology, vol. 35, no. 1, pp. 1–10, 2015, doi: 10.1002/jat.3004.
R. G. Kakerow, H. Kappert, E. Spiegel, and Y. Manoli, “Low-power Single-chip CMOS
Potentiostat,” in Proceedings of the International Solid-State Sensors and Actuators
Conference - TRANSDUCERS ’95, Stockholm, Sweden: IEEE, 1995, pp. 142–145. doi:
10.1109/SENSOR.1995.717115.
M. A. Kanchwala, G. A. McCallum, and D. M. Durand, “A Miniature Wireless Neural
Recording System for Chronic Implantation in Freely Moving Animals,” in 2018 IEEE
Biomedical Circuits and Systems Conference (BioCAS), Oct. 2018, pp. 1–4. doi:
10.1109/BIOCAS.2018.8584701.
Y. Kato, M. Nishino, I. Saito, T. Suzuki, and K. Mabuchi, “Flexible Intracortical Neural Probe
with Biodegradable Polymer for Delivering Bioactive Components,” in 2006

152
International Conference on Microtechnologies in Medicine and Biology, May 2006, pp.
143–146. doi: 10.1109/MMB.2006.251512.
S. Kaushik et al., “Lack of Pain Associated with Microfabricated Microneedles,” Anesthesia &
Analgesia, vol. 92, no. 2, p. 502, Feb. 2001, doi: 10.1213/00000539-200102000-00041.
S. Kaushik et al., “Lack of Pain Associated with Microfabricated Microneedles,” Anesthesia &
Analgesia, vol. 92, no. 2, p. 502, Feb. 2001, doi: 10.1213/00000539-200102000-00041.
T. K. L. Kiang, V. Schmitt, M. H. H. Ensom, B. Chua, and U. O. Häfeli, “Therapeutic Drug
Monitoring in Interstitial Fluid: A Feasibility Study Using a Comprehensive Panel of
Drugs,” Journal of Pharmaceutical Sciences, vol. 101, no. 12, pp. 4642–4652, Dec. 2012,
doi: 10.1002/jps.23309.
U. Kiistala, “Suction Blister Device for Separation of Viable Epidermis from Dermis*,” Journal
of Investigative Dermatology, vol. 50, no. 2, pp. 129–137, Feb. 1968, doi:
10.1038/jid.1968.15.
B. J. Kim et al., “3D Parylene sheath neural probe for chronic recordings,” J. Neural Eng., vol.
10, no. 4, p. 045002, Aug. 2013, doi: 10.1088/1741-2560/10/4/045002.
B. J. Kim and E. Meng, “Micromachining of Parylene C for bioMEMS,” Polymers for Advanced
Technologies, vol. 27, no. 5, pp. 564–576, 2016, doi: 10.1002/pat.3729.
K. Kim et al., “A tapered hollow metallic microneedle array using backside exposure of SU-8,” J.
Micromech. Microeng., vol. 14, no. 4, p. 597, Feb. 2004, doi: 10.1088/0960-
1317/14/4/021.
K. B. Kim, W.-C. Lee, C.-H. Cho, D.-S. Park, S. J. Cho, and Y.-B. Shim, “Continuous glucose
monitoring using a microneedle array sensor coupled with a wireless signal transmitter,”
Sensors and Actuators B: Chemical, vol. 281, pp. 14–21, Feb. 2019, doi:
10.1016/j.snb.2018.10.081.
Y. Kim et al., “A new MEMS neural probe integrated with embedded microfluidic channel for
drug delivery and electrode array for recording neural signal,” in 2013 Transducers &
Eurosensors XXVII: The 17th International Conference on Solid-State Sensors, Actuators
and Microsystems (TRANSDUCERS & EUROSENSORS XXVII), Jun. 2013, pp. 876–
879. doi: 10.1109/Transducers.2013.6626907.
Y.-C. Kim, J.-H. Park, and M. R. Prausnitz, “Microneedles for drug and vaccine delivery,”
Advanced Drug Delivery Reviews, vol. 64, no. 14, pp. 1547–1568, Nov. 2012, doi:
10.1016/j.addr.2012.04.005.
S. Kisban et al., “Microprobe Array with Low Impedance Electrodes and Highly Flexible
Polyimide Cables for Acute Neural Recording,” in 2007 29th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society, Aug. 2007, pp.
175–178. doi: 10.1109/IEMBS.2007.4352251.

153
D. C. Klonoff, D. Ahn, and A. Drincic, “Continuous glucose monitoring: A review of the
technology and clinical use,” Diabetes Research and Clinical Practice, vol. 133, pp. 178–
192, Nov. 2017, doi: 10.1016/j.diabres.2017.08.005.
E. B. Knudsen et al., “A thin-film optogenetic visual prosthesis.” Neuroscience, Feb. 03, 2023.
doi: 10.1101/2023.01.31.526482.
J. Koch, M. Schuettler, C. Pasluosta, and T. Stieglitz, “Electrical connectors for neural implants:
design, state of the art and future challenges of an underestimated component,” J. Neural
Eng., vol. 16, no. 6, p. 061002, Oct. 2019, doi: 10.1088/1741-2552/ab36df.
J. Kool et al., “Suction blister fluid as potential body fluid for biomarker proteins,”
PROTEOMICS, vol. 7, no. 20, pp. 3638–3650, 2007, doi: 10.1002/pmic.200600938.
T. D. Y. Kozai et al., “Mechanical failure modes of chronically implanted planar silicon-based
neural probes for laminar recording,” Biomaterials, vol. 37, pp. 25–39, Jan. 2015, doi:
10.1016/j.biomaterials.2014.10.040.
N. V. Kulagina and A. C. Michael, “Monitoring Hydrogen Peroxide in the Extracellular Space of
the Brain with Amperometric Microsensors,” Anal. Chem., vol. 75, no. 18, pp. 4875–
4881, Sep. 2003, doi: 10.1021/ac034573g.
C. A. Kuliasha and J. W. Judy, “In Vitro Reactive-Accelerated-Aging Assessment of Anisotropic
Conductive Adhesive and Back-End Packaging for Electronic Neural Interfaces,” in 2019
41st Annual International Conference of the IEEE Engineering in Medicine and Biology
Society (EMBC), Jul. 2019, pp. 3766–3769. doi: 10.1109/EMBC.2019.8856692.
M. Kuperstein and D. A. Whittington, “A Practical 24 Channel Microelectrode for Neural
Recording in Vivo,” IEEE Transactions on Biomedical Engineering, vol. BME-28, no. 3,
pp. 288–293, Mar. 1981, doi: 10.1109/TBME.1981.324702.
S. P. Lacour, G. Courtine, and J. Guck, “Materials and technologies for soft implantable
neuroprostheses,” Nat Rev Mater, vol. 1, no. 10, pp. 1–14, Sep. 2016, doi:
10.1038/natrevmats.2016.63.
F. Laermer and A. Schilp, “Method of anisotropically etching silicon,” US5501893A, Mar. 26,
1996 Accessed: Feb. 10, 2024. [Online]. Available:
https://patents.google.com/patent/US5501893A/en
M. Layani, M. Gruchko, O. Milo, I. Balberg, D. Azulay, and S. Magdassi, “Transparent
Conductive Coatings by Printing Coffee Ring Arrays Obtained at Room Temperature,”
ACS Nano, vol. 3, no. 11, pp. 3537–3542, Nov. 2009, doi: 10.1021/nn901239z.
P. Ledochowitsch, R. J. Félus, R. R. Gibboni, A. Miyakawa, S. Bao, and M. M. Maharbiz,
“Fabrication and testing of a large area, high density, parylene MEMS µECoG array,” in
2011 IEEE 24th International Conference on Micro Electro Mechanical Systems, Jan.
2011, pp. 1031–1034. doi: 10.1109/MEMSYS.2011.5734604.

154
D. Lee et al., “In vivo imaging of hydrogen peroxide with chemiluminescent nanoparticles,”
Nature Mater, vol. 6, no. 10, Art. no. 10, Oct. 2007, doi: 10.1038/nmat1983.
H. Lee, R. V. Bellamkonda, W. Sun, and M. E. Levenston, “Biomechanical analysis of silicon
microelectrode-induced strain in the brain,” J. Neural Eng., vol. 2, no. 4, pp. 81–89, Sep.
2005, doi: 10.1088/1741-2560/2/4/003.
J. H. Lee, M. A. Whittington, and N. J. Kopell, “Top-Down Beta Rhythms Support Selective
Attention via Interlaminar Interaction: A Model,” PLoS Comput Biol, vol. 9, no. 8, p.
e1003164, Aug. 2013, doi: 10.1371/journal.pcbi.1003164.
S. J. Lee, H. S. Yoon, X. Xuan, J. Y. Park, S.-J. Paik, and M. G. Allen, “A patch type nonenzymatic biosensor based on 3D SUS micro-needle electrode array for minimally
invasive continuous glucose monitoring,” Sensors and Actuators B: Chemical, vol. 222,
pp. 1144–1151, Jan. 2016, doi: 10.1016/j.snb.2015.08.013.
H. Li et al., “A Biomimetic Phosphatidylcholine-Terminated Monolayer Greatly Improves the In
Vivo Performance of Electrochemical Aptamer-Based Sensors,” Angewandte Chemie
International Edition, vol. 56, no. 26, pp. 7492–7495, 2017, doi: 10.1002/anie.201700748.
H. Li et al., “High frequency, calibration-free molecular measurements in situ in the living body,”
Chem. Sci., vol. 10, no. 47, pp. 10843–10848, Dec. 2019, doi: 10.1039/C9SC04434E.
W. Li, D. C. Rodger, E. Meng, J. D. Weiland, M. S. Humayun, and Y. Tai, “Wafer-Level
Parylene Packaging With Integrated RF Electronics for Wireless Retinal Prostheses,”
Journal of Microelectromechanical Systems, vol. 19, no. 4, pp. 735–742, Aug. 2010, doi:
10.1109/JMEMS.2010.2049985.
Y. C. Li et al., “An Easily Fabricated Low-Cost Potentiostat Coupled with User-Friendly
Software for Introducing Students to Electrochemical Reactions and Electroanalytical
Techniques,” J. Chem. Educ., vol. 95, no. 9, pp. 1658–1661, Sep. 2018, doi:
10.1021/acs.jchemed.8b00340.
L. Lin and A. P. Pisano, “Silicon-processed microneedles,” J. Microelectromech. Syst., vol. 8, no.
1, pp. 78–84, Mar. 1999, doi: 10.1109/84.749406.
S. Lin et al., “Wearable microneedle-based electrochemical aptamer biosensing for precision
dosing of drugs with narrow therapeutic windows,” Science Advances, vol. 8, no. 38, p.
eabq4539, Sep. 2022, doi: 10.1126/sciadv.abq4539.
M. Liu et al., “An amperometric biosensor based on ascorbate oxidase immobilized in poly(3,4-
ethylenedioxythiophene)/multi-walled carbon nanotubes composite films for the
determination of L-ascorbic acid,” Anal Sci, vol. 27, no. 5, p. 477, 2011, doi:
10.2116/analsci.27.477.
G. E. Loeb, R. A. Peck, W. H. Moore, and K. Hood, “BIONTM system for distributed neural
prosthetic interfaces,” Medical Engineering & Physics, vol. 23, no. 1, pp. 9–18, Jan. 2001,
doi: 10.1016/S1350-4533(01)00011-X.

155
D. Lu and C. Wong, Materials for advanced packaging, vol. 181. Springer, 2009.
L. Luan et al., “Ultraflexible nanoelectronic probes form reliable, glial scar–free neural
integration,” Science Advances, vol. 3, no. 2, p. e1601966, Feb. 2017, doi:
10.1126/sciadv.1601966.
M. J. Madou, Fundamentals of microfabrication: the science of miniaturization. CRC press, 2018.
A. Mahajan, C. D. Frisbie, and L. F. Francis, “Optimization of Aerosol Jet Printing for HighResolution, High-Aspect Ratio Silver Lines,” ACS Appl. Mater. Interfaces, vol. 5, no. 11,
pp. 4856–4864, Jun. 2013, doi: 10.1021/am400606y.
J. J. Mastrototaro, “The MiniMed Continuous Glucose Monitoring System,” Diabetes
Technology & Therapeutics, vol. 2, no. supplement 1, pp. 13–18, Dec. 2000, doi:
10.1089/15209150050214078.
B. H. McAdams and A. A. Rizvi, “An Overview of Insulin Pumps and Glucose Sensors for the
Generalist,” Journal of Clinical Medicine, vol. 5, no. 1, Art. no. 1, Jan. 2016, doi:
10.3390/jcm5010005.
D. V. McAllister, “Three-dimensional hollow microneedle and microtube arrays,” in Proc.
Transducers’ 99, 1999, pp. 1098–1101.
D. V. McAllister, M. G. Allen, and M. R. Prausnitz, “Microfabricated Microneedles for Gene
and Drug Delivery,” Annu. Rev. Biomed. Eng., vol. 2, no. 1, pp. 289–313, Aug. 2000,
doi: 10.1146/annurev.bioeng.2.1.289.
D. B. McCreery, T. G. H. Yuen, W. F. Agnew, and L. A. Bullara, “A characterization of the
effects on neuronal excitability due to prolonged microstimulation with chronically
implanted microelectrodes,” IEEE Trans. Biomed. Eng., vol. 44, no. 10, pp. 931–939,
Oct. 1997, doi: 10.1109/10.634645.
D. D. McManus et al., “A novel application for the detection of an irregular pulse using an
iPhone 4S in patients with atrial fibrillation,” Heart Rhythm, vol. 10, no. 3, pp. 315–319,
Mar. 2013, doi: 10.1016/j.hrthm.2012.12.001.
E. Meng, P.-Y. Li, and Y.-C. Tai, “Plasma removal of Parylene C,” J. Micromech. Microeng.,
vol. 18, no. 4, p. 045004, Feb. 2008, doi: 10.1088/0960-1317/18/4/045004.
J.-U. Meyer, “Retina implant—a bioMEMS challenge,” Sensors and Actuators A: Physical, vol.
97–98, pp. 1–9, Apr. 2002, doi: 10.1016/S0924-4247(01)00807-X.
J.-U. Meyer, T. Stieglitz, O. Scholz, W. Haberer, and H. Beutel, “High density interconnects and
flexible hybrid assemblies for active biomedical implants,” IEEE Transactions on
Advanced Packaging, vol. 24, no. 3, pp. 366–374, Aug. 2001, doi: 10.1109/6040.938305.

156
P. R. Miller et al., “Multiplexed microneedle-based biosensor array for characterization of
metabolic acidosis,” Talanta, vol. 88, pp. 739–742, Jan. 2012, doi:
10.1016/j.talanta.2011.11.046.
P. R. Miller et al., “Extraction and biomolecular analysis of dermal interstitial fluid collected
with hollow microneedles,” Commun Biol, vol. 1, no. 1, pp. 1–11, Oct. 2018, doi:
10.1038/s42003-018-0170-z.
R. K. Mishra, A. M. V. Mohan, F. Soto, R. Chrostowski, and J. Wang, “A microneedle biosensor
for minimally-invasive transdermal detection of nerve agents,” Analyst, vol. 142, no. 6,
pp. 918–924, Mar. 2017, doi: 10.1039/C6AN02625G.
K. M. Mitchell, “Acetylcholine and Choline Amperometric Enzyme Sensors Characterized in
Vitro and in Vivo,” Anal. Chem., vol. 76, no. 4, pp. 1098–1106, Feb. 2004, doi:
10.1021/ac034757v.
S. Mitragotri, M. Coleman, J. Kost, and R. Langer, “Analysis of ultrasonically extracted
interstitial fluid as a predictor of blood glucose levels,” Journal of Applied Physiology,
vol. 89, no. 3, pp. 961–966, Sep. 2000, doi: 10.1152/jappl.2000.89.3.961.
K. Mizuseki, S. Royer, K. Diba, and G. Buzsáki, “Activity dynamics and behavioral correlates of
CA3 and CA1 hippocampal pyramidal neurons,” Hippocampus, vol. 22, no. 8, pp. 1659–
1680, Aug. 2012, doi: 10.1002/hipo.22002.
F. Mizutani, T. Yamanaka, Y. Tanabe, and K. Tsuda, “An enzyme electrode forl-lactate with a
chemically-amplified response,” Analytica Chimica Acta, vol. 177, pp. 153–166, Jan.
1985, doi: 10.1016/S0003-2670(00)82947-5.
A. M. V. Mohan, J. R. Windmiller, R. K. Mishra, and J. Wang, “Continuous minimally-invasive
alcohol monitoring using microneedle sensor arrays,” Biosensors and Bioelectronics, vol.
91, pp. 574–579, May 2017, doi: 10.1016/j.bios.2017.01.016.
A. R. B. Moniz et al., “Minimally Invasive Enzyme Microprobes: An Alternative Approach for
Continuous Glucose Monitoring,” J Diabetes Sci Technol, vol. 6, no. 2, pp. 479–480,
Mar. 2012, doi: 10.1177/193229681200600239.
M. Mueller, N. de la Oliva, J. del Valle, I. Delgado-Martínez, X. Navarro, and T. Stieglitz,
“Rapid prototyping of flexible intrafascicular electrode arrays by picosecond laser
structuring,” J. Neural Eng., vol. 14, no. 6, p. 066016, Nov. 2017, doi: 10.1088/1741-
2552/aa7eea.
E. V. Mukerjee, S. D. Collins, R. R. Isseroff, and R. L. Smith, “Microneedle array for
transdermal biological fluid extraction and in situ analysis,” Sensors and Actuators A:
Physical, vol. 114, no. 2–3, pp. 267–275, Sep. 2004, doi: 10.1016/j.sna.2003.11.008.
A. C. Müller et al., “A Comparative Proteomic Study of Human Skin Suction Blister Fluid from
Healthy Individuals Using Immunodepletion and iTRAQ Labeling,” J. Proteome Res.,
vol. 11, no. 7, pp. 3715–3727, Jul. 2012, doi: 10.1021/pr3002035.

157
E. Musk and Neuralink, “An Integrated Brain-Machine Interface Platform With Thousands of
Channels,” J Med Internet Res, vol. 21, no. 10, p. e16194, Oct. 2019, doi: 10.2196/16194.
K. Najafi, K. D. Wise, and T. Mochizuki, “A high-yield IC-compatible multichannel recording
array,” IEEE Trans. Electron Devices, vol. 32, no. 7, pp. 1206–1211, Jul. 1985, doi:
10.1109/T-ED.1985.22102.
F. Narita et al., “A Review of Piezoelectric and Magnetostrictive Biosensor Materials for
Detection of COVID-19 and Other Viruses,” Advanced Materials, vol. 33, no. 1, p.
2005448, 2021, doi: 10.1002/adma.202005448.
J. Niu and J. Y. Lee, “Renewable-surface graphite–ceramic enzyme sensors for the determination
of hypoxanthine in fish meat,” Analytical Communications, vol. 36, no. 3, pp. 81–83,
1999, doi: 10.1039/A900896I.
M. C. Odenthal, V. Claar, O. Paul, and P. Ruther, “Hierarchical Bonding Yield Test Structure for
Flexible High Channel-Count Neural Probes Interfacing ASIC Chips,” in 2023 IEEE 36th
International Conference on Micro Electro Mechanical Systems (MEMS), Jan. 2023, pp.
409–412. doi: 10.1109/MEMS49605.2023.10052500.
K. Okabe, H. P. Jeewan, S. Yamagiwa, T. Kawano, M. Ishida, and I. Akita, “Co-Design Method
and Wafer-Level Packaging Technique of Thin-Film Flexible Antenna and Silicon
CMOS Rectifier Chips for Wireless-Powered Neural Interface Systems,” Sensors, vol. 15,
no. 12, pp. 31821–31832, Dec. 2015, doi: 10.3390/s151229885.
Y. Orii et al., “Ultrafine-pitch C2 flip chip interconnections with solder-capped Cu pillar bumps,”
in 2009 59th Electronic Components and Technology Conference, May 2009, pp. 948–
953. doi: 10.1109/ECTC.2009.5074127.
J. Ortigoza-Diaz, K. Scholten, and E. Meng, “Characterization and Modification of Adhesion in
Dry and Wet Environments in Thin-Film Parylene Systems,” Journal of
Microelectromechanical Systems, vol. 27, no. 5, pp. 874–885, Oct. 2018, doi:
10.1109/JMEMS.2018.2854636.
K. Pan et al., “A flexible retinal device with CMOS smart electrodes fabricated on parylene C
thin-film and bioceramic substrate,” Jpn. J. Appl. Phys., vol. 62, no. SC, p. SC1022, Jan.
2023, doi: 10.35848/1347-4065/acaca5.
S.-Y. Park et al., “A Miniaturized 256-Channel Neural Recording Interface With Area-Efficient
Hybrid Integration of Flexible Probes and CMOS Integrated Circuits,” IEEE
Transactions on Biomedical Engineering, vol. 69, no. 1, pp. 334–346, Jan. 2022, doi:
10.1109/TBME.2021.3093542.
J. Patel, S. Fujisawa, A. Berényi, S. Royer, and G. Buzsáki, “Traveling Theta Waves along the
Entire Septotemporal Axis of the Hippocampus,” Neuron, vol. 75, no. 3, pp. 410–417,
Aug. 2012, doi: 10.1016/j.neuron.2012.07.015.

158
E. Patrick, M. E. Orazem, J. C. Sanchez, and T. Nishida, “Corrosion of tungsten microelectrodes
used in neural recording applications,” Journal of Neuroscience Methods, vol. 198, no. 2,
pp. 158–171, Jun. 2011, doi: 10.1016/j.jneumeth.2011.03.012.
J. Perelaer et al., “Printed electronics: the challenges involved in printing devices, interconnects,
and contacts based on inorganic materials,” J. Mater. Chem., vol. 20, no. 39, p. 8446,
2010, doi: 10.1039/c0jm00264j.
K. E. Petersen, “Silicon as a mechanical material,” Proceedings of the IEEE, vol. 70, no. 5, pp.
420–457, May 1982, doi: 10.1109/PROC.1982.12331.
M. Pohanka, “Overview of Piezoelectric Biosensors, Immunosensors and DNA Sensors and
Their Applications,” Materials, vol. 11, no. 3, Art. no. 3, Mar. 2018, doi:
10.3390/ma11030448.
A. Prasad et al., “Comprehensive characterization and failure modes of tungsten microwire
arrays in chronic neural implants,” J. Neural Eng., vol. 9, no. 5, p. 056015, Oct. 2012, doi:
10.1088/1741-2560/9/5/056015.
A. Prasad et al., “Abiotic-biotic characterization of Pt/Ir microelectrode arrays in chronic
implants,” Front. Neuroeng., vol. 7, 2014, doi: 10.3389/fneng.2014.00002.
M. R. Prausnitz, “Microneedles for transdermal drug delivery,” Advanced Drug Delivery
Reviews, vol. 56, no. 5, pp. 581–587, Mar. 2004, doi: 10.1016/j.addr.2003.10.023.
P. Puthongkham and B. J. Venton, “Recent advances in fast-scan cyclic voltammetry,” Analyst,
vol. 145, no. 4, pp. 1087–1102, Feb. 2020, doi: 10.1039/C9AN01925A.
B. C. Raducanu et al., “Time Multiplexed Active Neural Probe with 1356 Parallel Recording
Sites,” Sensors, vol. 17, no. 10, p. 2388, Oct. 2017, doi: 10.3390/s17102388.
G. Rao, P. Glikfeld, and R. H. Guy, “Reverse Iontophoresis: Development of a Noninvasive
Approach for Glucose Monitoring,” Pharm Res, vol. 10, no. 12, pp. 1751–1755, Dec.
1993, doi: 10.1023/A:1018926215306.
S. Y. Rhee et al., “Clinical Experience of an Iontophoresis Based Glucose Measuring System,” J
Korean Med Sci, vol. 22, no. 1, pp. 70–73, Feb. 2007, doi: 10.3346/jkms.2007.22.1.70.
G. Rios, E. V. Lubenov, D. Chi, M. L. Roukes, and A. G. Siapas, “Nanofabricated Neural Probes
for Dense 3-D Recordings of Brain Activity,” Nano Lett., vol. 16, no. 11, pp. 6857–6862,
Nov. 2016, doi: 10.1021/acs.nanolett.6b02673.
J. G. Roberts and L. A. Sombers, “Fast-Scan Cyclic Voltammetry: Chemical Sensing in the
Brain and Beyond,” Anal. Chem., vol. 90, no. 1, pp. 490–504, Jan. 2018, doi:
10.1021/acs.analchem.7b04732.

159
D. L. Robinson, A. Hermans, A. T. Seipel, and R. M. Wightman, “Monitoring Rapid Chemical
Communication in the Brain,” Chem. Rev., vol. 108, no. 7, pp. 2554–2584, Jul. 2008, doi:
10.1021/cr068081q.
G. Rocchitta et al., “Enzyme Biosensors for Biomedical Applications: Strategies for
Safeguarding Analytical Performances in Biological Fluids,” Sensors (Basel), vol. 16, no.
6, p. 780, May 2016, doi: 10.3390/s16060780.
G. Rocchitta et al., “Development and characterization of an implantable biosensor for telemetric
monitoring of ethanol in the brain of freely moving rats,” Anal Chem, vol. 84, no. 16, pp.
7072–7079, Aug. 2012, doi: 10.1021/ac301253h.
D. C. Rodger, J. D. Weiland, M. S. Humayun, and Y.-C. Tai, “Scalable high lead-count parylene
package for retinal prostheses,” Sensors and Actuators B: Chemical, vol. 117, no. 1, pp.
107–114, Sep. 2006, doi: 10.1016/j.snb.2005.11.010.
P. J. Rousche and R. A. Normann, “Chronic recording capability of the Utah Intracortical
Electrode Array in cat sensory cortex,” Journal of Neuroscience Methods, vol. 82, no. 1,
pp. 1–15, Jul. 1998, doi: 10.1016/S0165-0270(98)00031-4.
A. A. Rowe et al., “CheapStat: An Open-Source, ‘Do-It-Yourself’ Potentiostat for Analytical and
Educational Applications,” PLOS ONE, vol. 6, no. 9, p. e23783, Sep. 2011, doi:
10.1371/journal.pone.0023783.
S. Royer et al., “Control of timing, rate and bursts of hippocampal place cells by dendritic and
somatic inhibition,” Nat Neurosci, vol. 15, no. 5, pp. 769–775, May 2012, doi:
10.1038/nn.3077.
P. Rustogi and J. W. Judy, “Electrical Isolation Performance of Microgasket Technology for
Implant Packaging,” in 2020 IEEE 70th Electronic Components and Technology
Conference (ECTC), Jun. 2020, pp. 1601–1607. doi: 10.1109/ECTC32862.2020.00251.
C. Sales-Carbonell, P. E. Rueda-Orozco, E. Soria-Gomez, G. Buzsaki, G. Marsicano, and D.
Robbe, “Striatal GABAergic and cortical glutamatergic neurons mediate contrasting
effects of cannabinoids on cortical network synchrony,” Proceedings of the National
Academy of Sciences, vol. 110, no. 2, pp. 719–724, Jan. 2013, doi:
10.1073/pnas.1217144110.
P. P. Samant and M. R. Prausnitz, “Mechanisms of sampling interstitial fluid from skin using a
microneedle patch,” PNAS, vol. 115, no. 18, pp. 4583–4588, May 2018, doi:
10.1073/pnas.1716772115.
S. Samavat et al., “Uniform sensing layer of immiscible enzyme-mediator compounds developed
via a spray aerosol mixing technique towards low cost minimally invasive microneedle
continuous glucose monitoring devices,” Biosensors and Bioelectronics, vol. 118, pp.
224–230, Oct. 2018, doi: 10.1016/j.bios.2018.07.054.

160
A. Samiei and H. Hashemi, “Energy efficient neural stimulator with dynamic supply modulation,”
Electronics Letters, vol. 57, no. 4, pp. 173–174, 2021, doi: 10.1049/ell2.12024.
M. Santos-Cancel, R. A. Lazenby, and R. J. White, “Rapid Two-Millisecond Interrogation of
Electrochemical, Aptamer-Based Sensor Response Using Intermittent Pulse
Amperometry,” ACS Sens., vol. 3, no. 6, pp. 1203–1209, Jun. 2018, doi:
10.1021/acssensors.8b00278.
E. M. Schmidt, M. J. Bak, and J. S. McIntosh, “Long-term chronic recording from cortical
neurons,” Experimental Neurology, vol. 52, no. 3, pp. 496–506, Sep. 1976, doi:
10.1016/0014-4886(76)90220-X.
E. M. Schmidt, J. S. Mcintosh, and M. J. Bak, “Long-term implants of Parylene-C coated
microelectrodes,” Med. Biol. Eng. Comput., vol. 26, no. 1, pp. 96–101, Jan. 1988, doi:
10.1007/BF02441836.
S. Schmidt, K. Horch, and R. Normann, “Biocompatibility of silicon-based electrode arrays
implanted in feline cortical tissue,” Journal of Biomedical Materials Research, vol. 27, no.
11, pp. 1393–1399, 1993, doi: 10.1002/jbm.820271106.
J. Scholvin et al., “Close-Packed Silicon Microelectrodes for Scalable Spatially Oversampled
Neural Recording,” IEEE Transactions on Biomedical Engineering, vol. 63, no. 1, pp.
120–130, Jan. 2016, doi: 10.1109/TBME.2015.2406113.
M. Schuettler, J. S. Ordonez, T. Silva Santisteban, A. Schatz, J. Wilde, and T. Stieglitz,
“Fabrication and test of a hermetic miniature implant package with 360 electrical
feedthroughs,” presented at the 2010 Annual International Conference of the IEEE
Engineering in Medicine and Biology, Aug. 2010, pp. 1585–1588. doi:
10.1109/IEMBS.2010.5626677.
R. L. Schultz and T. J. Willey, “The ultrastructure of the sheath around chronically implanted
electrodes in brain,” J Neurocytol, vol. 5, no. 6, pp. 621–642, Dec. 1976, doi:
10.1007/BF01181577.
E. B. Secor, “Principles of aerosol jet printing,” Flex. Print. Electron., vol. 3, no. 3, p. 035002,
Jul. 2018, doi: 10.1088/2058-8585/aace28.
K. Seidl, S. Herwik, T. Torfs, H. P. Neves, O. Paul, and P. Ruther, “CMOS-Based High-Density
Silicon Microprobe Arrays for Electronic Depth Control in Intracortical Neural
Recording,” Journal of Microelectromechanical Systems, vol. 20, no. 6, pp. 1439–1448,
Dec. 2011, doi: 10.1109/JMEMS.2011.2167661.
T. Seifert, M. Baum, F. Roscher, M. Wiemer, and T. Gessner, “Aerosol Jet Printing of Nano
Particle Based Electrical Chip Interconnects,” Materials Today: Proceedings, vol. 2, no. 8,
pp. 4262–4271, Jan. 2015, doi: 10.1016/j.matpr.2015.09.012.

161
J. P. Seymour and D. R. Kipke, “Neural probe design for reduced tissue encapsulation in CNS,”
Biomaterials, vol. 28, no. 25, pp. 3594–3607, Sep. 2007, doi:
10.1016/j.biomaterials.2007.03.024.
J. P. Seymour and D. R. Kipke, “Fabrication of Polymer Neural Probes with Sub-cellular
Features for Reduced Tissue Encapsulation,” in 2006 International Conference of the
IEEE Engineering in Medicine and Biology Society, Aug. 2006, pp. 4606–4609. doi:
10.1109/IEMBS.2006.260528.
S. Sharma, Z. Huang, M. Rogers, M. Boutelle, and A. E. G. Cass, “Evaluation of a minimally
invasive glucose biosensor for continuous tissue monitoring,” Anal Bioanal Chem, vol.
408, no. 29, pp. 8427–8435, Nov. 2016, doi: 10.1007/s00216-016-9961-6.
G. Shin et al., “Flexible Near-Field Wireless Optoelectronics as Subdermal Implants for Broad
Applications in Optogenetics,” Neuron, vol. 93, no. 3, pp. 509–521, Feb. 2017, doi:
10.1016/j.neuron.2016.12.031.
Shinko Electric Industries Co., Ltd., “Copper Pillar Bumping,” IC Assembly. Accessed: Oct. 29,
2021. [Online]. Available:
https://www.shinko.co.jp/english/product/package/assembly/cu-pillar.php
A. L. Simonian, E. I. Rainina, J. Wild, and P. F. Fitzpatrick, “A Biosensor for L-Tryptophan
Determination Based on Recombinant Pseudomonas savastanoi Tryptophan-2-
Monooxygenase,” Analytical Letters, vol. 28, no. 10, pp. 1751–1761, Jul. 1995, doi:
10.1080/00032719508000353.
W. H. Smart and K. Subramanian, “The Use of Silicon Microfabrication Technology in Painless
Blood Glucose Monitoring,” Diabetes Technology & Therapeutics, vol. 2, no. 4, pp. 549–
559, Dec. 2000, doi: 10.1089/15209150050501961.
M. A. Smith, X. Jia, A. Zandvakili, and A. Kohn, “Laminar dependence of neuronal correlations
in visual cortex,” Journal of Neurophysiology, vol. 109, no. 4, pp. 940–947, Feb. 2013,
doi: 10.1152/jn.00846.2012.
H. S. Sohal et al., “The sinusoidal probe: a new approach to improve electrode longevity,”
Frontiers in Neuroengineering, vol. 7, 2014, Accessed: Feb. 10, 2024. [Online]. Available:
https://www.frontiersin.org/articles/10.3389/fneng.2014.00010
M. D. Steinberg and C. R. Lowe, “A micropower amperometric potentiostat,” Sensors and
Actuators B: Chemical, vol. 97, no. 2, pp. 284–289, Feb. 2004, doi:
10.1016/j.snb.2003.09.002.
T. Stieglitz, H. Beutel, and J.-U. Meyer, “‘Microflex’—A New Assembling Technique for
Interconnects,” Journal of Intelligent Material Systems and Structures, vol. 11, no. 6, pp.
417–425, Jun. 2000, doi: 10.1106/R7BV-511B-21RJ-R2FA.

162
T. Stieglitz and M. Schuettler, “Material–tissue interfaces in implantable systems,” in
Implantable Sensor Systems for Medical Applications, Elsevier, 2013, pp. 39–67. doi:
10.1533/9780857096289.1.39.
B. Stoeber and D. Liepmann, “Fluid injection through out-of-plane microneedles,” in 1st Annual
International IEEE-EMBS Special Topic Conference on Microtechnologies in Medicine
and Biology. Proceedings (Cat. No.00EX451), Oct. 2000, pp. 224–228. doi:
10.1109/MMB.2000.893777.
J. Subbaroyan, D. C. Martin, and D. R. Kipke, “A finite-element model of the mechanical effects
of implantable microelectrodes in the cerebral cortex,” J. Neural Eng., vol. 2, no. 4, pp.
103–113, Oct. 2005, doi: 10.1088/1741-2560/2/4/006.
H. Sun et al., “IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence
estimates for 2021 and projections for 2045,” Diabetes Research and Clinical Practice,
vol. 183, p. 109119, Jan. 2022, doi: 10.1016/j.diabres.2021.109119.
T. Suzuki, K. Mabuchi, and S. Takeuchi, “A 3D flexible parylene probe array for multichannel
neural recording,” in First International IEEE EMBS Conference on Neural Engineering,
2003. Conference Proceedings., Mar. 2003, pp. 154–156. doi:
10.1109/CNE.2003.1196780.
I. Szabó, K. Máthé, A. Tóth, I. Hernádi, and A. Czurkó, “The application of elastomeric
connector for multi-channel electrophysiological recordings,” Journal of Neuroscience
Methods, vol. 114, no. 1, pp. 73–79, Feb. 2002, doi: 10.1016/S0165-0270(01)00515-5.
D. H. Szarowski et al., “Brain responses to micro-machined silicon devices,” Brain Research, vol.
983, no. 1–2, pp. 23–35, Sep. 2003, doi: 10.1016/S0006-8993(03)03023-3.
S. Takeuchi, D. Ziegler, Y. Yoshida, K. Mabuchi, and T. Suzuki, “Parylene flexible neural
probes integrated with microfluidic channels,” Lab Chip, vol. 5, no. 5, p. 519, 2005, doi:
10.1039/b417497f.
J. Tao, A. Mathewson, and K. M. Razeeb, “Study of fine pitch micro-interconnections formed by
low temperature bonded copper nanowires based anisotropic conductive film,” in 2014
IEEE 64th Electronic Components and Technology Conference (ECTC), Orlando, FL:
IEEE, May 2014, pp. 1064–1070. doi: 10.1109/ECTC.2014.6897420.
I. M. Taylor et al., “Aptamer-functionalized neural recording electrodes for the direct
measurement of cocaine in vivo,” J. Mater. Chem. B, vol. 5, no. 13, pp. 2445–2458, Mar.
2017, doi: 10.1039/C7TB00095B.
F. Tehrani et al., “An integrated wearable microneedle array for the continuous monitoring of
multiple biomarkers in interstitial fluid,” Nat. Biomed. Eng, vol. 6, no. 11, Art. no. 11,
Nov. 2022, doi: 10.1038/s41551-022-00887-1.

163
F. Tehrani et al., “An integrated wearable microneedle array for the continuous monitoring of
multiple biomarkers in interstitial fluid,” Nat. Biomed. Eng, May 2022, doi:
10.1038/s41551-022-00887-1.
J. K. Thompson, M. R. Peterson, and R. D. Freeman, “Single-Neuron Activity and Tissue
Oxygenation in the Cerebral Cortex,” Science, vol. 299, no. 5609, pp. 1070–1072, Feb.
2003, doi: 10.1126/science.1079220.
B. Q. Tran et al., “Proteomic Characterization of Dermal Interstitial Fluid Extracted Using a
Novel Microneedle-Assisted Technique,” J. Proteome Res., vol. 17, no. 1, pp. 479–485,
Jan. 2018, doi: 10.1021/acs.jproteome.7b00642.
E. M. Trautmann et al., “Large-scale high-density brain-wide neural recording in nonhuman
primates.” bioRxiv, p. 2023.02.01.526664, May 04, 2023. doi:
10.1101/2023.02.01.526664.
W. Trimmer et al., “Injection of DNA into plant and animal tissues with micromechanical
piercing structures,” in Proceedings IEEE Micro Electro Mechanical Systems. 1995,
Amsterdam, Netherlands: IEEE, 1995, p. 111. doi: 10.1109/MEMSYS.1995.472544.
J. Trzebinski, S. Sharma, A. R.-B. Moniz, K. Michelakis, Y. Zhang, and A. E. G. Cass,
“Microfluidic device to investigate factors affecting performance in biosensors designed
for transdermal applications,” Lab on a Chip, vol. 12, no. 2, pp. 348–352, 2012, doi:
10.1039/C1LC20885C.
K. Tsuchiya, N. Nakanishi, Y. Uetsuji, and E. Nakamachi, “Development of Blood Extraction
System for Health Monitoring System,” Biomed Microdevices, vol. 7, no. 4, pp. 347–353,
Dec. 2005, doi: 10.1007/s10544-005-6077-8.
C. Tuerk and L. Gold, “Systematic Evolution of Ligands by Exponential Enrichment: RNA
Ligands to Bacteriophage T4 DNA Polymerase,” Science, vol. 249, no. 4968, p. 505,
Aug. 1990.
R. F. B. Turner, D. J. Harrison, and H. P. Baltes, “A CMOS potentiostat for amperometric
chemical sensors,” IEEE J. Solid-State Circuits, vol. 22, no. 3, pp. 473–478, Jun. 1987,
doi: 10.1109/JSSC.1987.1052753.
G. Valdés-Ramírez et al., “Microneedle-based self-powered glucose sensor,” Electrochemistry
Communications, vol. 47, pp. 58–62, Oct. 2014, doi: 10.1016/j.elecom.2014.07.014.
A. Vanhoestenberghe and N. Donaldson, “Corrosion of silicon integrated circuits and lifetime
predictions in implantable electronic devices,” J. Neural Eng., vol. 10, no. 3, p. 031002,
Jun. 2013, doi: 10.1088/1741-2560/10/3/031002.
W. Verheecke, M. Van Dyck, F. Vogeler, A. Voet, and H. Valkenaers, “Optimizing aerosol jet
printing of silver interconnects on polyimide film for embedded electronics applications,”
in Eighth International DAAAM Baltic Conference Industrial Engineering, Tallinn,
Estonia, 2012, pp. 19–21.

164
E. Verrinder, K. K. Leung, M. Kaan Erdal, L. Sepunaru, and K. W. Plaxco, “Comparison of
voltammetric methods used in the interrogation of electrochemical aptamer-based
sensors,” Sensors & Diagnostics, 2023, doi: 10.1039/D3SD00083D.
P. A. Vieira et al., “Ultra-High-Precision, in-vivo Pharmacokinetic Measurements Highlight the
Need for and a Route Toward More Highly Personalized Medicine,” Frontiers in
Molecular Biosciences, vol. 6, 2019, Accessed: Feb. 08, 2024. [Online]. Available:
https://www.frontiersin.org/articles/10.3389/fmolb.2019.00069
J. Wang, “Electrochemical Glucose Biosensors,” Chem. Rev., vol. 108, no. 2, pp. 814–825, Feb.
2008, doi: 10.1021/cr068123a.
J. Wang, R. C. Y. Auyeung, H. Kim, N. A. Charipar, and A. Piqué, “Three-Dimensional Printing
of Interconnects by Laser Direct-Write of Silver Nanopastes,” Advanced Materials, vol.
22, no. 40, pp. 4462–4466, 2010, doi: 10.1002/adma.201001729.
P. M. Wang, M. Cornwell, and M. R. Prausnitz, “Minimally Invasive Extraction of Dermal
Interstitial Fluid for Glucose Monitoring Using Microneedles,” Diabetes Technology &
Therapeutics, vol. 7, no. 1, pp. 131–141, Feb. 2005, doi: 10.1089/dia.2005.7.131.
X. Wei et al., “Nanofabricated Ultraflexible Electrode Arrays for High‐Density Intracortical
Recording,” Adv. Sci., vol. 5, no. 6, p. 1700625, Jun. 2018, doi: 10.1002/advs.201700625.
R. J. White, N. Phares, A. A. Lubin, Y. Xiao, and K. W. Plaxco, “Optimization of
Electrochemical Aptamer-Based Sensors via Optimization of Probe Packing Density and
Surface Chemistry,” Langmuir, vol. 24, no. 18, pp. 10513–10518, Sep. 2008, doi:
10.1021/la800801v.
R. J. White and K. W. Plaxco, “Exploiting Binding-Induced Changes in Probe Flexibility for the
Optimization of Electrochemical Biosensors,” Anal. Chem., vol. 82, no. 1, pp. 73–76, Jan.
2010, doi: 10.1021/ac902595f.
N. J. Wilkinson, M. A. A. Smith, R. W. Kay, and R. A. Harris, “A review of aerosol jet
printing—a non-traditional hybrid process for micro-manufacturing,” Int J Adv Manuf
Technol, vol. 105, no. 11, pp. 4599–4619, Dec. 2019, doi: 10.1007/s00170-019-03438-2.
K. R. Williams, K. Gupta, and M. Wasilik, “Etch rates for micromachining processing-part II,” J.
Microelectromech. Syst., vol. 12, no. 6, pp. 761–778, Dec. 2003, doi:
10.1109/JMEMS.2003.820936.
J. R. Windmiller et al., “Microneedle array-based carbon paste amperometric sensors and
biosensors,” Analyst, vol. 136, no. 9, pp. 1846–1851, Apr. 2011, doi:
10.1039/C1AN00012H.
J. R. Windmiller et al., “Bicomponent Microneedle Array Biosensor for Minimally-Invasive
Glutamate Monitoring,” Electroanalysis, vol. 23, no. 10, pp. 2302–2309, 2011, doi:
10.1002/elan.201100361.

165
B. D. Winslow, M. B. Christensen, W.-K. Yang, F. Solzbacher, and P. A. Tresco, “A comparison
of the tissue response to chronically implanted Parylene-C-coated and uncoated planar
silicon microelectrode arrays in rat cortex,” Biomaterials, vol. 31, no. 35, pp. 9163–9172,
Dec. 2010, doi: 10.1016/j.biomaterials.2010.05.050.
K. D. Wise, J. B. Angell, and A. Starr, “An Integrated-Circuit Approach to Extracellular
Microelectrodes,” IEEE Transactions on Biomedical Engineering, vol. BME-17, no. 3, pp.
238–247, Jul. 1970, doi: 10.1109/TBME.1970.4502738.
A. Wolf et al., “Evaluation of Continuous Lactate Monitoring Systems within a Heparinized In
Vivo Porcine Model Intravenously and Subcutaneously,” Biosensors, vol. 8, no. 4, Art.
no. 4, Dec. 2018, doi: 10.3390/bios8040122.
U. Wollenberger, F. W. Scheller, A. Böhmer, M. Passarge, and H.-G. Müller, “A specific
enzyme electrode for l-glutamate-development and application,” Biosensors, vol. 4, no. 6,
pp. 381–391, Jan. 1989, doi: 10.1016/0265-928X(89)80004-5.
V. Woods et al., “Long-term recording reliability of liquid crystal polymer µ ECoG arrays,” J.
Neural Eng., vol. 15, no. 6, p. 066024, Dec. 2018, doi: 10.1088/1741-2552/aae39d.
Y. Wu et al., “Microneedle Aptamer-Based Sensors for Continuous, Real-Time Therapeutic
Drug Monitoring,” Anal. Chem., vol. 94, no. 23, pp. 8335–8345, Jun. 2022, doi:
10.1021/acs.analchem.2c00829.
Y. Wu, B. Midinov, and R. J. White, “Electrochemical Aptamer-Based Sensor for Real-Time
Monitoring of Insulin,” ACS Sens., vol. 4, no. 2, pp. 498–503, Feb. 2019, doi:
10.1021/acssensors.8b01573.
Y. Xiao, A. A. Lubin, A. J. Heeger, and K. W. Plaxco, “Label-Free Electronic Detection of
Thrombin in Blood Serum by Using an Aptamer-Based Sensor,” Angewandte Chemie,
vol. 117, no. 34, pp. 5592–5595, 2005, doi: 10.1002/ange.200500989.
C. Xie, J. Liu, T.-M. Fu, X. Dai, W. Zhou, and C. M. Lieber, “Three-dimensional macroporous
nanoelectronic networks as minimally invasive brain probes,” Nature Mater, vol. 14, no.
12, pp. 1286–1292, Dec. 2015, doi: 10.1038/nmat4427.
H. Xu, A. W. Hirschberg, K. Scholten, T. W. Berger, D. Song, and E. Meng, “Acute in vivo
testing of a conformal polymer microelectrode array for multi-region hippocampal
recordings,” Journal of neural engineering, vol. 15, no. 1, p. 016017, 2018.
X. Yang et al., “Bioinspired neuron-like electronics,” Nat. Mater., vol. 18, no. 5, pp. 510–517,
May 2019, doi: 10.1038/s41563-019-0292-9.
A. Ylinen et al., “Sharp wave-associated high-frequency oscillation (200 Hz) in the intact
hippocampus: network and intracellular mechanisms,” J. Neurosci., vol. 15, no. 1, pp.
30–46, Jan. 1995, doi: 10.1523/JNEUROSCI.15-01-00030.1995.

166
J. J. Yoo and E. Meng, “Bonding Methods for Chip Integration with Parylene Devices,” J.
Micromech. Microeng., Feb. 2021, doi: 10.1088/1361-6439/abe246.
J. J. Yoo and E. Meng, “ASIC Integration via Polymer Ultrasonic Bump Bonding to A 64-
Channel Penetrating Parylene Multielectrode Array,” in 2024 IEEE 37th International
Conference on Micro Electro Mechanical Systems (MEMS), Austin, TX, USA: IEEE,
Jan. 2024, pp. 392–395. doi: 10.1109/MEMS58180.2024.10439599.
Y. Yoon, G. S. Lee, K. Yoo, and J.-B. Lee, “Fabrication of a Microneedle/CNT Hierarchical
Micro/Nano Surface Electrochemical Sensor and Its In-Vitro Glucose Sensing
Characterization,” Sensors, vol. 13, no. 12, Art. no. 12, Dec. 2013, doi:
10.3390/s131216672.
D. Zhang et al., “Electrochemical aptamer-based microsensor for real-time monitoring of
adenosine in vivo,” Analytica Chimica Acta, vol. 1076, pp. 55–63, Oct. 2019, doi:
10.1016/j.aca.2019.05.035.
E. T. Zhao et al., “A CMOS-based highly scalable flexible neural electrode interface,” SCIENCE
ADVANCES, 2023.
Z. Zhao et al., “Ultraflexible electrode arrays for months-long high-density electrophysiological
mapping of thousands of neurons in rodents,” Nat Biomed Eng, vol. 7, no. 4, pp. 520–532,
Apr. 2023, doi: 10.1038/s41551-022-00941-y.
D. Ziegler, T. Suzuki, and S. Takeuchi, “Fabrication of Flexible Neural Probes With Built-In
Microfluidic Channels by Thermal Bonding of Parylene,” Journal of
Microelectromechanical Systems, vol. 15, no. 6, pp. 1477–1482, Dec. 2006, doi:
10.1109/JMEMS.2006.879681.
J. B. Zimmerman and R. Mark. Wightman, “Simultaneous electrochemical measurements of
oxygen and dopamine in vivo,” Anal. Chem., vol. 63, no. 1, pp. 24–28, Jan. 1991, doi:
10.1021/ac00001a005.
“Intan-RHX.” Intan Technologies, Jun. 25, 2021. Accessed: Aug. 04, 2021. [Online]. Available:
https://github.com/Intan-Technologies/Intan-RHX
“PIE Foundry.” Accessed: Sep. 09, 2023. [Online]. Available: https://piefoundry.usc.edu/
“Global Wearable Technology Market 2023-2027,” Technavio, Aug. 2023.
“analogdevicesinc/ad5940-examples: AD594x related application examples and block level
examples.” Accessed: Dec. 01, 2023. [Online]. Available:
https://github.com/analogdevicesinc/ad5940-examples

167
APPENDICES
Appendix A: BONDING RIBBON CABLE FABRICATION
Here follows the fabrication process for ribbon cable test devices (Mask sets 1, 2, and 3)
1. Bake 4” silicon wafer to remove moisture 110 °C, > 10 min
2. Deposit Parylene C layer 1 10 µm
3. Pattern AZ 5214-IR photoresist (PR) for metal liftoff Target: 1-2 µm
a. Pre spin 8 s, 500 rpm
b. Spin 45 s, 3200 rpm
c. Soft bake 110 °C, 60 s
d. Exposure 42 mJ/cm2
e. Image reversal (IR) bake 110 °C, 63 s
f. Global exposure 280 mJ/cm2
g. Development 18 s (AZ 340 1:4 dilution)
4. Descum, O2 plasma 100 W, 100 mTorr, 1 min
5. Metal deposition (Pt) 2000 Å
6. Lift off in acetone
7. Descum, O2 plasma 100 W, 100 mTorr, 1 min
8. Deposit Parylene C layer 2 10 µm
9. Pattern AZ 4620 15 µm
a. Pre spin 10 s, 500 rpm
b. Spin 45 s, 1000 rpm
c. Soft bake 90 °C, 5 min
d. Hydrate > 5 min
e. Exposure 480 mJ/cm2
f. Development 70 s (AZ 340 1:4 dilution)
10. Deep reactive ion etch (DRIE) 105 loops (until metal is exposed)
Table A-1: DRIE deposition and etch parameters for each loop
Parameter Deposition Etch
ICP Power (W) 700 700
RF Power (W) 10 20
O2 (ccm) 1 60
C4F8 (ccm) 35 1
Ar (ccm) 40 40
SF6 (ccm) 0 0
Pressure (mTorr) 23 23
Time (s) 3 10
11. Remove PR with acetone, IPA, DI water
12. Descum, O2 plasma 100 W, 100 mTorr, 1 min
13. Pattern AZ 4620 (same as previous)
14. DRIE
15. Remove PR with acetone, IPA, DI water
16. Release with water

168
Appendix B: POLYMER MEA FABRICATION
Here follows the fabrication process for pMEAs from the PIE Foundry:
1. Bake 4” silicon wafer to remove moisture 110 °C, > 15 min
2. Deposit Parylene C layer 1 10 µm
3. Thermal anneal 150 °C, 4 hr, vacuum
4. Descum, O2 plasma 100 W, 125 mTorr, 5 min
5. Pattern AZ 5214-IR for metal liftoff Target: 1.2 µm
a. Dry bake Parylene coated wafer 60 °C, > 15 min, 15 inHg, 15
sccm N2
b. Pre spin 5 s, 500 rpm
c. Spin 45 s, 3200 rpm
d. Soft bake 110 °C, 60 s
e. Exposure 42 mJ/cm2
f. Image reversal (IR) bake 110 °C, 63 s
g. Global exposure 280 mJ/cm2
h. Immerse in DI water bath 2 min
i. Development 18 s (AZ 340 1:4 dilution)
6. Descum, O2 plasma 100 W, 125 mTorr, 5 min
7. Metal deposition (Ti/Au/Pt) 20/155/25 nm
8. Lift off in acetone
a. Soak in hot NMP/PG remover bath 60 °C, 5-10 min
b. Ultrasonic soak in hot NMP/PG remover 60 °C, ~13 min
c. NMP spray and rinse 5 min
d. IPA spray and rinse 5 min
e. Soak in IPA 10 min
f. Rinse in DI water 3-5 times
g. Dry bake wafers 60 °C, > 15 min, 15 inHg, 15
sccm N2
9. Descum, O2 plasma 100 W, 125 mTorr, 5 min
10. Silanize Parylene C surface
a. Mix 900 mL DI water, 900 mL IPA, 9 mL A-174 and let
sit
2.5 hr
b. Soak wafer in solution 30 min
c. Dry 30 min
d. Spray with IPA and rinse 30 min
e. Dry with nitrogen
11. Deposit Parylene C layer 2 10 µm
12. Thermal anneal 150 °C, 4 hr, vacuum
13. Descum, O2 plasma 100 W, 125 mTorr, 5 min
14. Dry bake Parylene coated wafer 60 °C, > 15 min, 15 inHg, 15
sccm N2
15. Pattern AZ 12XT-20PL-15 14-15 µm
a. Pre spin 10 s, 500 rpm
b. Spin 45 s, 2000 rpm
c. Remove edge bead
d. Soft bake 110 °C, 3 min
e. Exposure 185 mJ/cm2

169
f. Post exposure bake 90 °C, 1 min
g. Development 75 s (726 MIF)
16. Deep reactive ion etch (DRIE) 105 loops (or until metal is
exposed)
Table A-2: DRIE deposition and etch parameters for each loop
Parameter Deposition Etch
ICP Power (W) 700 700
RF Power (W) 10 20
O2 (ccm) 1 60
C4F8 (ccm) 35 1
Ar (ccm) 40 40
SF6 (ccm) 0 0
Pressure (mTorr) 23 23
Time (s) 3 10
17. Reactive ion etch (RIE) 150 W, 150 mTorr, 1 hr
18. Remove PR with acetone, IPA, DI water
19. Dry bake Parylene coated wafer 60 °C, > 15 min, 15 inHg, 15
sccm N2
20. Descum, O2 plasma 100 W, 125 mTorr, 5 min
21. Dry bake wafer???
22. Pattern AZ 12XT-20PL-15 (same as previous)
23. DRIE
24. Remove PR with acetone, IPA, DI water
25. Release with water

170
Appendix C: OTHER PUB BONDING APPLICATIONS
We experimented with other TAB tools to limited success. The waffle tool used for all
devices has the part number 7045W-TI-10050-3/4-M TDF=046 W2=008 FX=01654
FR=BR=0015 (ref master # 490-00043-MA). It was a custom tool made by Gaiser Tools, which
was acquired by Coorstek and then sold to Small Precision Tools (SPT). SPT offers their own
TAB tools. The 7045W-C-9060-3/4-M is a waffle tool with an area of 90 × 60 µm2
, but it does
not produce the same quality bonds as the first tool. Parameter optimization would be necessary
for deploying this tool. We also tried using a round TAB tool with a flat fact (1152-010003-
750GM-TIC), Coorstek); its diameter was 25 µm. Using this tool, either no bonds would be
formed, or the tool would puncture through the Parylene C. No attempts to try other tools were
made.
Because printed circuit boards (PCBs) are nearly ubiquitous in all electronics, we attempted
to bond standard Parylene-Pt-Parylene devices (20 µm thick, 200 nm Pt) directly to gold-coated
PCB bond pads. The bonding parameters were identical to those found in early studies of PUB
bonding using gold/glass structures. No changes were necessary to achieve good bonds.

171
Figure A-1: The earliest test of PUB bonding a Parylene-Pt-Parylene device to a PCB
showed excellent mechanical adhesion. Scale bar is 0.5 mm.
The Polymer Implantable Electrode (PIE) Foundry has adopted the use of PUB bonding to
attach polymer microelectrode arrays (pMEAs) to PCBs [1]. Using bond pads with greater area
and pitch (200 × 350 µm2
), the manufacture and distribution of over 500 pMEAs has been
accomplished [2].

172
Figure A-2: The PIE foundry uses PUB bonding to attach Parylene C microelectrode
arrays directly to PCBs for compact packaging.
Another substrate that has been amenable to PUB bonding is Pt wire. A significant challenge
was establishing a ball bond on top of the wire itself. Pt wire (36 µm diameter) was flattened by
pressing it with the waffle tool against a non-metal surface. The wire was then secured to bond
pads using solder to minimize movement during ball bonding. Ball bonds were successfully
attached and then tamped down with the waffle tool. Bonding Parylene C devices with thin-film
Pt was straightforward, using the same PUB parameters as in previous work.

173
Figure A-3: Parylene C devices were directly PUB bonded to Pt wire, though soldering
the Pt wire to bond pads was necessary for gold ball bonding.
For finer pitch, PUB bonding will require a combination of thinner wire, smaller tool, and
possibly automated equipment. To investigate the first of these, preliminary work showed that
PUB bonding with 18 µm wire (down from 25 µm) is possible. Test devices were Parylene-PtParylene structures with a pair of connected contact pads. Both pads were PUB bonded to a gold
coated silicon wafer and then tested with a multimeter. Of four test structures, two were
successfully bonded, indicating that finer pitch bonding is feasible.

174
Figure A-4: Shorted contact pads were PUB bonded to gold. The remaining contact pads
were for practice. Scale bar is 0.5 mm
PUB bonding was also accomplished with thinner Parylene insulation. Using devices where
one side was < 5 µm, gold bumping was unchanged, but during the final PUB bonding step, we
placed a sheet of 10 µm between the device and waffle tool. This produced mechanical bonding,
but further studies are required to characterize the relationship between Parylene thickness and
PUB bond quality.
Lastly, we attempted to bond Parylene-Pt-Parylene devices to other Parylene-Pt-Parylene
devices. Pre-treatment of gold wire consisted of placing the wire atop a non-bonding surface
(e.g., FR4 substrate of a PCB) and using wedge bonding parameters with lower force and with
the waffle tool. Next, two Parylene devices sandwiched the gold wire so that Pt was in full
contact. Then ultrasonic energy was applied as in the usual PUB bonding. Preliminary results
showed excellent mechanical bonding; however, electrical continuity could not be ascertained
with these devices.

175
Figure A-5: Optical micrographs of a) a bird’s eye view and b) an oblique view of two
Parylene-Pt-Parylene devices that have been PUB bonded together. The PCB shown is
not bonded to the devices. Scale bars are 0.5 mm.
Indeed, this preliminary work goes to show that PUB bonding is a versatile interconnect that
functions wherever gold can be wire bonded. There is room for improving PUB bonding by
examining the effects of Parylene thickness and gold wire diameter. Though the circular TAB
tool punctured our devices likely due to its small footprint (490 um2
), there may be other
geometries (e.g., flat face, cross groove, etc.) with a footprint between 500 and 5000 um2
that
would fare better.

176
Appendix D: SILICON MICRONEEDLE FABRICATION
Recipe for buffered oxide etch (BOE):
1. Dehydrate wafers: heat wafer in oven 120 C for 15 minutes or more
2. Place in chamber with HMDS for 5 min
3. Spin AZ 4620 5 s acceleration, 45 s at 1800 RPM
4. Soft bake: 90 C, 10 min on hotplate
5. Rehydrate: 10 min
6. Expose 480 mW/cm2
7. Develop: 100 s in 1:4 AZ300
8. Hard bake: 90 C, 15 min
9. Place in 5:1 BOE, 10 min
10. Strip PR

177
Appendix E: EAB SENSOR FUNCTIONALIZATION PROCESS
1. Reduce the oxidized DNA Deprotecting
a. Prepare 10 mM tris(2-carboxyethyl)phosphine (TCEP) 10 mg TCEP:3.49 mL Millipore
water
b. Combine 16 µL of TCEP solution with 2 µL aptamer
c. Store in the dark 1 hour minimum
2. Electrochemically clean the electrode
a. Perform cyclic voltammetry (CV) on sensor in 0.5 M NaOH -1 to -1.8 V potential
1V/s scan rate
500 cycles
b. Perform CV on sensor in 0.5 M H2SO4 0 to 1.8 V potential
1 V/s scan rate
100 cycles
c. Roughen via chronoamperometry in H2SO4 0 to 2.2 V potential
3. Attach aptamer to electrode surface
a. Dilute DNA from 100 µM to 500 nM 18 µL DNA + 382 µL PBS
b. Rinse H2SO4 from sensors
c. Store in dark 1 hour minimum
4. Orient aptamer thiol ends to gold surface
a. Make 10 mM mercaptohexanol (MCH) 1.368 µL MCH:1 mL PBS
b. Transfer electrodes to MCH
c. Store in dark overnight

178
Appendix F: GOLD WIRE TESTING IN FERROCYANIDE
1. Prepare 5 mM ferrocyanide solution 63.35 mg potassium hexacyanoferrate(II) trihydrate
49.5 mg potassium hexacyanoferrate(III)
30 mL 1× PBS
2. Perform square-wave voltammetry using three-electrode setup V initial: -0.25 V
V final: 0.75 V
Step size: 1 mV
Amplitude: 25 mV
Quiet time: 2 s
Frequency: varies
3. Store ferrocyanide solution in the dark

179
Appendix G: EAB SENSOR TESTING IN VANCOMYCIN
1. Prepare 30 mL of PBS + 2 mM MgCl2 203.3 mg McCl2
0.5 L 1×PBS
2. Insert Ag/AgCl reference electrode, Pt counter electrode
3. Connect working electrode last
4. Prepare 0.01 M target solution 59.48 mg vancomycin
4 mL solvent
5. Run square wave voltammetry (SWV) V initial: -0.1 V
a. Include date, electrode ID, frequency, and concentration in the
file name
V final: -0.4 V
Step size: 1 mV
b. For vancomycin at room temperature, a strong “signal-off”
frequency is 10 Hz, and “signal-on” is 100 Hz
Amplitude: 25 mV
Quiet time: 2s
c. If there is no peak around -0.3 V, check for problems Frequency: varies
6. Add target solution in small quantities to obtain a titration curve See table
Table A-3: Titration amounts for target concentration
Concentration
Total (cumulative) vol
of 0.01 M target in
30 mL solvent
(µL)
Vol of 0.01 M target
added to beaker after
previous step
(8 titration points)
(µL)
Vol of 0.01 M target
added to beaker after
previous step
(5 titration points)
(µL)
0 M (no target) 0 0 0
10–6 M = 1 µM 3.0 3 3
10–5.5 M = 3.16 µM 9.49 6.49 -
10–5 M = 10 µM 30 20.5 27
10–4.5 M = 31.6 µM 94.9 64.9 -
10–4 M = 100 µM 300 205.1 273
10–3.5 M = 316 µM 949 648.7 -
10–3 M = 1000 µM 3000 2051.3 2727

Abstract (if available)

Abstract By bringing sensing and actuation to the microscale, microelectromechanical systems (MEMS) has profoundly shaped the modern world. Particularly in the biomedical sphere, MEMS sensors can operate in close proximity to the targets of choice, be they organs, cells, or biofluids. Still, there remain engineering challenges to the practical realization of these technologies into useable products. For neural interfaces, bulky electrical connections hamper higher resolution. For wearables such as microneedles, benchtop laboratory equipment must become portable. By focusing on electronics and packaging, future generations of MEMS devices can continue to improve lives.

In this work, chapter 1 discusses the role of MEMS in biological sensing and the challenges faced therein. Chapter 2 introduces bonding methods by which rigid electronic chips can be integrated with flexible polymer devices. Next, chapter 3 details the application of one such bonding method—polymer ultrasonic on bump (PUB)—to integrate a neural recording chip with a 64-channel Parylene C multielectrode array. Finally, chapter 4 captures the development of two key components of a wearable microneedle system: the microneedle array itself and the compact electronics necessary to take measurements.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Parylene-based biomems sensors for multiple physiological systems

PDF

Fabrication and packaging of three-dimensional Parylene C neural interfaces

PDF

Development of micromachined technologies for neural interfaces

PDF

Penetrating parylene neural probe array for dense, in vivo, chronic recordings

PDF

Bidirectional neural interfaces for neuroprosthetics

PDF

Battery-less detection and recording of tamper activity along with wireless interrogation

PDF

Parylene C bioMEMS for implantable devices with electrochemical interfaces

PDF

Development of implantable Parylene-based MEMS technologies for cortical applications

PDF

The electrochemical evaluation of Parylene-based electrodes for neural applications

PDF

Modular bio microelectromechanical systems (bioMEMS): intraocular drug delivery device and microfluidic interconnects

PDF

Implantable bioMEMS drug delivery systems

PDF

Parylene-based implantable interfaces for biomedical applications

PDF

Strategies for improving mechanical and biochemical interfaces between medical implants and tissue

PDF

A wireless implantable MEMS micropump system for site-specific anti-cancer drug delivery

PDF

High frequency ultrasonic phased array system and its applications

PDF

Zero-power sensing and processing with piezoelectric resonators

PDF

Liquid flow rate sensing via solution resistance for in vivo applications

PDF

Integrated wireless piezoelectric ultrasonic transducer system for biomedical applications

PDF

Highly integrated 2D ultrasonic arrays and electronics for modular large apertures

PDF

Biosensing and biomimetic electronics

Asset Metadata

Creator Yoo, James Jung (author)

Core Title High-resolution data acquisition with neural and dermal interfaces

School Andrew and Erna Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Degree Conferral Date 2024-05

Publication Date 03/19/2024

Defense Date 03/07/2024

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

Tag ASIC integration,electrochemical aptamer-based sensing,microneedles,miniature potentiostat,neural probes,OAI-PMH Harvest,Parylene C

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Meng, Ellis (committee chair), Kim, Eun Sok (committee member), Song, Dong (committee member), Zhou, Qifa (committee member)

Creator Email james.yoo@usc.edu,jamesjungyoo@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113857452

Unique identifier UC113857452

Identifier etd-YooJamesJu-12702.pdf (filename)

Legacy Identifier etd-YooJamesJu-12702

Document Type Dissertation

Format theses (aat)

Rights Yoo, James Jung

Internet Media Type application/pdf

Type texts

Source 20240325-usctheses-batch-1130 (batch), 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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

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

Repository Email cisadmin@lib.usc.edu