University of Southern California Dissertations and Theses

Catalytic oxidation and precursor identification of disinfection byproducts in recycled water

(USC Thesis Other)

Catalytic oxidation and precursor identification of disinfection byproducts in recycled water

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Catalytic Oxidation and Precursor Identification of
Disinfection Byproducts in Recycled Water
by
Euna Kim
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
(ENVIRONMENTAL ENGINEERING)
May 2024
Copyright 2024 Euna Kim

ii
Acknowledgements
Carrying a suitcase and leaving the beloved homeland to start anew, I believe all
international students share the same feelings. Will I truly find what I seek here? Can the time
spent apart from my family and friends be compensated by meaningful experiences? Looking
back now, I can assert that it was indeed a valuable investment. The person who contributed the
most to enabling me to have these experiences is my advisor, Dr. McCurry. Dan not only opened
the door for me to study at USC, but he also made every effort to ensure that my time here was
worthwhile. Countless meetings with him, often filled with lighthearted jokes, are some of my
happiest memories. Spending hours together discussing my data, I learned the joy of designing
the perfect puzzle piece to validate our hypothesis, coupled with the excitement of “adding a
touch of spice”. Sharing numerous moments with him was a great blessing for me. I also want to
express my gratitude to my dear labmate Marella Schammel, who always shared in our highs and
lows. Having a trustworthy friend during the PhD journey was truly a great fortune. BHE was
both a hub of learning and a haven for me. The opportunity to spend a chapter of my life in this
perfect place, surrounded by warm-hearted professors, their like-minded students, and
undergraduates was a profound privilege.
I am forever thankful to my parents for their unconditional support, fully aware that I
couldn't be who I am today without their love. Lastly, to my loving husband, Sunghyun Kim, and
our most wonderful creation, Aryne Kim: You consistently awaken me to the preciousness of
everyday happiness, fostering within me a profound gratitude for life. I am grateful for your
presence and hope that, as we embark on the next chapter of our lives, they remain as joyous as
they currently are.

iii
Table of Contents
Acknowledgements................................................................................................... ii
List of Tables .......................................................................................................... vii
List of Figures........................................................................................................ viii
List of Schemes....................................................................................................... xii
Abstract .................................................................................................................. xiii
CHAPTER 1. .............................................................................................................1
1 Introduction ............................................................................................................................................1
1.1 Background and Motivation............................................................................................... 1
1.2 Objectives and Scope of Work........................................................................................... 6
CHAPTER 2. ...........................................................................................................10
2 Catalytic Oxidation of Trace Aqueous Aldehydes with Ambient Dissolved Oxygen .....................10
2.1 Abstract ............................................................................................................................ 10
2.2 Introduction and Background........................................................................................... 11
2.3 Materials and Methods..................................................................................................... 13
2.3.1 Materials and Reagents.............................................................................................. 13
2.3.2 Catalyst Characterization........................................................................................... 14
2.3.3 Batch Experiments to Determine Aldehyde Oxidation Kinetics............................... 14
2.3.4 18O-labeled Water Experiments ................................................................................. 15
2.3.5 Reactant and Product Analysis .................................................................................. 16
2.3.6 Dissolved Oxygen Measurement ............................................................................... 19
2.3.7 Statistical Analysis..................................................................................................... 20
2.4 Results and Discussion..................................................................................................... 20
2.4.1 Catalyst Characterization........................................................................................... 20
2.4.2 Aldehyde Oxidation Reaction Rate Order and Rate Constants................................. 22

iv
2.4.2.1 Calculation of Hexanal Oxidation Rate Constant................................................ 28
2.4.2.2 Calculation of Reaction Rate Order of Dissolved Oxygen.................................. 29
2.4.3 Role of Dissolved Oxygen in Catalytic Oxidation of Aldehydes.............................. 30
2.4.4 Aldehyde Oxidation Mechanism by Pt/C in Aqueous Solution ................................ 35
2.5 Implications...................................................................................................................... 43
2.6 Acknowledgements.......................................................................................................... 43
CHAPTER 3. ...........................................................................................................44
3 Application of Heterogeneous Catalytic System to Real Wastewater.............................................44
3.1 Abstract ............................................................................................................................ 44
3.2 Introduction and Background........................................................................................... 45
3.3 Materials and Methods..................................................................................................... 46
3.3.1 Materials and Reagents.............................................................................................. 46
3.3.2 Sample Collection...................................................................................................... 47
3.3.3 Synthesis and Characterization of the Heterogeneous Catalysts............................... 48
3.3.4 Batch Reactor Experiments........................................................................................ 49
3.3.5 Flow-through Column Reactor Experiments............................................................. 49
3.3.6 Reactant and Product Analysis .................................................................................. 50
3.3.7 Statistical Analysis..................................................................................................... 51
3.4 Results and Discussion..................................................................................................... 51
3.4.1 Batch Experiments to understand the effect of ionic strength ................................... 51
3.4.2 Aldehyde Oxidation in a Flow-through Column ....................................................... 56
3.4.3 Characterization of Synthesized Catalyst .................................................................. 58
3.4.4 Formaldehyde Oxidation on Pt/GAC......................................................................... 59
3.5 Implications...................................................................................................................... 64
3.6 Acknowledgements.......................................................................................................... 64
CHAPTER 4. ...........................................................................................................65
4 Identifying Precursors of N-DBPs in Treated Wastewater by using Chemical Derivatization .......65

v
4.1 Abstract ............................................................................................................................ 65
4.2 Introduction and background ........................................................................................... 65
4.3 Materials and Methods..................................................................................................... 67
4.3.1 Materials and Reagents.............................................................................................. 67
4.3.2 Wastewater Sample Collection and Analysis............................................................ 69
4.3.3 Chemical Derivatization of Amines........................................................................... 69
4.3.4 Oxidation of Derivatized Amines and Controls to Measure Changes in DBP
Formation Potential............................................................................................................. 71
4.3.4.1 Standardization of ozone, free chlorine and chloramines.................................... 72
4.3.5 Product Analysis........................................................................................................ 73
4.3.5.1 Analytical Methods.............................................................................................. 74
4.4 Results and Discussion..................................................................................................... 75
4.4.1 Assessing Derivatization Efficiency with Model Amines......................................... 75
4.4.2 N-Nitrosodimethylamine Formation Potential of Derivatized Model Amines.......... 77
4.4.3 Haloacetonitrile Formation Potential of the Derivatized Amine Model Compounds 80
4.4.4 N-DBP Formation Potential of Derivatized Secondary Effluent............................... 82
4.5 Implications...................................................................................................................... 86
4.6 Acknowledgements.......................................................................................................... 87
CHAPTER 5. ...........................................................................................................88
5 Mechanistic Investigation and Kinetic Analysis of DBPs formation...............................................88
5.1 Abstract ............................................................................................................................ 88
5.2 Introduction and background ........................................................................................... 89
5.3 Materials and Methods..................................................................................................... 91
5.3.1 Analytical Details for NacTrp and Initial Byproducts............................................... 91
5.3.2 Chloramination Experiments of Nitromethane.......................................................... 91
5.3.3 Estimation of reaction rate constants for nitromethane chloramination reactions..... 92
5.3.3.1 Reaction Rate Laws for Nitromethane Chloramination Reactions ..................... 93
5.4 Results and Discussion..................................................................................................... 94

vi
5.4.1 High resolution MS analysis of novel DBPs............................................................. 94
5.4.2 Proposed Reaction Pathway..................................................................................... 104
5.4.3 Estimating Chloramination Rate Constants of Nitromethane.................................. 105
5.5 Implications.................................................................................................................... 111
5.6 Acknowledgements........................................................................................................ 112
CHAPTER 6. .........................................................................................................113
6 Conclusion .....................................................................................................................................113
References..............................................................................................................116

vii
List of Tables
Table 2.1. Chemical structure, calculated masses, measured masses, and mass errors of
DNPH-derivatized aldehydes measured by LC-IM-QTOF. ................................................. 18
Table 2.2. Chemical structures and masses of butyric acid and benzoic acid isotopes. .............. 18
Table 2.3. Retention times and mass transition used for butyraldehyde analysis........................ 19
Table 2.4. Calculated activation terms from data in Figure S18. The experiments were
conducted at 10, 25 and 40 °C in 10 mM pH 7 phosphate buffer with 400 mg/L of Pt/C
catalyst. ................................................................................................................................. 41
Table 3.1. List of chemical suppliers, abbreviations, and purities............................................... 46
Table 4.1. List of chemical suppliers, abbreviations, and purities............................................... 68
Table 4.2. Secondary effluent water quality data......................................................................... 69
Table 4.3. Chemical structures of amines and derivatives........................................................... 71

viii
List of Figures
Figure 1.1. Schematic diagram of indirect potable water reuse (IPR)........................................... 3
Figure 2.1. SEM images of Pt/C catalyst particles at magnification levels of: (a) 800,000×,
(b) 20,000×, and (c, d) 500,000× .......................................................................................... 20
Figure 2.2. The size distribution of a 0.4 g/L suspension of 5% Pt on C in Milli-Q water
measured by dynamic light scattering. The light scattering was measured after
equilibrating samples at 25 °C for 120 seconds with a refractive index of 1.63 and an
absorption of 0.001. .............................................................................................................. 21
Figure 2.3. STEM images of Pt/C catalyst particles at magnification levels of: (a) 250,000×,
(b) 120,000×, and (c) 800,000×, (d) TEM image of Pt/C in 800,000× ................................ 21
Figure 2.4. (a) Representative reaction profile for acetaldehyde oxidation including
acetaldehyde, acetic acid, and dissolved oxygen concentrations, and mass balance (sum
of aldehyde and acid). (b) Aldehyde oxidation first-order rate constants (h-1
) and rate
constants normalized by platinum atom concentration ([Pt]-1 s-1
)........................................ 23
Figure 2.5. Representative reaction profiles of replicate results for aldehyde oxidation
including aldehyde, carboxylic acid, and dissolved oxygen concentrations, and mass
balance (sum of aldehyde and acid concentrations). ............................................................ 23
Figure 2.6. Normalized logarithm of aldehyde concentrations in Figures 1 and S2 to
determine first-order rate constants. ..................................................................................... 24
Figure 2.7. Acetaldehyde oxidation to acetic acid under initial rate conditions (at three
different fixed aldehyde concentrations [panels a-c] to determine reaction rate order in
aldehyde. ............................................................................................................................... 27
Figure 2.8. Aldehyde oxidation to respective products under initial rate conditions at four
different catalyst doses to determine reaction rate order in the available Pt surface............ 28
Figure 2.9. Reaction profiles for aldehyde oxidation including aldehyde, carboxylic acid, and
dissolved oxygen concentrations, and mass balance (sum of aldehyde and acid
concentrations)...................................................................................................................... 31
Figure 2.10. (a) Concentrations of butyraldehyde and dissolved O2 in N2-purged solution. (b)
Normalized natural logarithm of butyraldehyde concentrations over time to determine
first-order rate constants of oxidation reactions in N2-purged solution................................ 31
Figure 2.11. (a) Reaction profiles for aldehyde oxidation including aldehyde, carboxylic
acid, and dissolved oxygen concentrations, and mass balance (sum of aldehyde and acid
concentrations), at intermediate initial aldehyde concentration. (b) Normalized
logarithm of aldehyde concentrations to determine first-order rate constant of oxidation
reactions. ............................................................................................................................... 32

ix
Figure 2.12. Aldehyde oxidation to respective products under initial rate conditions (at three
different fixed oxygen concentrations [panels (a)-(c)]) to determine reaction rate order
in oxygen [panel (d)]............................................................................................................. 33
Figure 2.13. Normalized logarithm of butyraldehyde concentrations to determine first-order
rate constant of oxidation reactions. ..................................................................................... 34
Figure 2.14. Acetaldehyde oxidation rate constants in 10 mM buffer solution (citrate for pH
4, 5, and 6; phosphate for pH 7, 8, and 12; and carbonate for pH 9, 10, and 11)................. 35
Figure 2.15. Normalized logarithm of aldehyde concentrations to determine first-order rate
constant of oxidation reactions at different pH values.......................................................... 36
Figure 2.16. Isotopic distribution of acid products from butyraldehyde and benzaldehyde
oxidation. .............................................................................................................................. 37
Figure 2.17. The ratio of 18O-labeled butyraldehyde (C4H8
18O) to total butyraldehyde
(C4H8
18O + C4H8
16O) in 1 mL of 5% v/v of 18O-labeled heavy water (H2
18O) at two
equilibration time intervals (2 hours and 5 days); T = 24 ± 0.5 °C. .................................... 39
Figure 2.18. (a) Reaction profiles for benzaldehyde oxidation including benzaldehyde,
benzoic acid, and dissolved oxygen concentrations, and mass balance (sum of aldehyde
and acid concentrations). (b) Normalized logarithm of aldehyde concentrations to
determine first-order rate constant of oxidation reactions. ................................................... 40
Figure 2.19. Normalized logarithm of aldehyde concentrations to determine first-order rate
constant of oxidation reactions at different temperature (Panels a-c)................................... 41
Figure 3.1. Schematic diagram of synthesizing Pt/GAC catalysts by using wet impregnation
techniques. ............................................................................................................................ 48
Figure 3.2. Schematic diagram of flow-through column experiment setting having a length
of 70 mm and an inner diameter of 5.6 mm.......................................................................... 50
Figure 3.3. (a-c) Normalized natural logarithm of formaldehyde concentrations over time to
determine first-order rate constants of oxidation reactions at different buffer
concentrations. ...................................................................................................................... 52
Figure 3.4. (a-c) Normalized natural logarithm of aldehyde concentrations to determine firstorder rate constant of oxidation reactions at different sodium chloride concentrations....... 53
Figure 3.5. (a-c) Normalized natural logarithm of aldehyde concentrations to determine firstorder rate constant of oxidation reactions at different sodium chloride concentrations....... 54
Figure 3.6. Normalized natural logarithm of aldehyde concentrations to determine first-order
rate constant of oxidation reactions in Milli-Q water. .......................................................... 55
Figure 3.7. The average first-order rate constant of oxidation reactions in various constitutes;
NaCl (5, 10, and 20 mM) with 10 mM phosphate buffer, NaNO3 (5, 10, 25 mM) with 10
mM phosphate buffer, phosphate buffer only (1, 5, and 10 mM), and Milli-Q water.......... 55
Figure 3.8. Reaction profile of flow-through reactor for aldehyde oxidation including
formaldehyde, formic acid, and dissolved oxygen concentrations and mass balance.......... 57

x
Figure 3.9. Concentration of platinum in effluent of flow-through column reactor.................... 57
Figure 3.10. Reaction profile of control (catalyst-free) flow-through reactor for aldehyde
oxidation including formaldehyde, formic acid, and dissolved oxygen concentrations,
and mass balance. ................................................................................................................. 58
Figure 3.11. SEM images of Pt/GAC catalyst particles at magnification levels of: (a)150×,
and (b) 2,000×....................................................................................................................... 59
Figure 3.12. (a) SEM images of Pt/GAC catalyst particles at magnification of 650×, and (b)
EDS image of the identical area shown in the SEM image, where yellow marking
signify the presence of platinum........................................................................................... 59
Figure 3.13. (a) Reaction profile of flow-through reactor for formaldehyde oxidation. (b)
Product analysis showing the formaldehyde, formic acid, and inorganic carbon
concentrations. ...................................................................................................................... 61
Figure 3.14. (a) Reaction profile of flow-through reactor for formate oxidation with Pt/GAC
catalyst and initial formate concentration of 500 μM. (b) Reaction profile of
formaldehyde oxidation including formaldehyde, formate and inorganic carbon with 40
mg/100 mL of Pt/GAC catalyst and initial formaldehyde concentration of 500 μM. .......... 62
Figure 3.15. Formate oxidation rate constants in 10 mM buffer solution (citrate for pH 3, 4,
5, and 6, and phosphate for pH 2, and 7). Error bars indicate the range of values from
duplicate experiments. .......................................................................................................... 62
Figure 3.16. Normalized logarithm of formate concentration to determine first-order rate
constants at different pH values............................................................................................ 63
Figure 4.1. Nitromethane formation potential changes after derivatization of model amines
with ortho-phthaldehyde (OPA), di-tert-butyl dicarbonate (Boc2O) and benzyl bromide
(BB) and subsequent ozonation: (a) benzyl amine (BA), (b) N-benzylmethylamine
(BMA), and (c) N,N-dimethylbenzylamine (BDMA). ......................................................... 76
Figure 4.2. Nitromethane formation potential changes after derivatization of dimethylamine
(DMA) with different reagents and subsequent ozonation................................................... 77
Figure 4.3. Nitromethane formation potential changes after derivatization of model amines
with ortho-phthaldehyde (OPA), di-tert-butyl dicarbonate (Boc2O) and benzyl bromide
(BB) and subsequent ozonation: (a) methyl amine (MA), (b) dimethylamine (DMA),
and (c) trimethylamine (TMA). ............................................................................................ 77
Figure 4.4. N-nitrosodimethylamine (NDMA) formation potential of N,Ndimethylbenzylamine with added NaBr as a control and derivatized N,Ndimethylbenzylamine with BB. ............................................................................................ 79
Figure 4.5. N-nitrosodimethylamine (NDMA) formation potential of N,Ndimethylbenzylamine and derivatized N,N-dimethylbenzylamine....................................... 79
Figure 4.6. N-nitrosodimethylamine (NDMA) formation potential of dimethylamine (DMA)
and derivatized DMA............................................................................................................ 80

xi
Figure 4.7. DCAN formation potential changes after derivatization of β-alanine (ALN),
phenethylamine (PEA), and tyramine (TRM) with OPA and subsequent chloramination
by either (a) 14N-chloramines or (b) 15N-chloramines. ......................................................... 81
Figure 4.8. DCAN formation potential changes of OPA and OPA-derivatized methylamine
after 15N-chloramination. ...................................................................................................... 82
Figure 4.9. Normalized nitromethane formation potential changes of secondary effluents
from three anonymous plants after being derivatized by OPA, Boc2O and BB................... 85
Figure 4.10. Normalized 15N-NDMA formation potential changes of secondary effluents
after being derivatized by OPA, Boc2O and BB................................................................... 85
Figure 4.11. Normalized 14N-DCAN formation potential changes of secondary effluents after
being derivatized by OPA, Boc2O and BB. .......................................................................... 86
Figure 5.1. Structures of parent and fragment ions for two proposed products with 261.0881
m/z that could form by anti-Markovnikov addition to the indole ring. Blue lines provide
fragmentation cleavage points to produce fragment ions. The fragments below each
structure correspond to the proposed structures of fragments with exact mass 131.0371
Da.......................................................................................................................................... 96
Figure 5.2. Proposed reaction pathway for the production of novel DBPs from NacTrp.
Circled letters correspond to specific products discussed in the text. .................................. 97
Figure 5.3. Dilution of the 261 m/z products from 50% acetonitrile/50% water ten-fold in
water increases the relative importance of 261-1 vs. 261-2, suggesting that the increase
in the solvent dielectric constant promotes conversion of the lactim tautomer to the
lactam.................................................................................................................................... 98
Figure 5.4. Structures of parent and fragment ions for two proposed products with 277.0380
m/z. ..................................................................................................................................... 100
Figure 5.5. Dilution of the 277 m/z products from 50% acetonitrile/50% water ten-fold in
water increases the relative importance of 277-1 vs. 277-2, suggesting that the increase
in the solvent dielectric constant promotes conversion of the Lactim tautomer to the
Lactam. ............................................................................................................................... 101
Figure 5.6. Structure of 2-acetamido-4-(2-formamidophenyl)-4-oxobutanoate, proposed to
be the parent compound for the product with 277.0380 m/z, corresponding to 277-3....... 102
Figure 5.7. Structures of parent and fragment ions for the proposed product with 329.0101
m/z. ..................................................................................................................................... 103
Figure 5.8. Total nitromethane decay (measured by GC/MS/MS) during chloramination, and
the formation of nitrate and nitrite (measured by IC) at pH values of 7, 9 and 11............. 107
Figure 5.9. Comparison of estimated concentrations of total nitromethane, nitrate, and nitrite
by the rate model to experimentally measured values........................................................ 109
Figure 5.10. Predicted concentrations of nitromethane and nitromethyl anion during
chloramination reactions depicted in Fig. 5.9..................................................................... 110

xii
List of Schemes
Scheme 1: Aldehyde equilibrium with hydrated gem-diol........................................................... 38
Scheme 2.2. Proposed aldehyde oxidation reaction pathway through β-hydride elimination. .... 42
Scheme 2.3. Proposed aldehyde oxidation reaction pathway through β-hydrogen atom
abstraction............................................................................................................................. 42
Scheme 5.1. Proposed Pathways of Reactions of Nitromethane with Chloramines.................. 107

xiii
Abstract
Amidst increasing water scarcity leading to the expansion of water reuse, this research
addresses critical aspects of water purification technology, particularly focusing on recycled
wastewater. The study first demonstrates the oxidation of trace aqueous aldehydes to their
corresponding acids using a heterogeneous catalyst (5% Pt on C) with ambient dissolved oxygen.
It reveals that low molecular weight aldehydes, significant in the organic carbon pool of recycled
wastewater, are efficiently converted, with the process primarily proceeding through a basepromoted beta-hydride elimination mechanism. Building upon these findings, the research
further delves into the efficient abatement of aldehydes, such as formaldehyde, using
heterogeneous Pt/GAC catalysts in a flow-through column reactor. The study investigates the
influence of various parameters like ionic strength and salt concentrations on the oxidation
process and assesses the scalability of this treatment method, using real wastewater samples to
validate conditions for future water reuse plants. In addition to aldehyde treatment, the study also
explores the formation of nitrogenous disinfection byproducts (N-DBPs) in recycled wastewater,
which pose a significant risk to human health. Utilizing chemical derivatization techniques, the
research evaluates how different amines contribute to N-DBP formation in secondary effluents
from wastewater treatment plants by transforming amines into less reactive compounds. The
effectiveness of this approach is validated using model amines, with findings showing that
primary and secondary amines contribute differently to the formation of N-DBPs such as
nitromethane, NDMA, and DCAN. Advancing our foundational knowledge, the study delves
into the formation of initial chlorine transformation products from amino acids and peptides,
hypothesizing that these lead to higher yields of nitrogenous DBPs compared to small molecules,

xiv
as demonstrated with N-acetyltryptophan (NacTrp). Additionally, the research studies on the
kinetics of halonitromethane formation, particularly nitromethane, to better understand
chloramination processes. This approach broadens our understanding of DBP formation,
contributing to the development of safer and more sustainable water treatment methods. Overall,
this comprehensive research emphasizes the need for customized wastewater treatment methods
to reduce these harmful byproducts, thereby enhancing water safety and compliance with
regulatory standards, especially in systems with diverse secondary effluent types. Significantly
contributing to the field of environmental engineering, by providing innovative solutions to
water treatment challenges, paving the way for safer and more sustainable water reuse practices
in the face of global water scarcity.

1
CHAPTER 1.
1 Introduction
1.1 Background and Motivation
A warming climate presents escalating risks globally, including in the United States,
where it affects life, property, economies, and ecosystems. These changes manifest in altered
weather patterns, melting ice, and variable water resources, impacting natural ecosystems and
amplifying extreme events.1 In urban areas, particularly large cities, these shifts exacerbate
existing challenges in water management. Such cities, already grappling with demands that often
surpass local precipitation levels, are now confronting intensified climate change-induced water
scarcity.2 Facing the reality of climate change, many large cities are increasingly reliant on water
sourced from faraway rivers and reservoirs, often situated hundreds of miles away. This reliance
brings significant economic and environmental costs, including the need for substantial
electricity to power pumps and maintain extensive infrastructure.3
For example, San Diego
obtains nearly 90% of its fresh water from the Colorado River, necessitating transport across
Southern California. This river is a shared resource for seven states and Mexico.2 With the city's
population growing rapidly and no additional water supply from the Colorado River, San Diego
is compelled to seek sustainable alternatives to satisfy its burgeoning water demands. This
scenario is prompting numerous cities to consider the potential of treating municipal wastewater
as a new source of fresh water.

2
Municipal wastewater typically undergoes a two-stage treatment process: primary
treatment, which focuses on sedimentation to remove solid substances, and secondary treatment,
which involves biological oxidation to reduce oxygen-depleting compounds before discharge
into the environment.4 Some utilities recycle a portion of this water for agricultural irrigation and
industrial use, and this secondary effluent contains a large amount of natural and anthropogenic
compounds.5,6 Therefore, it goes through several additional treatment processes through
advanced treatment, especially if being used for replenishment of ground water for use as
drinking water.7,8 Treatment steps vary depending on the facility, but most employ a combination
of 1) nano/microfiltration, 2) reverse-osmosis (RO) membranes, 3) UV photolysis of H2O2 to
generate hydroxyl radical as advanced oxidation processes (AOPs), and sometimes
ozone/biological activated carbon (O3/BAC) is added before nano/microfiltration.9 However,
even after passing through these multiple treatment steps, about 100 µg/L organic carbon is
found in typical final recycled water.10
Despite many studies to identify low molecular weight compounds in drinking water,
until recently, less than 35% of dissolved organic carbon in RO permeate was found.9 One
fraction, disinfection byproducts including trihalomethanes, haloacetic acids, and Nnitrosodimethylamine, accounts for approximately 5–10% of the TOC in the final effluent.16
Little of the remaining unknown carbon had been identified, until a recent study demonstrated
that carbonyl compounds account for 19–38% of the remaining DOC in recycled wastewater,
most of which were saturated and unsaturated aldehydes.9 Formaldehyde, which accounts for
most of the aldehydes found in finished recycled wastewater, is polar and hydrophilic, making
that the removal by RO is only 9% and the UV/H2O2 process does not remove formaldehyde at

3
all at neutral pH.17 Studies on the toxicity of aldehydes are being actively conducted, among
which saturated aldehyde is known to cause changes in the gene expression by randomly
alkylating DNA and mRNA.18 Although unsaturated aldehydes exist in relatively small
concentrations, they are very toxic and pose a threat to public health.19 These aldehydes were
detected in the full-scale advanced treatment plants,9 indicating that current treatment
technologies are unable to effectively degrade these compounds, and an additional process is
required to complement the existing treatment system.
To prevent the potential toxicity of these organic compounds from adversely affecting
human health, Orange County Water District (OCWD) recharges the reclaimed water to
groundwater through groundwater replenishment system (GWRS) operation to do indirect
potable reuse (IPR) using an environmental buffer, with the hope that passing water through the
ground will provide additional treatment benefits (Figure 1.1).8,11,12 Although this project has a
great advantage in terms of improving public acceptance, there is a growing interest in direct
potable reuse (DPR), in which treated water is transferred directly to drinking water treatment
facilities without going through environmental buffers.10,13,14
Figure 1.1. Schematic diagram of indirect potable water reuse (IPR)

4
The State Water Board’s Division of Drinking Water (DDW) is at the forefront of
implementing a series of legislative mandates pertaining to DPR. DPR involves the intentional
addition of recycled water directly into a public drinking water system (Figure 1.2), or into a raw
water supply just before a drinking water treatment plant.15 Originating from the former
California Department of Public Health’s Drinking Water Program, the DDW has embarked on
an exploration to develop uniform water recycling criteria for DPR. This investigation
culminated in December 2016 with a comprehensive report submitted to the Legislature,
detailing findings and recommendations. In 2017, the DDW was charged with the development
of uniform water recycling criteria for DPR, aimed at safeguarding public health. The State
Water Board is required to establish uniform water recycling criteria for DPR through raw water
augmentation by December 31, 2023. This involves utilizing information from the DPR research
outlined in the State Water Board’s Report to the Legislature and incorporating feedback from
stakeholders.
Figure 1.2. Schematic diagram of direct potable water reuse (DPR) with each treatment trains of
tertiary treatment
In the context of these legislative efforts, Section 64669.50, titled "Chemical Control,"
delineates a comprehensive framework for the continuous treatment of municipal wastewater in
Direct Potable Reuse (DPR) projects prior to its integration into a public water distribution

5
system. This regulation mandates the implementation of a treatment train, comprised of no fewer
than three distinct treatment processes, each utilizing a unique mechanism for chemical
reduction. The specified treatment train includes: (1) an ozonation process followed by
biologically activated carbon (ozone/BAC), subject to exemption under delineated conditions;
(2) a reverse osmosis membrane process; and (3) an advanced oxidation process. The designated
sequence of these processes is systematically arranged as ozone/BAC followed by reverse
osmosis, culminating in advanced oxidation. Elaborating further, the regulation stipulates
stringent design and validation requirements for the ozone/BAC process. The ozonation process
is required to achieve a minimum 1.0 log reduction of specified indicators such as
carbamazepine and sulfamethoxazole, while the BAC must be designed with a specified empty
bed contact time, similarly aiming for a 1.0 log reduction of indicators including formaldehyde
and acetone. These processes are required to undergo individual validation to ascertain their
efficacy and reliability in full-scale operations. Additionally, the reverse osmosis membrane is
subject to rigorous performance standards, encompassing sodium chloride rejection rates and
total organic carbon (TOC) concentration limits. The advanced oxidation process is mandated to
provide a minimum of 0.5 log reduction of the indicator 1,4-dioxane. Integral to the operational
protocol, the regulation necessitates continuous monitoring and documentation of various
operational parameters, alongside established critical limits to ensure the treatment processes'
effectiveness and compliance. In the context of advancing water treatment methodologies,
investigating the most effective strategies for the removal of Disinfection Byproducts (DBPs)
and identifying their precursors is crucial. This is imperative to enhance the safety and efficacy

6
of water treatment processes, ensuring the provision of high-quality potable water in line with
public health standards.
1.2 Objectives and Scope of Work
Chapter 2 is centered around the objective of oxidizing trace organic contaminants in
recycled water, specifically focusing on using heterogeneous catalysts and dissolved oxygen.
The chapter aims to delve into both mechanistic and kinetic analyses to understand and optimize
the rate-limiting steps involved in the catalytic transformation process. The initial goal is to
assess the feasibility of applying commercially available catalysts under mild environmental
conditions (25 ˚C, pH 6-8) for the oxidation of aldehydes that have been recently identified in
recycled wastewater. The study specifically targets six saturated aliphatic aldehydes, including
formaldehyde and acetaldehyde, and one unsaturated aliphatic aldehyde, crotonaldehyde, to
investigate the ability of these catalysts to oxidize these aldehydes under conditions that closely
resemble those in water treatment processes. An integral part of the study involves the use of
isotopic labeling experiments to discern the role of dissolved oxygen in the reaction, particularly
focusing on tracking the origin of the oxygen atom in the reaction products. These experiments
are crucial in revealing the transformation pathway of the contaminants and identifying the ratelimiting step, which is essential for further optimization of the process. Moreover, the chapter
sets out to design a proof-of-concept experiment to demonstrate the feasibility of the "Catalytic
Converters for Water Treatment" concept in a bench-scale flow-through reactor. This involves
applying the catalytic system to real water samples collected from advanced treatment facilities,
aiming to validate the practical applicability of the system and providing insights necessary for

7
optimizing and designing the process for scaling up. Through these objectives, the chapter
contributes to advancing the field of water treatment, focusing on the removal of harmful trace
contaminants from recycled wastewater using innovative catalytic methods.
Chapter 3 focuses on addressing the challenge of efficiently removing aldehydes,
especially formaldehyde, from recycled wastewater. The chapter is driven by the increasing
interest in potable water reuse due to intensifying water scarcity and the recognition that a
substantial portion of organic carbon in product water from water reuse plants comprises
carbonyl compounds, predominantly aldehydes. These compounds, due to their polarity, low
molecular weight, and neutral charge, are not easily removed by reverse osmosis (RO) or
UV/H2O2 processes. Building on recent findings that a heterogeneous catalyst (5% Pt on C) can
effectively convert aldehydes to carboxylic acids using ambient dissolved oxygen as an oxidizing
agent, this study aims to advance the treatment of formaldehyde using heterogeneous Pt/GAC
catalysts in a flow-through column reactor. The research expands upon prior work showing
significant conversion of formaldehyde to harmless carbonate species. A key aspect of this study
is exploring the influence of various parameters, such as ionic strength and salt concentrations,
on the effectiveness of the oxidation process. The scope of work also encompasses the scalability
of the aldehyde treatment method. The study plans to bridge the gap between bench-scale
experiments and real-world application by transitioning to a pilot-scale system. This progression
is crucial for ensuring the practical viability of the treatment method. Real wastewater samples
will be employed to validate and optimize the conditions for future water reuse plants, ensuring
that the findings are directly applicable to real-world scenarios. The expected outcomes of this
chapter are significant advancements in water purification technology, particularly for advanced

8
water treatment facilities. By optimizing the catalytic conversion of harmful aldehydes in
recycled wastewater, this research aims to contribute to the development of more efficient and
sustainable water treatment solutions, addressing the critical issue of water scarcity and the need
for reliable potable water reuse systems.
Chapter 4 of the study delves into a detailed exploration of the mechanisms behind
nitrogenous disinfection byproduct (N-DBP) formation in wastewater treatment, a crucial issue
in environmental chemistry and public health. The chapter is framed around three main
objectives, each contributing to a comprehensive understanding of N-DBP formation and
potential mitigation strategies. The first objective centers on the identification and adaptation of
chemical derivatization techniques from the field of synthetic chemistry. This process is vital as
it targets the selective protection of functional groups that are known to play a key role in the
formation of N-DBPs such as NDMA (N-nitrosodimethylamine), halonitromethane, and
haloacetonitrile. The focus is specifically on amines of various orders—primary (1º), secondary
(2º), and tertiary (3º)—which are believed to be significant contributors to N-DBP formation.
The second goal of the chapter is to thoroughly validate these derivatization techniques. This
involves conducting rigorous experiments to measure the reduction in N-DBP formation
potential following the derivatization of model amines. The purpose of this step is to ensure the
methods are not only effective but also reliable for application in further studies. It serves as a
foundational element in establishing the efficacy of the derivatization process in altering the
reactivity of amines, thus potentially influencing N-DBP formation. Finally, the chapter aims to
apply an orthogonal derivatization strategy to real-world scenarios by quantifying the
contributions of different amines to the precursor pool of N-DBPs in actual wastewater effluent.

9
This comprehensive approach is expected to yield valuable insights into the specific roles that
various amines play in N-DBP formation. By elucidating these roles, the study seeks to guide the
development of more targeted and effective wastewater treatment strategies. These strategies are
aimed at mitigating the formation of N-DBPs, thereby addressing a significant environmental
and public health concern. The outcomes of this research have the potential to significantly
advance our understanding of N-DBP chemistry and inform the design of safer and more
efficient water treatment processes.
Chapter 5 addresses the complexities of disinfection byproduct (DBP) formation in water
treatment processes with a dual objective approach. The first part focuses on the mechanistic
exploration of initial chlorine transformation products, particularly amino acids and peptides as
key precursors for nitrogenous DBPs, hypothesizing that these precursors yield higher DBP
concentrations compared to small molecule counterparts. This is explored through the
investigation of N-acetyltryptophan (NacTrp) as a model for peptide-bound tryptophan, aiming
to characterize its initial chlorine transformation products as an alternative method for
identifying high-yield DBPs. The second part of the study delves into the kinetic analysis of
halonitromethane formation, especially nitromethane, recognized as a crucial intermediate in
chloropicrin formation in scenarios involving ozonation followed by chlorination. This involves
elucidating the chloramination reaction mechanism of nitromethane and determining the
interaction dynamics of monochloramine and dichloramine with nitromethane. The study
combines these mechanistic and kinetic perspectives to enhance the understanding of DBP
formation and contribute to more effective water treatment and mitigation strategies.

10
CHAPTER 2.
2 Catalytic Oxidation of Trace Aqueous Aldehydes with Ambient Dissolved
Oxygen
This chapter includes work published at the journal Environmental Science and Technology.
E. Kim, G. B. Cardosa, K. E. Stanley, T. J. Williams, D. L. McCurry* (2022) Out of Thin Air? Catalytic
Oxidation of Trace Aqueous Aldehydes with Ambient Dissolved Oxygen. Environ. Sci. Tech. 56, 12,
8756-8764
2.1 Abstract
Water reuse is expanding due to increased water scarcity. Water reuse facilities treat
wastewater effluent to a very high purity level, typically resulting in a product water that is
essentially deionized water, often containing less than 100 µg/L organic carbon. However, recent
research has found that low molecular weight aldehydes, which are toxic electrophiles, comprise
a significant fraction of the final organic carbon pool in recycled wastewater in certain treatment
configurations. In this manuscript, we demonstrate oxidation of trace aqueous aldehydes to their
corresponding acids using a heterogeneous catalyst (5% Pt on C), with ambient dissolved oxygen
serving as the terminal electron acceptor. Mass balances were essentially quantitative across a
range of aldehydes, and pseudo-first order reaction kinetics are observed in batch reactors, with
kobs varying from 0.6 h-1 for acetaldehyde to 4.6 h-1 for hexanal, while they were low for
unsaturated aldehydes. Through kinetic and isotopic labeling experiments, we demonstrate that
while oxygen is essential for the reaction to proceed, it is not involved in the rate-limiting step,
and the reaction appears to proceed primarily through a base-promoted β-hydride elimination

11
mechanism from the hydrated gem-diol form of the corresponding aldehyde. This is the first
report we are aware of that demonstrates useful abiotic oxidation of a trace organic contaminant
using dissolved oxygen.
2.2 Introduction and Background
Increased water scarcity has motivated greater adoption of wastewater recycling to
augment potable water sources.1,2,20 The risk from pathogens and hazardous chemicals in
wastewater is mitigated by the use of multiple treatment steps including pre-oxidation (typically
with ozone or chlorine), reverse osmosis (RO), and advanced oxidation processes (AOPs).10
Despite the high level of treatment used during wastewater recycling, recent studies have
documented the presence of recalcitrant organic carbon in the final effluent, and the majority of
these chemicals have not yet been fully identified.9,21,22 Furthermore, some water reuse
operations have considered replacing RO-based treatment trains with ozone followed by
biological activated carbon (O3/BAC) due to treatment and disposal concerns for RO
concentrates,7,23–25 however higher total organic carbon (TOC) concentrations are expected in the
effluent of the O3/BAC process compared to current practice (e.g., 2.5 mg/L for O3/BAC and 0.2
mg/L for RO).9
Despite many studies on the low molecular weight compounds present in potable reuse
product water, until recently 35% of the dissolved organic carbon in reverse osmosis permeate
had been characterized.9,11,16 One fraction, disinfection byproducts including trihalomethanes,
haloacetic acids, and N-nitrosodimethylamine, accounts for approximately 5–10% of the TOC in
the final effluent.16 Little of the remaining unknown carbon had been identified, until a recent

12
study demonstrated that carbonyl compounds account for 19–38% of the remaining DOC in
recycled wastewater, most of which were saturated and unsaturated aldehydes.9 These aldehydes
are toxic electrophiles18,26 and are difficult to remove with existing reverse osmosis and
advanced oxidation treatment systems due to their polarity, low molecular weight, and neutral
charge.17,19,27–31
Catalytic transformation of contaminants with environmentally-relevant, mild, and
scalable conditions is a subject of considerable research interest.32–36 Many studies using
heterogeneous catalysts for environmental purposes in water have focused on the reduction of
inorganic contaminants such as bromate, nitrate, and perchlorate32,37,38 or hydrodehalogenation of
halogenated organics.38–40 Oxidative catalysis has been applied to carry out degradation of
contaminants on an electrode, or on a membrane surface with the addition of strong electron
acceptor such as hydrogen peroxide or persulfate.41–46 Photocatalytic oxidative degradation of
micropollutants, largely with TiO2, has focused on oxidation by surface reactions and generation
of hydroxyl radical initiated by UV irradiation.47–51 However, catalytic oxidation of organic
pollutants in water without electrochemistry, addition of electron acceptors, or photochemistry,
each of which challenge the potential scalability of the respective methods, has not been
demonstrated to the best of our knowledge.
In this study, we sought to extend the concept of “Catalytic Converters for Water
Treatment”36 from reductive treatment of oxyanions to oxidation of organic pollutants, by
oxidizing contaminants with dissolved molecular oxygen as the terminal electron acceptor. We
first identified a catalyst in the organic synthesis literature reported to be capable of oxidizing
aqueous alcohols and aldehydes to their corresponding acids in the presence of dissolved oxygen

13
and aimed to identify its substrate scope in batch experiments. We determined kinetic parameters
for oxidation of a range of low molecular weight aldehydes, aimed to elucidate the reaction
mechanism, in particular the role of dissolved oxygen. This study takes a first step toward
oxidation of trace aldehyde compounds in final recycled water effluent to generally non-toxic
organic acids, using heterogeneous catalysts and ambient molecular oxygen.
2.3 Materials and Methods
2.3.1 Materials and Reagents
Deuterium oxide (99.9%) was purchased from Cambridge Isotope Laboratories, Inc.
(Andover, MA). Dibasic (extra pure) and monobasic (99%) potassium phosphate were purchased
from Acros Organics (New Jersey, USA). Starch (99+%) was purchased from Alfa Aesar (Ward
Hill, MA). Sodium bicarbonate (99.7-100.3%) and sulfuric acid (95-98%), acetonitrile and
methanol (HPLC grade) were purchased from EMD Millipore Corporation (Burlington, MA).
Sodium iodide was purchased from Fisher Scientific (Geel, Belgium). Technical grade citric acid
was purchased from PTI Process Chemicals (Ringwood, IL). Pentanal (valeraldehyde) (97%),
2,4-dinitrophenylhydrazine (2,4-DNPH) (97%), hexanal (98%), crotonic acid (98%),
acetaldehyde (99%), sodium thiosulfate (99%), propionaldehyde (98%), boric acid (99.5%),
butyraldehyde (99.5%), propionic acid (99.5%), sodium azide (99.5%), crotonaldehyde
(predominantly trans) (99%), acetic acid (99%), butyric acid (99%), valeric acid (99%), and
hexanoic acid (99%), magnesium sulfate (97%), formaldehyde (100 µg mL in H2O), 5 wt.%
platinum on carbon, and 18O-water (97 atom% 18O) were purchased from Sigma Aldrich (St.

14
Louis, MO). Formic acid (99+%) was purchased from Thermo Scientific (Rockford, IL). 5 wt.%
platinum on carbon catalyst (Sigma Aldrich (St. Louis, MO)).
2.3.2 Catalyst Characterization
The morphology of the catalyst supporter was observed by scanning electron microscope
(SEM) using Thermo Scientific Helios G4 PFIB UXe, and the size distribution of catalyst
particles was measured by using dynamic light scattering (DLS) with a Zetasizer Ultra (Malvern
Instruments Ltd., UK). A 4-mL of 0.4 g/L suspension of Pt/C was prepared in Milli-Q water and
transferred to a polystryrol/polystyrene cuvette for analysis. The light scattering was measured
after equilibrating samples at 25 °C for 120 seconds with a refractive index of 1.63 and an
absorption of 0.001. The size and dispersion of the platinum supported on the carbon was
characterized by measuring scanning transmission electron microscopy (STEM) and
transmission electron microscopy (TEM) using JEOL 2100F at 200 kV with high angle annular
dark field detector.
2.3.3 Batch Experiments to Determine Aldehyde Oxidation Kinetics
Batch experiments were performed in 100 mL glass syringes to avoid creating headspace
while sampling periodically, because several target compounds were semi-volatile and mass
balance was not conserved in preliminary tests in open reactors. Within the syringes, 40 mg of
heterogeneous 5% Pt on C catalyst particles were suspended in 99 mL of 10 mM buffer solution
adjusted to pH 4–12: citrate for pH 4, 5 and 6, phosphate for pH 7, 8 and 12, and carbonate for
pH 9, 10 and 11. After adding a magnetic stir bar, the syringe was fitted tightly with a glass
plunger and secured to a stir plate. Once the catalyst was dispersed homogeneously in the

15
syringe, 1 mL of concentrated aqueous aldehyde solution was injected to initiate the reaction.
Aliquots of 4.5 mL were withdrawn periodically and filtered with a syringe filter (0.2 μm,
PTFE).
2.3.4 18O-labeled Water Experiments
Batch oxidation reactions were conducted in 18O-labeled water (H2
18O) to differentiate
water-derived oxygen atoms in the reaction products from O2-derived oxygen atoms. In 2 mL
HPLC vials, 1.6 mg of the catalyst was added along with 400 μL of 18O-labeled water containing
97 atom % 18O. To initiate the reaction, 0.36 μL of pure butyraldehyde or 0.41 μL of pure
benzaldehyde ([aldehyde]0 = 10 mM) was added to the vials along with a small stir bar. The
solutions were then stirred for six hours and filtered through a 0.2 µm syringe filter before
analysis. The hydration equilibrium of butyraldehyde was evaluated with 5% v/v of 18O-labeled
water (H2
18O) in 2 mL HPLC vials, and the m/z values were acquired by using GC/MS/MS
(Agilent 7890B/7010) at two reaction time intervals, 2 h and 5 d.

16
2.3.5 Reactant and Product Analysis
Depending on the reaction parameters, experiments were conducted at two different
ranges of initial concentration. When the initial concentration of reactants was over 100 M, the
analytes (aldehydes and acid products) were analyzed via high pressure liquid chromatography
(HPLC; Agilent 1260). Analytes were separated on an Aminex HPX-87H (300 mm × 7.8 mm ×
9 m) column without pretreatment. The mobile phase (30% acetonitrile and 70% aqueous
sulfuric acid (20 mM)) was pumped through the column at 0.5 mL/min at 60 °C with a sample
injection volume of 100 μL. The analytes were detected by a photodiode array (Agilent 1260) at
206 nm for aldehydes and 282 nm for carboxylic acids.
For initial reactant concentrations below 25 µM, aldehydes were derivatized by adding 20
μL of a 3 mg/mL 2,4-dinitrophenylhydrazine solution and 40 μL of 1 M citrate buffer solution
adjusted to pH 3 to a 1 mL of sample aliquot.52 After heating the solution at 50 °C for two hours,
the aldehyde derivatizes were separated via HPLC (Agilent 1290 or Agilent 1260) on a Kinetex
Biphenyl column (100 mm × 4.6 mm × 2.6 µm). The mobile phase consisted of an organic
channel (50:50 methanol:ethanol) which increased from an initial 60% to 75% over 10 minutes
at a constant flowrate of 1.3 mL/min at 40 °C, with the remainder of the mobile phase consisting
of Milli-Q (Millipore Advantage A10) water. The separated compounds were detected by either
high-resolution mass spectrometry (Agilent 6560 ion mobility quadrupole time-of-flight [LCIM-QTOF]) with negative mode electrospray ionization (details provided in Table 2.1; all mass
errors <3 ppm), or with a UV/visible photodiode array detector (Agilent 1260) at 360 nm for
derivatized saturated aldehydes and at 382 nm for derivatized crotonaldehyde. For product
analysis, carboxylic acids were separated by a Dionex IonPac AS11-HC column (250 mm × 2

17
mm × 9 µm) in an aqueous mobile phase of KOH increasing from 1 mM to 9 mM over 8 minutes
and measured via ion chromatography (Dionex ICS-2100) with DS6 heated conductivity cell.
Platinum was quantified in column experiment permeate via inductively-coupled plasma mass
spectrometry (ICP-MS; Agilent 8900).
When conducting 18O-labeled water experiments, butyric acid and benzoic acid were
separated by an Agilent Extend-C18 column (50 mm × 2.1 mm × 1.8 µm) in an isocratic mobile
phase of 30% acetonitrile and 70% aqueous formic acid (0.1%) at a constant flow rate of 0.5
mL/min at 40 °C, and detected by high-resolution MS (Agilent 6560) in positive mode. Mass
spectrometry details are provided in Table 2.2. All measured mass errors were < 10 ppm. For
butyraldehyde measurement, 1 mL aqueous samples were placed in 4 mL vials, and 0.4 g of
Na2SO4 was added. After adding 1 mL of dichloromethane, the vials were vigorously shaken for
2 minutes, and the organic solvent phase was transferred to a 2 mL of HPLC vial. Butyraldehyde
was analyzed via gas chromatography/triple quadrupole mass spectrometry (GC/MS/MS)
(Agilent 7890B/7010, Santa Clara, CA) with a DB-1701 column (60 m × 0.25 mm × 0.25 µm)
0.2 µL samples were injected into the inlet at a split ratio of 10:1 at 250°C. The GC oven held at
80 °C for 1 minute and increased to 250 °C at a rate of 20 °C/min. The analytes were ionized by
electron ionization (EI) at 230 °C and measured as listed in Table 2.3.

18
Table 2.1. Chemical structure, calculated masses, measured masses, and mass errors of DNPHderivatized aldehydes measured by LC-IM-QTOF.
FORDNPH
ACEDNPH
PRODNPH
BUTDNPH
PENDNPH
HEXDNPH
CRODNPH
Chemical structure
Calc m/z (M-H)- (Da) 209.0316 223.0473 237.0629 251.0786 265.0942 279.1099 249.0629
Detected mass (Da) 209.0311 223.0469 237.0623 251.0782 265.0939 279.1094 249.0631
Mass difference (ppm) 2.39 1.79 2.53 1.59 1.13 1.79 0.80
Retention time (min) 3.368 4.314 5.613 6.683 7.972 9.204 6.389
Table 2.2. Chemical structures and masses of butyric acid and benzoic acid isotopes.
Chemical structure
Butyric acid 18O-Butyric
acid
18O2-Butyric
acid Benzoic acid 18O- Benzoic
acid
2 18OBenzoic acid
Chemical formula C4H8O2 C4H8O[18O] C4H8[18O]2 C7H6O2 C7H6O[18O] C7H6[18O]2
Calc m/z (M-H)- (Da) 89.0597 91.0640 93.0682 123.0441 125.0483 127.0525
Detected mass (Da) 89.0594 91.0632 93.0678 123.0438 125.0479 127.0524
Mass difference (ppm) 3.37 8.79 4.30 2.44 3.20 0.79
Retention time (min) 0.476 0.476 0.476 0.710 0.710 0.710
Peak area Not detected 35669 201727 54599 666970 4410560

19
Table 2.3. Retention times and mass transition used for butyraldehyde analysis.
Butyraldehyde 18O-Butyraldehyde
Chemical structure
Chemical formula C4H8O C4H8
18O
Retention time (min) 7.745 7.748
Mass transition
Precursor à Product
72 à 57 74 à 59
Collision energy (eV) 5 5
2.3.6 Dissolved Oxygen Measurement
Dissolved oxygen concentrations were measured using a modified Winkler method.53 To
minimize headspace, 2 mL of sample was injected into a 2 mL HPLC vial. 20 µL of 3.55 M
manganese sulfate monohydrate was then added, followed by 20 µL of a solution composed of 8
M sodium hydroxide, 3.34 M potassium iodide, and 0.15 M sodium azide. Finally, 20 µL of pure
sulfuric acid was added to dissolve the precipitate. Dissolved oxygen was quantified as iodine
(produced at a 2:1 stoichiometric ratio) by titrating with 25 mM sodium thiosulfate and 10 mM
sodium hydroxide; starch was added midway through the experiment to visually indicate the
completion of the titration as the dark blue solution became transparent. One dissolved oxygen
molecule forms two stoichiometric equivalents of manganese dioxide (eq. (2.2)) which each
oxidize iodide to iodine (eq. (2.3)). Finally, the iodine is titrated back to iodide (eq. (2.4)), and
the added titrant is used to calculate the dissolved oxygen concentration.
2 Mn2+ (aq) + 4 OH- (aq) + O2 (aq) → 2 MnO2 (s) + 2 H2O (l) (2.2)
2 MnO2 (s) + 4 I- (aq) + 8 H+ (aq) → 2 Mn2+ (aq) + 2 I2 (aq) + 4 H2O (l) (2.3)
4 S2O3
2- (aq) + 2 I2 (aq) → 2 S4O6
2- + 4 I- (aq) (2.4)
O 18O

20
2.3.7 Statistical Analysis
Experimental results were statistically analyzed using GraphPad Prism 9, with a simple
linear regression model and a sum-of-squares F test with 95% confidence intervals. The slopes of
logarithm-transformed data sets were compared by analysis of variance (ANOVA).
2.4 Results and Discussion
2.4.1 Catalyst Characterization
Catalyst particles were rough, consistent with an activated carbon supporter (Figure 2.1,
panels a-d), with a mean particle size of 226 nm and a standard deviation was 31 nm (Figure
2.2). Platinum sites were approximately uniformly dispersed (Figure 2.3, panels a-c), with a size
of less than 5 nm on the carbon surface (Figure 2.3, panel d).
Figure 2.1. SEM images of Pt/C catalyst particles at magnification levels of: (a) 800,000×, (b)
20,000×, and (c, d) 500,000×

21
Figure 2.2. The size distribution of a 0.4 g/L suspension of 5% Pt on C in Milli-Q water
measured by dynamic light scattering. The light scattering was measured after equilibrating
samples at 25 ℃ for 120 seconds with a refractive index of 1.63 and an absorption of 0.001.
Figure 2.3. STEM images of Pt/C catalyst particles at magnification levels of: (a) 250,000×, (b)
120,000×, and (c) 800,000×, (d) TEM image of Pt/C in 800,000×

22
2.4.2 Aldehyde Oxidation Reaction Rate Order and Rate Constants
Based on previous research identifying the presence of certain toxic aldehydes in the final
recycled water effluent,9 six saturated aldehydes and one unsaturated aldehyde were chosen as
oxidation targets: formaldehyde, acetaldehyde, propionaldehyde, butyraldehyde, pentanal,
hexanal and crotonaldehyde. Initial aldehyde oxidation experiments were performed with
relatively low initial aldehyde concentrations to ensure approximately constant dissolved oxygen
concentrations for determination of rate constants. Aldehyde concentrations decreased while
corresponding acid product concentrations increased, following apparent first-order kinetics
(representative dataset in Figure 2.4, panel a; complete set in Figure 2.5; log-transformed data
to obtain rate constants in Figure 2.6). Mass balances were approximately complete except for
with hexanal (Figure 2.5, panel e), which is discussed below. The nearly 100% mass balance
indicates that aldehyde oxidation by Pt/C/O2 produces the corresponding acids as an exclusive
product under the evaluated experimental conditions.

23
Figure 2.4. (a) Representative reaction profile for acetaldehyde oxidation including
acetaldehyde, acetic acid, and dissolved oxygen concentrations, and mass balance (sum of
aldehyde and acid). (b) Aldehyde oxidation first-order rate constants (h-1
) and rate constants
normalized by platinum atom concentration ([Pt]-1 s-1
). Error bars indicate the standard deviation
of the rate constants determined by linear regression of replicate values. Experimental values
without error bars indicate that the error bars are smaller than the data marker. Experimental
conditions: 40 mg/100 mL of Pt/C catalyst, 20 μM nominal initial aldehyde concentration, pH 7,
10 mM phosphate buffer, T = 24 ± 0.5 ℃.
Figure 2.5. Representative reaction profiles of replicate results for aldehyde oxidation including
aldehyde, carboxylic acid, and dissolved oxygen concentrations, and mass balance (sum of
aldehyde and acid concentrations). Experimental conditions: 40 mg/100 mL of Pt/C catalyst,
[Aldehyde]0 = 20 μM, pH 7, 10 mM phosphate buffer, T = 24 ± 0.5 ℃. Panels correspond to (a)
formaldehyde, (b) propionaldehyde, (c) butyraldehyde, (d) pentanal, (e) hexanal, and (f)
crotonaldehyde.

24
Figure 2.6. Normalized logarithm of aldehyde concentrations in Figures 1 and S2 to determine
first-order rate constants. Symbol types (square, triangle, upside down triangle) represent
concentrations from individual replicate reactions. Dotted lines indicate 95% confidence
intervals resulting from a linear regression model of all replicate results. Panels correspond to (a)
formaldehyde, (b) acetaldehyde, (c) propionaldehyde, (d) butyraldehyde, (e) pentanal, (f)
hexanal, and (g) crotonaldehyde.

25
For saturated aldehydes, observed first-order oxidation rate constants (kobs) increased as
the length of the carbon chain increased: 0.61 ± 0.06 h-1 for acetaldehyde, 1.23 ± 0.32 h-1 for
propionaldehyde, 2.42 ± 0.19 h-1 for butyraldehyde, 3.09 ± 0.32 h-1 for pentanal, and 4.58 h-1 ±
0.37 h-1 for hexanal. Corresponding rate constants normalized by molar concentration of
platinum, as previously treated for other noble metal catalysts,32 ranged from 1.7 to 12.4 ([Pt]-1 s1
), and are provided on the secondary y-axis of Figure 2.4, panel b. When oxidizing hexanal, the
hexanal concentration in the first sample was measured to be less than half of the initial
concentration, resulting in a poor mass balance at that time point (Figure 2.5, panel e). As the
reaction progressed, the mass balance recovered as the concentration of the product increased
and eventually plateaued. Because hexanal is relatively hydrophobic (e.g., Kow = 1.80 for
hexanal; Kow = 0.82 for butyraldehyde54) the hexanal rapidly absorbed to the syringe filter used
for separating the catalyst and aqueous solution, resulting in an apparent rapid decrease in
aqueous concentration. As sorbed hexanal was oxidized to produce hexanoic acid, the mass
balance recovered. To account for the phenomenon, the reaction rate of hexanal oxidation was
calculated by using the concentration of hexanoic acid assuming a single rate-limiting step
(derivation of integrated rate law provided in section of 2.4.2.1 Calculation of hexanal
oxidation rate constant).
Longer chain saturated aldehydes generally reacted faster, with observed first order rate
constants increasing monotonically from acetaldehyde to hexanal, potentially suggesting a ratelimiting mass transfer step aided by the higher hydrophobicity of the longer aldehydes. However,
formaldehyde was oxidized approximately five times faster than acetaldehyde with an observed
rate constant of 2.86 ± 0.20 h-1. Crotonaldehyde, which contains the same number of carbons as

26
butyraldehyde but is unsaturated at the 2-position, reacted with the slowest observed rate
constant of 0.0102 h-1 ± 0.0012 h-1. We sought explanations for the apparent relationship
between structure and reactivity for oxidation of aldehydes by Pt/C/O2, by probing the reaction
mechanism, as discussed below.
First, to determine the reaction rate order of aldehydes in the rate-limiting step, initial rate
kinetics experiments were performed with acetaldehyde. Acetic acid concentration was
monitored during approximately the first 10% of the reaction, during which the concentrations of
acetaldehyde and dissolved oxygen changed negligibly. These reactions were performed at three
different acetaldehyde concentrations, and logarithmic transformation of a generic rate law
(2.4.2.2 Calculation of reaction rate order of dissolved oxygen), gives an equation which can
be plotted to give the rate order as the slope of a linear regression (Figure 2.7). In this case, the
slope was statistically significantly indistinguishable from 1.0 (p = 0.5966), indicating that the
rate-limiting step of the reaction is first-order in aldehyde concentration. This observation is
consistent with the results of experiments performed at low initial concentration, in which the
reaction appeared to remain first order, even at concentrations below 1 μM, close to the aldehyde
concentration observed in the product water of some reuse facilities (Figure 2.6).

27
Figure 2.7. Acetaldehyde oxidation to acetic acid under initial rate conditions (at three different
fixed aldehyde concentrations [panels a-c] to determine reaction rate order in aldehyde.
Reactions were performed at pH 7 in 10 mM phosphate buffer with 400 mg/L of Pt/C catalyst,
temperature: 24 ± 0.5 ℃. Dotted lines indicate 95% confidence intervals resulting from a linear
regression.
Next, we evaluated the rate order of the catalyst in the reaction. While catalytic reactions
are often zero-order in catalyst (if the catalyst is saturated), a first-order rate dependence on
catalyst dose may occur if the reactants do not saturate the catalyst. To determine the reaction
rate order for the available Pt sites, initial rate experiments were conducted at different catalyst
loadings, analogously to the aldehyde rate order determination above. The reaction order for
available Pt sites also followed first-order kinetics, with a slope statistically indistinguishable
from 1.0 (p = 0.2139) (Figure 2.8).

28
Figure 2.8. Aldehyde oxidation to respective products under initial rate conditions at four
different catalyst doses to determine reaction rate order in the available Pt surface. Reactions
were performed at pH 7 in 10 mM phosphate buffer, temperature: 24 ± 0.5 ℃. Dotted lines
indicate 95% confidence intervals resulting from a linear regression.
2.4.2.1 Calculation of Hexanal Oxidation Rate Constant
Calculation of hexanal oxidation rate constant using product rather than reactant
concentration data.
≡ []; ≡ [ ]
Typically, the rate constant would be calculated by using the concentration of reactant over time:
[]
= −[]
Which in a reaction with no significant accumulation of intermediates is equal to minus one
times the rate of change of product concentration:
[]
= −[] = − []

29
As the last concentration of hexanoic acid remained constant over the last three data points of the
experiment, they were used to approximate the total mass balance () for the aldehyde-acid
reaction, assuming that the reaction was complete. This average final concentration of hexanoic
acid is represented by [].
= [] + [] ≈ []
[] = [] − []
This can then be substituted into the original rate equation to give:
[]
= �[] − []�
Lastly, the equation is integrated to provide the final rate equation:
� 1
[] − []
[]
[]
0
= �

0
− ln�[] − []��
0
[]
= |0

ln �1 − []
[]
� = −
2.4.2.2 Calculation of Reaction Rate Order of Dissolved Oxygen
Calculation of reaction rate order of dissolved oxygen using initial rate experiment data.
The initial rates of acid formation were measured over a short reaction period assuming aldehyde
and dissolved oxygen concentrations were held at steady state.
≡ [ℎ]; ≡ [ ]; ≡ [ ]

30
[]
= − []
= −[]
[]

[]
= log + [] + []
2.4.3 Role of Dissolved Oxygen in Catalytic Oxidation of Aldehydes
To confirm that oxygen is required for the reaction to proceed (e.g., ruling out evolution
of hydrogen gas or Cannizzaro disproportion as the mechanism of oxidation, that are known to
occur in late-metal catalyzed alcohol-to-carboxylate oxidation),55 superstoichiometric initial
acetaldehyde concentrations were applied in a closed system containing the catalyst, to
deliberately deplete dissolved oxygen and observe whether the reaction continued in the absence
of oxygen. Once oxygen was depleted, aldehyde oxidation ceased (Figure 2.9, panel a). In a
second experiment performed under similar conditions but with a much lower propionaldehyde
concentration, once the aldehyde oxidation was complete, the dissolved oxygen concentration
remained approximately constant (Figure 2.9, panel b), indicating that the catalyst does not
consume oxygen on its own. Both experiments suggest that molecular oxygen is essential for
oxidizing aldehydes with the Pt/C catalyst. To further verify that oxygen is essential for the
reaction to occur, a batch experiment was performed butyraldehyde and the catalyst in N2-purged
solution to minimize the dissolved oxygen concentration, and the reaction rate slowed
dramatically (0.17 ± 0.08 h-1 versus 2.42 ± 0.13 h-1 with ambient dissolved oxygen levels)
(Figure 2.10). In a final experiment at an intermediate starting aldehyde concentration, the
aldehyde oxidation rate remained approximately constant, continuing to follow first-order
kinetics, while oxygen concentration was decreasing (Figure 2.11), suggesting that while oxygen

31
is essential for the reaction to proceed, the reaction rate does not depend on the dissolved oxygen
concentration, as further investigated below.
Figure 2.9. Reaction profiles for aldehyde oxidation including aldehyde, carboxylic acid, and
dissolved oxygen concentrations, and mass balance (sum of aldehyde and acid concentrations).
Experimental conditions: 40 mg/100 mL of Pt/C catalyst, initial concentrations of 3 mM and 1
mM, pH 7, 10 mM phosphate buffer, T = 24 ± 0.5 ℃. (a) Acetaldehyde, (b) Propionaldehyde.
Figure 2.10. (a) Concentrations of butyraldehyde and dissolved O2 in N2-purged solution. (b)
Normalized natural logarithm of butyraldehyde concentrations over time to determine first-order
rate constants of oxidation reactions in N2-purged solution. Dotted lines indicate 95% confidence
intervals resulting from a linear regression of a result. Experimental conditions: 40 mg/100 mL
of Pt/C catalyst, 50 μM nominal initial aldehyde concentration, pH 7, 10 mM phosphate buffer,
temperature: 24 ± 0.5 ℃.

32
Figure 2.11. (a) Reaction profiles for aldehyde oxidation including aldehyde, carboxylic acid,
and dissolved oxygen concentrations, and mass balance (sum of aldehyde and acid
concentrations), at intermediate initial aldehyde concentration. (b) Normalized logarithm of
aldehyde concentrations to determine first-order rate constant of oxidation reactions. Dotted lines
indicate 95% confidence intervals resulting from a linear regression. Experimental conditions: 40
mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 1.4 mM at pH 7 in 10 mM phosphate buffer,
temperature: 24 ± 0.5 ℃.
Puzzlingly, in both oxygen consumption experiments, the oxygen consumption was
approximately one fourth of the concentration of aldehyde oxidized, i.e., 4:1 aldehyde:O2
stoichiometry, where a 2:1 ratio is expected to mediate a 2 e- oxidation with a 4 e- oxidant. One
possible explanation is that in a batch reactor, the catalyst is retaining electronic holes, either in
the carbon support or in the form of high-valent Pt, that remain from air exposure prior to
introduction to the reaction.
To test directly whether oxygen is involved in the rate-determining step of aldehyde
oxidation, initial rate kinetics experiments were performed to determine the reaction rate order in
oxygen. Formation of acetic acid during aldehyde oxidation was monitored during approximately
the first 10% of conversion, during which the concentrations of acetaldehyde and dissolved

33
oxygen change negligibly. These reactions were performed at three different dissolved oxygen
concentrations (Figure 2.12, panels a-c) and analyzed as described previously to determine the
rate order in oxygen, which was statistically indistinguishable from zero (p = 0.4079) (Figure
2.12, panel d), indicating that dissolved oxygen is not involved in the rate-limiting step of the
reaction.
Figure 2.12. Aldehyde oxidation to respective products under initial rate conditions (at three
different fixed oxygen concentrations [panels (a)-(c)]) to determine reaction rate order in oxygen
[panel (d)]. Reactions were performed at pH 7 in 10 mM phosphate buffer with 400 mg/L of Pt/C
catalyst, temperature: 24 ± 0.5 ℃.
Last, to evaluate whether reactive oxidative species (ROS), such as hydroxyl radical and
singlet oxygen, are involved in the reaction pathway, kinetics experiments were conducted in the

34
presence of ROS scavengers: 1 mM of tert-butanol for scavenging hydroxyl radical, and 1 mM
of 2-furoic acid for scavenging singlet oxygen, respectively. Furoic acid was chosen over
furfuryl alcohol56 to avoid scavenger oxidation by the catalyst, which is capable of oxidizing
primary alcohols.57 Butyraldehyde oxidation was not slowed by the presence of t-butanol (p =
0.3552), ruling out a role for hydroxyl radical (Figure 2.13). 2-furoic acid slowed the
butyraldehyde oxidation rate slightly but statistically significantly (from 0.57 h-1 to 0.40 h-1
; p =
0.0021). However, the slight decline suggests that singlet oxygen generation is not the primary
oxidation pathway, and it is possible that the addition of the 2-furoic acid scavenger to the
system affected the reaction rate in some other way, e.g., by blocking catalyst active sites.
Figure 2.13. Normalized logarithm of butyraldehyde concentrations to determine first-order rate
constant of oxidation reactions. Symbol types (circle, square, and rhombus etc.) represent
concentrations from individual replicate reactions. Dotted lines indicate 95% confidence
intervals resulting from a linear regression of all replicate results. Red data points acquired in the
presence of 1 mM of 2-furoic acid; blue data points acquired in the presence of 1 mM of tertbutanol; black data points acquired in absence of any scavenger. Other experimental conditions:
40 mg/100 mL of Pt/C catalyst, [Butyraldehyde]0 = 20 μM at pH 7 in 10 mM phosphate buffer,
temperature: 24 ± 0.5 ℃.

35
2.4.4 Aldehyde Oxidation Mechanism by Pt/C in Aqueous Solution
To begin probing the reaction mechanism, acetaldehyde oxidation experiments were
conducted at a range of pH values (4 - 12) different buffer concentrations to determine whether
the rate-limiting step might be acid- or base-catalyzed. As pH increased, the observed oxidation
rate of aldehydes increased from 0.14 ± 0.06 h-1 at pH 4 to 7.76 ± 0.26 h-1 at pH 12 (Figure 2.14
and 2.15). These experimental results suggest the intuitive conclusion that aldehyde oxidation is
base-promoted but not buffer-catalyzed.
Figure 2.14. Acetaldehyde oxidation rate constants in 10 mM buffer solution (citrate for pH 4, 5,
and 6; phosphate for pH 7, 8, and 12; and carbonate for pH 9, 10, and 11). Error bars indicate the
range of values from duplicate experiments, except for pH 7 (triplicate). Experimental values
without error bars indicate that the error bars are smaller than the data marker. Experimental
conditions: 40 mg/100 mL of Pt/C catalyst, [acetaldehyde]0 = 20 μM, T = 24 ± 0.5 ℃.

36
Figure 2.15. Normalized logarithm of aldehyde concentrations to determine first-order rate
constant of oxidation reactions at different pH values. Symbol types (circle, triangle) represent
concentrations from individual replicate reactions. Dotted lines indicate 95% confidence
intervals resulting from a linear regression of all replicate results. Experimental conditions: 40
mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 20 μM in 10 mM buffer solution: citrate for pH 4, 5
and 6, phosphate for pH 8 and 12, and carbonate for pH 9, 10 and 11, temperature: 24 ± 0.5 ℃.
To further investigate the mechanism, butyraldehyde and benzaldehyde, representatives
of aliphatic and aromatic aldehydes respectively, were oxidized by Pt/C in 18O-labeled water to

37
differentiate oxygen atoms originating from dissolved oxygen (16O2) from those derived from
water (H2
18O). The majority of the carboxylic acid products (85.0% of butyric acid and 85.9% of
benzoic acid) were doubly-labeled with two 18O atoms, meaning that both oxygens were derived
from water (Figure 2.16). 15.0% of butyric acid and 13.0% of benzoic acid are labeled with one
18O atom and one 16O atom, suggesting that one oxygen atom from H2
16O, the product of Ptcatalyzed 16O2 reduction, incorporated into the carboxylic acid product. While it might not fully
explain 14% 16O incorporation, 18O atom exchange with 16O silicates in the glass may have
contributed. Regardless, only a small portion of unlabeled acids was found, indicating that
dissolved oxygen did not react directly with dissolved organics in this reaction. In the gas phase,
platinum is known to dissociate molecular oxygen into two oxygen atoms that incorporate into
the aldehyde to form a carboxylate on the surface,58–61 but this reaction is inconsistent with our
observation on the origin of the oxygen atoms in the product.
Figure 2.16. Isotopic distribution of acid products from butyraldehyde and benzaldehyde
oxidation. Reactions were performed in 97 atom % 18O-labeled water with 4 g/L of Pt/C catalyst
for 6 hours, temperature: 24 ± 0.5 ℃.

38
Aldehydes in aqueous solution rapidly hydrate to the corresponding gem-diol in a
reversible equilibrium (Scheme 1), replacing the oxygen atom in the carbonyl group with oxygen
from water. The replacement of oxygen atoms in aldehydes by water was confirmed by
measuring m/z of butyraldehyde in 1 mL of 5% v/v of 18O-labeled heavy water (H2
18O) at two
reaction time intervals, 2 hours and 5 days. GC/MS/MS analysis indicated that the m/z
distribution of 18O-labeled butyraldehyde (C4H8
18O) relative to total butyraldehyde was 5% at
both reaction times (Figure 2.17), indicating that aldehyde had fully exchanged its oxygen atom
with water within 2h. The hydration equilibrium constant between the two species (Khyd =
RCH(OH)2/RCHO) is a function of chemical structure, generally decreasing for primary
aldehydes as chain length increases (2000 for formaldehyde, 1.20 for acetaldehyde, 0.85 for
propionaldehyde, 0.60 for butyraldehyde, 0.55 for pentanal, 0.50 for hexanal, 1.13 for
crotonaldehyde).62,63 In contrast, aromatic aldehydes such as benzaldehyde are predominantly
present as the aldehydic form (Khyd = 0.008).64 Regardless of the value of the equilibrium
constant, all aldehydes dissolved in 18O-labeled water with sufficient time to reach equilibrium
will have fully-labeled oxygens in both forms: aldehydes with one 18O atom and gem-diols with
two 18O atoms.
Scheme 2.1: Aldehyde equilibrium with hydrated gem-diol.

39
Figure 2.17. The ratio of 18O-labeled butyraldehyde (C4H8
18O) to total butyraldehyde (C4H8
18O
+ C4H8
16O) in 1 mL of 5% v/v of 18O-labeled heavy water (H2
18O) at two equilibration time
intervals (2 hours and 5 days); T = 24 ± 0.5 ℃.
Given that aldehydes exist in aqueous solution as both the free carbonyl and the gem-diol,
we anticipated that each form must either proceed through a pre-equilibrium scenario or through
a different reaction mechanism. We find the latter unlikely, because equal isotope ratios of the
products of butyraldehyde (large Khyd) and benzaldehyde (small Khyd) imply that 16O2, or an
R16OS therefrom, is not reacting directly, as might be proposed for transformation of the free
carbonyl form of the aldehyde. Moreover, alcohol oxidation on Group 8-10 metal catalysts in
aqueous solution is known to proceed through a dehydrogenation mechanism.65–68 A similar
mechanism explains oxidation of primary alcohols to carboxylic acids (or esters), via a
Tishchenko-like pathway involving hydration (or alcoholysis) of the intermediate aldehyde.55 A
mechanism proceeding through the diol is also consistent with kinetic observations: 1) the
reaction rate of formaldehyde, which is overwhelmingly hydrated, is five times faster than
acetaldehyde which is hydrated to a lesser degree at equilibrium. 2) benzaldehyde, despite a
similar hydrophobicity (Kow = 1.71) to hexanal (Kow = 1.80),54 is oxidized much more slowly

40
(0.19 ± 0.14 h-1 (Figure 2.18) compared to 4.58 ± 0.37 h-1)). This is consistent with the reaction
proceeding primarily though the gem-diol, and the extent of hydration driving reactivity, as
benzaldehyde is present primarily as the aldehyde (Khyd = 0.008),64 while hexanal is hydrated to a
greater extent (Khyd = 0.50).62
Figure 2.18. (a) Reaction profiles for benzaldehyde oxidation including benzaldehyde, benzoic
acid, and dissolved oxygen concentrations, and mass balance (sum of aldehyde and acid
concentrations). (b) Normalized logarithm of aldehyde concentrations to determine first-order
rate constant of oxidation reactions. Dotted lines indicate 95% confidence intervals resulting
from a linear regression. Experimental conditions: 40 mg/100 mL of Pt/C catalyst, [Aldehydes]0
= 10 μM at pH 7 in 10 mM phosphate buffer, temperature: 24 ± 0.5 ℃.
To further understand the rate-limiting step of the reaction, aldehyde oxidation
experiments were conducted at different temperatures, and the reaction rates were calculated to
determine activation parameters. Transition state enthalpy (ΔH‡
), entropy (ΔS‡
), and activation
energy were calculated using Eyring and Arrhenius plots (Table 2.4, Figure 2.19).69 Negative
values of ΔS‡
suggest that two molecules are combining in the rate-limiting step (i.e., it is a
bimolecular reaction).70

41
Table 2.4. Calculated activation terms from data in Figure S18. The experiments were conducted
at 10, 25 and 40 ℃ in 10 mM pH 7 phosphate buffer with 400 mg/L of Pt/C catalyst.
Formaldehyde Acetaldehyde Propionaldehyde
ΔH‡ (kJ/mol) 22.18 ± 3.81 19.77 ± 5.53 26.54 ± 5.98
ΔS‡ (kJ/mol-K) -0.23 ± 0.01 -0.24 ± 0.02 -0.22 ± 0.02
Ea (kJ/mol) 24.63 22.22 28.97
Figure 2.19. Normalized logarithm of aldehyde concentrations to determine first-order rate
constant of oxidation reactions at different temperature (Panels a-c). Symbol types represent
concentrations from individual replicate reactions. Dotted lines indicate 95% confidence
intervals resulting from a linear regression of all replicate results. Experimental conditions: 40
mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 20 μM at pH 7 in 10 mM phosphate buffer,
temperature: 10 ℃, 25 ℃, and 40 ℃. Eyring (Panels d-f) and Arrhenius (Panels g-i) plots of
aldehyde oxidation rate constants to calculate activation parameters.

42
Based on the results from Figures 2.12, 2.14, and Table 2.4, two possible reaction
pathways of aldehyde oxidation are suggested in Scheme 2.2 and Scheme 2.3. Each incorporates
oxygen from either H2O (major pathway) or O2 (minor pathway) into the carboxylate,
respectively involving a β-hydride elimination from gem-diol or platinum insertion into aldehyde
C–H bond. For the first pathway, we suspect that platinum binds a hydroxyl group of the gemdiol. The aldehyde C–H group is then cleaved by β-hydride elimination to form the product
carboxylate and an intermediate platinum hydride. The latter is oxidized by O2 in a subsequent
step. We suspect that if a second, direct aldehyde oxidation pathway occurs, it involves aldehyde
C-H insertion as shown in Scheme 2.3. The apparent rate-limiting step is consistent with the
slow reaction rates of crotonaldehyde and benzaldehyde in which the compound contains
multiple carbons with low electron density, potentially serving as alternative targets for the
Lewis acid sites of platinum rather exclusively reacting with the gem-diol.26,71 The significantly
faster reaction rate for formaldehyde relative to acetaldehyde is also consistent with the reaction
proceeding through the gem-diol form, as almost all formaldehyde exists in aqueous solution as
the hydrated geminal diol form.
Scheme 2.2. Proposed aldehyde oxidation reaction pathway through β-hydride elimination.
Scheme 2.3. Proposed aldehyde oxidation reaction pathway through β-hydrogen atom
abstraction.

43
2.5 Implications
Platinum and palladium catalysts, while costly, are present at gram scale in virtually
every truck and automobile in the United States in a catalytic converter, the function of which is
to fully oxidize dilute, partially-oxidized organic compounds, especially CO, in a fluid
containing oxygen. A recent article used the term “Catalytic Converters for Water Treatment”
somewhat figuratively to refer to oxyanion reduction in water using rare metal catalysts and an
electron source. Herein, we have taken the first step toward applying this term literally and
demonstrated oxidative catalytic water treatment with noble metal catalysts and ambient
molecular oxygen. Future work should examine catalyst robustness in dirtier matrices,
regeneration of spent catalysts, whether similar reactivity to Pt could be obtained with cheaper
materials (e.g., Ni, Cu), whether the poor reactivity for unsaturated aldehydes could be
improved, and whether practical scale-up is achievable in a flow-through passive treatment step
at the end of a reuse treatment train.
2.6 Acknowledgements
E.K. was partially supported by a USC Provost Fellowship. We acknowledge additional
funding from the National Science Foundation (Award Nos. CBET-1944810 (D.L.M.) and CHE1856395 (T.J.W.)).

44
CHAPTER 3.
3 Application of Heterogeneous Catalytic System to Real Wastewater
3.1 Abstract
As the water scarcity intensifies, interest in potable water reuse is also increasing. Many
researchers are trying to identify the remaining organic compounds in product water from water
reuse plants, and recent studies have found that a large proportion of the organic carbon is
carbonyl compounds. Among them, aldehydes occupying the largest amount are not easily
removed by RO or UV/H2O2 process because of their polarity, low molecular weight, and neutral
charge. A recent study showed that a heterogeneous catalyst (5% Pt on C) effectively converts
aldehydes to carboxylic acids by using ambient dissolved oxygen as an oxidizing agent. This
study addresses the efficient abatement of aldehydes in recycled wastewater. Emphasizing the
treatment of formaldehyde, we explore the use of heterogeneous Pt/GAC catalysts in a flowthrough column reactor. Our research builds on previous findings, demonstrating a significant
conversion of formaldehyde to benign carbonate species. The study delves into the impact of
various parameters such as ionic strength and salt concentrations on the oxidation process. It
further investigates the scalability of the treatment method, transitioning from bench-scale
experiments to a pilot-scale system. By employing real wastewater samples, we aim to validate
the optimized conditions for future water reuse plants. The results hold promise for significant
advancements in water purification technology, particularly in the context of advanced water
treatment facilities.

45
3.2 Introduction and Background
The increasing water scarcity has intensified the focus on recycling wastewater,
employing treatment steps such as pre-oxidation (with ozone or chlorine), reverse osmosis (RO),
and advanced oxidation processes (AOPs) to mitigate pathogen and chemical hazards.1,2,20
Despite these efforts, recent studies have uncovered that a significant portion of organic carbon,
primarily in the form of aldehydes and ketones, remains in the final effluent, accounting for over
30% of the dissolved organic carbon.9,21,22 These findings are alarming because these aldehydes,
which can be generated during the ring opening of aromatic compounds by strong oxidants like
ozone and chlorine, are known for their toxic electrophilic nature and ability to alkylate DNA
and mRNA, leading to genotoxicity.18 Despite constituting 19-38% of the remaining DOC in
recycled wastewater, they pose a significant challenge for removal with existing RO and AOP
systems due to their polarity, low molecular weight, and neutral charge, highlighting the need for
more effective treatment methods.
As formaldehyde, the most abundant aldehyde found in reuse effluent, is polar and
hydrophilic, only 9% removal of formaldehyde was achieved by RO, and the UV/H2O2 process
did not oxidize the formaldehyde at neutral pH.17 More broadly, the study reporting aldehyde
concentrations after each treatment step indicated that considerable concentrations of aldehydes
were still present after UV/H2O2, demonstrating the limitation of the current advanced treatment
processes.17 Increased formaldehyde removal efficiency by UV/H2O2 process has been observed
at pH 3.
31 However, such acidic pH values are not observed in current treatment systems, in
which aldehydes treat RO permeate that is typically mildly acidic (pH ~5.5-6).
72 Therefore, we
believe that current AOP technology is not sufficient to adequately abate aldehydes in recycled

46
water. Furthermore, aldehydes passing through RO and UV/AOP in direct potable reuse systems
can react with during secondary disinfection with chlorine, producing chlorinated species, which
are potentially even more toxic than their parent aldehydes.
19
In this study, we aim to advance the treatment of aldehydes in wastewater by refining the
use of heterogeneous Pt/GAC catalysts within a flow-through column reactor.
32–36 Building on
our previous research that demonstrated the successful transformation of aldehydes to carboxylic
acids using Pt/C catalysts under mild conditions, we seek to optimize and scale up this process.
Our focus includes understanding the impact of various parameters such as ionic strength and
salt concentrations on aldehyde oxidation and improving treatment efficiency through benchscale experiments. By integrating these insights, we plan to transition from batch experiments to
a flow-through column configuration that more accurately simulates real water treatment
conditions. The goal is to validate these optimized conditions using samples from actual water
reuse treatment systems and to evaluate the feasibility of applying this catalytic treatment system
on a pilot scale for future water reuse plants, marking a significant step towards effective and
environmentally friendly aldehyde abatement.
3.3 Materials and Methods
3.3.1 Materials and Reagents
Chemical suppliers, abbreviations, and purities are listed in Table 3.1.
Table 3.1. List of chemical suppliers, abbreviations, and purities.
Chemical name Acronym Purity Supplier
Dibasic potassium phosphate Extra pure Acros Organics

47
Monobasic potassium phosphate 99% Acros Organics
Sodium bicarbonate 99.7-100.3% EMD Millipore Corp.
Sulfuric acid 95-98% EMD Millipore Corp.
Acetonitrile HPLC grade EMD Millipore Corp.
Methanol HPLC grade EMD Millipore Corp.
Citric acid Technical grade PTI Process Chemicals
2,4-dinitrophenylhydrazine 2,4-DNPH 97% Sigma Aldrich
Sodium sulfate 99% Sigma Aldrich
Formaldehyde
100 µg/mL in
H2O
Sigma Aldrich
Formic acid 99% Thermo Scientific
Granular activated carbon GAC
Platinum(IV) chloride 96% Sigma Aldrich
Sodium borohydride 95%
3.3.2 Sample Collection
Water samples were collected full-scale wastewater reuse plants at sampling points of RO
permeate and final product water. Samples were collected in 4 L collapsible containers and
stored at 4°C before analyzing water quality including pH, total organic carbon (TOC)
concentration, and each aldehyde concentration after being derivatized with 2,4-
dinitrophenylhydrazine.

48
3.3.3 Synthesis and Characterization of the Heterogeneous Catalysts
A catalytic material was synthesized by impregnating platinum onto the Granular
activated carbon (GAC) support, utilizing the wet impregnation technique as drawn in Figure
3.1. In a 150 mL flask filled with 125 mL of Milli-Q water, 0.6 grams of GAC, having a particle
size range of 0.425 - 1.13 mm, were added, sonicated for one minute, and subsequently drained,
a procedure that was repeated three times for thorough rinsing. The rinsed GAC was then
transferred to flasks containing 100 mL of Milli-Q water and 0.0546 grams of platinum(IV)
chloride, and it was stirred for one hour. After adding 0.08 grams of sodium borohydride to the
solution, the mixture was stirred for another 10 to 15 minutes until the solution appeared clear.
The catalysts were drained and baked for 5 hours at 60 ℃. The GAC control were prepared
using the same method, excluding the addition of platinum(IV) chloride. Scanning electron
microscopy (SEM) images were utilized to observe the surface morphology of GAC, while
Energy dispersive X-ray spectroscopy (EDS) was employed to analyze the dispersion of
platinum on a GAC support.
Figure 3.1. Schematic diagram of synthesizing Pt/GAC catalysts by using wet impregnation
techniques.

49
3.3.4 Batch Reactor Experiments
Batch experiments were performed in 100 mL glass syringes to avoid creating headspace
while sampling periodically, because several target compounds were semi-volatile and mass
balance was not conserved in preliminary tests in open reactors. Within the syringes, 40 mg of
heterogeneous catalyst particles were suspended in 99 mL of a solution having various
constituents; different phosphate buffer concentrations (1, 5, and 10 mM) adjusted to pH 7,
different sodium chloride concentrations (5, 10, and 20 mM) with 10 mM phosphate buffer
adjusted to pH 7, and different sodium nitrate concentrations (5, 10, and 25 mM) with 10 mM
phosphate buffer adjusted to pH 7, and pure Milli-Q water. After adding a magnetic stir bar, the
syringe was fitted tightly with a glass plunger and secured to a stir plate. Once the catalyst was
dispersed homogeneously in the syringe, 1 mL of concentrated aqueous aldehyde solution was
injected to initiate the reaction. Aliquots of 2.5 mL were withdrawn periodically and filtered with
a syringe filter (0.2 μm, PTFE).
3.3.5 Flow-through Column Reactor Experiments
A small-scale column test was conducted to determine the feasibility of aldehyde
oxidation by Pt/C catalysts in a flow-through configuration. The borosilicate glass column had a
length of 70 mm and an inner diameter of 5.6 mm for a total bed volume of 1.72 cm3 as drawn in
Figure 3.2. The column was packed with a heterogenous mixture of 40 mg of platinum on
carbon dispersed in 3 g of Ottawa sand (to increase hydraulic conductivity) and was plugged on
each end with glass fiber. A 100 µM formaldehyde solution was pumped from an amber
borosilicate bottle with a peristaltic pump at a flow rate of 0.256 mL/min, which led to an empty
bed contact time (EBCT) of 6.7 minutes as shown below (eq. (3.1)):

50
= ()
(/) = 1.72
0.256 / = 6.7 (3.1)
Figure 3.2. Schematic diagram of flow-through column experiment setting having a length of 70
mm and an inner diameter of 5.6 mm.
3.3.6 Reactant and Product Analysis
Analyte (aldehydes, acid products and dissolved oxygen concentrations) were analyzed
via HPLC (Agilent 1260) or a modified Winkler method as described in section 2.3.5 and 2.3.6.
Platinum was quantified in column experiment permeate via inductively-coupled plasma mass
spectrometry (ICP-MS; Agilent 8900). Total inorganic carbon was analyzed using a Shimazu
TOC-LCPN with a TNM-L unit at 718 ºC furnace temperature, utilizing NDIR sensors for
detection.

51
3.3.7 Statistical Analysis
Experimental results were statistically analyzed using GraphPad Prism 9, with a simple
linear regression model and a sum-of-squares F test with 95% confidence intervals. The slopes of
logarithm-transformed data sets were compared by analysis of variance (ANOVA).
3.4 Results and Discussion
3.4.1 Batch Experiments to understand the effect of ionic strength
To determine the effect of buffer concentrations on oxidation rate of aldehyde, first-order
kinetic experiments were performed. The reaction rate was not significantly affected by buffer
concentration (p = 0.7422) (Figure 3.3), meaning that the reaction is not affected by buffer
concentration. Reaction rate dependence on ionic strength was evaluated by varying salt
concentrations with NaCl, NaNO3 and pure water (Milli-Q) while maintaining otherwise
identical conditions. No significant relationship was found in different NaCl and NaNO3
concentrations (Figure 3.4 and 3.5), but the significantly decreases in the average reaction rate
were observed compared to the reaction rate of buffer-only; 2.21 h-1 for NaCl, 3.79 h-1 for
NaNO3, 5.20 h-1 for buffer-only. However, the formaldehyde oxidation rate in pure water was
calculated as 11.2 h-1 (Figure 3.6), which was considerably faster than the reaction rate acquired
with buffered solution. These results show that different types of ions in solution have a greater
effect on the reaction rate than the change of ionic strength as summarized in Figure 3.7.

52
Figure 3.3. (a-c) Normalized natural logarithm of formaldehyde concentrations over time to
determine first-order rate constants of oxidation reactions at different buffer concentrations.
Symbol types (circle, triangle) represent concentrations from individual replicate reactions.
Dotted lines indicate 95% confidence intervals resulting from a linear regression of all replicate
results. (d) Formaldehyde oxidation rate constants obtained from panels (a-c) plotted against
buffer concentration. Experimental conditions: 40 mg/100 mL of Pt/C catalyst; [formaldehyde]0
= 10 μM; pH = 7; [PO4]TOT = 1, 5 and 10 mM; temperature: 24 ± 0.5 ℃.

53
Figure 3.4. (a-c) Normalized natural logarithm of aldehyde concentrations to determine firstorder rate constant of oxidation reactions at different sodium chloride concentrations. Symbol
types (rhombus, triangle) represent concentrations from individual replicate reactions. Dotted
lines indicate 95% confidence intervals resulting from a linear regression of all replicate results.
(d) Formaldehyde oxidation rate constants in different sodium chloride concentrations.
Experimental conditions: 40 mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 10 μM in 5, 10 and 20
mM sodium chloride with 10 mM phosphate buffer solution, temperature: 24 ± 0.5 ℃.

54
Figure 3.5. (a-c) Normalized natural logarithm of aldehyde concentrations to determine firstorder rate constant of oxidation reactions at different sodium chloride concentrations. Symbol
types (rhombus, triangle) represent concentrations from individual replicate reactions. Dotted
lines indicate 95% confidence intervals resulting from a linear regression of all replicate results.
(d) Formaldehyde oxidation rate constants in different sodium chloride concentrations.
Experimental conditions: 40 mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 10 μM in 5, 10 and 25
mM sodium nitrate with 10 mM phosphate buffer solution, temperature: 24 ± 0.5 ℃.

55
Figure 3.6. Normalized natural logarithm of aldehyde concentrations to determine first-order
rate constant of oxidation reactions in Milli-Q water. Symbol types (circle, triangle) represent
concentrations from individual replicate reactions. Dotted lines indicate 95% confidence
intervals resulting from a linear regression of all replicate results. Experimental conditions: 40
mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 10 μM in Milli-Q water, temperature: 24 ± 0.5 ℃.
Figure 3.7. The average first-order rate constant of oxidation reactions in various constitutes;
NaCl (5, 10, and 20 mM) with 10 mM phosphate buffer, NaNO3 (5, 10, 25 mM) with 10 mM
phosphate buffer, phosphate buffer only (1, 5, and 10 mM), and Milli-Q water. Experimental
conditions: 40 mg/100 mL of Pt/C catalyst, [Aldehydes]0 = 10 μM, temperature: 24 ± 0.5 ℃.

56
3.4.2 Aldehyde Oxidation in a Flow-through Column
To make a preliminary evaluation of the ability of the Pt/C/O2 system to serve as a flowthrough “catalytic convertor” for oxidizing aldehydes in recycled wastewater, a column
experiment was performed continuously pumping 100 µM formaldehyde, which was the most
abundant aldehyde found in RO permeate at reuse plants,9 through glass tubes filled with
homogenously mixed Pt/C catalyst and Ottawa sand. Formaldehyde concentrations in the
aliquots collected from the outlet of the glass tubes indicated that the initial conversion rate of
the formaldehyde was > 99% at the catalyst loading and EBCT evaluated (23 mg/cm3 and 6.7
min), and remained at 90% after treating 2162 bed volumes, corresponding to oxidation of 6.4
mmol C/g-Pt-1 (24.8 mol-formaldehyde mol-Pt-1
) (Figure 3.8). At the end of the experiment after
6239 bed volumes, the conversion rate was 68.5%. The consistent mass balance suggests that
oxidation to formic acid was the dominant removal mechanism of formaldehyde. The incomplete
mass balance relative to nominal initial formaldehyde concentration may be attributable to loss
of formaldehyde, which is semi-volatile, to the atmosphere during feed solution preparation. In a
second column experiment, platinum concentration and particle size distribution were measured
in column permeate by ICP-MS and DLS over 1600 bed volumes, to assess whether catalyst loss
may explain declining reactivity. The permeate contained a small amount of Pt (~1 ppb) (Figure
3.9), and DLS measurements found particles in the permeate with an average size of 226 nm,
consistent with our prior measurements of catalyst particle size. These results suggest that
particle escape, rather than Pt leaching or loss of catalyst potency, may explain declining
reactivity. Following work evaluate the possibility of better retaining the Pt/C particles. Finally,
the dissolved oxygen concentration at the outlet of the column with Pt/C catalyst was

57
consistently lower than the control column filled only with the sand (Figure 3.10), confirming
that oxygen is consumed during the reaction in the column.
Figure 3.8. Reaction profile of flow-through reactor for aldehyde oxidation including
formaldehyde, formic acid, and dissolved oxygen concentrations and mass balance.
Experimental conditions: 40 mg of Pt/C catalyst mixed with 3 g of Ottawa sand, pH 7 in 10 mM
phosphate buffer, temperature: 24 ± 0.5 ℃.
Figure 3.9. Concentration of platinum in effluent of flow-through column reactor. Experimental
conditions: 40 mg of Pt/C catalyst mixed with 3 g of Ottawa sand, Milli-Q water, temperature:
24 ± 0.5 ℃.

58
Figure 3.10. Reaction profile of control (catalyst-free) flow-through reactor for aldehyde
oxidation including formaldehyde, formic acid, and dissolved oxygen concentrations, and mass
balance. Experimental conditions: 3 g of Ottawa sand, pH 7 in 10 mM phosphate buffer,
temperature: 24 ± 0.5 ℃.
3.4.3 Characterization of Synthesized Catalyst
To enhance the retention of catalysts in the flow-through column, Pt/GAC catalyst was
synthesized by supporting platinum on granular activated carbon (GAC). SEM images of the
GAC supporter revealed a rough surface, indicative of a high surface area favorable for platinum
loading (Figure 3.11 and 3.12, panel a). The EDS image displayed a uniform dispersion of
platinum across the GAC surface, visually represented as a yellow color (Figure 3.12, panel b),
which suggests an effective synthesis process and a well-supported platinum catalyst.

59
Figure 3.11. SEM images of Pt/GAC catalyst particles at magnification levels of: (a)150×, and
(b) 2,000×.
Figure 3.12. (a) SEM images of Pt/GAC catalyst particles at magnification of 650×, and (b) EDS
image of the identical area shown in the SEM image, where yellow marking signify the presence
of platinum.
3.4.4 Formaldehyde Oxidation on Pt/GAC.
The column reactor with the Pt/GAC catalyst demonstrated effective performance,
achieving a 70% conversion rate of formaldehyde concentration without any noticeable decline
in efficiency (Figure 3.13, panel a). However, the significant discrepancy in the mass balance

60
was observed, particularly when assuming formate as an exclusive oxidation product, a result
reported in the previous study. Notably, the observed increase in total inorganic carbon
concentration in the aliquots suggests that formate, the primary oxidized product of
formaldehyde, further oxidizes into carbonate species (Figure 3.13, panel b). To ascertain the
origin of the carbon atoms within the carbonate species, an additional experiment introducing
formate as a reactant was performed with the same experimental conditions with a flow-through
column reactor. Comparative analysis indicates a reduction in formate concentration in the
effluent of the Pt/GAC column than to the GAC control, supporting the hypothesis that formate
undergoes serial oxidation to form carbonate species (Figure 3.14, panel a). In batch
experiments measuring the concentrations of formaldehyde, formate, and inorganic carbon over
time, the result reveals a pronounced decline in formaldehyde levels, while formate
concentrations initially surge, reaching a peak between 20-40 minutes (Figure 3.14, panel b).
The pattern observed suggests that formate is transiently produced as an intermediate, which then
undergoes further oxidation into carbonate species. The consistent increase in carbonate
concentration indicates that carbonate species are the end products of the oxidation of both
formaldehyde and formate. This finding holds substantial implications, particularly for advanced
water treatment facilities, where there is a stringent requirement to remove specific percentages
of total organic content. The ability of the Pt/GAC catalyst to effectively convert organic carbons
into less problematic carbonate species via serial oxidation represents a significant advancement
in the domain of water purification technology.
To identify the formate species that undergoes oxidation on the Pt/GAC catalyst, formate
oxidation experiments were conducted at a range of pH values (4-7). As pH decreased, the

61
observed oxidation rate of formaldehyde increased from 0.53 ± 0.04 h-1 at pH 7 to 2.40 ± 0.23 h-1
at pH 4 (Figure 3.15 and 3.16). These experiment results suggest that the formate oxidation
primarily proceeds with the acid form rather than its basic form.
Figure 3.13. (a) Reaction profile of flow-through reactor for formaldehyde oxidation. (b)
Product analysis showing the formaldehyde, formic acid, and inorganic carbon concentrations.
Experimental conditions: 400 mg of Pt/GAC catalyst, pH 7 in 10 mM phosphate buffer,
temperature: 24 ± 0.5 ℃.

62
Figure 3.14. (a) Reaction profile of flow-through reactor for formate oxidation with 400 mg of
Pt/GAC catalyst and initial formate concentration of 500 μM. (b) Reaction profile of
formaldehyde oxidation including formaldehyde, formate and inorganic carbon with 40 mg/100
mL of Pt/GAC catalyst and initial formaldehyde concentration of 500 μM. Experimental
conditions: [formaldehyde]0 = pH 7, 10 mM phosphate buffer, T = 24 ± 0.5 ℃.
Figure 3.15. Formate oxidation rate constants in 10 mM buffer solution (citrate for pH 3, 4, 5,
and 6, and phosphate for pH 2, and 7). Error bars indicate the range of values from duplicate
experiments. Experimental values without error bars indicate that the error bars are smaller than
the data market. Experimental conditions: 40 mg/100 mL of Pt/GAC catalyst, [formate]0 = 500
μM, temperature: 24 ± 0.5 ℃.

63
Figure 3.16. Normalized logarithm of formate concentration to determine first-order rate
constants at different pH values. Symbol types (circle, triangle) represent concentrations from
individual replicate reactions. Dotted lines indicate 95% confidence intervals resulting from a
linear regression model of all replicate results. Experimental conditions: 40 mg/100 mL of
Pt/GAC catalyst, [formate]0 = 500 μM in 10 mM buffer solution: citrate for pH 3, 4, 5, and 6,
and phosphate for pH 2, and 7, temperature: 24 ± 0.5 ℃.

64
3.5 Implications
The implications of this research are profound for environmental engineering and water
reuse technology. First, it provides a scalable and effective solution for the abatement of toxic
aldehydes, which are challenging to remove in conventional treatment systems. The ability of the
Pt/GAC catalyst to convert organic carbons into less problematic compounds like carbonate
species through serial oxidation is a notable achievement. This approach not only improves the
safety of recycled wastewater but also aligns with the stringent requirements for organic content
removal in advanced water treatment facilities. Furthermore, the study's findings can guide the
development of new standards and practices in water treatment, emphasizing the removal of
recalcitrant organic pollutants. The successful scaling of this method from laboratory to pilotscale reinforces its potential for real-world application, setting a precedent for future innovations
in water reuse and purification.
3.6 Acknowledgements
We acknowledge funding from the National Science Foundation (Award Nos. CBET1944810 (D.L.M.)).

65
CHAPTER 4.
4 Identifying Precursors of N-DBPs in Treated Wastewater by using
Chemical Derivatization
4.1 Abstract
Faced with the increasing necessity for water reuse due to recurrent droughts and
diminishing freshwater supplies, this study investigates the formation of nitrogenous disinfection
byproducts (N-DBPs) in recycled wastewater. These N-DBPs, typically more toxic than
trihalomethanes and haloacetic acids, present a substantial risk to human health. Utilizing
chemical derivatization techniques, the study evaluates how different amines contribute to NDBP formation in secondary effluents from three wastewater treatment plants by transforming
amines into compounds with lower reactivity. The effectiveness of this derivatization was first
verified using model amines to measure the impact on N-DBP formation potential following
ozonation or chloramination, and then applied to actual secondary effluent samples. The findings
show that em, underscoring the importance of customized wastewater treatment methods to
reduce these harmful byproducts, thereby improving water safety and meeting regulatory
standards, particularly in systems with diverse secondary effluent types.
4.2 Introduction and background
Recurring droughts and declining freshwater supply have motivated increasing adoption
of water reuse in the American Southwest and elsewhere. Despite a high level of treatment in

66
typical water reuse processes, a small amount of organic carbon remains in the final product
water, including low molecular weight disinfection byproducts (DBPs). DBPs in recycled
wastewater are thought to pose a greater threat to human health than other commonly measured
organic contaminants,16,73,74 including pharmaceuticals and endocrine disrupting compounds.2
Among the >700 disinfection byproducts identified in drinking water and recycled wastewater,
toxicological assessments indicate that nitrogenous disinfection byproducts (N-DBPs) are
typically much more toxic than currently-regulated trihalomethanes and haloacetic acids,75,76 and
considerable effort has been made to understand their precursors and formation pathways,77 to
facilitate treatment modifications to minimize their formation.72,78–81
Three classes of N-DBPs receiving substantial attention due to their high toxicity and
frequent detection in drinking water and recycled wastewater are N-nitrosamines,
halonitromethanes (HNMs), and haloacetonitriles (HANs). N-Nitrosamines, especially Nnitrosodimethylamine (NDMA), have long been associated with water reuse, and the discovery
of NDMA in product water led to the closure and retrofit of a major water reuse facility in
California.82 Halonitroalkanes and haloacetonitriles frequently occur in recycled wastewater,83–86
both appear on the most recent USEPA Candidate Contaminant List (CCL5),87 and have both
been implicated as drivers of calculated toxicity in DBP mixtures.88 Although the causative agent
of DBP mixture toxicity is not yet settled (and may not be dominated by a single byproduct
class), the frequent occurrence and high toxic potency of these compounds has led to substantial
efforts to better understand their precursors and formation pathways.
Chloramination or ozonation of wastewater effluent ubiquitously forms NDMA,89,90 and
recent research has established that dichloramine is the chlorine species responsible for NDMA

67
formation both from model compounds91 and in real wastewater.72,79,80,92,93 Extensive model
compound research has identified an array of chloramine-reactive NDMA precursors, most
prominently dimethylamine77,94 and N,N-dimethyl-alpha-arylamines including ranitidine,
dimethylbenzylamine91 and methadone.95 Halonitroalkanes are associated with ozonation
followed by secondary disinfection with chlorine,83 and were shown to be derived from 1º and/or
2º amine precursors when formed from relatively pristine natural organic matter.83 However,
their precursors in wastewater are unknown, despite increasing popularity of ozone as a primary
disinfectant in recycled wastewater schemes. While mechanistic work with model organic
precursors has led to considerable progress on understanding which types of molecules can serve
as precursors of these byproducts, relatively little is yet known about which types of compounds
do lead to N-DBP formation in drinking water and recycled wastewater.
The objectives of this study were to (1) Identify and adapt chemical derivatization
techniques from the synthetic chemistry literature for selectively protecting functional groups
thought to be responsible for NDMA, halonitromethane, and haloacetonitrile formation (3º, 2º,
and 1º amines, respectively), (2) to validate these derivatization techniques by measuring loss of
N-DBP formation potential from model amines after derivatization, and (3) to apply this
orthogonal derivatization strategy to quantify the contribution of 1º, 2º, and 3º amines to the
precursor pool of these three classes of byproducts in real wastewater effluent.
4.3 Materials and Methods
4.3.1 Materials and Reagents
Chemical suppliers, abbreviations, and purities are listed in Table 4.1.

68
Table 4.1. List of chemical suppliers, abbreviations, and purities.
Chemical name Acronym Purity Supplier
Benzylamine BA 98% Sigma Aldrich
N-Benzylmethylamine BMA 97% Sigma Aldrich
N,N-Dimethylbenzylamine BDMA >99% Sigma Aldrich
β-Alanine ALN 99% Sigma Aldrich
Methylamine hydrochloride MA 97% Spectrum Chemical
Dimethylamine hydrochloride DMA 99% Sigma Aldrich
Trimethylamine hydrochloride TMA 98% Sigma Aldrich
Nitromethane NM >99% Acros Organics
Dichloroacetonitrile DCAN 5.0 mg/mL in acetone AccuStandard
N-Nitrosodimethylamine NDMA 1000 μg/mL in
Dichloromethane AccuStandard
o-Phthalaldehyde OPA >99% Sigma Aldrich
Benzyl bromide BB 98% Sigma Aldrich
Di-tert-butyl dicarbonate Boc2O >99% Sigma Aldrich
Sodium carbonate anhydrous >99.5% VWR Life Science
Potassium phosphate
(monobasic) 99% Acros Organics
Potassium phosphate (dibasic) Extra pure Acros Organics
Sodium hydroxide >97.0% Millipore Sigma
Sulfuric acid 95.0-98.0% Sigma Aldrich
Sodium bicarbonate >99.7% Sigma Aldrich
Sodium bromide >99.0% Sigma Aldrich
Phenethylamine PEA 99% Sigma Aldrich
Tyramine TRM >98.0% Sigma Aldrich
Ammonium chloride >99.5% Sigma Aldrich
Ammonium-15N chloride >98 atom% 15N Sigma Aldrich
Sodium hypochlorite solution 4.00-4.99% Sigma Aldrich

69
4.3.2 Wastewater Sample Collection and Analysis
Secondary effluents from three anonymous wastewater treatment facilities in Southern
California were collected in collapsible polyethylene containers after rinsing them with the
effluent twice. 45 mL of samples were aliquoted into 50 mL conical tubes, stored at -80 ℃, and
defrosted in a 25 ℃ water bath for 2 hours before use. Basic water quality data are provided in
Table 4.2.
Table 4.2. Secondary effluent water quality data
Secondary Effluent A Secondary Effluent B Secondary Effluent C
pH 7.35 6.96 7.26
TN (mg/L) 13.1 54.2 51.0
TOC (mg/L) 55.9 108.0 94.3
Ammonia-N (mg/L) 4.7 41.6 39.2
Nitrate-N (mg/L) 7.03

Abstract (if available)

Abstract Amidst increasing water scarcity leading to the expansion of water reuse, this research addresses critical aspects of water purification technology, particularly focusing on recycled wastewater. The study first demonstrates the oxidation of trace aqueous aldehydes to their corresponding acids using a heterogeneous catalyst (5% Pt on C) with ambient dissolved oxygen. It reveals that low molecular weight aldehydes, significant in the organic carbon pool of recycled wastewater, are efficiently converted, with the process primarily proceeding through a base-promoted beta-hydride elimination mechanism. Building upon these findings, the research further delves into the efficient abatement of aldehydes, such as formaldehyde, using heterogeneous Pt/GAC catalysts in a flow-through column reactor. The study investigates the influence of various parameters like ionic strength and salt concentrations on the oxidation process and assesses the scalability of this treatment method, using real wastewater samples to validate conditions for future water reuse plants. In addition to aldehyde treatment, the study also explores the formation of nitrogenous disinfection byproducts (N-DBPs) in recycled wastewater, which pose a significant risk to human health. Utilizing chemical derivatization techniques, the research evaluates how different amines contribute to N-DBP formation in secondary effluents from wastewater treatment plants by transforming amines into less reactive compounds. The effectiveness of this approach is validated using model amines, with findings showing that primary and secondary amines contribute differently to the formation of N-DBPs such as nitromethane, NDMA, and DCAN. Advancing our foundational knowledge, the study delves into the formation of initial chlorine transformation products from amino acids and peptides, hypothesizing that these lead to higher yields of nitrogenous DBPs compared to small molecules, as demonstrated with N-acetyltryptophan (NacTrp). Additionally, the research studies on the kinetics of halonitromethane formation, particularly nitromethane, to better understand chloramination processes. This approach broadens our understanding of DBP formation, contributing to the development of safer and more sustainable water treatment methods. Overall, this comprehensive research emphasizes the need for customized wastewater treatment methods to reduce these harmful byproducts, thereby enhancing water safety and compliance with regulatory standards, especially in systems with diverse secondary effluent types. Significantly contributing to the field of environmental engineering, by providing innovative solutions to water treatment challenges, paving the way for safer and more sustainable water reuse practices in the face of global water scarcity.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The formation, fate and transformation of nitromethane in potable reuse processes

PDF

Integrating systems of desalination and potable reuse: reduced energy consumption for increased water supply

PDF

Fate of antibiotic resistance in anaerobic membrane bioreactors

PDF

Bioelectrochemical treatment of anaerobic process effluents: mitigation of dissolved methane and sulfide

PDF

Water desalination: real-time membrane characterization for performance prediction and system analysis for energetic enhancement

PDF

Integrated technologies, blending schemes, and reuse practices to address contaminant and energy challenges in water reclamation

PDF

Advancing energy recovery from food waste using anaerobic biotechnologies: performance and microbial ecology

PDF

Wastewater-based epidemiology for emerging biological contaminants

PDF

Thermally driven water treatment with membrane distillation: membrane performance, waste heat integration, and cooling analysis

PDF

Optimizing biomembrane reactor systems for water reclamation and reuse applications

PDF

Chemical recycling of amine/epoxy composites at atmospheric pressure

PDF

The roles of surface and pore properties in wetting resistance for membrane distillation membranes

PDF

Lithium-ion batteries (LIB) industrial recycling and policies

PDF

Developing high-resolution spatiotemporal methods to model and quantify water use for energy

PDF

Identifying and mitigating the effects of urban heat islands in California

PDF

Analysis of sources and profiles of organic carbon in ambient particulate matter across fine and coarse sizes and introducing an optical technique for real-time urban dust measurement

PDF

Groundwater contaminant transport predictions in a sustainable water management scenario

PDF

Electrokinetic transport of Cr(VI) and integration with zero-valent iron nanoparticle and microbial fuel cell technologies for aquifer remediation

PDF

Development of novel techniques for evaluating physical, chemical and toxicological properties of particulate matter in ambient air

PDF

Hybrid physics-based and data-driven computational approaches for multi-scale hydrologic systems under heterogeneity

Asset Metadata

Creator Kim, Euna (author)

Core Title Catalytic oxidation and precursor identification of disinfection byproducts in recycled water

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Environmental Engineering

Degree Conferral Date 2024-05

Publication Date 04/08/2024

Defense Date 01/10/2023

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

Tag catalytic oxidation,disinfection byproducts,OAI-PMH Harvest,water reuse

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor McCurry, Daniel L. (committee chair), Childress, Amy E. (committee member), Smith, Adam L. (committee member), Williams, Travis J. (committee member)

Creator Email eunakim@usc.edu,mprydmsdk@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113871199

Unique identifier UC113871199

Identifier etd-KimEuna-12766.pdf (filename)

Legacy Identifier etd-KimEuna-12766

Document Type Thesis

Format theses (aat)

Rights Kim, Euna

Internet Media Type application/pdf

Type texts

Source 20240408-usctheses-batch-1136 (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