Download Figure 2

Modeling Advanced Oxidation Processes for Water Treatment
Ashley N. Anhalt, A. Eduardo Sáez, Robert G. Arnold, and Mario R. Rojas
Department of Chemical and Environmental Engineering
The University of Arizona
• Many conventional wastewater treatment processes only
partially remove trace organics that result from human use,
including hormones and pharmaceuticals.
• Advanced oxidation processes (AOPs) can be used to remove
the chemicals that remain. Ultraviolet Photolysis of H2O2
(UV/H2O2) is one of the most common AOPs used in practice.
• In this work, we propose a kinetic model to simulate the
UV/H2O2 process taking into account the destruction of trace
organics by radicals generated in multiple reactions.
• This research predicts the degradation of organic
contaminants over a wide range of conditions and illustrates
the potential for polishing conventionally treated
wastewater with AOPs.
CHEMICAL REACTIONS
METHODOLOGY
ABSTRACT
Learning the Model:
• In order to validate the efficacy of the UV/H2O2 model, data from previously published research
were successfully reproduced.
• Accurate comparisons between these graphs and those previously developed demonstrate an
understanding of the model.
Fixing the Model:
• Originally, the UV/H2O2 model functioned as one reactor (Figure 2), in which the concentrations of
the radicals are assumed to be uniform throughout the reactor.
This assumption of uniformity throughout the reactor is inaccurate.
• We adjusted the model by dividing the reactor into multiple equivalent sections (Figures 3 and 4).
• By considering two or three individual reactors, the model better accounts for light intensity
effects and spatial variations of radical concentrations.
• Comparisons between the one-reactor setup, the two-reactor setup, and the three-reactor setup
allow for insight as to how depth influences the chemical degradations.
INTRODUCTION
• Typical wastewater treatment processes do not completely
remove organics, such as pharmaceuticals and endocrine
disrupters.
• An advanced treatment method which removes these
unwanted chemicals in a cost-efficient manner is highly
desirable.
• This research simulates and analyzes a UV/H2O2 AOP, which
converts organic contaminants into carbon dioxide (CO2),
instead of transporting the contaminants across different
treatment phases, such as in adsorption processes.
Figure 2: This one-reactor
setup assumes uniformity of
all chemical concentrations.
Figure 3: This two-reactor setup
considers each reactor1 and reactor2
as separate reactors and assumes
uniformity of all chemical
concentrations in each reactor.
Figure 4: This three-reactor setup
considers each reactor1, reactor2, and
reactor3 as separate reactors and
assumes uniformity of all chemical
concentrations in each reactor.
Figure 5: Light intensity
attenuated by absorption.
RESULTS
Preliminary model results demonstrate that the UV/H2O2 model was successful in reproducing
previously published results (Figure 6).
Below are the elementary chemical reactions involved in the
UV/H2O2 model. Kinetic and equilibrium constants are at 25°C.
Reactions E1 to E4 are considered to equilibrate instantaneously.
No. Reaction
Rate constant, k (M-1s-1) or
equilibrium constant, K
R1
R2
R3
R4
R5
R6
R7
R8
R9
R10
R11
R12
R13
R14
R15
R16
R17
R18
R19
R20
R21
E1
E2
E3
E4
f=0.5 (primary quantum yield)
k2 = 2.7  107
k3 = 7.5  109
k4 = 8.5  106
k5 = 3.9  108
k6 = 6.6  109
k7 = 8.0  109
k8 = 5.5  109
k9 = 3.0  109
k10 = 1.3  10-1
k11 = 6.5  108
k12 = 9.7  107
k13 = 8.6  105
k14 = 3.7
k15 = 8.0  105
k16 = 3.0  107
k17 = 2.0  107
kPC = 1.2  1010
kNP = 1.33  1010
kIPOH = 1.9  109
kEtOH = 1.9  109
Ka1 = 10-11.60
Ka2 = 10-4.86
Ka3 = 10-6.30
Ka4 = 10-10.36
H2O2 + hn → 2HO·
·OH + H2O2 → O2·- + H2O + H+
·OH + HO2- → O2·- + H2O
·OH + HCO3- → CO3·- + H2O
·OH + CO32- → CO3·- + OH·OH + HO2· → O2 + H2O
·OH + O2·- → O2 + OH·OH + ·OH → H2O2
·OH + CO3·- → Products
O2·- + H2O2 → ·OH + O2 + OHO2·- + CO3·- → O2 + CO32O2·- + HO2· + H2O → H2O2 + O2 + OHHO2· + HO2· → H2O2 + O2
HO2· + H2O2 → ·OH + O2 + H2O
CO3·- + H2O2 → HCO3- + O2·- + H+
CO3·- + HO2- → HCO3- + O2·CO3·- + CO3·- → 2CO32PC + ·OH → Products
NP + ·OH → Products
C3H8O + ·OH → Products
C2H6O + ·OH → Products
H2O2  HO2- + H+
HO2·  O2·- + H+
H2CO3  HCO3-+ H+
HCO3-  CO32-+ H+
CONCLUSIONS
• The adjusted UV/H2O2 models, which take into account the
spatial variations of radical concentrations, are improved
models.
• Expanding this multiple-reactor approach to other data sets
and different conditions would be rewarding future research.
• As originally hypothesized, the degradation of organic
contaminants is predictable over a wide range of conditions.
Figure 1: Student conducting UV/H2O2 AOP experiments.
• This project applies an innovative approach to the UV/H2O2
model, taking into account spatial variations of radical
concentrations in the reactor.
• This UV/H2O2 model illustrates the promise for effectively and
efficiently removing potentially harmful contaminants in
water.
• By improving an already robust UV/H2O2 AOP model, there is
obvious potential for polishing conventionally treated
wastewater.
• The implications of these results are significant.
UV/H2O2 MODEL OBJECTIVES
• To oxidize unwanted compounds remaining in wastewater.
Figure 6: The decomposition of p-cresol for three initial H2O2 concentrations using the UV/H2O2 model.
[PC]0=240μM and λ=250nm. (Left) Figure from Rojas et al. (2010). (Right) From this research.
The new multiple-reactor approaches take into account spatial variations of radical concentrations.
The figure below (Figure 7) demonstrates slight improvement in the accuracy of the multiple-reactor
models compared to the original single-reactor model.
• To characterize the mechanism and kinetics behind the
decomposition of nonylphenol (NP) and p-cresol (PC), two
chemicals in wastewater that serve as surrogates for
endocrine disruptors.
• To improve an already robust UV/H2O2 AOP model by taking
into account spatial variations of radical concentrations.
• To predict the degradation of organic contaminants over a
wide range of conditions, thus broadening the model’s
applications.
Figure 7: (Left) The decomposition of PC using the original single-reactor UV/H2O2 model. (Middle) The decomposition of PC using
the improved two-reactor UV/H2O2 model. (Right) The decomposition of PC using the improved three-reactor UV/H2O2 model.
• Determining a consistently successful and cost-effective
method for the removal of these pollutants is essential.
ACKNOWLEDGEMENTS
I would like to thank:
• Dr. Maria Teresa Velez, the Director of the University of
Arizona Undergraduate Research Opportunities Consortium
(UROC)
• Donna Treloar, the Director of the University of Arizona
Summer Research Institute (SRI)
This research was supported by the Western Alliance to Expand
Student Opportunities (WAESO) ‘Senior Alliance’ Louis Stokes
Alliance for Minority Participation (LSAMP) National Science
Foundation (NSF).