Embed Size (px)
Citation preview
i
TOWARDS DECOMPOSITION OF PHOSPHONATES IN WATER USING CATALYTIC MEMBRANES: PARAQUAT ON SILICA AS A PROXY
BY
SHAMA FARABI BARNA
THESIS
Submitted in partial fulfillment of the requirements for the degree of Master of Science in Mechanical Engineering
in the Graduate College of the University of Illinois at Urbana-Champaign, 2011
Urbana, Illinois
Master’s Committee:
Professor Mark A. Shannon Professor Alexander Scheeline Dr. Glennys A. Mensing
ii
Abstract
Phosphonate compounds are acutely toxic and have strong interactions with surfaces, which
makes the detection and cleanup of such compounds necessary, but difficult, in a water
treatment plant. If materials capable of catalyzing phosphonate decomposition are identified,
the combined application of catalytic materials and membranes can lead to the effective
treatment of wastewater with phosphonates and ensure subsequent removal of degradation
by-products. The focus of this MSc thesis is to conduct a feasibility study of using silica
supported catalysts as membrane materials to treat wastewater containing phosphonates.
Because adsorption confounds the ability to study phosphonate decontamination chemistry,
an equally important but easier to study compound, paraquat dichloride, has been selected for
initial investigation. Both adsorption and decomposition kinetics have been explored utilizing
different analytical techniques such as liquid chromatography mass spectrometry and UV-Vis
spectrometry. The results demonstrate strong adsorption by the contaminant onto silica
surfaces. Competitive binding model predictions are found to be consistent with experimental
adsorption data. Addition of an exogenous agent (e.g. H2O2) was also attempted to assist
paraquat decomposition; however, investigation of paraquat-H2O2 interaction both in
presence and absence of the catalysts indicate that peroxide functionality was not effective
for chemical decomposition of paraquat.
iii
Acknowledgement
At first, I would like to thank almighty Allah for enabling me to complete this thesis. I have
always been influenced by Prof. Mark Shannon’s research vision and deeply acknowledge his
contribution towards my growth as a graduate student in UIUC. I am also grateful to Prof.
Alexander Scheeline for the unconditional support and motivation he has given me during
this project. Their dedication and affection had been crucial in instilling in me the confidence
and skills to promote my understanding of scientific research. I would also like to thank Dr.
Glennys Mensing for being a great source of assistance and nurturing me as a graduate
student when I was new to the Shannon group. Special thanks to Edward Chainani from Prof.
Scheeline’s Lab and Furong Sun from the School of Chemical Sciences Mass Spectrometry
lab for helping me with the UV-Vis and LC-MS experiments. The work presented in this
thesis is funded by PPG Industries and done in collaboration with Elizabeth Ott from the
University of Illinois and Nguyen Hoai Thu from Hanoi University of Science. Special
thanks to both of them for their noteworthy role in this research.
On a personal level, I will be forever indebted to my extraordinary parents for what I am
today and to Tahmina Haque and Zubaer Hossain for their consistent moral support through
good and bad times in the US.
iv
Table of Contents
Chapter 1: Introduction … … … … … … … ……..… … …… … … … … … … …… ..1
1.1 Objective … … … … … … … … … … … … … …. … … … … … … … … … ... 1
1.2 Phosphonate Contamination: Sources and Impacts … … … … … … … … … … …1
1.3 Mechanism of Intoxication … … … … … … ……... … … … … … … … … … .. ..5
1.4 Detection and Decontamination Challenges … … …... … … … … … … … … ... .. .. 5
1.5 Mechanisms of Phosphonate Degradation … … … ….. … … … … … … … … ... … 7
1.6 Silica as a Catalyst Support… … … …..… … … …. … … … … … … … … … ….8
1.7 Overview of the Research… … … … … … … … …... … … … … … … … … … …9
Chapter 2:Methods of Measurement … … … … … … ….. … … … … … … … ……….11
2.1 Spectrometric Determination of Paraquat … … … … … … … … … … … … ……11
2.1.1 Spectrometer Principle… … … … … … … … … … … … … … … … …...11
2.1.2 UV Absorption of Paraquat … … … … … …. … … … … … … … … …...12
2.1.3 Validation … … … … … …. … … … … ….. … … … … … …. … … ......13
2.1.4 Calibration Results … … … … … …. … … … … … … … …. … … …...13
2.2 Calibration Curve for H2O2 solutions … … … … … … … … … … …. … … ….. 15
2.3 LC-MS Analysis … … … … … … … … … … ….. … … … … … … … … ……16
2.3.1 LC-MS Principles … … … … … … … ….. … … … … … … … … ……16
2.3.2 Instrumentation … … … … … … … ……. … … … … … … … … ……18
2.3.3 Validation … … … … … … … ……. … ... … … … … … … … … ……18
Chapter 3: Adsorption Kinetics of Paraquat … … … … … … … … … … … … ……….20
3.1 Background … … … … … … … ……. …... …... … … … … … … … … ……20
3.2 Methodology… … … … … … … ……. …... …...... … … … … … … … … ……20
3.3 Spectrometric Results… … … … … … … ……. …. … … … … … … … …….....21
3.4 LC-MS Results … … … … … … … ……. …. ….... … … … … … … … …….....24
3.5 Modeling of Adsorption Kinetics… … … … … ….. … … … … … … … …….....27
3.5.1 Langmuir Adsorption Model… … … … … … … … … … … ….……....27
3.5.2 Competitive Binding Assay… … … … … …. … … … … … … … …….....30
v
Chapter 4: Role of H2O2 in Degradation of Paraquat… … … … … … … … … ……….32
4.1 Background … … … … … … … ……... … … … … … … … … … … … ……….32
4.2 Methodology… … … … … … … ……... … … … …... … … … … … … … ……….32
4.3Spectrometric Results … … … … … … … ……... …… … … … … … … … ……….33
4.4 LC-MS Results… … … … … … … ……... …… … … … … … … … … … ……….37
Chapter 5: Conclusion and Future Work … … … … … … … … … … … … … ……….40 References… … … … … … … …….. … … … … … … … … … … … … … … ……….41
1
Chapter 1: Introduction
1.1 Objective
The goal of this project is to explore the functionality of different silica-supported materials
for decontaminating phosphonate-containing wastewater, through an understanding of
degradation (e.g. adsorption, decomposition reaction) kinetics of the simulant, leading to
development of a catalytic membrane-based water treatment system.
1.2 Phosphonate Contamination: Sources and Impacts
Organophosphate compounds are extremely toxic and known to cause both acute and chronic
effects to humans, as well as wildlife, by inhibiting acetylcholinesterase, an enzyme
responsible for the breakdown of the neurotransmitter acetylcholine.1-7 About half (by
amount sold) of all insecticides used in the U.S. are organophosphates and classified by the
U.S. Environmental Protection Agency as highly or moderately toxic. Furthermore,
approximately 25,000 tons of chemical weapons including organophosphates are stockpiled
in the United States alone (Fig.1.1). 8,9,10 The organophosphate nerve agents (listed in Table
1.1) GA (tabun), GB (sarin), GD (soman), cyclosarin (GF) and VX are among the most toxic
chemical warfare agents known, with acute toxicity three to four orders of magnitude greater
than that of organophosphate pesticides.1,11 Approximately 30,000 tons of phosphonates are
consumed every year in the US and 16,000 tons in Europe as peroxide bleach stabilizers in
textile industries, detergents in industrial and household settings, and scale and corrosion
inhibitors in oil production, cooling water, boiler, and desalination systems.12 Although
several bacteria have the ability to metabolize phosphonates, biodegradation of the
compounds is not (or only slowly) observed in natural systems due to the presence of a large
number of natural adsorbents in the enviroment.13, 14
Because organophosphates share a common toxicity mechanism, the toxic effects from
multiple exposures to the compounds are cumulative; therefore, trace amount of phosphonate
residues from different sources, when combined, can be detrimental to humans and the
environment. Depending on the type and dosage, phosphonate poisoning can cause
contraction of pupils, profuse salivation, convulsions, involuntary urination, defecation and
eventual death by asphyxiation, due to uncontrollable activity of respiratory muscles.15
Additionally, being strong chelating agents14, high concentrations of phosphonates
2
Fig. 1.1: Unitary chemical weapons stockpile distribution throughout US.10 Green regions indicate states with Chemical Weapons Stockpiles and red regions indicate states that had chemical weapons stockpiles. About 75% of the original stockpile is now reported to be destroyed. The figure is used with permission from US Army Chemical Materials Agency.
can also result in remobilization of toxic metals in sediments and aquifers, posing a threat to
ground water and drinking water. Moreover, the likelihood of intentional contamination of
the drinking water system with phosphonate contaminants cannot be eliminated, and
disruption of the water system using phosphonates would produce severe public health and
safety risks. Therefore, a reliable and energy-efficient degradation mechanism, capable of
effectively removing phosphonates from natural water systems, would certainly be of interest
to prevent intrusion of toxicants into the human food chain and thus form a safeguard against
bioterrorism or environmental contamination.
3
Table 1.1: List of Organophosphate Nerve Agents1
Nerve Agent IUPAC name Molecular Structure
GA Ethyl N,N-Dimethylphosphoramidocyanidate
GB propan-2-yl methylphosphonofluoridate
GD 3,3-Dimethylbutan-2-yl
methylphosphonofluoridate
GF Cyclohexyl methylphosphonofluoridate
VX O-ethyl S-[2(diisopropylamino)ethyl]
methyl phosphonothioate
4
b.
Fig. 1.2: a. Function of the enzyme, acetylcholinesterase, in absence of phosphonates: positively charged nitrogen in acetylcholine is attracted to the ionic site on acetylecholinesterase, and hydrolysis is catalyzed at the esteric site to form choline and acetic acid; b. Inhibition of acetylcholinesterase: partially electropositive phosphorus is attracted to partially electropositive serine (first diagram). The second diagram represents the transition state showing formation of short lived phosphonate-enzyme intermediate, followed by a stable bond between acetylcholinesterase inhibitor part of phosphonate and the enzyme itself (third diagram) preventing interaction of acetylcholine with the enzyme. Fig. 1.2a and Fig. 1.2b are modified (with permission from the copyright holder, Sage Publications and Centers for Disease Control and Prevention) from diagrams available in the literature.2,7
a.
5
1.3 Mechanism of Intoxication
Nerve agents act by inhibiting the hydrolysis of acetylcholine by acetylcholinesterase.
Acetylcholine is formed and released from the nerve cell, it binds to a muscle cell receptor
for a split second, stimulates a muscle function (e.g. moving an arm, taking a breath) and then
degrades (hydrolyzes) into choline and acetic acid (Fig. 1.2a). 1,3-7 Once an organophosphate
nerve agent is absorbed through skin or inhalation, the phosphorus atom binds to a serine
hydroxyl group in the active site of acetylcholinesterase and inactivates the enzyme
responsible for breaking down acetylcholine (Fig. 1.2b). The result is excessive accumulation
of acetylcholine at synapses, resulting in sustained and uncontrolled stimulation of the
nervous system.
1.4 Detection and Decontamination Challenges
Due to the extreme level of toxicity and usage restriction, research on chemical behavior of
organophosphosphate pesticides or nerve agents are conducted by using simulants that are
less toxic but have chemical and physical properties similar to that of the actual compound.
However, it is less likely for a proxy compound to satisfactorily reflect all the chemical and
physical behavior of a given agent, and therefore much less is known about the
environmental fate and behavior of phosphonate contaminates. Although depending on the
physical-chemical properties of interest, a number of different phosphonate simulants are
(e.g. Prophos, DMMP, DIMP, TMP, TEP) 9, 16, 17 proposed and used in the literature to study
environmental behavior of phosphonates, experimental study on phosphonate degradation in
a water treatment plant is difficult due to strong adsorption of the compound onto the pipe
surfaces. Asymptotic uptake of prophos on different pipe materials is reported in the
literature18,19 Transport model predictions based on hydraulic movement assumptions are
often found to be inconsistent with field experience of actual contamination events as the
models do not consider the interaction of the contaminants with pipe surfaces and other
species present in the water distribution system.20-21 A molecular computational approach to
quantitatively characterize the sorption of phosphonates on pipe material for determining the
fate and transport constants is relatively recent, and the results suggest that organophosphate
uptake on the pipe surface can be controlled by manipulating the hydrophilic or hydrophobic
character of the surface.18
6
To date, none of the phosphonate simulants we know of are reported to exhibit light
absorption within the detectable range of convenient UV-Vis absorption-spectroscopy based
instruments. Also, the very strong interaction of phosphonates with surfaces results in low
dissolved concentration of the compounds in natural water, whereas available analytical
methods for phosphonate determination either have detection limits well above the expected
natural concentrations or suffer from interferences from natural samples. The orthodox
method that has a detection limit of about .05–10µM for the determination of phosphonates
in natural water samples, is ion-chromatography followed by post-column reaction with Fe
(III) and then spectrometric detection of the Fe(III)-complexes at a wavelength of 260nm, the
peak of an absorption band extending from 220–330nm.22 Phosphonates are also measured
by capillary electrophoresis with detection limits of 1–11 µM. 23 Another reported method is
to probe the amount of phosphates generated from post column oxidation of phosphonates,
using the molybdenum blue method. 24 Ion chromatography with pulsed amperometric
detection of amine-containing phosphonates 25 and ion-chromatography with indirect
photometric detection26 have also been detailed in the literature.24 Another analytical
approach is the formation and subsequent detection of phosphonic acid derivatives by LC-
MS.27. However, due to the cost and complexity associated with available chromatographic
methods, the techniques are also not convenient for screening large numbers of samples.
Although several enzymatic methods based on the concentration-dependent activation or
inhibition (e.g. acetylcholinererase or organophosphorus hydrolase) of the enzyme by
organophosphorus compounds are currently being explored28-30 as cheaper and reliable
alternatives, when compared to the traditional chromatographic approaches, the methods still
lack the selectivity and sensitivity required for accurate identification and quantification of
phosphanate analytes in a sample.
The usual problems faced in a phosphonate-contaminated water treatment facility are as
follows:
1. Adsorption of phosphonates onto surrounding surfaces can contaminate the whole
experimental facility.
2. Unless a cost-effective and feasible surface passivation technique is employed, the
adsorptive removal of phosphonates from water by an adsorption-based technique
cannot be distinguished separately from adsorptive loss due to phosphonate
7
contamination of surrounding surfaces. As a result, the actual phosphonate removal
efficiency of the system cannot be estimated.
3. If phosphonate concentration in the effluent is below the detection limit of the
analytical technique, removal efficiency is usually reported based on the lower limit
of the instrument and therefore, post-treatment fate of the compounds remains
unknown.
1.5 Mechanisms of Phosphonate Degradation
One primary pathway to remove phosphonates from the aqueous phase is adsorption onto
surfaces of which metal oxides are most important. For several metal oxides (SiO2, 31 Al 2O3,
32-33 Fe2O3, 12, 34-36 ZNO36) adsorption of phosphonate is reported in the literature. Presence of
different metal ions Fe(III), Zn2+, Cu(I), Cu(II), or Ca2+ in solution also influences the
adsorption behavior, and considerable increases in adsorption are reported once Ca and Zn
concentrations are added in excess of the phosphonate concentration. Natural adsorbents such
as coconut, charcoal, mussel shell, laterite stone, and clays also have good adsorption
capacity towards phosphonate residues in water.13, 37 Therefore, it is likely that large amounts
of contaminants can be removed simply by filtering phosphonates through suitable
adsorbents, and then, toxic products generated from the physisorption process can be roasted
to destroy organophosphates. However, incineration of the filtered waste may result in
signification emissions of phosphorus-containing particulates, ultimately causing a
eutrophication problem due to the reentrance of emitted phosphorus into the aquatic
environment.38-40 Fortunately, studies of nano-scale oxides of zinc, magnesium, 41-42
calcium43 and aluminum44 reacting at ambient temperatures with VX, GD, and paraxion, via
a dissociative chemisorption process, produce nontoxic phosphonates and thus, have been
promising as alternative solutions to degrade phosphonates. Morever, several basic oxide
additives (VxOy, Fe2O3), when used as dopants or coating materials, can further catalyze the
destructive phosphonate-adsorption onto the oxide surface (e.g. MgO, CaO). 45-47 Thermal
activation of the nano-scale catalysts in decomposing the chemisorbed phosphonates is also
evident in the literature.48-52 An alternative approach to avoid temperature-induced
complications of thermally catalyzed systems may be photo-activation of TiO2 to oxidize
phosphonates at room temperatures.9
8
1.6 Silica as a Catalyst Support
One major problem in catalytic/photo-catalytic decomposition of phosphonates is
deactivation of both the catalysts and support due to poisoning by P2O5 and accumulation of
reaction products, resulting in a drastic loss of available surface area. Although thermal
decomposition studies of phosphonates over metal oxides48-49 found VxOy and MnO to be
very active as catalysts, 50 sustained oxidative catalysis of DMMP can only be maintained at
temperatures >600oC for most catalyst supports.50
Fig. 1.3: Effect of metal oxide supports on DMMP conversion by V2O5 catalysts over time. 49(the figure is used with permission from the copyright owner, Elsevier
Limited)
Fig. 1.3 shows results of thermo-catalytic decomposition by V2O5 catalysts supported on
different substrates including Al2O3, SiO2, and TiO2. Of all the supports tested, silica was
found to be the most sustainable catalyst support, capable of resisting P2O5 poisoning, and
giving a protection time of 25h for 100% DMMP conversion.49 Infrared spectroscopic study
of silica-phosphonate interaction using nerve gas simulants DMMP, TMP and TCP reveals
that when heated, all the adsorbed phosphonates molecularly desorbs and the adsorbent, silica
returns to the original state.30 The inherent reusability and resistivity of silica against
phosphonate poisoning, along with strong phosphonate adsorption capability, can be one of
the reasons to select the material as a suitable support for phosphonate-degrading catalysts.
9
1.7 Overview of the Research
The current focus of the ongoing research is to characterize the functionality of different
silica-supported catalysts in decontaminating water from toxic substances, including
phosphonates. All silica samples used are proprietary to PPG Industries, and are referred to
using a sample number throughout the thesis to protect sample identities. If materials capable
of decomposing phosphonates can be synthesized, combined application of the material with
a membrane support or filter can ensure effective destruction and removal of the
contaminants.
Due to the complexity of conducting experiments with phosphonates (stated earlier in section
1.2 and 1.3), an equally important but easier to study compound, paraquat dichloride
(properties listed in Table. 1.2) has been chosen as the proxy contaminant in an initial
approach to test the activity of different functionalized and non-functionalized silica particles.
The primary reasons for selecting paraquat dichloride are as follows:
1. Although there is no similarity in molecular structure and intoxication mechanism of
paraquat and phosphonates, both the compounds are known as toxic contaminants and
have the potential to be used as weapons of mass destruction. Due to extreme toxicity,
risk of environmental contamination resulting from heavy agricultural use of paraquat
as herbicide, removal of the compound from the aquatic environment is a task equally
as important as removing phosphonates, and therefore, finding a silica-supported
catalyst capable of effectively destroying paraquat will certainly aid in achieving the
broader goal of the research.
2. In contrast to phosphonates, paraquat dichloride is a quaternary ammonium salt that is
highly soluble in water and can be easily detected with a spectrophotometer, thereby
making investigation of silica-paraquat interaction easier.
10
Table 1.2: Properties of Paraquat Dichloride
IUPAC name :1,1'-dimethyl-4,4'-bipyridinium
dichloride Molecular Formula : C12H14Cl2N2·xH2O
Synonyms: Gramoxone Methyl Viologen Dichloride Molar Mass : 257.16 g/mol Appearance : Light yellow Form: Crystalline Melting Point:>300 oC Detection : UV -258 nm NFPA Rating:4-0-0
The thesis begins with a presentation of the analytical techniques used to apprehend the role
of silica particles in paraquat degradation chemistry (Chapter 2). Research is then presented
which experimentally investigates the adsorption phenomenon of paraquat onto the silicas
(Chapter 3). Nature of the adsorption was further characterized by modeling the experimental
data with well-known mathematical adsorption equations. Peroxide functionality in
degradation of the proxy contaminant has been explored with an understanding of H2O2-
paraquat interaction. Details of the results and reasons behind inactivity of the oxidizing
agents are explained (Chapter 4) in light of the information available from existing literature.
Development of an actual catalytic membrane system ensuring easy handling and efficient
decomposition of both phosphonates and paraquat will continue to be the subject of ongoing
research.
11
Chapter 2: Methods of Measurement
2.1 Spectrometric Determination of Paraquat
2.1.1 Spectrometer Principle
Absorption spectroscopy exploits the light absorption propensity of chemical species as an
analytical tool to quantitatively determine the presence of a particular species in a sample
solution.15,53 Many substances absorb or transmit certain wavelengths of radiant energy, and
in absorption spectroscopy, the sought-for information is the fraction of externally generated
radiant power (absorbance, A) absorbed by the analyte species. Most often in absorption-
spectroscopy based instruments, the absorbance is determined by measuring the transmittance
(T).
T=IT/I0 … … … … … (2.1)
TA log−= … … … … … (2.2)
Where,
Io =Intensity of incident light IT = Intensity of transmitted light
Then again, as per the Beer-Lambert law, the absorbed radiation depends on the thickness of
the medium (e.g. path length of the sample container) and concentration of the adsorbing
species. The relation can be expressed as follows,
lcTA ε=−= log … … … … … (2.3)
Where,
Light Source
Sample with C concentration
of analyte Detector
Io I
T
l
c=Concentration of the sample ε= Molar extinction coefficient of the analyte l = Path length of the cuvette
Fig. 2.1: Measurement Principle of a UV-Vis Spectrometer
12
As the molar extinction coefficient of the analyte is constant at fixed chemical conditions
(pH, ionic strength, solvent etc.), for a sample placed in a cuvette of fixed path length,
absorbance is a function of analyte concentration only; hence concentration of the analyte can
be determined using Eq. (2.3) once spectrometric absorbance measurements are known.
So far, two different approaches are available for spectrometric determination of paraquat
residues in water:54-56
1. Measuring UV absorption of paraquat ions in aqueous solution: As there is a greater
tendency of formulation additives to interfere at the lower wavelengths, the method
can only be considered reliable for very pure solutions of paraquat.
2. Generation of colored paraquat radicals by alkaline dithionite reduction, followed by
determination of paraquat residues from the peak in the visible absorption spectrum.
This method is considered convenient for determination of paraquat in the formulated
products.
For our research, measurement of the paraquat ion in the UV region was chosen to be the
appropriate method as
(i) The calibration for the method provided a reproducible detection limit
(discussed in section 2.1.4 )
(ii) The transience of paraquat radicals generated by dithionite reduction may
interfere with estimating of catalyst-induced degradation.
The paraquat solutions prepared in the lab are free of additives. Further, any
degradation might interfere chemically or spectroscopically with the assay, so
the possibility of interference by any undesired compound other than
degradation products is moot.
2.1.2 UV Absorption of Paraquat
Calibration of the spectrophotometer for UV absorption of paraquat was performed at pH
=4.96 (0.2 M acetate buffer) and at pH=8.06 (a 0.025 M borate buffer). The idea was to
determine the reliable working range of concentrations of the instrument as well as to
investigate the pH effects on absorbance measurements. Paraquat dichloride was purchased
from Sigma Aldrich, MO, USA. Since paraquat salts are highly hygroscopic, the powders
were dried at 100oC for 5 hours immediately before use to ensure that it reached a constant
weight before use as a standard. After drying, the powders were cooled and stored in a
desiccator. To prepare paraquat solutions of stock concentrations, the dried paraquat salts
13
were weighed using an adequately sensitive microbalance, and then sample parqaut solutions
of various concentrations were prepared by diluting aliquots of the stock solutions. pH of the
solutions were measured using a pH meter, and absorbance measurements of the solutions
were made by a HP 8542A diode array spectrophotometer.
2.1.3 Validation
Fig. 2.2 shows the absorbance spectrum of paraquat at pH=4.96 (acetate buffer). The
wavelength of maximum absorbance is 258 nm, as was expected from literature54-55
Fig. 2.2: The absorption spectrum of paraquat dichloride at pH=4.96 (acetate buffer). Concentration
of the solution=0.22 ppm.
2.1.4 Calibration Results
UV absorption of paraquat was found to be highly pH-dependent. The absorbance of
paraquat solutions in the alkaline media (Fig. 2.4, for pH = 8.06) was found to be
approximately four times higher than absorbance measured with solutions in an acidic
medium (Fig. 2.3 for pH = 4.96). For the alkaline paraquat solution (Fig. 2.4, for pH = 8.06),
the working range of concentrations was from 0.25 ppm to 25.0 ppm.
14
Fig. 2.3: Calibration curve for UV absorption of paraquat dichloride at 258 nm, pH=4.96 (acetate buffer). Regression analysis was conducted for 95% confidence interval. The plot is poorly linear, showing a detection limit of greater than0.05 ppm.
Fig. 2.4: Calibration curve for UV absorption of paraquat dichloride at 258 nm, pH=8.06 (borate buffer). Regression analysis was conducted for 95% confidence interval. The plot is linear with a high R2 value (0.996). Only data represented by the dark blue points were used in calculating the linear regression. These dark points show the working range of the spectrophotometer.
0.000
0.005
0.010
0.015
0.020
0.025
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Ab
sorb
ance
at
258
nm
Concentration (ppm)
Linear regression lineat 95% confidenceinterval
Lower 95% confidenceinterval
0.000
0.500
1.000
1.500
2.000
2.500
3.000
0 10 20 30 40 50
Ab
sorb
ance
at
258
nm
Concentration (ppm)
Linear regression line at95% confidence intervalLower 95% confidenceintervalUpper 95% confidenceintervalExperimental Data
15
2.2 Calibration Curve for H2O2 solutions
During this research, addition of H2O2 solutions into both silica-paraquat heterogeneous
mixtures and homogeneous paraquat solutions (no silica) were attempted to investigate
peroxide functionality in decomposing paraquat. Therefore, calibration of the
spectrophotometer for absorbance of H2O2 was also conducted (Fig. 2.5) to effectively
interpret the absorbance data of H2O2-paraquat solutions. Before calibration, H2O2 solutions
were standardized using cerium (IV) titration.57, 58 Initially titration was attempted by
preparing a standard 0.1M ammonium cerium (IV) nitrate, (NH4)2Ce(NO3)6 solution.
Fig. 2.5: Calibration curve for the UV absorption of hydrogen peroxide (H2O2) at 258 nm, pH of the solution was 5.85 (in acetate buffer).
However, due to the low solubility of the ceric ammonium salt in aqueous solution, the
titration procedure became complicated. Alternatively, commercially available standard 0.1M
ceric sulfate, Ce(SO4)2 solution was used to complete the standardization process. The
concentration of the reagent-grade H2O2 used was found to be 30.5% w/v.
0.000
0.200
0.400
0.600
0.800
1.000
0 5 10 15 20 25 30 35 40 45 50
Ab
sorb
ance
at
258
nm
Concentration (ppm)
Linear regression lineat 95% confidenceintervalLower 95% confidenceinterval
16
2.3 LC-MS Analysis
Although spectrometry offers the ease of measuring the concentration of paraquat,
quantitative analysis based on the method is only reliable if multiple analytes do not absorb
light in the same wavelength region. Therefore, to study paraquat degradation chemistry
without any interference and accurately identify any potential degradation byproducts that
may form if paraquat decomposition is prominent, investigation using a more selective
analysis (e.g. Liquid Chromatography-Mass Spectrometry (LC-MS)) than spectrophotometry
was deemed necessary.
2.3.1 LC-MS Principles
Liquid chromatography–mass spectrometry (LC-MS, or alternatively HPLC-MS) is an
analytical chemistry technique that combines the physical separation capabilities of liquid
chromatography (or HPLC) with the mass analysis capabilities of mass spectrometry.
In chromatographic separation, a high pressure pump is used to generate flow of a sample-
free phase, called the mobile phase, over a second sample-free stationary phase, which is
usually a solid or liquid film coated on the inner substrate of an HPLC column. An injector
(auto-sampler or sample manager) can inject the sample into the continuously flowing mobile
phase stream that transports the sample into the HPLC column (Fig. 2.6). The columns are
designed to employ individual solute-stationary phase interactions to separate the solution
components. Depending on the strength of interaction of the components with the stationary
phase, different component elutes at different times (Fig. 2.7) from the column, and then the
isolated compounds are detected by distinct signals of a detector (e.g. photodiode array
detectors). A mass spectrometer can analyze a sample solution based on the mass to charge
ratio of the particles present in the solution. After passing the HPLC instrument and diode
After passing the HPLC instrument and diode array detectors, the sample is loaded into the
mass spectrometer, ionized and finally separated according to mass-to-charge ratios in an
analyzer by applying electromagnetic fields.15
17
Fig. 2.6: Schematic Diagram of an HPLC instrument
a.
b.
c.
Fig. 2.7: (a) Progress of a Column Chromatographic mixture with two solutes: As mobile phase passes through the column, the sample separates into two solute bands and elutes from the column59; (b) concentration of each solute as a function of distance travelled down the column59; (c) Individual detector’s response of the isolated solutes.59(Figure is reused with permission from the copyright owner, Prof. David T. Harvey)
18
2.3.2 Instrumentation
Chromatographic separation was performed using a Waters 2795 separation module (reverse-
phase HPLC) equipped with a quarternary solvent delivery system, auto-sampler and column
heater. The column used was Eclipse XDB C18 (2.1mm×50mm). While testing the samples,
the column was first decontaminated from previously used solutions by running a 100 ppm
unbuffered aqueous solution of paraquat through the system.
Gradient elution was used for separation of the aliquots by varying the polarity of the mobile
phase. Solvent A was a mixture of 95% water and 5% acetonitrile, and solvent B was a
mixture of 95% acetonitrile and 5% water. The elution solvent used was initially 100%
solvent A and 0% solvent B. Over nine minutes, the solvent mixture was gradually changed
to 100% solvent B and 0% solvent A. From nine to ten minutes, the solvent mixture was
rapidly changed to 100% solvent A.
Mass spectrometry was carried out using a Waters Quattro Ultima equipped with electrospray
ionization and operated using MassLynxMS software. Ionization takes place in the source at
atmospheric pressure. The ions are sampled through a series of orifices into the first
quadrupole where they are filtered according to their mass to charge ratio (m/z). Data
acquisition was performed from m/z=100 to m/z=500.
2.3.3 Validation
Fig. 2.8 shows the mass spectrum of paraquat dichloride obtained experimentally from a 100
ppm solution by our LC-MS analysis. The ion at m/z =186 represents the singly-charged
paraquat ion [PQ]+. However, the most intense peak was observed at m/z = 185, indicating
the abundance of deprotonated paraquat cation [PQ-H]+. The peak at m/z= 171 can be
attributed to the loss of a methyl group [PQ-CH3] + from a singly-charged paraquat cation.
It should be noted that the mass spectra of aqueous paraquat solution should only exhibit the
dication, [PQ] 2+ unless single charged free radical cations, [PQ] +, are generated by using a
reduction mechanism such as sodium dithionite addition.53,54 However, reports of abundance
of both single charged and double charged paraquat ions in the mass spectra are found in the
literature. Several factors such as buffer composition,60 presence of acetonitrile as a weak
19
Fig. 2.8: Experimentally-obtained ESI Mass Spectrum for Paraquat Dichloride.
electron donor,61 ionization techniques and parameters may enhance the formation of singly-charged cations when 100 ppm sample solution of paraquat is transported into the LC-MS system. Our experimental data (Fig. 2.8) was found to accord with one such published result.62
20
Chapter 3: Adsorption Kinetics of Paraquat
3.1 Background
Paraquat adsorption behavior on as-received silica samples from PPG was characterized
using both an experimental approach and a mathematical modeling approach. The idea was to
identify the best silica candidate for adsorption as well as to use knowledge of adsorption to
simplify further analysis of any silica-assisted paraquat-decomposition chemistry.
3.2 Methodology
For spectrometric analysis of paraquat adsorption kinetics, each of the functionalized and
control silica samples was mixed separately with buffered proxy (paraquat dichloride)
solution. The initial concentration of paraquat used in each trial was 25 ppm. All the silica
samples except sample 13-16 were tested as received. Samples 13-16 were dried at 100o C
before use as recommended by PPG Industries. Each silica sample was added to the paraquat
solution to give a final silica concentration of 1.66 g/L. Because we had no knowledge of the
specific surface area, this constant mass concentration approach confounded some theoretical
understanding, as we could not maintain constant surface area or constant active site count.
The system was kept in suspension using a magnetic stirrer. Sample aliquots were initially
collected at regular intervals of 5 seconds for the first 30 seconds and then at intervals of 30
seconds for the next 120 seconds. To determine the total adsorption capacity of silica
particulates, three additional data were also collected for each of the samples over 48 hours.
All the solution aliquots were filtered using 0.45µm Millipore filters. The first few drops of
the filtrate were used to decontaminate the cuvette from previous solutions and the rest was
used for absorbance measurement with the spectrophotometer.
LCMS analysis of the samples was also conducted to investigate if the contaminant
adsorption onto the silica particles is accompanied by any chemical dissociation or
degradation phenomena. To reduce noise in the mass spectra from buffer components,
paraquat solutions were prepared using only deionized water. Each silica sample (1.3 g/L)
was added to 5 mL of the paraquat dichloride solution and the system was kept in suspension
using a magnetic stirrer. Samples were collected after two minutes of stirring using 0.45 µm
syringe filters to filter out the silica particles.
21
As the LCMS results confirmed physorption to be the only active degradation mechanism,
UV-Vis assay was repeated for commercially-available activated carbon products (Darco
G60 from Fisher Scientific, USA), a frequently used material for physorption based
applications, to have a basis of comparison for paraquat removal efficiency of the silica
samples.
Also, in a theoretical approach to understand the adsorbate transport mechanism into surface
sites of the particles, adsorption data were fit to both Langmuir isotherm and competitive
adsorption site binding models. The details are presented in section 3.4.
3.3 Spectrometric Results
For all the tested samples, except samples 1, 9, 11 and 15, paraquat adsorption tends to reach
equilibrium within the maximum observation time- of 48 hours (Fig. 3.1). Paraquat
adsorption capacity was found to be the greatest for activated carbon (Table. 3.1) powders;
however, more than 96% of paraquat was removed from solution with silica samples 3, 4, 5,
6, 8,13 and 16. Out of the silica samples for which the adsorption kinetics did not reach
equilibrium, paraquat removal was maximum (90% of initial concentration) for sample-1.
Significant decrease in paraquat concentration was noticed within the first 5 seconds after
adding any of the silicas (Fig. 3.2 and Fig. 3.3). Silica samples 9-16 were supplied by PPG
Industries at a later stage of this research, and among the new samples, an apparently
exponential trend in the adsorbance time series was evident for samples 10, 11, and 13 where
degradation rate for sample 13 was relatively rapid (Fig. 3.3). Paraquat adsorption rate by
sample-10 was the slowest; however, total adsorptive paraquat removal by the sample was
found to be higher than that by samples 11, 12 and 15 (Fig. 3.1 and Table. 3.1). As sample
10 is granulated and has low surface area by configuration, the sample probably offers fewer
available adsorption sites but more particles in a solution compared to the same quantity of
other silica particles with larger surface areas.
22
Fig. 3.1: Absorption kinetics of paraquat onto the tested particles over 48 hours, with time plotted on
a log scale. Initial concentration concentration of paraquat was 25 ppm, and pH of the solution was 8.06 (using borate buffer).
Fig. 3.2: The measured paraquat concentrations for silica samples 1-8 up to 150 seconds after adding
the particles. Sample aliquots were collected at intervals of 5 seconds. pH of the solution was 8.04 in borate buffer.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
1 10 100 1,000 10,000 100,000 1,000,000
[par
aqu
at]
(pp
m)
time (sec)
Sample-1 Sample-2Sample-3 Sample-4Sample-5 Sample-6Sample-7 Sample-8Sample-9 Sample-10Sample-11 Sample-12"Sample-13 (dried)" "Sample-14 (dried)""Sample-15( dried)" "Sample-16 (dried)"Activated Carbon
0.00
5.00
10.00
15.00
20.00
25.00
30.00
0 5 10 15 20 25 30
[par
aqu
at]
(pp
m)
Time (sec)
Sample 1 Sample 2 Sample 3 Sample 4
Sample 5 Sample 6 Sample 7 Sample 8
23
Table 3.1: Paraquat adsorption data 48 hours after adding the silica particles
Type of Particles
Initial paraquat concentration
(ppm)
Paraquat concentration after
48 hours (ppm)
Percent paraquat adsorbed
(%) Silica sample-1 25 2.5 90 Silica sample-2 25 1.7 93 Silica sample-3 2 0.5 98 Silica sample-4 25 0.5 98 Silica sample-5 25 0.6 97 Silica sample-6 25 0.4 98 Silica sample-7 25 1.2 95 Silica sample-8 25 0.9 96 Silica sample-9 25 6.1 75 Silica sample-10 25 1.3 95 Silica sample-11 25 9.8 61 Silica sample-12 25 13 50
Silica sample-13 (dried) 25 0.8 97 Silica sample-14 (dried) 25 1.1 96 Silica sample-15 (dried) 25 4.7 82 Silica sample-16 (dried) 25 0.3 99
Activated Carbon 25 0.1 100 Considering adsorption capacity and paraquat removal rate, sample-3 exhibited the best
performance among all the tested silica as 93% of paraquat was removed from the
heterogeneous solution within the first 150 seconds after adding the silica.
24
Fig. 3.3: The measured paraquat concentrations after adding the eight silica samples 8-16. Samples
were collected up to 150 seconds after adding the silica particles. Sample aliquots were collected at intervals of 5 seconds. pH of the solution was 8.04 in borate buffer.
3.4 LC-MS Results
Fig. 3.4(a) exhibits the chromatograms for the control sample (100 ppm paraquat solution).
For all the initially-received silica samples except sample 5 (treated with polyethylene glycol,
PEG), the chromatograms were similar to Fig. 3.4(a). The chromatogram for sample 5 (Fig.
3.4(b)) displayed additional elution peaks from retention time t = 2.54 minutes to 4.08
minutes. The compounds, which eluted between 2.54 minutes and 4.08 minutes, can be
identified from an analysis of the corresponding mass spectrum shown in Fig. 3.5. Two
significant features can be noted:
(i) The combined mass spectrum, taken from t = 3.61minutes to 4.08 minutes,
displays mass peaks at m/z values higher than that of the precursor ion,
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
0 5 10 15 20 25 30
[par
aqu
at]
(pp
m)
Time (sec)
Sample-9 Sample-10
Sample-11 Sample-12
"Sample-13 (dried)" "Sample-14 (dried)"
"Sample-15 (dried)" "Sample-16 (dried)"
Activated Carbon
25
Fig. 3.4a. Chromatogram of aqueous solution of paraquat dicholride; b. Chromatogram of
paraquat solution mixed with sample 5. The initial concentration of parquat dichloride
solution was 100 ppm.
[PQ]+, and thus indicates the inclusion of additional compounds into the solution,
most likely from the silica surface (sample 5).
(ii) The presence of the mass peaks at regular m/z intervals of 44.0 further confirms
that oligomers of PEG are present, since the monomer of polyethylene
glycol, (-CH2-O-CH2), has a fragment mass of 44.0 Dalton.
t(min)
t(min)
a)
b)
26
Fig. 3.5. Combined mass spectrum of 100 ppm paraquat solution degraded by silica sample 5, from retention time t=3.61 min to 4.08 min.
Fig. 3.6. Mass spectrum of 100 ppm paraquat solution mixed with silica sample 9, at retention time
t=7.74 min.
m/z
27
Among samples 9-12, additional mass peaks were identified at m/z = 305.3, 273.3, and 327.3
for sample 9 (Fig. 3.6) at a retention time of 7.74 minutes. Similar mass spectra and
chromatograms were obtained for samples 11 and 12 at retention times of 7.85 minutes and
7.98 minutes, respectively. However, the identity of these peaks cannot be determined
from the information at hand.
3.5 Modeling of Adsorption Kinetics
3.5.1 Langmuir Adsorption Model
According to Langmuir, adsorbate molecules are assumed to bind to distinct empty sites on
the surfaces of the adsorbent molecules through the following reversible process:
Where, A represents an adsorbate molecule, S is an adsorption site on the surface of the
adsorbate, and A-S is the complex formed due to adsorption of the solute molecule onto the
adsorbent surface. ak and dk represent adsorption and desorption rate constants respectively.
The basic idea behind the Langmuir adsorption model is the coverage of the homogeneous
adsorbent surface by a mono-molecularic layer of adsorbate molecules. As per this adsorption
model, the fraction of the adsorbent surface occupied by the solute (θ ) can be expressed as a
nonlinear function of initial solute concentration (C) by the following equation:
θ = K cC
1+ K cC K c = ka
kd
… … …… (3.1)
Furthermore, by balancing the relative rates of uptake and release, the Langmuir adsorption
isotherm modeling can be related to the rate of change of surface coverage (dθ/dt) [Eq. 3.2]
and then can be integrated to predict the adsorption kinetics on a time scale [Eq. 3.3].
ka
A-S A+S
k d
28
dθdt
= ka (1− θ)C − kdθ … … … … … (3.2)
Integrating Eq. 3.2 and taking K ' = C
C + kd
ka
andK '' = kaC + kd , the fraction of the
surface coverage can be calculated as a function of time.
θ(t) = K ' [1− exp(−K ''t)] … … … … (3.3)
A comparison between the Langmuir model predictions based on [Eq. 3.3] and experimental
adsorption data for paraquat uptake onto sample 10 is demonstrated in Fig. 3.7. The results
found Langmuir isotherm modeling to be a poor fit to the experimental data. The potential
reasons may be as follows:
1. The Langmuir modeling was attempted with the assumption that paraquat is the only
adsorbing molecule and that all sites are initially free (i.e. none of surface sites are
pre-adsorbed with water). However, when a species is adsorbed from solution, there
should be accompanying desorption of another species for mass as well as charge
balance and entropy consideration.
2. The surface of the silica may be heterogeneous which contradicts the important
Langmuir premise that the adsorption sites must be equivalent.
29
Fig. 3.7: Comparison between Langmuir model predictions and experimentally obtained adsorption data over 150 seconds for sample 10. Concentration of silica in 25ppm paraquat solution (initial concentration) was 1.67 g/L. The exponential Langmuir model fit decays more slowly than the experimental data.
Fig. 3.8: Comparison between competitive binding Model predictions and experimentally obtained
adsorption data over 150 seconds for sample 10, with time plotted in logarithmic scale. Concentration of silica in 25ppm paraquat solution (initial concentration) was 1.67 g/L. Within the timeframe shown, model predictions are in good agreement with the experimental adsorption data
5.00
7.00
9.00
11.00
13.00
15.00
17.00
19.00
21.00
23.00
25.00
0 20 40 60 80 100 120 140
[par
aqu
at]
(pp
m)
Time (sec)
Experimental
Langmuir Model
5.00
7.00
9.00
11.00
13.00
15.00
17.00
19.00
21.00
23.00
25.00
1 10 100
[par
aqu
at]
(pp
m)
Time (sec)
Experimental
Competitive Binding Method
30
Fig. 3.9: Comparison between Langmuir Model predictions and experimentally obtained adsorption data over 48 hours for sample 10, with time plotted in logarithmic scale. Concentration of silica in 25ppm paraquat solution (initial concentration) was 1.67 g/L. The model fits well with experimental data collected for the first 2 hours but the actual paraquat concentration is much lower than the model predicts after 48 hours.
3.5.2 Competitive Binding Assay
Competitive binding assays are based on the idea that two molecules A and B compete for
the same adsorption site, and adsorption of one molecule (A) onto an adsorption site is
accompanied by the desorption of the previously adsorbed molecule (B) from that particular
site. The adsorption process can be represented by the following interaction:
Assuming all the surface sites are initially saturated with molecule B before adsorption of A
occurs, fractional coverage of the surface by adsorption of molecule A (θA ) can be expressed
1
10
100
1 10 100 1000 10000 100000
[par
aqu
at]
(pp
m)
Time (sec)
Competitve Binding ModelExperimental Data
A+ B-S
ka
k d
A-S+ B
31
as a function of the solution concentration (C) by balancing the adsorption and desorption
rates of A at equilibrium.
θA = KC
1+ KCθ … … … … … (3.4)
Where,
θ = θA + θB = Fraction of surface sites occupied by the adsorbates A and B at equilibrium
K = ka
kd
=Ratio of adsorption constant to desorption constant of A
On a time scale, the adsorption kinetics of the model is governed by the following equation,
θA (t) = K1t
1+ K1tθ… … … … (3.5)
The competitive binding model was attempted to predict paraquat adsorption kinetics based
on the idea that all the surface adsorption sites are occupied with water before paraquat
adsorption occurs. The assay was found to be in good agreement with the experimental data
for sample 10 through the first 1200 seconds (Fig. 3.8) assuming 70% surface coverage at
equilibrium (θ=0.7), and constant K1=0.4 sec-1. The consistency of the modeling and
experimental results suggests that the adsorption kinetics of the silica catalyst are governed
by competitive adsorption of water and paraquat. However, comparison of competitive
binding model prediction with experimental data over 48 hours exhibits that the model
deviates from experimental data over time. The final paraquat concentration was found to be
much lower than the extrapolated model fit after 48 hours (Fig. 3.9). One potential
rationale explaining such contradiction of our experimental observation with model
calculations is that the competitive binding model neglects heterogeneity of binding sites
and porosity of the silica particles. Therefore the molecules adsorbed onto the surface of the
silica particles may diffuse into the adsorbents over time, and thus the particles actually
offer available surface sites for more adsorption than the model estimates. Once surface
configuration of the silica particles are determined, improved predictions of paraquat
adsorption kinetics onto silica can be achieved, based on branched pore diffusion model, by
incorporating the influence of internal structure of the adsorbents on the adsorption
characteristics and iteratively selecting best combination of mass transfer parameters
which accounts for different adsorption phenomena occurring at the solute particle
interface. 63
32
Chapter 4: Role of H2O2 in Degradation of Paraquat
4.1 Background
As adsorption was found to be the only active sequestration or degradation mechanism during
paraquat-silica interactions, addition of exogenous oxidant (H2O2) to the heterogeneous proxy
solutions (with silica) was attempted to provide a reactant so that silica might be able to
catalytically decompose paraquat. To discriminate the contribution of silica catalysis from
homogeneous, non-catalytic paraquat decomposition by H2O2, paraquat- H2O2 interaction was
also investigated without adding silica particulates into homogeneous solution mixtures.
4.2 Methodology
Initially, spectrophotometric analysis of homogenous solutions in which H2O2-was added to
paraquat was conducted to identify formation of any light-absorbing reaction by products.
The results (Fig. 2.5 and Fig. 4.1) indicate the presence of additional light absorbing
compounds as the total absorbance of H2O2-paraquat solution was found to be higher than the
summation of individual absorption of both the participating components. For having further
insights into the degradation chemistry, LC-MS analysis was adopted to identify the
degradation products present in the solution. The LC-MS assays of paraquat decomposition
were conducted by adding 50 mM hydrogen peroxide to unbuffered paraquat solutions
(pH=5.85) in the presence of 2 g/L of presaturated silica particles (sample 1-8). Because UV-
Vis analysis confirmed rapid adsorptive removal of paraquat by all the available silica
supports (chapter-3), paraquat solutions were separately mixed with the silica samples for
two minutes before adding H2O2 to the solutions. The aim of this method was to saturate the
active adsorption sites on the surface of the adsorbents and thus distinguish between the
contributions of simple adsorption and H2O2 assisted decomposition in degrading paraquat.
As no oxidative decomposition of paraquat was detected in LC-MS analysis, the interaction
between the compounds (e.g.H2O2 and paraquat) was further investigated spectrometrically
by introducing varying concentrations of H2O2 to buffered paraquat solutions
(pH=5.87.acetate buffer). The idea was to look for unstable paraquat-H2O2 complex
formation and better understand the nature and stoichiometry of interaction, so viable
techniques to catalyze paraquat-decomposition could later be attempted.
33
Fig. 4.1 UV-Vis absorption spectra of the 25 ppm unbuffered paraquat solution before and after adding 25.1 mM and 50.2 mM hydrogen peroxide solution. The pH of the solution was measured to be 5.85.
4.3 Spectrometric Results
The light absorption behavior of the H2O2- paraquat solution was found to have a dependency
on H2O2 concentration present in the solution. Fig. 4.2 shows the absorption spectra of the
0.48 mM buffered paraquat solution at pH 5.87 after adding varying concentrations of H2O2
solutions. The absorbance of the solution was found to decrease at 258 nm, suggesting a
corresponding decrease of paraquat in the homogeneous H2O2- paraquat solution and reaction
to form an as-yet unidentified product. Also, the absorbance peak of the solution shifted to
lower wavelengths for higher concentrations of H2O2 in solution, possibly signifying the
presence of light absorbing unknown compounds. However, simplification of data presented
in Fig. 4.2 was necessary to completely understand the chemistry behind the continuous shift
of the absorption peak due to addition of H2O2 in the homogenous solution, and thus to
identify the unknown products. Therefore, the absorption spectra in Fig. 4.2 were re-plotted,
subtracting the putative absorbance of H2O2 (Fig. 4.3). Neither an additional absorbance peak
was identified nor was the position of the absorption peak corresponding to paraquat changed
with increasing H2O2 concentrations in solutions.
34
Fig. 4.2 Absorbance spectrum of H2O2-added paraquat solutions for varying H2O2 concentrations.
Concentration of paraquat solution was 0.48mM. The labels indicate the concentration of H2O2 added. pH of the solution was 5.87 (acetate buffer)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
210 220 230 240 250 260 270 280 290 300 310
Ab
sorb
ance
Wavelength (nm)
2.4 mM hydrogen peroxide
14.2 mM hydrogen peroxide
25.9 mM hydrogen peroxide
37.5 mM hydrogen peroxide
48.9 mM hydrogen peroxide
60.1 mM hydrogen peroxide
71.2 mM hydrogen peroxide
82.1 mM hydrogen peroxide
93.0 mM hydrogen peroxide
98.3 mM hydrogen peroxide
35
Fig. 4.3 Corrected absorbance spectrum (i.e. corrected absorbance = absorbance of H2O2-added
paraquat solution minus computed, scaled absorbance of H2O2 in that solution) of H2O2-added paraquat solutions for varying H2O2 concentrations. Concentration of paraquat solution was 0.48mM. The labels indicate the concentration of H2O2 added. pH of the solution was 5.87 (acetate buffer). The only shift is due to offset in the baseline, which may be due to changes in solution refractive index or instrument drift.
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
200 220 240 260 280 300 320 340 360 380 400
Co
rrec
ted
Ab
sorb
ance
Wavelength (nm)
2.4 mM hydrogen peroxide 8.3 mM hydrogen peroxide 14.2 mM hydrogen peroxide20.1 mM hydrogen peroxide 25.9 mM hydrogen peroxide 31.7 mM hydrogen peroxide 37.5 mM hydrogen peroxide 43.2 mM hydrogen peroxide 48.9 mM hydrogen peroxide 54.5 mM hydrogen peroxide 60.1 mM hydrogen peroxide 65.7 mM hydrogen peroxide71.2 mM hydrogen peroxide76.7 mM hydrogen peroxide 82.1 mM hydrogen peroxide87.6 mM hydrogen peroxide93.0 mM hydrogen peroxide98.3 mM hydrogen peroxide
36
Fig. 4.4 Continuous Variation (Job’s) plot data for varying mole fractions of paraquat (cA),
maintaining a total concentration of 0.48 mM. pH of the solutions were 5.87 (acetate buffer). The results confirm that no reaction is occurring between H2O2 and paraquat, and the
apparent shift observed in Fig. 4.2 is actually a measurement artifact that arises due to
interference from H2O2.
To further verify the results, the possibility of paraquat- H2O2 complex formation was also
explored based on Job’s method of continuous variation. Absorbance spectra were collected
with a total concentration C of 0.48 mM (C = [PQ] + [H2O2]). Several different H2O2-added
paraquat mixtures were prepared with varying mole fractions of paraquat and H2O2. Fig. 4.4
shows the Job’s plot data collected at 258 nm for H2O2-added paraquat solution with
increasing mole fractions of paraquat. The plot does not show any maxima or minima
corresponding to the formation of a stoichiometric paraquat- H2O2 complex. Similar trends
were also observed in the Job’s plots at wavelengths of 240 nm and 226 nm (results not
shown). The results confirm that H2O2 alone does not have any significant interaction with
paraquat.
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.0 0.2 0.4 0.6 0.8 1.0
Ab
sorb
ance
at
240
nm
cA/C
37
4.4 LC-MS Results
The LC-MS analysis of H2O2-added paraquat solutions with or without the catalysts (Fig.
4.5) further corroborates the spectrometric results. No decomposition was evident for any of
the aliquots tested. A silica sample which was prepared as described above (sample 2) was
also tested against a paraquat and H2O2 mixture to identify if the inactivity of the catalysts
was due to the reduction of surface active sites by pre-adsorbed paraquat. The results (Fig.
4.5 (b)) indicate that the heterogeneous solution remains nonreactive, even in the presence of
putatively-reactive surface sites.
Although H2O2 alone was found to be nonreactive with solutions of paraquat, the evidence
and mechanism of paraquat oxidation by hydrogen peroxide is already reported in the
literature.64-66 Degradation of paraquat by hydrogen peroxide requires the presence of single
charged radical cations (PQ+) or catalysts that may induce generation of hydroxyl radicals.
The generated hydroxyl radicals can further react with doubly-charged paraquat cations and
mineralize the contaminants. In presence of free radical cations (PQ+), produced by the
dithionite reduction of paraquat solution, hydroxyl radical generates through the following
pathway.64
However, during our research, all the investigations were conducted using buffered solutions
containing PQ2+ for simplification of the analysis, and radical generation was expected to be
favored by functionality added to the silica. Hence, based on the LC-MS, it can be concluded
that the functionalities currently added to particles are not yet effective for decomposing
paraquat.
H2O2 + e- OH- + OH
.
PQ+ PQ2+ + e-
38
Fig 4.5 Chromatograms of 100 ppm paraquat solution when mixed with approximately 50 mM H2O2
and a) sample 3, b) sample 7, c) sample 8, d) sample 9. Similar chromatograms were observed during adsorption studies of paraquat for samples 9, 11, and 12.
d)
c)
b)
a)
39
Fig 4.6 Chromatograms of 100 ppm paraquat solutions when mixed with approximately a) 50 mM
H2O2 and no catalyst; b) 6.25 mM H2O2 in the presence of 1.3 g/L sample 2.
t(min)
t(min)
b)
a)
40
Chapter 5: Conclusion and Future Work
The notable finding during the MSc research is the strong adsorption and exchange kinetics
of paraquat onto surfaces of silica. Although the adsorption capacity of the materials may not
always be as high as that of activated carbon products, more than 90% of paraquat removal
was achieved for almost all the silica samples tested (except Sample-9,11,12 and 15).
Furthermore, more than 80% of paraquat was removed within 30 seconds of first paraquat-
silica interaction by sample 3 (91%), sample 5(89%), sample 6 (90%) and sample 7 (77%).
Once effective functionalization techniques are implemented, such paraquat adsorption rates
can certainly aid reaction of the contaminant with reagents incorporated into the silica
supports, and thus it is crucial to optimize the efficiency of membrane based water treatment
systems. Agreement between experimental adsorption data and the competitive binding
model predictions also indicates competitive adsorption of both paraquat and water molecules
onto silica particles. However, none of the silica or any of the functionalities added to silica
reacted with paraquat, either in the presence or the absence of H2O2 in solution. As radical
generation is key to mineralization of organic contaminants, the findings suggest that
alternative techniques triggering generation of hydroxyl or single charged paraquat radicals
need to be explored.
In the next phase of this ongoing research, conformability of the adsorption results in
suspension systems to adsorptive removal data in actual membrane configuration will be
investigated. The idea is to identify the best silica candidate for adsorption and then
incorporate new functionalization techniques that may induce radical generation and paraquat
decomposition. Photo-activation of the embedded catalysts may also be attempted to further
catalyze the degradation process. The ultimate goal of the research would be to a devise
surface passivation technique facilitating use of phosphonate compounds as proxy
contaminants and subsequently tests the efficiency of engineered silica samples when coated
to membranes.
41
References
1 Bajgar, J. Organophosphates nerve agent poisoning: mechanism of action, diagnosis,
prophylaxis, and treatment. Adv. Clinic. Chem. 2004, 38,151-216.
2 Wiener, S. W.; Hoffman, R. S. Nerve agents: a comprehensive review. J. Intensive
Care Med. 2004, 19, 22-37.
3 Raushel, F. M. Bacterial detoxification of organophosphate nerve agents. Curr. Opin.
Microbiol. 2002, 5, 288-295.
4 Tafuri, J.; Roberts, J. Organophosphate poisoning. Ann. Emerg. Med. 1987, 16, 193-
202.
5 http://sc-ems.com/ems/NuclearBiologicalChemical/Sarin/sarin.htm
6 http://www.ndhealth.gov/epr/public/chemical/nerveeffects.htm
7 http://www.atsdr.cdc.gov/csem/cholinesterase/docs/cholinesterase.pdf
8 Longwell J.; Committee on Alternative Chemical National Research Council and
National Research Council. Alternative technologies for the destruction of chemical
agents and munitions. 1994 World Scientific Pub. Co. Inc.
9 Obee, T. N.; Satyapal, S. Photocatalytic decomposition of DMMP on titania. J.
Photochem. Photobiol. A. 1998, 118, 45-51.
10 http://www.cma.army.mil/map.aspx
11 Munro, N. Toxicity of the organophosphate chemical warfare agents GA, GB, and
VX: Implications for Public Protection. Environ. Health Perspect. 1994, 102(1),18-
37
12 Nowack, B.; Stone, A. T. Competitive adsorption of phosphate and phosphonates
onto goethite. Water Res. 2006, 40, 2201-2209.
13 Kumar, R. A.; Velayudhan, K. T.; Ramachandran, V.; Bhai, R. S.; Unnikrishnan, G.;
Vasu, K. Adsorption and removal kinetics of phosphonate from water using natural
adsorbents. Water Environ. Res. 2010, 82, 62-68.
14 Nowack, B. Environmental chemistry of phosphonates. Water Res. 2003, 37, 2533-
2546.
15 Wikipedia.org
16 Bartelt-Hunt, S. L.; Barlaz M. A.; Knappe D. A review of chemical warfare agent
simulants for the study of environmental behavior. Critic. Rev. in Environ. Sci. Tech.
2008, 38,112–136.
42
17 Paukku, Y.; Michalkova, A.; Leszczynski, J. Adsorption of dimethyl
methylphosphonate and trimethyl phosphate on calcium oxide: an ab initio study.
Struct. Chem. 2008, 19, 307-320
18 Ginsberg, M., D.,Hurley,;M., Aksimentiev A.; Hill F,; Hock V. Computational
models to determine transport and hydrolysis rate parameters of contaminants in a
water distribution system. Proceedings of the Army Science Conference 2008, Florida.
19 Guy, K; Ginsberg, M; Beckman, A; Rivera-Sustache; Carr, R; Comer, J; Scheeline,
A; Aksimentiev A; Ab initio comparison of the uptake of a hydrophobic material at a
water/solid interface. Proceedings of the 27th Army Science Conference 2010,
Florida,USA.
20 Ginsberg, M.D; Hock, V.F. Terrorism and security of water distribution systems: a
primer, Defense Security A. 2004, 20(4), 373-380.
21 Ginsberg, M.D; Hock, V.F. Secure water supply. Proceedings of the 27th Army
Science Conference 2005, Florida, USA.
22 Nowack, B. Determination of phosphonates in natural waters by ion-pair high-
performance liquid chromatography. J.Chrom. A. 1997, 773, 139-146.
23 Shamsi S. A.; Danielson, N. Ribonucleotide electrolytes for capillary electrophoresis
of polyphosphates and polyphosphonates with indirect photometric detection. Anal.
Chem. 1995, 67, 1845-1852
24 Merian E.; Frei, R. W.; Lawrence, J.F.; Brinkman, U.A. Th. Chemistry and fate of
organophosphorus compounds, 1987, Gordon and Brach Science Publishers.
25 Mahabir-Jagessar Tewari, K. van Stroe-Bieze, S.A.M. Analysis of amine-containing
phosphonates in detergent powders by anion-exchange chromatography with pulsed
amperometric detection. J. Chrom. A. 1997, 771, 155-161
26 Thompson R.; Grinberg N.; Perpall H.; Bicker G; Tway P. Separation of
organophosphonates by ion chromatography with indirect photometric detection. J.
Liq. Chromatogr. 1994,17 (11), 2511-2531.
27 Klinger, J.; Sacher F.; Brauch H.; Maler, D. Determination of organic phosphonates
in aqueous samples using liquid chromatography/particle beam mass spectrometry.
Acta Hydrochim. Hydrobiol. 1997, 25, 79.
28 Dyk J. S. V.; Pletschke B. Review on the use of enzymes for the detection of
organochlorine, organophosphate and carbamate pesticides in the environment.
Chemosphere 2011, 82(3), 291-307.
43
29 Varfolomeyev, S.; Kurochkin, I.; Eremenko, A.; Efremenko, E. Chemical and
biological safety. Biosensors and nanotechnological methods for the detection and
monitoring of chemical and biological agents. Pure Appl. Chem. 2002, 74, 2311-
2316.
30 Varfolomeyev, S.; Kurochkin, I.; Eremenko, A.; Rainina, E.; Gachok, I. In biosensors
and nanotechnological Immunochips for the detection and monitoring of Chemical
and Biological Agents. Proceedings of the NATO Advanced Research Workshop on
Defense against Bioterror: Detection Technologies, Implementation Strategies and
Commercial Opportunities 2004, Madrid, Spain.
31 Kanan, S.M.; Tripp, C.P. An Infrared Study of adsorbed organophosphonates on
silica: A prefiltering strategy for the detection of nerve agents on metal oxide
sensors. Langmuir 2001, 2213-2218.
32 Laiti, E.; Ohman, L. Acid/base properties and phenylphosphonic acid complexation
at the boehmite/water interface. J. Colloid Interface Sci. 1996, 183, 441-452.
33 Liu, Y.; Gao, L.; Yu, L.; Guo, J. Adsorption of PBTCA on alumina surfaces and its
influence on the fractal characteristics of sediments. J. Colloid Interface Sci. 2000,
227, 164-170.
34 Nowack, B.; Stone, A. T. Adsorption of phosphonates onto the goethite-water
interface. J. Colloid Interface Sci. 1999, 214, 20-30.
35 Nowack, B. Aminopolyphosphonate removal during wastewater treatment. Water
Res. 2002, 36, 4636-4642.
36 Nowack, B.; Stone, A. T. The influence of metal ions on the adsorption of
phosphonates onto goethite. Environ. Sci. Tech. 1999, 33, 3627-3633.
37 Bowen, J. M. Fourier transform infrared and Raman spectra of dimethyl
methylphosphonate adsorbed on montmorillonite. Environ. Sci. Tech. 1988, 22, 1178-
1181.
38 Correll, D. L. The role of phosphorus in the eutrophication of receiving waters: A
review. J. Environ. Qual. 1998, 27, 261-266.
39 Heil, D. M.; Barbarick, K. A. Water treatment sludge influence on the growth of
sorghum-sudangrass. J. Environ. Qual. 1989, 18, 292-298.
40 Archer, J. Phosphorus emissions from the incineration of chemical sludge. 1916,
Ontario Ministry of the Environment.
44
41 Rajagopalan, S.; Koper, O.; Decker, S.; Klabunde, K. J. Nanocrystalline metal oxides
as destructive adsorbents for organophosphorus compounds at ambient temperatures.
Chemistry 2002, 8, 2602-2607.
42 Michalkova, A.; Ilchenko, M.; Gorb, L.; Leszczynski, J. Theoretical study of the
adsorption and decomposition of sarin on magnesium oxide. J. Phys. Chem. B 2004,
108, 5294-5303.
43 Wagner, G. W.; Koper, O. B.; Lucas, E.; Decker, S.; Klabunde, K. J. Reactions of
VX, GD, and HD with nanosize CaO: autocatalytic dehydrohalogenation of HD. J.
Phys. Chem. B. 2000, 104, 5118-5123.
44 Wagner, G. W.; Procell L. R.;, O'Connor R. J.; Munavalli S.; Carnes C. L.; Kapoor P.
N.; Klabunde, K. J. Reactions of VX, GB, GD, and HD with nanosize Al2O3:
formation of aluminophosphonates. J. Am. Chem. Soc. 2001,123(8), 1636-1644
45 Decker, S. P.; Klabunde, J. S.; Khaleel, A.; Klabunde, K. J. Catalyzed destructive
adsorption of environmental toxins with nanocrystalline metal oxides. Fluoro-,
chloro-, bromocarbons, sulfur, and organophosophorus compounds. Environ. Sci. and
Tech. 2002, 36, 762-768.
46 Jiang, Y.; Decker, S.; Mohs, C.; Klabunde, K. J. Catalytic solid state reactions on the
surface of nanoscale metal oxide particles. J. Catal. 1998, 180, 24-35.
47 Oh, S. W.; Kim, Y. H.; Yoo, D. J.; Oh, S. M.; Park, S. J. Sensing behaviour of
semconducting metal oxides for the detection of organophosphorus compounds.
Sensors Actuators B: Chem. 1993, 13, 400-403.
48 Segal, S. R.; Cao L.; Suib, S. L.; Tang, X.; Satyapal, S. Thermal decomposition of
dimethyl Methylphosphonate over manganese oxide Catalysts. J. Catal. 2001, 198,
66-76.
49 Cao L; Segal S.R.; Suib S. L.; Tang X.;and Satyapal S. Thermocatalytic oxidation of
dimethyl methylphosphonate on supported metal oxides. J. Catal. 2000, 194(1), 61-
70
50 Cao, L.; Suib, S. L.; Tang, X.; Satyapal, S. Thermocatalytic decomposition of
dimethyl methylphosphonate on activated carbon. J. Catal. 2001, 197, 236-243.
51 Smentkowski V. S.; Hagans P.; Yates Jr. J. T. Study of the catalytic destruction of
dimethyl methylphosphonate(DMMP): oxidation over molybdenum(110). J. Phys.
Chem. 1988, 92 (22),6351–6357
45
52 Henderson M. A.; White J. M. Adsorption and decomposition of dimethyl
methylphosphonate on platinum (111). J. Am. Chem. Soc. 1988, 110 (21), 6939–
6947.
53 Ingle J. D.; Crouch S.R. Spectrochemical Analysis. 1988, Prentice Hall.
54 Ashley, M. G. Collaborative study of the determination of paraquat and diquat in
formulated products. Pestic. Sci. 1970, 1, 101–106.
55 Yuen S. H.; Bagness J. E. ; Mylest D. Spectrophotometric determination of diquat
and paraquat in aqueous herbicide formulations. Analyst 1967, 92,375-381.
56 Kuo T. L.; Lin D.L.; Liu R.H.; Moriya F.; Hashimoto Y. Spectra interference between
diquat and paraquat by second derivative spectrophotometry. Foren. Sci. Int., 2001,
121(12):134-9.
57 Laitinen H. A. Chemical Analysis, 1960, McGraw-Hill Book Company Inc.
58 http://www.h2o2.com/technicallibrary/analyticalmethods/default.aspx?pid=69&name
=Ceric-Sulfate-Titration
59 Harvey D., Analytical Chemistry 2.0 (Electronic Version)
60 Song X.; Budde W.L. Capillary electrophoresis––electrospray mass spectra of the
herbicides Paraquat and Diquat. J. Am. Soc. Mass Spectrom. 1996,7, 981–986
61 Marr J.C.; King J.B.; A simple High Performance Liquid Chromatography/Ionspray
Tandem Mass Spectrometry method for the direct determination of paraquat and
diquat in water. Rapid Commun. in Mass Spectrom. 1997, 11(5), 479-483.
62 http://www.chem.agilent.com/Library/applications/5988-7220EN.pdf
63 Yang, X.; Al-Duri, B. Application of branched pore diffusion model in the adsorption
of reactive dyes on activated carbon. Chem. Eng. J. 2001, 83, 15-23.
64 Levey, G.; Rieger, A. L.; Edwards, J. O. Rates and mechanism for oxidation of
paraquat and diquat radical cations by several peroxides. J. Org. Chem. 1981, 46,
1255-1260.
65 Levey Methyl viologen radical reactions with several oxidizing agents. J. Phys.
Chem, 1983, 87, 829-832.
66 Dhaouadi, A.; Adhoum, N. Heterogeneous catalytic wet peroxide oxidation of
paraquat in the presence of modified activated carbon. App. Catal. B: Environ. 2010,
97, 227-235.
