The electrochemical oxidation of diethylstilbestrol (DES) in acetonitrile

Accepted Manuscript The electrochemical oxidation of diethylstilbestrol (DES) in acetonitrile

Jazreen H.Q. Lee, Yu Rong Koh, Richard D. Webster PII: DOI: Reference:

S1572-6657(17)30405-8 doi: 10.1016/j.jelechem.2017.05.044 JEAC 3320

To appear in:

Journal of Electroanalytical Chemistry

Received date: Revised date: Accepted date:

12 December 2016 18 May 2017 23 May 2017

Please cite this article as: Jazreen H.Q. Lee, Yu Rong Koh, Richard D. Webster , The electrochemical oxidation of diethylstilbestrol (DES) in acetonitrile, Journal of Electroanalytical Chemistry (2017), doi: 10.1016/j.jelechem.2017.05.044

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

The Electrochemical Oxidation of Diethylstilbestrol (DES) in Acetonitrile

PT

Jazreen H. Q. Lee, Yu Rong Koh and Richard D. Webster*

RI

Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences,

NU

SC

Nanyang Technological University, Singapore 637371, Singapore

AC

CE

PT E

D

MA

*E-mail: [email protected], Telephone: +65 6316 8793; Fax: +65 6791 1961

1

ACCEPTED MANUSCRIPT Abstract The electro-oxidation of the former drug diethylstilbestrol (DES) was examined in detail by means of cyclic voltammetry (CV) and controlled potential electrolysis (CPE) techniques. Analogous to the electrochemistry of the structurally-related compound bisphenol A (BPA), DES undergoes two successive oxidation processes at ca. 0.64 and 0.81 vs. (Fc/Fc+)/V at glassy carbon electrodes in

PT

acetonitrile. The former anodic process was determined to involve up to four moles of electrons,

RI

whereas the latter anodic peak was assigned as the oxidation of a secondary product formed after the initial oxidation of DES. Results from variable voltammetric scan rate experiments showed that

SC

the secondary process can be outrun at high scan rates, while simultaneously detecting a new

NU

reduction wave (at ca. 0.50 vs. (Fc/Fc+)/V). Additionally, at fast scan rates, this reduction process was found to be chemically reversible; whereby a corresponding anodic peak was recorded at ca.

MA

0.53 vs. (Fc/Fc+)/V on the second cycle. Following a similar mechanistic pathway as BPA, under prolonged periods of time (e.g. slow scan rates), DES is proposed to undergo an overall four-

D

electron oxidation, whereas at fast scan rates the oxidation of DES was found to only involve two

PT E

electrons. The CV data were modeled using digital simulations that allowed an estimation of the

Keywords

CE

electrochemical and kinetic parameters associated with the electrode reactions.

AC

Diethylstilbestrol; Electrochemical Oxidation; Proton-coupled electron transfer (PCET); Bisphenol A; Digital simulation

2

ACCEPTED MANUSCRIPT 1.

Introduction

Once considered as a useful drug for the treatment of pregnancy- and miscarriage-related complications [1-3], diethylstilbestrol (DES, Scheme 1), a synthetic estrogen, has since been associated with an array of pathophysiologies including a rare cancer (clear-cell adenocarcinoma) disorder [4-6], and abnormalities in children who were exposed to DES in utero [3, 7-10].

PT

Furthermore, these deleterious effects were exacerbated by the use of DES as growth promoters in

RI

meat-producing animals in order to improve on the feed conversion efficiency [11-13]. As such, the use of DES for miscarriage prevention was banned by the Food and Drug Administration in 1971.

SC

Consequently, this has also led to numerous studies conducted on the cumulative effects of DES on

NU

affected women and their offspring (exposed prenatally) [3, 7-10], its mode of mechanism in human bodies [2, 10, 14, 15], and the control and monitoring of its illegal usage in livestock [16-18] and

MA

their derived food products (e.g. milk) [19-23]. Hence, it is not surprising that research into various analytical methods (e.g. high performance liquid chromatography [20, 23], gas chromatography-

D

mass spectrometry [17, 18, 24], chemiluminescence [25]) for the fast and accurate detection of DES

PT E

in food products, patients’ biological matrices (serum, blood, urine) has also been actively pursued

AC

CE

in recent years.

Scheme 1. Chemical structure of Diethylstilbestrol (DES). While the sensitive determination of DES has led to the development of a number of impressive works, those that have focused on the mechanisms of its degradation have remained sparse. In this regard, electrochemical methods remain highly pertinent because they can be used to determine the number of electrons involved, the order in which they occur, the identities of the associated 3

ACCEPTED MANUSCRIPT intermediates [38, 39], and therefore possibly provide insights into the metabolic pathways partaken by these molecules inside biological systems. In addition, the knowledge gained from these studies might be useful for the rationale design of electrochemical detection methodologies [40, 41]. In the area of voltammetric detection [26-29], most of the accounts were conducted in aqueous solutions, with some reports [30-33] disclosing the electro-oxidation of DES through a two-

PT

electron, two-proton process to give a 4',4"-diethylstilbestrol (DES) quinone, a product which was

RI

also reported to be one of the oxidative metabolites of DES [34-36]. In this regard, it would be interesting to supplement the existing redox information on DES with new mechanistic insights in a

SC

different medium (aprotic solvents). This is because the decrease in proton activities in aprotic

NU

solvents may sometimes aid in diminishing the propensity for chemical species, particularly for transient intermediates, to undergo other reactions (e.g. hydrolysis), and therefore might enable the

MA

observation of phenomena that were not observed in protic systems [37]. Bisphenol A (BPA, Scheme 2), a monomer that is extensively used in the production of plastics and

PT E

D

epoxy resin, has likewise received increasing attention due to its implication with numerous health problems [42-45]. Consequently, this has also led to the outlaw of BPA in consumer products including baby bottles in countries and territories such as the European Union, Canada and by the

AC

CE

Food and Drug Administration agency in the USA.

Scheme 2. Chemical structure of Bisphenol A (BPA). Interested in uncovering information regarding the possible metabolic reactions that BPA can undergo, we have previously delineated the electrochemical study of BPA in aprotic organic solvents and its associated redox transformations [46]. Besides being structurally similar compounds (both are bi-phenolic compounds linked by an alkyl bridge), BPA and DES are also

4

ACCEPTED MANUSCRIPT well-established endocrine disruptors [47-49] that are known to interfere with normal hormonal activities. In a continuous effort to examine the redox properties of synthetic industrial compounds that have raised concern regarding their potential toxicity, we report herein an in-depth electrochemical study on the oxidation of DES. In addition, a comparative study between the

Experimental

2.1.

Chemicals and reagents

RI

2.

PT

voltammetric behavior of DES and BPA is also described.

SC

Unless otherwise stated, all chemicals and reagents were purchased from commercial sources and

NU

used as received. DES (> 98%), BPA (≥ 99%) and HPLC grade acetonitrile (CH3CN) were obtained from Tokyo Chemical Industry Co., Ltd., Sigma-Aldrich and Macron respectively. Molecular sieves

MA

that are made up of 1/16 inch rods with pore size 3 Å (CAS: 308080-99-1) were bought from Fluka. Tetrabutylammonium hexafluorophosphate (Bu4NPF6), the supporting electrolyte, was prepared

D

according to a literature procedure [50] by reacting equal molar amounts of a 40% aqueous solution

PT E

of Bu4NOH (Alfa Aesar) and a 65% aqueous solution of HPF6 (Alfa Aesar), washing the resulting precipitate with ultrapure water and recrystallizing three times with hot ethanol, with subsequent

Voltammetry

AC

2.2.

CE

drying under vacuum at 140 °C for 6 hours.

Cyclic Voltammetry (CV) experiments were carried out using a computer controlled Eco Chemie Autolab PGSTAT302N potentiostat with a three-electrode system. The working electrode was a 1 mm diameter planar disk glassy carbon (GC) working electrode (eDAQ Pty Ltd), used along with a platinum (Pt) counter electrode (Metrohm) and a silver (Ag) wire miniature reference electrode (eDAQ Pty Ltd) that was separated from the sample solution via a salt bridge containing 0.5 M Bu4NPF6 in CH3CN. Prior to each scan, all sample solutions used for the voltammetric experiments were deoxygenated by purging with high purity argon gas, with the measurements performed in a

5

ACCEPTED MANUSCRIPT Faraday cage. Accurate potentials were obtained by measuring the potential of ferrocene (Fc) under the same conditions but in a separate solution, as well as by adding Fc to the test solutions containing DES at the completion of the measurements (the oxidation potential and chemical reversibility of the Fc/Fc+ system were not affected by the presence of DES). Variable temperature voltammetric experiments (20 to -30 °C) were conducted in a Metrohm jacketed glass cell, with the

Measurement of Water Content in Sample Solutions

RI

2.3.

PT

temperatures controlled by a Julabo FP89-HL ultralow refrigerated ethanol circulating bath.

SC

Karl Fischer titrations was conducted inside a humidity control box (122cm × 61cm × 61cm, Coy Laboratory Products Inc) using a Mettler Toledo DL32 coulometer where the HYDRANAL-

NU

coulomat CG (Riedel-deHaën) and HYDRANAL- coulomat AG were employed in the cathode and

MA

anode compartment respectively. Measurements were performed by injecting analyte solutions that were contained in a 5 mL vacuum syringe (SGE Analytical Science) into the coulometer through a

Preparations for Dry Voltammetry Experiments

PT E

2.4.

D

silicon/teflon septum.

Voltammetric experiments with lower water content were performed in dried CH3CN (over 3 Å

CE

molecular sieves) that was transferred into an electrochemical cell that was heated at 100 °C in an oven for at least 1 hour prior to the experiment and allowed to cool under an argon atmosphere.

AC

Aliquots of the analyte solution were measured by Karl Fischer titration, before subjecting to any CV analysis, to obtain an estimate of the initial water content directly inside the electrochemical cell (initial [H2O] = 14-17 mM). 2.5.

Controlled Potential Electrolysis

Controlled potential electrolysis (CPE) measurements were performed in a two-compartment electrolysis cell divided by a sintered glass frit of porosity no. 5 (1.0 – 1.7µm) [51]. Two identically sized GC cylinders were used as the working and counter electrodes, which were arranged 6

ACCEPTED MANUSCRIPT symmetrically with respect to each other, and used in conjunction with the aforementioned reference electrode that was positioned within 2 mm of the surface of the working electrode. Sample solutions (25 ml each) in both compartments of the cell were simultaneously deoxygenated and stirred using bubbles of argon gas throughout the entire experiment. The number of electrons transferred during the bulk electrolysis process was calculated using the Faraday’s law of

PT

electrolysis. (1)

RI

n = Q/NF

SC

where, N = number of moles of starting material used, Q = charge (in coulombs), n = number of

2.6.

NU

moles of electrons, and F is the Faraday constant (96 485 C/mol). Digital Simulations

MA

Digital simulations of the cyclic voltammograms were carried out using a DigiElch 7 simulation

Results and Discussion

PT E

3.

D

software purchased from Gamry Instruments.

3.1 Electrochemical Oxidation of BPA and DES in Acetonitrile

CE

We had previously found that BPA, when examined in CH3CN using a GC electrode and a scan rate

AC

0.1 V s-1, displays two consecutive oxidation peaks on the forward scan at ca. 0.99 and 1.44 vs. (Fc/Fc+)/V (wave I and II), with no corresponding reduction peaks when the scan direction was reversed (Figure 1(a)). A more thorough examination of the anodic process in wave I further revealed the involvement of a four-electron, two-proton (-4e-/-2H+) transfer which is eventually followed by a hydrolysis step to generate a hydroxylated bisdienone product [46]. With BPA as a reference compound, attention was turned to assessing the voltammetric behavior of DES under the same conditions (Figure 1(b)). Preliminary studies disclosed a comparable phenomenon to that of BPA whereby two oxidation peaks at ca. 0.64 and 0.81 vs. (Fc/Fc+)/V 7

ACCEPTED MANUSCRIPT (waves 1 and 2) were recorded on the forward scan. More interestingly, however, a diminutive reduction peak was also detected on the return sweep at ca. 0.50 vs. (Fc/Fc+)/V (wave 3); a cathodic

PT E

D

MA

NU

SC

RI

PT

process that was previously unobserved in the system using BPA.

Figure 1. Cyclic voltammograms of 2 mM (a) BPA (b) DES in CH3CN with 0.2 M Bu4NPF6,

CE

recorded at a 1 mm diameter planar GC electrode at 22 (±2) °C and at a scan rate of 0.1 V s-1.

AC

3.2 Cyclic Voltammetry of BPA and DES using Variable Scan Rates In order to assess the effects of shorter experimental time domains on the electrochemical behavior of BPA and DES, varying voltage sweep rate experiments were subsequently conducted on the aforementioned compounds. As shown in Figure 2(a), increasing scan rates led to the diminishing magnitude of BPA’s second oxidation step (wave II) until almost no discernible peak current could be detected at 20 V s-1. Furthermore, it was found that a restriction of the potential range to allow only the first oxidation process (wave I) to occur, together with the employment of faster scan rates,

8

ACCEPTED MANUSCRIPT did not appear to improve the chemical reversibility of wave I (Figure 2(b)). Additionally, regardless of the potential range used, the voltammetry of BPA was found to suffer from the effects of adsorption, based on the marked decrease in peak current upon continuous cycling, as shown in the second cycle (dotted line) of both figures (Figure 2(a) and 2(b)). This adsorption process may be responsible for the unusual peak shape observed during the first scan where the first process appears

AC

CE

PT E

D

MA

NU

SC

RI

PT

to consist of a shoulder before the main peak.

Figure 2. Variable scan rate cyclic voltammograms of 2 mM BPA in CH3CN (Initial [H2O] = 70 mM) with 0.2 M Bu4NPF6, recorded at a 1 mm diameter planar GC electrode at 22 (±2) °C. Current data were scaled by multiplying by (scan rate)-0.5. (—) First scan. (····) Second scan. (a) Switching potential ≈ 1.80 vs. (Fc/Fc+)/V. (b) Switching potential ≈ 1.17 to 1.33 vs. (Fc/Fc+)/V.

9

ACCEPTED MANUSCRIPT In comparison, DES similarly showed a decrease in peak current magnitude for wave 2 as the scan rates were increased (Figure 3(a)). Concomitantly, wave 3 became more voltammetrically noticeable at larger scan rates and was also observed to give a corresponding anodic peak at ca. 0.53 vs. (Fc/Fc+)/V (wave 4) upon continuous cycling. Likewise, CV experiments that limit the switching potential range to only the first oxidation peak of DES were carried out and are depicted

PT

in Figure 3(b). It is worth highlighting, however, that it is very difficult to set and maintain a constant switching potential for the first wave, as depicted in Figure 3(b), because wave 1 becomes

RI

more drawn out as the scan rates are increased and since the reaction involved in wave 2 is outrun at

SC

higher scan rates.

NU

Nevertheless, the results obtained in Figure 3(b) clearly shows that the species associated with waves 3 and 4 are formed after the initial oxidation at wave 1 (rather than wave 2). For instance,

MA

over the entire range of scan rates examined, as compared to the cyclic voltammograms recorded from the larger potential range (Figure 3(a)), almost identical voltammograms were obtained in the

D

narrower potential range (Figure 3(b)). Furthermore, the large peak to peak separation between

PT E

waves 3 and 4 with respect to wave 2 discounts the possibility that they are related, and the redox couple (wave 3 and 4) is more likely to be associated with wave 1. However, the close proximity of

CE

the redox couple waves 3 and 4 with wave 1 (for DES) negates the likelihood that wave 3 is associated with the regeneration of the starting material, and instead is more probable to be due to

AC

the conversion of the oxidized product back into one of its intermediates. Notably, this oxidized species must also only be moderately stable as evidenced by the more marked presence of wave 3 only at larger scan rates (≥ 0.5 V s-1). It can also be observed that unlike BPA, repetitive voltammetric scanning of solutions containing DES does not appear to result in significant adsorption effects (dotted lines in Figure 3(a) and 3(b)). The peak shape observed for the first oxidation process of DES was simpler than observed for BPA, possibly reflective of less surface interactions occurring during the oxidation of DES (Figures 1 – 3).

10

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

CE

Figure 3. Variable scan rate cyclic voltammograms of 2 mM DES in CH3CN (Initial [H2O] = 80 mM) with 0.2 M Bu4NPF6, recorded at a 1 mm diameter planar GC electrode at 22 (±2) °C. Current

AC

data were scaled by multiplying by (scan rate)-0.5. (—) First scan. (----) Second scan. (a) Switching potential ≈ 0.94 vs. (Fc/Fc+)/V. (b) Switching potential ≈ 0.74 to 0.87 vs. (Fc/Fc+)/V.

11

ACCEPTED MANUSCRIPT Based on the above experimental results, it is posited that waves II and 2 (for BPA and DES respectively) are likely related to the oxidation of secondary product(s) formed after the oxidation of BPA and DES at waves I and 1 respectively. That is, when higher scan rates are employed, the homogeneous reactions that form the secondary products are outrun and therefore are not detected at fast scan rates. Since more information can be obtained using the wider potential range (Figure

PT

2(a) and 3(a)), such as the outrun of wave II and 2 following the increase in scan rates, subsequent

RI

discussions are focused on using the larger potential window.

SC

3.3 Cyclic Voltammetry of DES under Variable Temperature Conditions Because lowering of the temperature can sometimes help in improving the stability and lifetime of

NU

transient species, thereby allowing their voltammetric detection [52], variable temperature

MA

voltammetry experiments ranging from 20 to -30°C were also conducted to further investigate the intermediates involved in the electrode reactions of DES (Figure 4). In doing so, it was found that waves 3 and 4 appear to become more discernible at ≤ 0 °C (even at a slow scan rate of 0.1 V s-1),

PT E

D

and was accompanied by the progressive decrease of wave 2. Hence, it is apparent that the electrode reactions induced at wave 2 will result in a decrease in the amount of species responsible for waves 3 and 4. Additionally, the necessity of lowered temperatures (≤ 0 °C) or higher scan rates (≥ 0.5 V s) in order for the detection of waves 3 and 4 implies a limited lifetime of the oxidized product

CE

1

AC

formed at wave 1. Nonetheless, a reduction of temperature only provided modest improvements in the stability of the oxidized product, as evidenced by the limited peak current achieved in waves 3 and 4.

12

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 4. Variable temperature cyclic voltammograms of 2 mM DES in CH3CN with 0.2 M Bu4NPF6, recorded at a 1 mm diameter planar GC electrode at a scan rate of 0.1 V s-1. (—) First

CE

scan. (----) Second scan.

AC

3.4 Cyclic Voltammetry of DES under Dry Conditions It was proposed that a hydrolysis reaction is one of the secondary processes partaken by the product formed after the oxidation of BPA at wave I, thereby leading to the redox process being observed as chemically irreversible [46]. As such, it was reasoned that the use of drier conditions might help to mitigate the further reaction of the oxidized form of DES and possibly aid in improving its stability. With this in mind, DES was subjected to a lower amount of water but this revealed only a minimum change in the voltammetric behavior of DES. More specifically, the redox couple of waves 3 and 4 could be more clearly observed at scan rates ≥ 0.2 V s-1 (Initial [H2O] = 14 mM, Figure 5), instead 13

ACCEPTED MANUSCRIPT of requiring higher scan rates of ≥ 0.5 V s-1 when examined under a higher trace water content condition (Initial [H2O] = 80 mM, Figure 3). As only modest improvements could be attained, this signifies that the oxidized product of DES is likely to be highly reactive in nature and thus relatively short-lived even in a low water environment. Similar attempts to reduce the reactivity of the anodic product of BPA by means of decreasing the concentration of trace water were also found to effect

AC

CE

PT E

D

MA

NU

SC

RI

PT

no substantial change (Figure A3 in the Appendix).

Figure 5. Variable scan rates cyclic voltammograms of 2 mM DES in CH3CN (Initial [H2O] = 14 mM, final [H2O] = 35 mM) with 0.2 M Bu4NPF6, recorded at a 1 mm diameter planar GC electrode at 22 (±2) °C. Current data were scaled by multiplying by (scan rate)-0.5. (—) First scan. (----) Second scan.

14

ACCEPTED MANUSCRIPT Nevertheless, consistent with the earlier findings, waves 3 and 4 in Figure 5 were observed to grow at the expense of wave 2 as the scan rate increased. Thus, it stands to reason that the electrode reactions initiated at waves 2 and 3 demonstrate a competing relationship. The dichotomy between waves 2 and 3 arises at higher scan rates as there is less time available for the species responsible for wave 2 to occur which consequently allows more of the oxidized species (formed at wave 1) to

PT

undergo the reduction reaction instead, resulting in an increase in magnitude of wave 3 (and

RI

corresponding wave 4).

SC

3.5 Controlled Potential Electrolysis of DES

By changing the composition of the solution appreciably, controlled potential electrolysis (CPE)

NU

experiments were performed in order to obtain information about the intermediates and long-term

MA

oxidation products. In addition, CPE also enables the determination of the number of electrons involved in a redox process. As wave 1 is the first heterogeneous step that DES has to undergo, the potential applied was made sufficiently positive (50 mV more positive than that of wave 1) to allow

PT E

D

only a full conversion of the electron transfer process in wave 1 while preventing the simultaneous occurrence of wave 2. Figure 6 outlines the coulometric and voltammetric graphs logged after an elapse of 4500 seconds. Verification of the complete oxidation was shown by the absence of wave 1

CE

in the voltammogram obtained after the bulk conversion (Figure 6(a), dotted line) as well as the

AC

observation of the current decaying from an initial ca.10 mA to an almost zero value, as illustrated in the coulometry data (Figure 6(b)). The slight upward slope in the current – time curve seen at the very early stages of the electrolysis shown in Figure 6(b) is possibly an artefact possibly caused by incomplete mixing in the electrolysis cell, as it is not reproducible between different experiments (Figure A4 in the Appendix). The phenomenon may arise from a mass transfer issue, since the sample solutions undergoing electrolysis were continuously deoxygenated and stirred using bubbles of argon gas throughout the entire experiment (as described in section 2.5) which does not give uniform mixing

15

ACCEPTED MANUSCRIPT rates. Nevertheless, the plateau shape observed at the beginning of the current-time curve could also be the result of a mechanistic feature due to the electrolysis reaction occurring in two distinct stages, although the data is not sufficiently reproducible to be quantifiably analysed to confirm this possibility. According to the Faraday's law of electrolysis (Eq. 1), approximately four moles of electron (per

PT

mole of DES) were found to be involved in the electron transfer reaction at wave 1. Furthermore,

RI

upon the loss of wave 1, a new anodic peak at ca. 0.68 vs. (Fc/Fc+)/V, together with the previously observed wave 2 were seen (Figure 6(a), dotted line). The appearance of wave 2 after the exhaustive

SC

electrolysis further corroborates the premise that wave 2 is the subsequent oxidation of the reaction

NU

product generated by wave 1. Although the oxidation peak current of the oxidized product is much smaller than that of the starting material, it does not necessarily mean that the product is present in

AC

CE

PT E

D

MA

very low yield, since its own oxidation may occur by fewer than four electrons.

Figure 6. Voltammetric and coulometric data recorded during the controlled potential electrolysis of 2 mM DES in CH3CN with 0.2 M Bu4NPF6 at 22 (±2) °C. (a) Cyclic voltammograms obtained at a scan rate of 0.1 V s-1 with a 1 mm diameter planar GC electrode. (—) Prior to bulk oxidation (····)

16

ACCEPTED MANUSCRIPT After exhaustive oxidation. (b) Current vs. time data logged during the exhaustive oxidation of DES at 0.69 vs. (Fc/Fc+)/V, with the number of electron transferred calculated using Eq. 1. 3.6 Proposed Mechanism and Digital Simulations for the Electrochemical Oxidation of DES On the basis of the above results, a tentative mechanism for the electrochemical oxidation of DES is illustrated in Scheme 3, which is similar to the mechanism that has been proposed for BPA in

PT

aprotic conditions [46]. Although the two phenolic groups in BPA and DES differ in their degree of

RI

conjugation that separates them, the overall electron count supports a similar sequence of electron

SC

transfers that are typical of phenolic compounds. Despite multiple attempts to isolate the oxidized product of DES after electrolysis, thin layer chromatography analysis revealed that the end product

NU

appeared to form several other products that were very difficult to separate and purify using preparative flash column chromatography, and appeared to degrade further on the column. The

MA

purification is made more difficult given the need to remove the electrolyte, which is usually added in large quantities (at least 10-100 fold excess of the analyte) during electrochemical experiments.

D

Thus the proposed mechanism relates to the initially formed compounds that undergo further

PT E

degradation to multiple products. Therefore, the mechanism in Scheme 3 represents a probable sequence of electron transfer and proton transfer steps on the short voltammetric time scale (≤

CE

seconds) that is consistent with the presently obtained voltammetric data and current knowledge of

AC

oxidation mechanisms of phenols [53]. Consistent with the proposed mechanism of BPA, the oxidation of DES is surmised to proceed via an overall loss of four electrons, two protons in an ECE manner (where E and C represent an electron transfer and chemical step respectively), with a stepwise transfer of an electron in each heterogeneous reaction and a proton involve in each homogeneous process. Notably, however, because of the highly conjugated structure of DES, as well as the involvement of a large number of steps, it is very difficult to voltammetrically confirm the exact sequence of the electron transfer and

17

ACCEPTED MANUSCRIPT homogeneous reactions, thus an alternative order of the electrochemical and chemical transformations cannot be completely ruled out. Furthermore, with the aim of gaining a better understanding on the mechanistic premise put forward in Scheme 3, digital simulations were also conducted to supplement the experimental data. Generally, the simulation work was performed in a trial and error manner that includes the

PT

systematic addition of the reaction parameters for the various heterogeneous and homogeneous

RI

processes. Additionally, in order to increase the credibility of the simulation results, rather than modeling and fitting to only a single voltammogram, a fixed set of reaction parameters was used for

SC

the entire range of scan rates (0.1 to 20 V s-1) examined in the current study [54]. As illustrated in

NU

Figure 7, a reasonably close fit between the experimental and simulated voltammograms was obtained, and the various optimized electrochemical and chemical kinetic parameters used are

MA

tabulated in Tables 1 and 2.

In line with the behavior of a typical phenol, the initial removal of an electron from DES (oxidation)

PT E

D

will produce a cation radical that is more acidic than the parent, thereby allowing the loss of a proton easily (compound 2 to 3). Consequently, this structural change also enables the second electron transfer process to proceed at a less positive potential thereby producing a multi-electron

CE

peak (wave 1) [55]. Due to the presence of two phenolic functional groups in DES, it is conceivable

AC

to propose that two ECE processes occurred, with a diamagnetic cation (compound 4 and 8) formed after every ECE reaction. This was ascertained by the results gathered from the electrolysis experiment where four moles of electrons were involved during the complete oxidation of DES (Figure 6). Bearing a positive charge, phenoxonium cations (e.g. compound 4 and 8 in Scheme 3) are generally highly electrophilic and can therefore readily engage in follow-up reactions (e.g. hydrolysis) with nucleophiles like water, which are commonly found in trace amounts in aprotic organic solvents (compound 4 to 5 and 8 to 9) [46, 56]. The possibility of the hydrolysis reactions occurring only

18

ACCEPTED MANUSCRIPT after the manifestation of two consecutive ECE processes (i.e. reaction of two molecules of water with the dicationic form of compound 8) instead of the reaction of one molecule of water with each cationic product that has been formed after every ECE process (compound 4 to 5 and 8 to 9, as shown in Scheme 3) was considered but thought to be less likely based on the simulation analysis. This can be reasoned by the observation of a comparatively lowered peak height of wave 1 at high

PT

scan rates, thus implying the occurrence of a less than four-electron process. Therefore, it is rationalized that the hydrolysis reactions take place in a stepwise fashion, with the first hydrolysis

RI

process (compound 4 to 5) being outrun at faster scan rates that will result in only two moles of

SC

electrons being transferred, which in turn leads to a dwindling magnitude of wave 1 at high scan rates. As such, the rate constants associated with the first hydrolysis step (Eq. (iv)) are expected to

NU

be relatively small in value (forward rate constants (kf) = 17, backward rate constants (kb) = 0.017).

MA

With the above information in mind, the species associated with wave 3 which appears near the vicinity of wave 1 is ascribed to be due to the reduction of compound 4 to 3. The possible

D

involvement of wave 3 as the regeneration of the starting material (DES) (compound 4 to 1) is

PT E

discounted based on the observation of the ratio of the peak current of wave 1 and 3 not reaching unity even with the employment of high scan rate such as 20 V s-1. Moreover, this was supported by

CE

the presence of a reverse oxidation peak (wave 4) on the second cycle of the CV experiment (i.e. waves 3 and 4 are a redox couple, as shown in Figure 3(b)). Therefore, the equilibrium constant for

AC

the proton loss in Eq. (ii) would be in favor of the formation of the product (compound 3) which accounts for only the last E step (Eq. (iii)) of the first ECE sequence (Eqs. (i)-(iii)) of the mechanism to be chemically reversible. Another argument in support of wave 3 not being associated with the regeneration of the starting material relates to the rarity of such processes being detected during the oxidation of phenols. There are only two reported examples of phenols that undergo multiple oxidative electron transfer steps being able to be reversibly regenerated from their oxidised forms during voltammetric scanning; that is for -tocopherol [38] and 2,6-di-tert-butyl-4(4-dimethylaminophenyl)phenol [57]. However, both of these chemically reversible systems are 19

ACCEPTED MANUSCRIPT simpler than the DES system as they only involve the transfer of two-electrons and one-proton, and their phenolic aromatic rings are substituted leaving less possibility for rapid radical coupling reactions that are a major reaction route for oxidized phenols [53]. For the variable scan rates experiments carried out under less dry conditions (Figure 3(a)), it was observed that the redox couple (waves 3 and 4) became more experimentally detectable at scan

PT

rates ≥ 0.5 V s-1, in conjunction with the decrease in peak current of wave 2. Notably, a similar

RI

trend was also observed when the solution temperature was lowered as displayed in Figure 4. Lastly, it is posited that wave 2 is a secondary process that takes place via an oxidation of

SC

compound 9 to 10 which can subsequently decompose into other products. The follow-up chemical

NU

reaction therefore also accounts for the chemically irreversible nature of wave 2.

MA

In aqueous systems, DES has been disclosed to undergo electro-oxidation through a two-electron, two-proton process to give a 4',4"-diethylstilbestrol (DES) quinone, a product which was also reported to be one of the oxidative metabolites of DES [34-36]. However, under the conditions

PT E

D

employed for the present study, it has been shown that the oxidation of DES occurs via four moles of electrons instead (Figure 6) and this vastly different number of electron count clearly indicates that the oxidized product formed in the current aprotic system must be very different from that of

CE

the aqueous media. Nevertheless, the effect of varying levels of water on the electrochemical

AC

oxidation of DES was conducted where accurately known volumes of water was carefully added into the sample solution (Figure A5 in the Appendix). It was found that wave 3 decreases and become almost undetectable with increasing water content and this is expected because the increase in water content would promote the hydrolysis reaction (conversion of compound 4 to 5) which would then negate the reduction of compound 4 back to 3. In addition, the ratio of wave 1 to 2 increases with higher water concentration and this can be reasoned by greater generation of the hydrolyzed secondary product that is associated with wave 2.

20

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 7. Variable scan rates cyclic voltammograms of 2 mM DES in CH3CN with 0.2 M Bu4NPF6, recorded at a 1 mm diameter planar GC electrode at 22 (±2) °C. (—) Experimental

CE

voltammograms. (·····) Simulated voltammograms generated based on the mechanism outlined in

AC

Scheme 3. Current data were scaled by multiplying by (scan rates)-0.5.

21

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

Scheme 3. Proposed mechanism for the electrochemical oxidation of DES.

22

ACCEPTED MANUSCRIPT Table 1. Electrochemical parametersa obtained by digital simulations of CV datab for the reaction mechanisms shown in Scheme 3. Electrochemical

Eq. i

Eq. iii

Eq. v

Eq. vii

Eq. ix

E0 (V)

0.74

0.49

0.71

0.46

0.92

ks (cm s-1)

0.70

0.60

0.70

0.60

0.20

Parameters

E = formal oxidation potential vs. (Fc/Fc+)/V, ks = heterogeneous electron transfer rate constant. The

PT

a 0

diffusion coefficients were estimated via simulation techniques to be 2.20 x 10 -5 cm2 s-1 for DES and its

RI

oxidized forms and 3.50 x 10-5 cm2 s-1 for H2O and H+. The transfer coefficient (α) values were assumed to be 0.5. bCV data were logged using a 1 mm diameter GC electrode for solutions of 2 mM DES with 0.2 M

SC

Bu4NPF6 in CH3CN at 22 (±2) °C. The water and proton concentrations used were 0.08 M.

NU

Table 2. Equilibrium and rate constantsa obtained by digital simulations of CV datab for the

Kinetic

MA

reaction mechanisms shown in Scheme 3. Eq. ii

Eq. iv

Eq. vi

Eq. viii

Eq. x

Keq

8.00 × 10-7

1.00 × 103

1.00 × 10-5

2.50 × 105

1.00 × 104

kf

1.00 × 100

1.70 × 101

5.00 × 10-1

5.00 × 105

1.00 × 107

kb

1.25 × 106

1.70 × 10-2

5.00 × 104

2.00 × 100

1.00 × 103

PT E

a

D

Parameters

Keq = equilibrium constant, kf = forward rate constant, kb = backward rate constant. The homogeneous rate

CE

constants have units of s−1 and L mol−1 s−1 for the first- and second-order reactions, respectively. bCV data were logged using a 1 mm diameter GC electrode for solutions of 2 mM DES with 0.2 M Bu 4NPF6 in

AC

CH3CN at 22 (±2) °C. The water and proton concentrations used were 0.08 M.

23

ACCEPTED MANUSCRIPT 4.

Conclusion

Overall, the electro-oxidation of DES in CH3CN revealed comparable voltammetric behavior to structurally similar BPA, in terms of the numbers of electrons transferred under short and long timescales. The concurrence of these results suggests a similar mechanistic premise for the oxidation of DES. Interestingly, in contrast with the chemically irreversible oxidative process

PT

recorded for BPA, a reduction peak (wave 3) was observed after the initial oxidation of DES. The

RI

cathodic process was reasoned to be due to the reduction of an intermediate diamagnetic cation (compound 4) back to an intermediate neutral radical compound 3 (wave 3), which gave a

SC

corresponding anodic peak (wave 4) on the subsequent cycle. However, the short-lived redox

NU

couple (waves 3 and 4) is only experimentally detected at fast scan rates (≥ 0.5 V s -1) during the CV measurements, thus implying the limited stability of compound 4. Conversely, at low scan rates, an

MA

additional anodic peak was seen at ca. 0.81 vs. (Fc/Fc+)/V (wave 2) on the forward scan. The latter oxidation process was observed to be outrun at high scan rates, while waves 3 and 4 concomitantly

D

increased in magnitude. Likewise, a similar trend was found on reducing the solution temperature.

PT E

In light of the aforementioned results, wave 2 is rationalized to be the oxidation of a secondary product. Through careful analysis of the gathered voltammetric responses, a tentative mechanism

CE

for the electrochemical oxidation of DES was postulated and digital simulation studies were performed in order to verify the validity of the proposed mechanism as well as to extract kinetic

5.

AC

information regarding its involved pathways. Acknowledgement

This work was supported by a Singapore Government MOE Academic Research Fund Tier 1 Grant (RG109/15). J.H.Q.L. thanks Nanyang Technological University (NTU) for the award of a NTU Research Scholarship.

24

ACCEPTED MANUSCRIPT 6.

References

[1] R.N. Hoover, M. Hyer, R.M. Pfeiffer, E. Adam, B. Bond, A.L. Cheville, T. Colton, P. Hartge, E.E. Hatch, A.L. Herbst, B.Y. Karlan, R. Kaufman, K.L. Noller, J.R. Palmer, S.J. Robboy, R.C. Saal, W. Strohsnitter, L. Titus-Ermstoff, R. Troisi, Adverse Health Outcomes in Women Exposed in Utero to Diethylstilbestrol, New Engl. J. Med. 365(14) (2011) 1304–1314.

PT

[2] J. Lewis-Wambi, V.C. Jordan, Diethylstilbestrol: A Tragedy In Reproductive Endocrinology But A Pioneering Cancer Treatment, in: V.C. Jordan (Ed.), Estrogen Action, Selective Estrogen

RI

Receptor Modulators and Women's Health. Progress and Promise, Imperial College Press, London,

SC

2013, pp. 31–54.

NU

[3] S. Schrager, B.E. Potter, Diethylstilbestrol Exposure, Am. Fam. Physician 69(10) (2004) 2395–

MA

2400.

[4] A.L. Herbst, H. Ulfelder, D.C. Poskanzer, Adenocarcinoma of The Vagina. Association of Maternal Stilbestrol Therapy with Tumor Appearance in Young Women, N. Engl. J. Med. 284(15)

PT E

D

(1971) 878–881.

[5] A.L. Herbst, Stilbestrol and Vaginal Cancer in Young Women, CA: Cancer J. Clin. 22(5) (1972)

CE

292–295.

[6] A.L. Herbst, R.J. Kurman, R.E. Scully, D.C. Poskanzer, Clear-Cell Adenocarcinoma of The

AC

Genital Tract in Young Females. Registry Report, N. Engl. J. Med. 287(25) (1972) 1259–1264. [7] R.M. Giusti, K. Iwamoto, E.E. Hatch, Diethylstilbestrol Revisited: A Review of The Long-Term Health Effects, Annals of Internal Medicine 122(10) (1995) 778–788. [8] S.H. Swan, Intrauterine Exposure to Diethylstilbestrol: Long-term effects in humans, APMIS 108(12) (2000) 793–804. [9] R.R. Newbold, Lessons Learned from Perinatal Exposure to Diethylstilbestrol, Toxicol. Appl. Pharmacol. 199(2) (2004) 142–150.

25

ACCEPTED MANUSCRIPT [10] C.E. Reed, S.E. Fenton, Exposure to Diethylstilbestrol During Sensitive Life Stages: A Legacy of Heritable Health Effects, Birth Defects Res. C. Embryo Today Rev. 99(2) (2013) 134–146. [11] K.E. McMartin, K.A. Kennedy, G. P., S.N. Alam, P. Greiner, J. Yam, Diethylstilbestrol: A Review of Its Toxicity and Use as a Growth Promotant in Food-Producing Animals, J. Environ. Pathol. Toxicol. 1(3) (1978) 279–313.

PT

[12] A.G. Rico, Metabolism of Endogenous and Exogenous Anabolic Agents in Cattle, J. Anim.

RI

Sci. 57(1) (1983) 226–232.

SC

[13] R.L. Preston, Hormone Containing Growth Promoting Implants in Farmed Livestock, Adv.

NU

Drug Del. Rev. 38(2) (1999) 123–138.

[14] J.G. Liehr, A.M. Ballatore, J.A. McLachlan, D.A. Sirbasku, Mechanism of Diethylstilbestrol

Cancer Res. 43(6) (1983) 2678–2682.

MA

Carcinogenicity as Studied with The Fluorinated Analog E-3',3",5',5"-Tetrafluorodiethylstilbestrol,

D

[15] T. Tokumoto, M. Tokumoto, P. Thomas, Interactions of Diethylstilbestrol (DES) and DES

PT E

Analogs with Membrane Progestin Receptor-α and The Correlation with Their Nongenomic Progestin Activities, Endocrinology 148(7) (2007) 3459–3467.

CE

[16] W.N. Sawaya, K.P. Lone, A. Husain, B. Dashti, S. Al-Zenki, Screening for Estrogenic Steroids in Sheep and Chicken by The Application of Enzyme-Linked Immunosorbent Assay and A

563–569.

AC

Comparison with Analysis by Gas Chromatography-Mass Spectrometry, Food Chem. 63(4) (1998)

[17] L.C. Dickson, J.D. MacNeil, J. Reid, A.C.E. Fesser, Validation of Screening Method for Residues of Diethylstilbestrol, Dienestrol, Hexestrol, and Zeranol in Bovine Urine using Immunoaffinity Chromatography and Gas Chromatography/Mass Spectrometry, J. AOAC Int. 86(4) (2003) 631–639.

26

ACCEPTED MANUSCRIPT [18] L.C. Dickson, R. Costain, D. McKenzie, A.C.E. Fesser, J.D. MacNeil, Quantitative Screening of Stilbenes and Zeranol and Its Related Residues and Natural Precursors in Veal Liver by Gas Chromatography-Mass Spectrometry, J. Agric. Food Chem. 57(15) (2009) 6536–6542. [19] X.-B. Chen, Y.-L. Wu, T. Yang, Simultaneous Determination of Clenbuterol, Chloramphenicol and

Diethylstilbestrol

in

Bovine

Milk

by

Isotope

Dilution

Ultraperformance

Liquid

PT

Chromatography-Tandem Mass Spectrometry, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci.

RI

879(11–12) (2011) 799–803.

SC

[20] Y. Yang, J. Chen, Y.-P. Shi, Determination of Diethylstilbestrol in Milk using Carbon Nanotube-Reinforced Hollow Fiber Solid-Phase Microextraction Combined with High-Performance

NU

Liquid Chromatography, Talanta 97 (2012) 222–228.

MA

[21] X. Du, Y. Wu, H. Zhou, N. Cheng, M. Feng, J. Zheng, W. Cao, Development of a HPLC-ECD Method for The Simultaneous Determination of Three Synthetic Estrogens in Milk, Anal. Methods

D

5(11) (2013) 2822–2826.

PT E

[22] L. Wang, Y. Zhang, G. Liu, C. Zhang, S. Wang, A Time-Resolved Fluorescence Immunoassay for The Ultrasensitive Determination of Diethylstilbestrol Based on the Double-Codified Gold

CE

Nanoparticles, Steriods 89 (2014) 41–46.

[23] J. Wang, C. Cheng, Y. Yang, Determination of Estrogens in Milk Samples by Magnetic-Solid-

AC

Phase Extraction Technique Coupled With High-Performance Liquid Chromatography, J. Food Sci. 80(12) (2015) C2655–C2661. [24] Q. Qiao, N. Shi, X. Feng, J. Lu, Y. Han, C. Xue, Diethylstilbestrol in Fish Tissue Determined through Subcritical Fluid Extraction and with GC-MS, J. Ocean Univ. China 15(3) (2016) 489–494. [25] Q.-L. Zhang, J. Li, T.-T. Ma, Z.-T. Zhang, Chemiluminescence Screening Assay for Diethylstilbestrol in Meat, Food Chem. 111(2) (2008) 498–502.

27

ACCEPTED MANUSCRIPT [26] X. Xie, D. Sun, G. Liu, Q. Zeng, Sensitive and Rapid Determination of Diethylstilbestrol Using Mesoporous SBA-15 Modified Electrode, Anal. Methods 6(6) (2014) 1640–1644. [27] X. Zhu, L. Lu, X. Duan, K. Zhang, J. Xu, D. Hu, H. Sun, L. Dong, Y. Gao, Y. Wu, Efficient Synthesis of Graphene–Multiwalled Carbon Nanotubes Nanocomposite and its Application in Electrochemical Sensing of Diethylstilbestrol, J. Electroanal. Chem. 731 (2014) 84–92.

PT

[28] X. Ma, M. Chen, Electrochemical Sensor Based on Graphene Doped Gold Nanoparticles

RI

Modified Electrode for Detection of Diethylstilboestrol, Sensors Actuators B: Chem. 215 (2015)

SC

445–450.

[29] Z. Yan, P. Xiong, N. Gan, J. He, N. Long, Y. Cao, F. Hu, T. Li, A Novel Sandwich-Type

NU

Noncompetitive Immunoassay of Diethylstilbestrol using β-cyclodextrin Modified Electrode and

MA

Polymer-Enzyme Labels, J. Electroanal. Chem. 736 (2015) 30–37. [30] J.-M. Séquaris, J. Fritz, Voltammetric Detection and Analysis of the Behavior of

D

Diethylstilbestrol Oxidation Products, Electroanalysis 4(1) (1992) 121–127.

PT E

[31] I. Biryol, B. Salci, E. Erdik, Voltammetric Investigation of Diethylstilbestrol, J. Pharm. Biomed. Anal. 32(6) (2003) 1227–1234.

CE

[32] D. Lu, S. Lin, L. Wang, X. Shi, C. Wang, Y. Zhang, Synthesis of Cyclodextrin-Reduced Graphene Oxide Hybrid Nanosheets for Sensitivity Enhanced Electrochemical Determination of

AC

Diethylstilbestrol, Electrochim. Acta 85 (2012) 131–138. [33] L. Peng, S. Dong, H. Xie, G. Gu, Z. He, J. Lu, T. Huang, Sensitive Simultaneous Determination of Diethylstilbestrol and Bisphenol A based on Bi2WO6 Nanoplates Modified Carbon Paste Electrode, J. Electroanal. Chem. 726 (2014) 15–20. [34] M. Metzler, The Metabolism of Diethylstilbestrol, Crit. Rev. Biochem. 10(3) (1981) 171–212. [35] J.G. Liehr, B.B. DaGue, A.M. Ballatore, Reactivity of 4',4"-diethylstilbestrol quinone, a metabolic intermediate of diethylstilbestrol, Carcinogenesis 6(6) (1985) 829–836. 28

ACCEPTED MANUSCRIPT [36] D. Roy, J.G. Liehr, Metabolic Oxidation of Diethylstilbestrol to Diethylstilbestrol-4’,4”quinone in Syrian Hamsters, Carcinogenesis 10(7) (1989) 1241–1245. [37] H. Lund, Practical Problems in Electrolysis, in: H. Lung, O. Hammerich (Eds.), Organic Electrochemistry, Marcel Dekker, New York, USA, 2001, pp. 223–292. [38] L.L. Williams, R.D. Webster, Electrochemically Controlled Chemically Reversible

PT

Transformation of α-Tocopherol (Vitamin E) into Its Phenoxonium Cation, J. Am. Chem. Soc.

RI

126(39) (2004) 12441–12450

SC

[39] S. Chen, K.Y. Tai, R.D. Webster, The Effect of the Buffering Capacity of the Supporting Electrolyte on the Electrochemical Oxidation of Dopamine and 4-Methylcatechol in Aqueous and

NU

Nonaqueous Solvents, Chem. Asian J. 6(6) (2011) 1492–1499.

MA

[40] H. Beitollahi, S.G. Ivari, M. Torkzadeh-Mahani, Voltammetric Determination of 6Thioguanine and Folic Acid Using A Carbon Paste Electrode Modified with ZnO-CuO Nanoplates

D

and Modifier, Mater. Sci. Eng., C 69 (2016) 128–133.

PT E

[41] S. Jahani, H. Beitollahi, Selective Detection of Dopamine in the Presence of Uric Acid Using NiO Nanoparticles Decorated on Graphene Nanosheets Modified Screen-printed Electrodes,

CE

Electroanalysis 28(9) (2016) 2022–2028.

[42] P. Fenichel, N. Chevalier, F. Brucker-Davis, Bisphenol A: An Endocrine and Metabolic

AC

Disruptor, Ann. Endocrinol. 74(3) (2013) 211–220. [43] L.L. Ferreira, R. Couto, P.J. Oliveira, Bisphenol A as Epigenetic Modulator: Setting the Stage for Carcinogenesis?, Eur. J. Clin. Invest. 45(1) (2015) 32–36. [44] F. Rancière, J.G. Lyons, V.H.Y. Loh, J. Botton, T. Galloway, T. Wang, J.E. Shaw, D.J. Magliano, Bisphenol A and The Risk of Cardiometabolic Disorders: A Systematic Review with Meta-Analysis of The Epidemiological Evidence, Environ. Health (London, U. K.) 14(46) (2015) 1–34.

29

ACCEPTED MANUSCRIPT [45] D.D. Seachrist, K.W. Bonk, S.-M. Ho, G.S. Prins, A.M. Soto, R.A. Keri, A Review of the Carcinogenic Potential of bisphenol A, Reprod. Toxicol. 59 (2016) 167–182. [46] Y.Y. Chan, Y. Yue, Y. Li, R.D. Webster, Electrochemical/Chemical Oxidation of Bisphenol A in a Four-electron/Two-proton Process in Aprotic Organic Solvents, Electrochim. Acta 112 (2013) 287–294.

PT

[47] C.A. Frye, E. Bo, G. Calamandrei, F. Dessì-Fulgheri, M. Fernández, L. Fusani, O. Kah, M.

RI

Kajta, Y. Le Page, H.B. Patisaul, A. Venerosi, A.K. Wojtowicz, G.C. Panzica, Endocrine

SC

Disrupters: A Review of Some Sources, Effects, and Mechanisms of Actions on Behaviour and Neuroendocrine Systems, J. Neuroendocrinol. 24(1) (2012) 144–159.

NU

[48] C. Preda, M.C. Ungureanu, C. Vulpoi, Endocrine Disruptors in the Environment and Their

MA

Impact on Human Health, Environ. Eng. Manage. J. 11(9) (2012) 1697–1706. [49] A.C. Gore, V.A. Chappell, S.E. Fenton, J.A. Flaws, A. Nadal, G.S. Prins, J. Toppari, R.T.

D

Zoeller, EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting

PT E

Chemicals, Endocr. Rev. 36(6) (2015) E1–E150. [50] A.J. Fry, W.E. Britton, Solvents and Supporting Electrolytes, in: P.T. Kissinger, W.R.

CE

Heineman (Eds.), Laboratory Techniques in Electroanalytical Chemistry, Marcel Dekker, New York, USA, 1984, pp. 375–378.

AC

[51] R.D. Webster, A.M. Bond, T. Schmidt, Electrochemical Reduction of Some 2,6-disubstituted Pyridine-Based Esters and Thioic S-esters in Acetonitrile, J. Chem. Soc., Perkin Trans. 2 (7) (1995) 1365–1374. [52] S. Creager, Solvents and Supporting Electrolytes, in: C.G. Zoski (Ed.), Handbook of Electrochemistry, Elsevier, Amsterdam, The Netherlands, 2007, pp. 57–72.

30

ACCEPTED MANUSCRIPT [53] R. Francke, T. Quell, A. Wiebe, S. R. Waldvogel, Oxygen-Containing Compounds, in: O. Hammerich, B. Speiser (Eds.), Organic Electrochemistry: Revised and Expanded, CRC Press, Boca Raton, Florida, 2016, pp. 982–1029. [54] B. Speiser, Application of Digital Simulation, in: O. Hammerich, B. Speiser (Eds.), Organic Electrochemistry: Revised and Expanded, CRC Press, Boca Raton, Florida, 2016, pp. 205–227.

PT

[55] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd ed.,

RI

Wiley, New York, USA, 2001.

SC

[56] H.M. Peng, R.D. Webster, Investigation into Phenoxonium Cations Produced During the Electrochemical Oxidation of Chroman-6-ol and Dihydrobenzofuran-5-ol Substituted Compounds,

NU

J. Org. Chem. 73(6) (2008) 2169–2175.

MA

[57] B. Speiser, A. Rieker, Electrochemical Oxidations. Part 1. Phenoxy cations in the Anodic

AC

CE

PT E

D

Oxidation of Phenolic Compounds. J. Chem. Res. (S) (1977) 314–315.

31

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Graphical abstract

32

ACCEPTED MANUSCRIPT Highlights

CE

PT E

D

MA

NU

SC

RI

PT

Diethylstilbestrol (DES) undergoes electrochemical oxidation in acetonitrile. DES exhibited comparable voltammetric behaviour to bisphenol A (BPA). The oxidation process was posited to involve up to four moles of electrons.

AC

  

33