Electrochemical Flow Injection Analysis of Hydrazine in an Excess of

Aug 24, 2015 - Electrochemical Flow Injection Analysis of Hydrazine in an Excess of an Active Pharmaceutical Ingredient: Achieving Pharmaceutical Dete...
8 downloads 13 Views 2MB Size
Subscriber access provided by UNIV OF CAMBRIDGE

Article

Electrochemical Flow Injection Analysis of Hydrazine in an Excess of an Active Pharmaceutical Ingredient: Achieving Pharmaceutical Detection Limits Electrochemically Robert B Channon, Maxim B Joseph, Eleni Bitziou, Anthony W.T. Bristow, Andrew D. Ray, and Julie V. Macpherson Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b02719 • Publication Date (Web): 24 Aug 2015 Downloaded from http://pubs.acs.org on September 5, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Electrochemical Flow Injection Analysis of Hydrazine in an Excess of an Active Pharmaceutical Ingredient: Achieving Pharmaceutical Detection Limits Electrochemically Robert B. Channon,† Maxim B. Joseph,† Eleni Bitziou,† Anthony W. T. Bristow,‡ Andrew D. Ray,‡ and Julie V. Macpherson*† †

Department of Chemistry, University of Warwick, Coventry, CV4 7AL, United Kingdom



Pharmaceutical Development, AstraZeneca, Macclesfield, SK10 2NA, United Kingdom

The quantification of genotoxic impurities (GI) such as hydrazine (HZ) is of critical importance in the pharmaceutical industry in order to uphold drug safety. HZ is a particularly intractable GI and its detection represents a significant technical challenge. Here we present for the first time, the use of electrochemical analysis to achieve the required detection limits by the pharmaceutical industry for the detection of HZ in the presence of a large excess of a common active pharmaceutical ingredient (API); acetaminophen (ACM) which itself is redox active, typical of many APIs. A flow injection analysis approach with electrochemical detection (FIA-EC) is utilized, in conjunction with a co-planar boron doped diamond (BDD) microband electrode, insulated in an insulating diamond platform for durability and integrated into a two piece flow cell. In order to separate the electrochemical signature for HZ such that it is not obscured by that of the ACM (present in excess), the BDD electrode is functionalized with Pt nanoparticles (NPs) to significantly shift the half wave potential for HZ oxidation to less positive potentials. Microstereolithography was used to fabricate flow cells with defined hydrodynamics which minimize dispersion of the analyte and optimize detection sensitivity. Importantly, the Pt NPs were shown to be stable under flow and a limit of detection of 64.5 nM or 0.274 parts per million for HZ with respect to the ACM, present in excess, was achieved. This represents the first electrochemical approach which surpasses the required detection limits set by the pharmaceutical industry for HZ detection in the presence of an API and paves the wave for on-line analysis and application to other GI and API systems. ________________________________________________________________________________________________________________

Submitted to Analytical Chemistry, July 2015; revised August 2015

ACS Paragon Plus Environment 1

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

INTRODUCTION Hydrazine (HZ) is a carcinogenic and mutagenic compound which is used as a reducing agent,1-3 pharmaceutical precursor4,5 and as a direct liquid fuel cell.6 In a pharmaceutical process, HZ or HZ derivatives usually occur as reagents, intermediates or degradation products and because of their carcinogenicity are classified as genotoxic impurities (GI). In order to uphold drug safety and adhere to pharmaceutical guidelines, GIs must be quantified down to parts-permillion (ppm) levels with respect to the active pharmaceutical ingredient (API), which can be present at high concentration.7,8 This necessitates quantification of GIs down to sub-µM concentrations in the presence of excess API, up to molar concentrations. Amongst the many chemicals classified as GIs,9 HZ is particularly difficult to quantify due to its high volatility, high polarity, low molecular weight and lack of chromophore, exacerbated in the presence of non-volatile, high-molecular weight APIs.10 A wide variety of techniques have been evaluated for HZ quantitation,11 as shown in Table 1, with the resulting limits of detection (LOD) displayed in ppm versus the amount of API present. Derivatization is often employed in the analysis process (Table 1), whereby HZ is reacted with a species such as acetone12 or benzaldehyde13-15 to enable detection. However, the yields of the derivatization reaction are varied (~85-95%), leading to uncertainty in the amount of HZ present.14 Some approaches also utilize a pre-concentration step, for example liquid-liquid extraction.13 Many of these methods require long analysis times, high cost of instrumentation and are difficult to operate in-line, adding to the overall cost of drug development. Furthermore, whilst the required detection limits are sometimes less stringent for shorter periods of exposure to the GI, many approaches do not

ACS Paragon Plus Environment 2

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

reach the required LOD (~ 1 ppm), compared to the amount of API, as stipulated by pharmaceutical guidelines.8 The various approaches commonly taken are shown in Table 1.

Table 1: Exemplar methods for HZ detection in the presence of APIs Technique

Application

Derivatization

LOD with respect Ref. to API / ppm

GC-MS

Analysis with API

acetone-d6

0.03†

12

HPLC-liquidliquid extraction

Analysis in a range of Benzaldehyde excipients

0.04‡

13

HPLC

Analysis in hydralazine

0.27‡

14

GLC

Analysis in hydralazine Benzaldehyde and isoniazid

3

15

Fluorimetric detection

Analysis in isoniazid 2-hydroxy-1formulations naphthaldehyde

6.4

16

Fluorescence detection

Analysis in water, 3,6plasma and isoniazid diacetoxyfluoran

2340

17

CE-IPD

Analysis with API

n/a

2

18

CE-EC

Analysis in a range of excipients

n/a

29

19

IC-EC

Analysis with API

n/a

50

20

HILIC-CLND

Analysis with API

n/a

200

21

CE-EC

Analysis in a range of excipients

n/a

233

22

Benzaldehyde

GC-MS: gas chromatography mass spectrometry; HPLC: high performance liquid chromatography; GLC: gas liquid chromatography; CE: capillary electrophoresis; EC: electrochemical detection; IPD: indirect photometric detection; IC: ion chromatography; HILIC: hydrophilic interaction liquid chromatography and CLND: chemiluminescent nitrogen detection. †Based on quoted limit of quantitation; ‡smallest ppm of HZ detected for a range of batches of different APIs.

ACS Paragon Plus Environment 3

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

Electrochemical methods, by contrast, typically exhibit fast analysis times, inexpensive instrumentation and are adaptable for on-line analysis. Many APIs show an electrochemical signature23 and electrochemical (EC) detection of the API has been demonstrated, in conjunction with capillary electrophoresis22 and ion chromatography,20 within the pharmaceutical industry. Boron doped diamond (BDD) electrodes exhibit several beneficial electrochemical properties over more traditional electrodes, such as a wide solvent window, low background currents and high resistance to fouling.24 Many of these properties arise as a result of the catalytically inactive nature of the BDD surface. HZ, due to its inner sphere nature, undergoes irreversible oxidation on BDD at very high overpotentials.25 Note, inner sphere species (or reactant intermediates) must adsorb on the electrode surface in order to transfer electrons; the nature of the adsorption process plays a role in determining the efficiency of electron transfer. Despite this, HZ can still be determined down to µM levels (0.032 µg ml-1) at BDD electrodes, although these studies were conducted in the absence of an API.25 As a large majority of APIs are electroactive in the oxidative potential range,23,26-28 and the API is always present in large excess, the electrochemical signature of the API will always be significantly greater than that of the GI. It is therefore essential that the GI detection signal is electrochemically separable from that of the API and can be quantified down to ppm levels with respect to the API. By changing the elemental identity of metal nanoparticles (NP) electrodeposited on a BDD electrode, it is possible to shift the EC half wave potential of inner sphere redox couples, such as HZ.29 Metal NPs on BDD electrodes offer a high EC signal to noise ratio.24 However, as the vast majority of experiments are typically carried out under stationary conditions30 this can constrain the achievable LOD’s. There are limited reports of metal NPs on carbon electrodes under hydrodynamic (flow or rotation) conditions, however the metal NPs are often held in place by a

ACS Paragon Plus Environment 4

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

support or mask to stabilize the particles in the flow environment.31-36 Pt37 and Au38 NPs have been employed in conjunction with rotating BDD electrodes, however, the mechanical stability of the NPs, under these forced convection conditions was not discussed. Flow injection analysis (FIA) is a well-established technique39,40 which has seen increasing use in the pharmaceutical industry.41-44 In short, a plug of analyte is injected into a carrier stream which flows through a detector. Minimal quantities of reagent are required for analysis and high sampling rates are often used, making FIA a particularly attractive approach for analysis of pharmaceuticals. Also, when combined with EC, the convective contribution to mass transport, compared to diffusive stationary experiments enhances the analytical signal without significantly increasing the background signal. Co-planar microband electrodes, coupled to FIA and deployed in specifically designed microfluidic flow cells are particularly attractive in EC analysis; offering trace level detection of the analyte of interest under well-defined hydrodynamic conditions (laminar Poiseuille flow).45 In this paper, we report for the first time, EC detection of a GI, HZ, in the presence of large excess of an electrochemically active API (acetaminophen: ACM), at the required LOD stipulated by pharmaceutical guidelines for GIs in the presence of an API.8 This approach importantly does not require derivatization or liquid-liquid extraction of the GI, as in e.g. HPLC or GC-MS analysis. We employ a co-planar insulating diamond sealed BDD microband electrode operated under hydrodynamic flow conditions and functionalized with metal NPs, to separate the HZ and ACM electrochemical signals. We also report on the stability of the metal NPs under FIA-EC conditions and show that through optimization of the flow cell and the electrode geometry, HZ can be quantified down to sub-ppm levels, with respect to ACM, present

ACS Paragon Plus Environment 5

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

in excess (50 mM). This approach holds considerable promise for on-line analysis and is applicable to a wide range of electroactive GIs, in the presence of excess API.

EXPERIMENTAL Solutions: HZ sulfate, ACM, potassium nitrate (KNO3), potassium hexachloroplatinate(IV) (PtK2Cl6), sulfuric acid (H2SO4), sodium dihydrogen orthophosphate (NaH2PO4), sodium phosphate dibasic heptahydride (Na2HPO4.7H2O) and sodium hydroxide (NaOH) were purchased

from

Sigma-Aldrich

(UK).

Ferrocene

methyltrimethylammonium

hexafloruophosphate (FcTMA+) was prepared in house as previously described.46 Phosphate buffer solutions (PBS) were prepared from NaH2PO4 and Na2HPO4.7H2O with 0.1 M KNO3, then adjusted to pH 6.9 via drop-wise addition of 1 M NaOH. All solutions were prepared using MilliQ distilled water (18.2 MΩ resistivity, Millipore). Flow detection system: A piston pump (305, GILSON, USA) was used in conjunction with a manometric module (806, GILSON, USA) and a manual sample injector (7725i, Rheodyne UK) with a 50 µL sample loop (PEEK, VICI, USA). PEEK tubing (0.18 mm ID, VICI, USA) was used to connect the sample injector to the flow cell and Teflon tubing (0.5 mm ID, VWR, NI) for the flow cell outlet. The channel flow cell was designed using SolidWorks (Dassault Systems, FR), as shown in Figure 1a, in order to minimize channel dimensions and dead volumes compared to previous designs (see electronic supporting information ESI1).47 It was manufactured using microstereolithography (Perfactory Mini, EnvisionTec, UK) from a methacrylate based photoactive resin (R11, EnvisionTec, UK), as depicted in Figure 1b. The channel dimensions were 22.5 µm height (2h, as determined by interferometry, ContourGT, Bruker, UK) by 3 mm width (w) and 6 mm length (l). All electrochemical measurements within the flow cell were carried out versus a homemade Ag/AgCl quasi-reference electrode (1 mm

ACS Paragon Plus Environment 6

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

diameter) and a Pt wire counter electrode (1 mm diameter), both placed in the outlet of the flow cell (Figure 1b). The Ag/AgCl electrode was fabricated by chloridising an Ag wire, by holding at -1 V in 3 M KCl for 1 minute. BDD Electrode: The insulating diamond sealed BDD microband electrode used herein was fabricated according to previously described procedures for the production of co-planar all diamond electrodes.48,49 In short, two trenches, 89 µm and 500 µm wide, were lasered (E-355H3-ATHI-O, Oxford Lasers, UK) into an insulating diamond substrate (thermal grade, Element Six, Harwell, UK), 1 mm thick and freestanding, grown by microwave chemical vapor deposition (MW-CVD). High quality electroanalysis grade BDD (Element Six, Harwell, UK);50 doped with sufficient boron to be metal-like (~ 3 × 1020 boron atoms cm-3), containing negligible sp2 carbon, with a characteristic low capacitance and extended solvent window was then overgrown into the trenches. The sample was polished flat (~ nm surface roughness) to reveal two co-planar band electrodes (Figure 1c). The small black spots on the insulating diamond are micro-fractal defects (~ nm in size).51 These may contain small amounts of non-diamond-carbon, but as they are electrically isolated from the BDD will not contribute to the electrochemical response of the band electrode. For this study only the 89 µm wide band electrode was employed in order to achieve the highest mass transport rates and thus the lowest LOD. Note this device can also be operated in two electrode mode e.g. for electrochemical generation-collection type experiments.52 The whole structure was then acid cleaned in concentrated H2SO4 supersaturated with KNO3. Ohmic contacts were made by sputtering (Edwards E606 sputter/evaporator) Ti (~20 nm) and Au (~300 nm) and annealing for four hours at 400 °C.53 Each band was then connected to Cu wire using conducting silver epoxy (RS components, UK) and the connections sealed using non-

ACS Paragon Plus Environment 7

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

conductive epoxy (Robnor, UK). Before use, the electrode was polished with alumina slurry (0.05 µm, Buehler, DE), rinsed with water and air dried. Deposition of Pt NPs was achieved by holding the BDD electrode at -1 V versus a silver/silver chloride electrode (Ag/AgCl) for 1.5 s in a solution of 1 mM PtK2Cl6 in 0.1 M H2SO4 using a procedure adapted from reference 24. An optimized cyclic voltammetric (CV) cleaning step was utilized in-between each injection for the FIA-EC determination of HZ in the presence of ACM, by cycling once between -0.40 V and +1.20 at 0.5 V s-1 in the carrier stream (0.2 M PBS).

Figure 1. FIA-EC analysis system: (a) CAD schematic of the FIA flow cell, (b) the key components of the flow-cell, the all diamond BDD microband electrode (bottom) the microstereolithographically fabricated FIA flow cell (middle) and the connecting tubes (top) and (c) the all diamond BDD dual band electrode (only the 89 µm wide band electrode was employed in these studies).

ACS Paragon Plus Environment 8

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Software: An Ivium potentiostat (CompactStat, Alvatek, UK) connected to a desktop PC was used for electrochemical measurements. A high-resolution Supra 66 VP Field-emission scanning electron microscope (FE-SEM, Zeiss, DE) was used to obtain the FE-SEM images in conjunction with a secondary electron detector, 2 kV accelerating voltage and 5 mm working distance. The FE-SEM images were analyzed using MATLAB (R2013a, MathWorks, UK). An Enviroscope atomic force microscope (AFM, Veeco, USA) with a NanoScope IV controller (Veeco, USA) was used to profile the electrode topography in tapping mode.TM

RESULTS and DISCUSSION Characterization of the flow detection system: The experimental set-up used for this study is shown in Figure 1b. Encapsulation of the BDD band electrode in an insulating diamond substrate results in a robust, well-defined electrode that does not change electrode geometry with repeated use (Figure 1c). Furthermore, the BDD is essentially co-planar with the insulating diamond substrate; AFM reveals the BDD is recessed by only 1.69 nm ± 0.75 (n = 3) compared to the insulating surround. The substrate forms the base of a two part flow cell (Figures 1b) where the cell is clamped to the electrode substrate54 with minimal force, facilitating easy assembly/disassembly. The BDD electrode can also be easily cleaned, using, for example, alumina slurry, without modifying the electrode area. This contrasts with lithographic approaches where thin film Au or Pt microband electrodes on glass cannot be abrasively cleaned due to mechanical removal of the film. In order to characterize the experimental set-up, CVs recorded under continuous flow at a potential scan rate of 0.1 V s-1 were recorded for the fast outer sphere electron transfer couple; ferrocenylmethyltrimethylammonium hexafluorophosphate (10 µM FcTMA+ in 0.1 M KNO3,

ACS Paragon Plus Environment 9

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

Figure 2a) as a function of volume flow rate, Vf over the range 0.1 – 1.5 ml min-1. The flow system was designed such that laminar flow was fully developed in the channel before reaching the electrode, diffusion along the direction of flow was negligible, and the concentration gradient was confined adjacent to the electrode (Lévêque approximation). Flow in the channel was characterized by a Reynolds number (Re) = 22.0 for Vf = 1 ml min-1, corresponding to laminar flow, as calculated from equation (1); ܴ௘ =

ഥ‫ܦ‬௛ ܷ ߥ

(1)

where Ū, is the mean fluid velocity (cm s-1) = Vf / 2hw, Dh is the hydraulic diameter (cm) = 8hw /2h + w, and ν is the kinematic viscosity = 0.01 cm2 s-1.47 Under these conditions the theoretical limiting current, ilim, as a function of Vf is given by the Levich equation;

ilim = 0.925nFcb D2/3V f 1/3w2/3h−2/3 xe2/3 (2) where n is the number of electrons transferred, F is Faraday’s constant, cb is the bulk analyte concentration, D is the diffusion coefficient (DFcTMA+ = 6.71 × 10-6 cm2 s-1, determined by CV with a 14.5 µm radius Pt ultramicroelectrode in a solution containing 1 mM FcTMA+ and 0.1 M KNO3), w is the channel width, 2h is the channel height and xe is the electrode width (dimensions given in the experimental section). As shown in the inset to Figure 2a, there is a very good correlation between the experimental ilim data (•) and the theoretical ilim as predicted by equation 2 (▬), confirming the flow characteristics of the cell, for the given electrode and cell dimensions.

ACS Paragon Plus Environment 10

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 2 (a) CVs recorded under continuous flow with 10 µM FcTMA+ in 0.1 M KNO3, at 0.1 V s-1 with Vf of 0.1 (lower), 0.25, 0.5, 0.75, 1, 1.25 and 1.5 (upper) ml min-1. Inset: Experimentally recorded ilim vs. Vf1/3 plotted against the Levich theory line (equation 2), for the cell and electrode dimensions defined (w = 3 mm, h = 22.5 µm, xe = 89 µm). (b) FIA-EC detection for 50 µL injection of 10 µM FcTMA+ in 0.1 M KNO3 at Vf = 1 ml min-1, electrode held under transport limited conditions at +0.6 V vs. Ag/AgCl. Flow injection analysis of FcTMA+: In FIA-EC systems, the detection capabilities are governed by the residence time, tr (the time between sample injection and peak signal), dispersion coefficient, Dc (how much the injected analyte is diluted before detection) and by the channel and electrode dimensions. Optimization of the flow cell to minimize dispersion compared to previous designs47,48 is described in the Electronic Supporting Information 1 (ESI 1). Note in our system due to the small cell height and fast volume flow rates convective (Pouiseulle) transport dominates over radial (Taylor-like) diffusion.55 Furthermore, given the injection volume (50 µL) is significantly larger than the volume of the cell (0.405 µL), the dispersion coefficient, Dc, can be given by the following equation;56

ACS Paragon Plus Environment 11

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Dc =

ilim ip

Page 12 of 23

(3)

where ip is the experimental peak current. As shown in Figure 2b, ip for an 50 µL injection of 10 µM FcTMA+ in 0.1 M KNO3, at a Vf of 1 ml min-1 (used for all FIA-EC measurements described herein) is 138.5 nA. The BDD electrode was biased at a potential of 0.6 V sufficient to oxidize FcTMA+ at a transport-controlled rate. When comparing this value to that predicted from the Levich equation (equation 2, Figure 2a) there is excellent correlation, with a calculated Dc = 1.00. This indicates minimal dispersion of the plug of injected analyte between injection and detection. tr, which controls the maximum frequency of measurement was determined as 1.83 seconds at Vf = 1 ml min-1. The peak shape is non-symmetrical, with a tailing current on the right hand side which we believe reflects difficult to avoid dead space in the cell design acting to trap analyte solution. NP-BDD electrodes and stability in flow: To differentiate the electrochemical oxidation signatures for HZ and the model API employed herein, ACM (present in excess), it is important that the significantly smaller HZ electrochemical signal is found at a potential less positive than that of the API. Using BDD, as shown in ESI 2, the oxidative signal for HZ is significantly more positive than that for ACM (half wave potential, E1/2 for HZ is +1.12 V vs. +0.59 V for ACM at a 1 mm diameter BDD electrode in pH 6.9 PBS at 0.1 V s-1). Hence BDD electrodes alone cannot be used to electrochemically resolve the two species. ACM is known to undergo a two electron oxidation mechanism and the oxidation potential has been previously shown to be sensitive to pH.57 As shown in Figure S-3a in ESI 3, by moving away from BDD, the E1/2 for ACM oxidation is shifted negatively to +0.42 V, +0.38 V and +0.37 V vs. SCE, for Pt, glassy carbon and Au electrodes respectively. For HZ, the shift in E1/2 in the negative direction compared to BDD is far more dramatic, as shown Figure S-3b in ESI 3 (glassy

ACS Paragon Plus Environment 12

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

carbon: +0.51 V; Au: +0.27 V and Pt: -0.16 V) with the largest shift seen for Pt, where ET is the most kinetically facile. Hence by changing the identity of the electrode, it should be possible to shift E1/2 such that the HZ signature occurs well before that of ACM. To achieve the required LOD it is preferable to work with Pt NP modified BDD electrodes rather than Pt electrodes, due to the inherently higher background signals of the latter.24 To verify electrochemical resolution of the HZ and ACM EC signals, linear sweep voltammograms (LSVs) were recorded at 0.1 V s-1 in a stationary solution (housed within the flow cell) containing 1 mM HZ in 0.2 M PBS (black line) and 50 mM ACM in 0.2 M PBS (dotted line), as shown in Figure 3. A BDD microband electrode (xe = 89 µm, w = 3 mm as dictated by width of channel) was employed, functionalized with Pt NPs, electrochemically deposited as described in the Experimental Section. For HZ oxidation, the current reaches a steady state at ~ 0.2 V, whilst a current does not begin flowing for ACM oxidation, present in much higher concentration, until ~ 0.15 V.

Figure 3. LSV of 1 mM HZ (black line) and 50 mM ACM (dotted black line), both in 0.2 M PBS at 0.1 V s-1, recorded at a Pt NP-BDD microband electrode in stationary solution (confined within the microfluidic flow cell).

ACS Paragon Plus Environment 13

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

It is essential that under the flow conditions utilized, the electrochemically deposited NPs are stable and not prone to detachment, as this could lead to decreasing signals and erroneous analysis with extended use.58 Some researchers have claimed a weak attachment between a BDD surface and metal NPs,59 however it is essential to take into account surface morphology and wettability. For example, on hydrophilic polished microcrystalline surfaces of the type employed here, Pt NPs on BDD have been found to be extremely stable in quiescent solutions.24 To explore NP stability on the BDD microband electrode under the relatively high Vf conditions of the microfluidic flow cell and over reasonable experimental timescales (1 - 2 hours), CVs were recorded in 0.1 M H2SO4 at Vf = 1 ml min-1 and 10 V s-1, every 10 minutes for 2 hours. Figure 4 shows the first CV recorded. Clearly evident are two typical hydrogen adsorption peaks (Hads) at -0.4 and -0.7 V vs. Ag/AgCl, as well as a broad hydrogen desorption peak (Hdes) at ca. -0.5 V (blue shaded region). As has previously been shown, these features can be used to characterize the surface, as the area under these peaks corresponds to the absorption/desorption of a monolayer of hydrogen atoms.58 Note that at the potential extremes, resistive current signatures for both hydrogen evolution (< -0.8 V) and oxygen evolution (> 1.2 V) are seen, due to a combination of the high currents passed in conjunction with the placement of the counter and reference within the flow cell outlet and the small cross sectional area of the channel. For the eleven subsequent CVs recorded over 2 hours, the area under the Hdes peak was found to remain relatively constant, 0.208 ± 0.014 µC, suggesting that the Pt NPs were stable under flow. Additionally, assuming that one hydrogen atom adsorbs on one atom of the Pt surface and that the density of atoms on a Pt surface is 1.31 × 10-15 atom cm-2, the charge per unit area is 210 µC cm-2.60 Therefore, the density of Pt NP on the surface of the BDD band electrode is 301 ± 20 NP

ACS Paragon Plus Environment 14

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

µm-2, which is reasonable considering previous studies of Pt NP-BDD under similar deposition conditions (130 NP µm-2 - 340 NP µm-2 between more and less active grains of BDD).24

Figure 4. CV of the Pt NP-BDD microband electrode recorded in 0.1 M H2SO4 at 10 V s-1 and Vf = 1 ml min-1. To further verify Pt NP stability under channel flow, FE-SEM was employed to characterize the electrode surface before and after continuous flow. EC was carried out using a droplet (15 µL) of 1 mM PtK2Cl6 in 0.1 M H2SO4, placed on top of the BDD band electrode. The electrode was then prepared for FE-SEM by washing gently with water and drying in air. A typical (n = 4 on the same substrate) 9 µm × 6 µm image of the surface area prior to solution flow is shown in Figure 5a. After imaging, the electrode was placed in the flow cell and 0.2 M PBS flown over at Vf = 1 ml min-1 for one hour, before repeat FE-SEM imaging using the same magnification. A representative image (n = 4 on the same substrate) after flow is shown in Figure 5b.

ACS Paragon Plus Environment 15

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

Figure 5. Representative FE-SEM images of the Pt NP-BDD band electrode (i) immediately after electrochemical deposition, under stationary conditions and (ii) after one hour of flowing continuously over the electrode at Vf = 1 ml min-1 with 0.2 M PBS. Note the image has been reverse color contrasted to enhance visualization of the Pt NPs (black spots). It is interesting to note that in both images slightly different densities of Pt NP are observed across different areas of the BDD surface, as has been observed previously.24,61 This is due to the differently doped regions of the polycrystalline surface, where slightly higher densities of NPs are observed at the more highly doped grains.62 The FE-SEM images clearly show that a significant number of NPs are still present after being subject to hydrodynamic flow. The FESEM images were analyzed in MATLAB using a circle finding script, in order to obtain NP density for each image. Quantitatively, it was found that the NP density does not change after flow (i.e. before 54.9 ± 3.7 NP µm-2, n = 4 and after flow 55.6 ± 3.6 NP µm-2, n = 4). Both the

ACS Paragon Plus Environment 16

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

CV and microscopic data indicate that the NPs are stable under the flow conditions used in this study. It is interesting to note that this method determines a lower NP density in comparison to the CV approach. This is most likely due to the circle finding script not being able to identify the smallest NPs, which are difficult to resolve using FE-SEM.

Hydrazine detection via FIA-EC analysis: From the LSV data in Figure 3, for FIA-EC detection of HZ in the presence of ACM, an electrode potential of +0.10 V versus Ag/AgCl was chosen to maximize the signal to noise ratio of the HZ signal, whist avoiding any current contributions from ACM oxidation. As recently discussed,63 HZ oxidation on Pt is sensitive to electrode pretreatment and the presence of residual oxides on the electrode surface. In order to improve the reproducibility of the HZ oxidative wave when repeat measurements are made, cleaning steps are often employed, for example sweeping between positive and negative potentials64,65 or holding the surface at a constant potential.66,67 Here, running a CV between 0.40 V and +1.20 (0.5 V s-1 in the carrier stream) immediately after each injection was found to give minimal change in HZ signal between repeat measurements and was thus adopted for HZ determination in the presence of ACM. HZ concentrations in the range 0 – 100 µM in the presence of 50 mM ACM were injected (50 µL, n = 6 per concentration) into the continuous phase flow stream (0.2 M PBS) as shown in Figure 6. The peak current varies linearly with HZ concentration in the range 1 - 100 µM, with a corresponding sensitivity of 0.337 µA µM-1. The LOD can be calculated from equation (4);68 LOD = µ + 3σ

(4)

where µ is the mean signal for a blank injection and σ is the standard deviation of the blank signal. This resulted in an LOD of 64.5 nM. In order to put this into context with the amount of ACM present, equation 5 was employed, where;

ACS Paragon Plus Environment 17

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

‫= ݉݌݌‬

௠௔௦௦ ௛௬ௗ௥௔௭௜௡௘ (௠௚) ௠௔௦௦ ௔௖௘௧௔௠௜௡௢௣௛௘௡ (௞௚)

Page 18 of 23

(5)

Thus for 50 mM ACM (mw = 151.163 g mol-1), 64.5 nM of HZ (mw = 32.045 g mol-1) equates to 0.274 ppm of HZ with respect to the API, which is well within the most stringent pharmaceutical guidelines for long term GI exposure.8 Furthermore, this result compares favorably with alternate methods for HZ detection in the presence of APIs as depicted in Table 1, particularly as the approach utilized in this study does not require derivatization or liquid-liquid extraction. There is scope for reducing LODs further by simply moving to higher Vf’s, smaller h flow channels, thinner bands etc.

Figure 6. FIA-EC with an analyte solution containing 50 mM ACM and 0, 1, 5, 10, 25, 50, 75 and 100 µM of HZ. The Pt NP BDD microband electrode was held at a potential of +0.1 V versus Ag/AgCl, and 50 µL volumes were injected into a 0.2 M PBS solution flowing at Vf = 1 ml min-1, (n = 6). Inset plot of peak current against concentration, R2 = 0.993, sensitivity = 0.337 µA µM-1, LOD = 64.5 nM (0.274 ppm).

CONCLUSIONS In this work we have demonstrated for the first time an electrochemical approach for the quantitation of a GI (HZ) in the presence of excess API (ACM) in aqueous solvents, which is of

ACS Paragon Plus Environment 18

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

key interest to the pharmaceutical industry. This approach paves the way for on-line analysis in both aqueous and organic solvents, as extraction or derivatization procedures which take place off-line, are no longer required. The experimental flow system allows for FIA-EC detection of species with minimal dispersion of the analyte in the carrier stream. The two part cell, with the bottom component comprising a co-planar BDD microband electrode, insulated in insulating diamond platform is also easy to assemble and disassemble repeatedly. Moreover, the diamond based substrate means the electrode never changes its geometry through repeated use. Using metal NP BDD electrodes, which offer higher signal to noise ratios that metal electrodes, it is also possible to resolve the electrochemical signatures for the HZ and the ACM simply through choice of the most appropriate metal (Pt was found appropriate in this study); this is not possible using BDD alone. It was also shown that the Pt NPs were stable under the flow conditions used herein, over the timescale of typical experiments, when electrochemically deposited onto a ~ nm flat BDD microcrystalline surface. No evidence of NP detachment was observed using either FE-SEM or adsorption/desorption of hydrogen electrochemistry. Finally, the proposed approach allows HZ detection down to sub-ppm levels with respect to the API in accordance with pharmaceutical guidelines and represents a fast, cost effective alternative to current methods for GI detection in the pharmaceutical industry.

ASSOCIATED CONTENT Supporting Information Optimization of the flow cell geometry (ES1) and voltammetric data for HZ and ACM oxidation at glassy carbon, Pt and Au electrodes (ESI 2 and 3) are provided in the supporting information. This material is available free of charge via http://pubs.acs.org

ACS Paragon Plus Environment 19

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

AUTHOR INFORMATION Corresponding Author * E-mail: [email protected]. Tel: +44 (0) 2476 573886 Notes The authors declare no competing financial interest ACKNOWLEDGEMENTS JVM thanks the Royal Society for an Industry Fellowship. RC thanks the ESPRC (iCase) and AstraZeneca for funding. We thank Element Six (Harwell, Oxford) for growth and polishing of the all-diamond electrode structures and for funding EB and MJ. We also thank Dr Jonathan Newland (Warwick Chemistry) for taking photos of the flow cell, Lingcong Meng (Warwick Chemistry) for assistance with FE-SEM and Sophie Kinnear (Warwick Chemistry) for help with MATLAB. Finally we thank Frank Courtney (Warwick University School of Engineering) for use of the MSL facility and Dr Stanley Lai (University of Twente) for useful discussions. REFERENCES (1) Dewanji, A.; Murarka, S.; Curran, D. P.; Studer, A. Org. Lett. 2013, 15, 6102-6105. (2) Graham, M. A.; Bethel, P. A.; Burgess, J.; Fairley, G.; Glossop, S. C.; Greenwood, R. D. R.; Jones, C. D.; Lovell, S.; Swallow, S. Org. Lett. 2013, 15, 6078-6081. (3) Uygun, Y.; Bayrak, H.; Ozkan, H. Turk. J. Chem. 2013, 37, 812-823. (4) Farghaly, T.; Abdallah, M.; Aziz, M. Molecules 2012, 17, 14625-14636. (5) Sycheva, T. P.; Pavlova, T. N.; Shchukina, M. N. Pharm. Chem. J. 1972, 6, 696-698. (6) Sakamoto, T.; Asazawa, K.; Sanabria-Chinchilla, J.; Martinez, U.; Halevi, B.; Atanassov, P.; Strasser, P.; Tanaka, H. J. Power Sources 2014, 247, 605-611. (7) Teasdale, A., US Department of Health and Human Services, F. a. D. A., Center for Drug Evaluation and Research (CDER), Ed., 2008. (8) Müller, L.; Mauthe, R. J.; Riley, C. M.; Andino, M. M.; Antonis, D. D.; Beels, C.; DeGeorge, J.; De Knaep, A. G. M.; Ellison, D.; Fagerland, J. A.; Frank, R.; Fritschel, B.; Galloway, S.; Harpur, E.; Humfrey, C. D. N.; Jacks, A. S.; Jagota, N.; Mackinnon, J.; Mohan, G.; Ness, D. K.; O’Donovan, M. R.; Smith, M. D.; Vudathala, G.; Yotti, L. Regul. Toxicol. Pharmacol. 2006, 44, 198-211.

ACS Paragon Plus Environment 20

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(9) Robinson, D. I. Org. Process Res. Dev. 2010, 14, 946-959. (10) Elder, D. P.; Snodin, D.; Teasdale, A. J. Pharm. Biomed. Anal. 2011, 54, 900-910. (11) Teasdale, A. Genotoxic Impurities Stratagies for Identification and Control; John Wiley & Sons: New Jersey, 2010, p 428. (12) Sun, M. J.; Bai, L.; Liu, D. Q. J. Pharm. Biomed. Anal. 2009, 49, 529-533. (13) Kean, T.; Miller, J. H. M.; Skellern, G. C.; Snodin, D. Pharmeuropa scientific notes 2006, 2, 23-33. (14) Gyllenhaal, O.; Grönberg, L.; Vessman, J. J. Chromatogr. A 1990, 511, 303-315. (15) Matsui, F.; Robertson, D. L.; Lovering, E. G. J. Pharm. Sci. 1983, 72, 948-951. (16) Mañes, J.; Gimeno, M. J.; Moltó, J. C.; Font, G. J. Pharm. Biomed. Anal. 1988, 6, 10231027. (17) Qu, D.-Y.; Chen, J.-L.; Di, B. Anal. Methods 2014, 6, 4705-4709. (18) Khan, M.; Kumar, S.; Jayasree, K.; Krishna Reddy, K. V. S. R.; Dubey, P. K. Chromatographia 2013, 76, 801-809. (19) You, T.; Niu, L.; Gui, J. Y.; Dong, S.; Wang, E. J. Pharm. Biomed. Anal. 1999, 19, 231237. (20) Jagota, N. K.; Chetram, A. J.; Nair, J. B. J. Pharm. Biomed. Anal. 1998, 16, 1083-1087. (21) Liu, M.; Ostovic, J.; Chen, E. X.; Cauchon, N. J. Chromatogr. A 2009, 1216, 2362-2370. (22) Liu, J.; Zhou, W.; You, T.; Li, F.; Wang, E.; Dong, S. Anal. Chem. 1996, 68, 3350-3353. (23) Ozkan, S. A. Electroanalytical Methods in Pharmaceutical Analysis and Their Validation; HNB publishing: New York, USA, 2012, p 350. (24) Hutton, L.; Newton, M. E.; Unwin, P. R.; Macpherson, J. V. Anal. Chem. 2009, 81, 10231032. (25) Sun, H.; Dong, L.; Yu, H.; Huo, M. Russ. J. Electrochem. 2013, 49, 883-887. (26) Gupta, V. K.; Jain, R.; Radhapyari, K.; Jadon, N.; Agarwal, S. Anal. Biochem. 2011, 408, 179-196. (27) Siangproh, W.; Apilux, A.; Chantarateepra, P.; Chailapakul, O. In Synthetic Diamond Films: Preparation, Electrochemistry, Characterization and Applications, Brillas, E.; MartinezHuitle, C. A., Eds.; John Wiley & Sons: Hoboken, New Jersey, 2011, pp 156-159. (28) Sartori, E. R.; Mansano, G. R.; Eisele, A. P. P. Anal. Methods 2014. (29) Channon, R. B.; Newland, J. C.; Bristow, A. W. T.; Ray, A. D.; Macpherson, J. V. Electroanal. 2013, 25, 2613-2619. (30) Toghill, K. E.; Compton, R. G. Electroanal. 2010, 22, 1947-1956. (31) Zen, J. M.; Hsu, C. T.; Kumar, A. S.; Lyuu, H. J.; Lin, K. Y. Analyst 2004, 129, 841-845. (32) Li, X.-H.; Xie, Z.-H.; Min, H.; Xian, Y.-Z.; Jin, L.-T. Electroanal. 2007, 19, 2551-2557. (33) de Oliveira, P. R.; Oliveira, M. M.; Zarbin, A. J. G.; Marcolino, L. H.; Bergamini, M. F. Sens. Actuator B-Chem. 2012, 171, 795-802. (34) Wiaderek, K. M.; Cox, J. A. Electrochim. Acta 2011, 56, 3537-3542. (35) Chakraborty, S.; Raj, C. R. Biosens. Bioelectron. 2009, 24, 3264-3268. (36) Norouzi, P.; Faridbod, F.; Larijani, B.; Ganjali, M. R. Int. J. Electrochem. Sci. 2010, 5, 1213-1224. (37) Hu, J.; Wisetsuwannaphum, S.; Foord, J. S. Faraday Discussions 2014, 172, 457-472. (38) Zhang, Y.; Suryanarayanan, V.; Nakazawa, I.; Yoshihara, S.; Shirakashi, T. Electrochim. Acta 2004, 49, 5235-5240. (39) Růžička, J.; Hansen, E. H. Anal. Chem. 2000, 72, 212A-217A. (40) de Castro, M. D. L.; Valcárcel, M. J. Pharm. Biomed. Anal. 1989, 7, 1291-1300.

ACS Paragon Plus Environment 21

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 23

(41) Tzanavaras, P. D.; Themelis, D. G. Anal. Chim. Acta 2007, 588, 1-9. (42) Trojanowicz, M. Anal. Chim. Acta 2009, 653, 36-58. (43) Felix, F. S.; Angnes, L. J. Pharm. Sci. 2010, 99, 4784-4804. (44) Hartwell, S. K.; Kehling, A.; Lapanantnoppakhun, S.; Grudpan, K. Anal. Lett. 2013, 46, 1640-1671. (45) Sansuk, S.; Bitziou, E.; Joseph, M. B.; Covington, J. A.; Boutelle, M. G.; Unwin, P. R.; Macpherson, J. V. Anal. Chem. 2013, 85, 163-169. (46) Lemay, S. G.; van den Broek, D. M.; Storm, A. J.; Krapf, D.; Smeets, R. M. M.; Heering, H. A.; Dekker, C. Anal. Chem. 2005, 77, 1911-1915. (47) Snowden, M. E.; King, P. H.; Covington, J. A.; Macpherson, J. V.; Unwin, P. R. Anal. Chem. 2010, 82, 3124-3131. (48) Bitziou, E.; Joseph, M. B.; Read, T. L.; Palmer, N.; Mollart, T.; Newton, M. E.; Macpherson, J. V. Anal. Chem. 2014, 86, 10834-10840. (49) Joseph, M. B.; Bitziou, E.; Read, T. L.; Meng, L.; Palmer, N. L.; Mollart, T. P.; Newton, M. E.; Macpherson, J. V. Anal. Chem. 2014, 86, 5238-5244. (50) Diafilm EA: Enabling new electroanalytical applications. www.e6.com/sensors [accessed 23/06/14]. (51) Pickles, C. S. J. Diamond and Related Materials 2002, 11, 1913-1922. (52) Dumitrescu, I.; Yancey, D. F.; Crooks, R. M. Lab Chip 2012, 12, 986-993. (53) Wilson, N. R.; Clewes, S. L.; Newton, M. E.; Unwin, P. R.; Macpherson, J. V. J. Phys. Chem. B 2006, 110, 5639-5646. (54) Bitziou, E.; Rudd, N. C.; Unwin, P. R. Journal of Electroanalytical Chemistry 2007, 602, 263-274. (55) Bae, A. J.; Beta, C.; Bodenschatz, E. Lab Chip 2009, 9, 3059-3065. (56) Růžička, J.; Hansen, E. H. Anal. Chim. Acta 1980, 114, 19-44. (57) Li, Y.; Chen, S. M. Int. J. Electrochem. Sci. 2012, 7, 2175-2187. (58) Kleijn, S. E. F.; Lai, S. C. S.; Koper, M. T. M.; Unwin, P. R. Angew. Chem., Int. Ed. 2014, 53, 3558-3586. (59) Wang, J.; Swain, G. M. J. Electrochem. Soc. 2003, 150, E24-E32. (60) Biegler, T.; Rand, D. A. J.; Woods, R. Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 1971, 29, 269-277. (61) Hu, J.; Lu, X.; Foord, J. S. Electrochem. Commun. 2010, 12, 676-679. (62) Janssen, G.; van Enckevort, W. J. P.; Vollenberg, W.; Giling, L. J. Diamond Relat. Mater. 1992, 1, 789-800. (63) Aldous, L.; Compton, R. G. Phys Chem Chem Phys 2011, 13, 5279-5287. (64) Dias, F. X.; Jaselskis, B. Analyst 1983, 108, 76-80. (65) Jannakoudakis, A. D.; Kokkinidis, G. J. Electroanal. Chem. Interfacial Electrochem. 1982, 134, 311-324. (66) Nishihara, C.; Raspini, I. A.; Kondoh, H.; Shindo, H.; Kaise, M.; Nozoye, H. J. Electroanal. Chem. 1992, 338, 299-316. (67) Bard, A. J. Anal. Chem. 1963, 35, 1602-1607. (68) Miller, J. N.; Miller, J. C. Statistics and Chemometrics for Analytical Chemistry, 5th ed.; Pearson Education Limited: London, 2005.

ACS Paragon Plus Environment 22

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Analytical Chemistry

million × excess acetaminophen

Eapplied = +0.1 V versus Ag/AgCl

Pt nanoparticle modified boron doped diamond flow electrodes select and detect hydrazine in the presence of 106 excess electroactive drug at the levels stipulated by the pharmaceutical industry ACS Paragon Plus Environment