Electrochimica Acta 130 (2014) 577582

Contents lists available at ScienceDirect

Electrochimica Acta

j ourna l ho me page: www.elsev ier .com/ locate /e lec tac ta

An elec oly

Reza B. MDepartment of

a r t i c l

Article history:Received 29 JaReceived in reAccepted 12 MAvailable onlin

Keywords:PolycarbazoleImpedanceIon transportElectron transVoltammetry

l impolycaixturn, whn trarimen

of th technpoten

1. Introdu

Polycarbazoles [14] have been attracting increasing atten-tion for a variety of applications including electronic [5] andelectrochromic [6] devices, solar cells [7], sensors [3,810] electro-catalysis [11,12], charge storage [1316], and ion extraction [17].However, the properties and electrochemistry of this class of poly-mers has repolymers su

This woports and celectrochemacid oxidatthe indirec(mainly adscomplete oissue repormatically dare used [1inary impethe turn-onelectrolyte.

The in sithas been in

CorresponE-mail add

a nand Sarac [20] reported impedance data for polycarbazole-coatedcarbon bers in propylene carbonate and acetonitrile with LiClO4and Et4NClO4 electrolytes. The results were analyzed by tting tothe circuit shown in Fig. 1, where Ru is the solution resistance, Cdl isthe double layer capacitance, R1 is electrolyte resistance (it is notclear to us how this is distinct from Ru), CPE is a constant phase ele-

http://dx.doi.o0013-4686/ ceived limited attention relative to other conductingch as polypyrrole and polyaniline.rk was prompted by reports that polycarbazole sup-oatings exhibit remarkable synergy with Pt for theical oxidation of formic acid [11,18,19]. While formic

ion at pure Pt is known to proceed primarily throught pathway with production of blocking intermediatesorbed CO), addition of PCZ diverts the reaction towardsxidation of FA to CO2 via a direct pathway. One majorted for PCZ is its intrinsic high resistance, which dra-ecreases the FA oxidation activity when thicker lms8]. The purpose of this work is to extend the prelim-dance study reported in [18], in order to investigate

of the PCZ conductivity and the inuence of the

u conductivity and capacitive behavior of polycarbazolevestigated by electrochemical impedance spectroscopy

ding author. Tel.: +1 709 864 8657; fax: +1 709 864 3702.ress: ppickup@mun.ca (P.G. Pickup).

ment representing the capacitance at the lm|electrolyte interface,R2 is a charge transfer resistance, W is the Warburg impedance ofthe polymer, CCF is the capacitance of the carbon ber electrode, andRCF is its resistance. It was found that the electrolyte/solvent combi-nation used in the electrochemical deposition of the polymers, thedeposition method, and the electrolyte/solvent combination usedin the EIS all inuenced the properties of the polycarbazole lay-ers. This methodology was also applied to a range of polycarbazolederivatives [15].

Gupta et al. [21] found that amorphous polycarbazole lmsprepared in acetonitrile containing Bu4NClO4 exhibited faster ionexchange and high doping levels than more ordered lms pre-pared with Bu4NBF4. Near-Ohmic impedance behaviour has beenreported for polycarbazole lms in KCl(aq), which is useful for bio-electric sensing [8].

Here we present voltammetric and EIS results obtained in var-ious combinations of aqueous H2SO4 and KNO3 for glassy carbonelectrodes galvanostatically coated with polycarbazole. H2SO4 wasused here because it was the electrolyte used for our electrocat-alytic studies [18]. Comparing its effect with that of KNO3 providesinformation on possible pH effects.

rg/10.1016/j.electacta.2014.03.0592014 Elsevier Ltd. All rights reserved.trochemical impedance study of thin p

oghaddam, Peter G. Pickup

Chemistry, Memorial University, St. Johns, Newfoundland, Canada A1B 3X7

e i n f o

nuary 2014vised form 12 March 2014arch 2014e 22 March 2014

port

a b s t r a c t

A voltammetric and electrochemicacoated with approximately 23 nm of paqueous H2SO4 and KNO3 and their mthe electrolyte type and concentratioThis was shown to be due to strong iothe small amplitude impedance expepolycarbazole layer. Thus combinationpolycarbazole electrochemistry. Boththe range of electroactivity to lower H2SO4 were present.

ction (EIS) incarbazole lms

edance study of glassy carbon electrodes galvanostaticallyrbazole (GC/PCZ) was conducted in various concentrations ofes. Cyclic voltammetry showed strong dependencies on bothile the impedance characteristics were inuenced much less.nsport effects in voltammetry which became insignicant ints, which were dominated by electron transport though thee two techniques provided a more complete description of theiques provided evidence of cation insertion which extendedtials at high KNO3 concentrations and when both KNO3 and

2014 Elsevier Ltd. All rights reserved.

umber of previous studies [8,11,13,16,18,20,21]. Ates

578 R.B. Moghaddam, P.G. Pickup / Electrochimica Acta 130 (2014) 577582

Fig. 1. Equivalfor electrocoat

2. Experim

2.1. Chemic

Carbazopotassium reagent, 9(Fluka; eleused as rectemperatur15 min. Dei

2.2. Workin

Glassy cpolished wiuse. Polycadeposited ammoniumpolymerizabazole eleclm (mPCZ)efciency a

mPCZ =QPo

Where Qmolar mas(96485 C mmerization mass of PCZca. 23 nm bawas prepar

2.3. Instrum

An EG&through M2measuremewire formeimpedance was also usmercial softperformed of 10 mV. Bheld for 2 mequilibriumformed witvoltammogat +0.1 V bewere very s

yclic voltammograms at 100 mV s1 for GC/PCZ electrodes in various con-ons of KNO3.

ults and discussion

clic voltammetry

2 shows cyclic voltammograms of GC/PCZ electrodes in var-ncenmog

wasn anitionper pe du7 V atratied reina

an bctionic petenttratie 1 Mnce ange

3 coH2SOent circuit used by Ates and Sarac [20] for modeling impedance resultsed polycarbazole lms on carbon ber microelectrodes.

ental

als

le (AlfaAesar; 95%), sulfuric acid (Fisher Scientic),nitrate (BDH), dichloromethane (Sigma Aldrich; ACS9.9%), tetrabutylammonium hexauorophosphatectrochemical grade, 99.0%), (Alfa Aesar; 95%) wereeived. All measurements were recorded at ambiente under a nitrogen atmosphere following purging foronized water was used throughout the experiments.

g electrode preparation

arbon electrodes (GC; CH Instruments; 0.071 cm2) wereth 0.05 m alumina and rinsed well with water beforerbazole lms were galvanostatically (0.28 mA cm2)from dichloromethane containing 0.1 M tetrabutyl-

hexauorophosphate and 0.01 M carbazole using ation charge density of 28.2 mC cm2. These polycar-trodes are designated as GC/PCZ. The mass of the PCZ

can be estimated by assuming 100% polymerizationnd using eq. 1 [11].

lymMPCZ

F.Z(1)

Polym, MPCZ, F, and Z are polymerization charge, thes of carbazole (167.2 g mol1), the Faraday constantol1), and number of electrons associated with poly-of one monomer (2.3) [11], respectively. The calculated

was 212 ng, which correspondes to a lm thickness ofsed on an assumed density of ca. 1.3 g cm3. A new lmed for each electrolyte employed.

entation and impedance tting

G Model 273A Potentiostat/Galvanostat run by a PC70 commercial software was used for voltammetricnts. A saturated calomel electrode (SCE) and a platinumd the reference and counter electrode, respectively. Formeasurements an EG&G Model 5210 Lock-in Ampliered and the system was run through Power-Suite com-

Fig. 2. Ccentrati

3. Res

3.1. Cy

Fig.ious covoltamtrationpositioIn addthe upmust bca. 0.4concenincreastor domeffect c(see secathodnan poconcenover thresistathe cha

Fig.KNO3, ware. Electrochemical impedance measurements wereover the range of 10 kHz to 0.1 Hz using an ac amplitudeefore any impedance measurement, the electrode wasin at the dc offset potential to reach electrochemical. Data analysis and equivalent circuit tting was per-h ZView2 software (Scribner Associates Inc.). All cyclicrams shown are for the 2nd cycle, with the scan stoppedtween the 1st and 2nd cycles. The CVs for the two cyclesimilar in all cases.

Fig. 3. Cyclic centrations oftrations of KNO3(aq). Although the general form of therams did not change greatly as the electrolyte concen-

decreased, there were clear changes in the anodic peakd height that can be attributed to increased resistance., the speed of the current response to scan reversal atotential limit reveals an increase in time constant thate to resistance effects. In contrast, the cathodic peak atctually shifted to more positive potentials as the KNO3on was decreased. This is opposite to the effect thatsistance would produce, indicating that some other fac-ted changes in the position of this peak. The resistancee attributed largely to changes in the solution resistance

3.4; Table 1). The positions of both the anodic andaks would also be inuenced by changes in the Don-ial [22], which would increase with decreasing KNO3on, resulting in ca. 77 mV positive shifts for both peaks

to 0.05 M range. The combined effects of the solutionnd Donnan potential can therefore adequately explains with KNO3 concentration seen in Fig. 2.mpares voltammograms of GC/PCZ lms in solutions of4 and mixtures of these two electrolytes. Addition ofvoltammograms at 100 mV s1 for GC/PCZ electrodes in various con- KNO3 and H2SO4 and some of their combinations.

R.B. Moghaddam, P.G. Pickup / Electrochimica Acta 130 (2014) 577582 579

Table 1Selected impedance modeling results for the GC/PCZ electrodes.

Ru/a RW(0.4 V)/k RW(0.6V)/k RW(0.8 V)/

0.5 M H2SO4 18 1 12 0.42 261 M KNO3 + 0 0.64 101 M KNO3 + 0 0.62 101 M KNO3 0.25 770.1 M KNO3 0.42 190.5 M KNO3 0.23 870.1 M H2SO4 0.45 440.1 M KNO3 0.26 80

a. Average and

H2SO4 to tha slightly lopeak at 0.4height. The0.45 to 0.62ent with Hto the decrthe changeare suggestIn particulacathodic shwhile the anproton and carbazole v

The effeprising, par0.1 M KNO3tial seen foresistance etive than elikely that rpolymer maobtained byincreased ththe cathodi

3.2. Impeda

Fig. 4 disGC/PCZ elechigh frequepensated sowhile no feadetected attative Bodeseen at high1 kHz, whicthe GC subs[26], there ica. 1000 toaccompanylm. This is

3.3. Modeli

In modegave acceptclearly desystem weof PCZ coacant contriimpedance was attribu

ng electron microscopy [18]. Here we therefore begin withing of the impedance of a bare GC electrode in 0.5 M H2SO4.as found that the equivalent circuit shown in Fig. 5A pro-ood ts to the impedance of the GC electrode at all potentials. Here Ru is the uncompensated solution resistance, Cdl is

uble layer capacitance of the GC electrode, and the short cir-arburg element (WS) represents a background current. The

of this background process is unknown. It has been mod-ere as a short circuit Warburg impedance because it was theement found to produce an acceptable t and is appropriateel a dc background current. An open Warburg (WO) did notch good ts, suggesting that the impedance is indeed due tockground current, and a resistor (see below and circuit 5 C)t provide acceptable ts..1 M H2SO4 20 4 19

.5 M H2SO4 22 10 12 29 14 44

+ 0.1 M H2SO4 37 4 13 46 20 26 49 2 15

143 24 21 standard deviation of values at 0.0, 0.2, 0.4, 0.6, and 0.8 V.

e KNO3 electrolyte solution shifted the anodic peak tower potential and decreased its height. The cathodic7 V also shifted to lower potential, but increased inre was also a slight increase in anodic current over the

V range (shoulder at ca. 0.55 V) that was also appar-2SO4 alone. The peak shifts can be attributed largelyease in solution resistance (see section 3.4). However,s in peak heights and the increased shoulder at 0.55 Vive of some minor cation (proton) involvement [2325].r, the increase in the cathodic peak and its signicantift are suggestive of proton insertion during reduction,odic shoulder could be due to proton expulsion. Mixedanion transport has previously been reported for poly-oltammetry in 9 M HClO4(aq) [24].cts of 0.1 M H2SO4 alone seen in Fig. 3 are quite sur-ticularly the overall decrease in current compared toand shift of the anodic potential to the highest poten-r any of the electrolytes. This is clearly not a solutionffect since the 0.1 M H2SO4 solution was more conduc-ither the 0.05 or 0.1 M KNO3 solution. It is thereforeestricted mobility of HSO4 and/or SO42 ions in thetrix was responsible. Evidence that this is the case was

increasing the H2SO4 concentration to 0.5 M, whiche anodic peak current considerably and also increasedc currents, indicating a higher level of anion insertion.

nce spectroscopy

plays Nyquist (4A) and series capacitance (4B) plots for atrode in 0.5 M H2SO4 at various dc-offset potentials. Thency region for all the plots is dominated by the uncom-lution resistance, Ru, and double later capacitance, Cdl,ture attributable to a charge transfer resistance can be

any potential. This is further evident in the represen- plot (Fig. 4B, inset) at +0.4 V. Capacitive behavior was

frequencies in Bode plots, typically in the range of 10 toh we attribute to charging the electrical double layer oftrate [18]. Typical for porous conducting polymer lmss a Warburg-type impedance at midrange frequencies,

10 Hz, indicative of signicant redox activity and theing ionic and electronic transport within the polymer

followed by capacitive behavior at lower frequencies.

ng of the impedance data

scannimodel

It wvided gstudiedthe docuit Wnatureelled honly elto modgive sua dc badid noling of the impedance data, the simplest circuits thatable ts were used, and only elements that representned physical parameters for the GC|polymer|solutionre employed. A preliminary study of the impedanceted GC electrodes showed that there was a signi-bution from the underlying GC surface, and the PCZappeared in parallel with the GC impedance [18]. Thisted to the high porosity of the PCZ lms observed by

Fig. 4. NyquisH2SO4 at vario0.4 V.t (A) and series capacitance (B) plots of a GC/PCZ electrode in 0.5 Mus dc-offset potentials. Insets show Nyquist (A) and Bode (B) plots at

580 R.B. Moghaddam, P.G. Pickup / Electrochimica Acta 130 (2014) 577582

Fig.

Values odid not vardeviations respectivelyboth the reseffective difcient) decreHowever, th4% to 39%),nicant. Coof RW = 2.5 are given he

For the Gtrode (Fig. 5in 0.1 M H2ever, in 0.5be modeled0.73, respecresistance (impedance KNO3 and 1an Rct (4.7 kas a CPE (CPEs were uas Cdl = (TW

The needdata at 0 V sutions to thelectrolyte capacitanceapparent frothe variationot vary sigeight electrative to thethe partial proposed [1the GC/PCZwere no coand standaRW = 2.9 2distinguishing that thenot be reso

Impedanusing the saonly small c

ode pd 0.1

ce oons int re0.4 Vs chanse

cleae da

ct omapacancets inent w

all c) wa

Fig. 5ivelyd siRu. Ct (Ccand be

modtainveral cases at 0.6 V (0.1 M, 0.5 M, and 1 M KNO3) replace-f the CPE with a resistor (Rdc; see Fig. 5 C) was required. Rdcmed to represent the effect of a very small dc backgroundt. This is consistent with Bode plots (see below), where a 5. Equivalent circuits used for modeling impedance data.

f Ru and Cdl obtained from equivalent circuit ttingy with potential, with average values and standardof 21.1 0.5 and 0.50 0.01 F being obtained,. The WS did show some potential dependence, withistance (RW) and T parameter (TW = L2/D, where L is thefusion thickness, and D is the effective diffusion coef-asing as the potential was raised, and then decreasing.ese parameters had large uncertainties (ranging from

and so it is not clear whether these changes are sig-nsequently, only the averages and standard deviations,

1.2 M, TW = 14 7 s, and the power ( = 0.66 0.02)re.C/PCZ electrodes, the model used for the bare GC elec-A) provided adequate ts to the impedance data at 0 VSO4, 0.1 M KNO3, and 1 M H2SO4 + 0.1 M KNO3. How-

M H2SO4 and 0.1 M H2SO4 + 0.1 M KNO3, Cdl had to as a constant phase element of power () 0.81 andtively. In 0.5 M H2SO4 + 0.1 M KNO3, a charge transferRct) of 120 was required in series with the Warburg(Cdl was modeled as a pure capacitor), while in 0.5 M

M KNO3 an open Warburg element was required with and 1.6 k, respectively) in series and Cdl modeled

= 0.78 and 0.80, respectively) (Fig. 5B). In cases wheresed, values of Cdl were estimated by the Brug method

/Ru(-1))1/ [27]. for a variety of different circuits to model the GC/PCZuggests that there are unresolved (see below) contrib-e impedance from the PCZ, and that these vary with theemployed. This is supported by increases in the series, which are discussed in section 3.4, but is not readilym the tting parameters (see below). However, despitens in the modeling circuits employed, values of C did

Fig. 6. BKNO3 an

evidenvariatiPertine

At trolytea respoThis is0.4 V thwith Rlayer cimpedelemenconsist[18]. Into 0.48cuit inextensrequirefor by elemena signirequire

Thedata oband sement ois assucurrendl

nicantly. The average and standard deviation for theolyte solutions were 0.23 0.06 F. The decrease rel-

value for bare GC (0.50 0.01 F) is consistent withblocking of the GC surface by PCZ that was previously8]. As for the bare GC electrode, RW and TW values for

electrodes at 0 V exhibited large uncertainties. Thererrelations with the nature of the electrolyte. Averagesrd deviations for the eight electrolyte solutions were.2 M, TW = 14 6 s, and = 0.73 0.05. These are notable from the values for the bare GC electrode, indicat-

contributions of the PCZ to the impedance at 0 V canlved from the background currents at the GC surface.ce data for GC/PCZ electrodes at 0.2 V were modeledme circuits as those used for the 0 V data, and there werehanges in parameters in most cases. As at 0 V, there was

clear resist( below 0Fig. 5 C becathe effect oelement didthe t parafor Cdl.

Bode ploare exhibiteInspection 1 Hz), wherior started shows evidior. Such lo+0.8 V in 0.1lots (phase shift as a function of frequency) at +0.6 and +0.8 V in 0.1 MM H2SO4.

f contributions to the impedance by the PCZ from then models required and increases in series capacitance.sults are presented and discussed in section 3.4., the impedance of the GC/PCZ electrodes in all elec-nged notably from essentially the bare GC response todominated by the Warburg impedance of the PCZ layer.r from observation of the data in Fig. 4. In all cases atta could be adequately modeled by the circuit in Fig. 5Bitted (negligible). Here, the CPE represents the double

itance (Cdl) of the GC electrode, while WO is the Warburg (open) of the polymer lm. Placing the CPE and Warburg

parallel was required to obtain acceptable ts, and isith the highly porous appearance of the polymer lms

ases, a non-ideal Warburg element ( ranged from 0.39s required to provide acceptable ts to the data. The cir-B is a simplication of that in Fig. 1 that has been used

by Ates and coworkers [16]. From that circuit, R1 is notnce the electrolyte resistance is adequately accounteddl and CPE then become a single unresolved capacitivePE). R2 is not required because there is no evidence oft charge transfer resistance, while CCF and RCF are notcause we did not use carbon ber electrodes.el in Fig. 5B (with Rct omitted) also worked well for

at 0.6 V in most electrolytes, while for all cases at 0.8 Vive behavior at the lowest frequencies can be detected.5 Hz). The CPE representing Cdl has been omitted fromuse the polymer capacitance is so large that it swampsf Cdl. Including a CPE or Cdl in parallel with the (RdcWO)

not signicantly increase the quality of the t, changemeters, or provide an acceptable value or uncertainly

ts at +0.6 and +0.8 V in 0.1 M H2SO4 and 0.1 M KNO3d in Fig. 6 in order to further explore the effects of Rdc.of these plots in 0.1 M KNO3 at low frequencies (belowe the phase shift of ca. 80 indicative of capacitive behav-to decrease signicantly with decrease in frequency,ence of a transition from capacitive to resistive behav-w frequency resistive behavior is seen at both +0.6 and

M KNO3, while it only appeared at +0.8 V in 0.1 H2SO4.

R.B. Moghaddam, P.G. Pickup / Electrochimica Acta 130 (2014) 577582 581

Fig. 7. Log(RW

3.4. Analysi

Table 1 sin all eight oRW values adominated so do not pr(however, ttrolyte weranomalouslKNO3, and deviations high valuesdata, they anicant. ThRu in Table uncertaintie

The Warwith increato be any detion (as reeeither the ioof the two to decreaseof charge caaccording tohere, the sain resistancconcentratithe concentever, deviatconcentrati

In this csolution haplots all exwith slopesobserved fothere is a sulevel off in would be exHowever, thhigh in manare likely tovalues maybecause the

eries cCZ ele

d 0.2y condica

and exper wielectctroe wentia

condions

carric res

a hiobseis thapedaheorrbure are not aware of a procedure to do this for values of.5. Therefore, in order to compare the capacitances of thems as functions of potential and electrolyte, we use seriestances obtained at the lowest frequency employed (C ).) vs. potential plots for GC/PCZ electrodes in various electrolytes.

s and discussion of the impedance parameters

hows Ru and RW values obtained for GC/PCZ electrodesf the electrolyte solutions used for the impedance work.t potentials below 0.4 V are not given because they areby the background impedance of the GC electrode andovide useful quantitative information on the PCZ layerhey are plotted in Fig. 7). Ru values obtained in each elec-e independent of potential, although there were somey high values at 0 V in 0.1 M KNO3, 0.5 M KNO3, 1 M1 M KNO3 + 0.5 M H2SO4, which led to large standardof the average values reported in Table 1. Since these

resulted from the lowest quality ts to the impedancere assumed to be due to tting errors, and not to be sig-e electrolytes have been arranged in order of increasing1, and this is compatible with the expected order if thes are considered.burg resistance (RW) values in Table 1 decrease greatlysing potential for all electrolytes. There does not appearpendence on the conductivity of the electrolytes solu-cted by Ru). These Warburg resistances could representnic or electronic resistance of the PCZ, or a combination[28]. For a permselective polymer, both are expected

exponentially at low potentials as the concentrationrriers increases with oxidation of the polymer chains

the Nernst equation. At higher potentials, not reachedturation of oxidizable sites would cause the decreasee to level off. For a permselective polymer lm, the

Fig. 8. Sfor GC/P

at 0 anroughlmay inat 0.2 V

Thetogethof the the eleIf therexponelectiveof the chargean ionlead tois not analysthe im

In tthe Waever, w /= 0PCZ lcapacion of charge compensating anions would be equal toration of electronic charge carriers (holes) [29]. How-ions are likely at low doping levels and high electrolyteon which would cause a breakdown of permselectivity.ontext, the resistances for the lms in each electrolyteve been plotted as log(RW) vs. potential in Fig. 7. Thesehibit an approximately linear region from 0.2 to 0.8 V,

close to 120 mV/decade. This is within the typical ranger p-doping of a conjugated polymer [30,31]. At 0.8 Vggestion that the decrease in resistance is beginning tosome cases, while the resistances at 0 V are lower thanpected from extrapolation from the 0.2 to 0.8 V region.e uncertainly in the RW values at 0 and 0.2 V was veryy cases, and so inferences drawn from data in this region

be unreliable. In addition, as mentioned above, these not accurately represent the resistance of the PCZ layer

impedance is dominated by the GC background current

Fig. 8 show(Clow = -1/in capacitancarrier conc

At 0 V th3.5 F in 0lower valuebare glassythese casesever, the hig1 M KNO3) due to the Fig. 3.

A small the potentifrom 0.2 tocides with apacitances calculated from the impedance at 0.1 Hz (Clim = 1/Z)ctrodes at various potentials in various electrolytes.

V. They are only included in Fig. 7 to show that they aresistent with the trend over the 0.4 to 0.8 V region. Thiste that the PCZ layer does contribute to the impedance

possibly at 0 V.onential dependences of RW values shown in Fig. 7,th the lack of any signicant dependence on the naturerolyte, provide convincing evidence that RW representsnic resistance of the PCZ, and not its ionic resistance.re signicant contribution from ion transport, such anl dependence would only be expected under permse-itions, which should break-down as the concentrationin the supporting electrolyte was increased, and at lowier levels (low potentials). Furthermore, the effects ofistance, in addition to the electronic resistance, wouldgh frequency offset of the PCZ impedance [28] whichrved in any of the data. It can be concluded from thist ion transport in the PCZ is too fast to be resolved fromnce data.y, capacitances of the PCZ lms could be obtained fromg parameters from tting of the impedance data. How-lows capacitances calculate from the impedance at 0.1 HzZ) for all electrolytes. These data show the sharp risece that is expected for a Nernstian increase in chargeentration.e capacitances of the PCZ coated electrodes ranged from.1 M KNO3 to 10.3 F in 1 M KNO3 + 0.5 M H2SO4. Thes are similar to the low frequency capacitance of the

carbon (e.g. 3.8 F in 0.5 M H2SO4), indicating that in the PCZ exhibited negligible electroactivity at 0 V. How-her values in the mixed KNO3 + H2SO4 electrolytes (andsuggest that there is residual PCZ redox. This could becation insertion suggested by the voltammograms in

increase in capacitance was observed in all cases whenal was increased to 0.2 V followed by larger increases

0.8 V. This turn-on of the PCZ electroactivity coin-the sharp decreases in RW that accompany oxidation

582 R.B. Moghaddam, P.G. Pickup / Electrochimica Acta 130 (2014) 577582

of the lm. The differences between the various electrolytes can beattributed in part to cation insertion, which increases the capaci-tance at lower potentials at the expense of anion insertion at higherpotentials. For example, this is seen for the KNO3 series as the con-centration wcomplex, wbination wipotentials. and/or bettmore sites t

4. Conclus

Variatiomixtures) avoltammetrelectrodes, Voltammetcontaining edation (e.g.H2SO4), whfor 1 M KNOAs for the cto +0.43 anddifferencespopulationsthe Donnanthe solution

These prtrodes (i.e. data, wherpolycarbazopotential. Tcarbazole rfunctions obazole resisnot signicin the polycdata.

From thport, the mpotentials. Fappears at (DFAFC) wothan this. Pto be activeSCE, althouof the polymin the electrin this regipolycarbazomay arise fbile anionssupports.

Acknowled

This woring Researc

References

[1] J.F. MorinMacromo

[2] N. BlouinAccounts

[3] M. Ates, A.S. Sarac, Conducting polymer coated carbon surfaces and biosensorapplications, Progress in Organic Coatings 66 (2009) 337.

[4] J.P. Lellouche, R.R. Koner, S. Ghosh, N-Substituted carbazole heterocycles andderivatives as multipurpose chemical species: at the interface of chemical engi-neering, polymer and materials science, Reviews in Chemical Engineering 29

13) 413..T. Bouymer . SappctrochancedLi, Aices, C. Kayiting pymer-es 4 (2. Forceura, K

PEDOterialsAtes, rs, Ma(2013. Zho

electrernati. Jiangy[polyforma12) 56araswole baBaibary(N-vlicatioogy 67Ates, edan

ctrodeAtes, Rnt elecanic CFedorceous mical. Mogporte. MogycarbaAtes,ylcarbrnal oupta

ycarbaaegeltratioorpora86) 62. Dufhysicnzelt,polymface, Jaudseerminfonate

Rubinights f9.arsount, an. Albe

elect90) 15en, Ppositao, P

y-[1-mmediahe Amfer, Righlys of h9.as increased. However, the effects of H2SO4 are moreith a higher concentration (i.e. 0.5 M H2SO4) and com-th KNO3 leading to generally higher capacitances at allThis could be due to a break-down of permselectivityer hydration of the polymer layer, which could allowo become electroactive.

ions

n of the aqueous electrolyte nature (KNO3, H2SO4 andnd concentration produced notable differences in theic characteristics of polycarbazole coated glassy carbonbut surprisingly small variations in their impedance.ric observations (i.e. Figs 2 and 3) showed that KNO3lectrolytes typically gave lower peak potentials for oxi-

0.73 V for 0.5 M KNO3 compared with 0.78 V for 0.5 Mile addition of H2SO4 led to even lower values (e.g. 0.70 V3 containing 0.5 M H2SO4 vs. 0.72 V for 1 M KNO3 alone).athodic scan, the peak at +0.47 V for 1 M KNO3 shifted

+0.38 V when 0.1 and 0.5 M H2SO4 were added. These have been attributed to differences in ion mobilities and

within the polycarbazole lm, including the effects of potential and cation insertion, as well as the effects of

resistance.onounced effects on the dynamic response of the elec-voltammetry) are relatively minor in the impedancee signicant changes in the bulk composition of thele lm are absent during experiments at a xed dche solution resistance is easily resolved, and the poly-esistance and limiting capacitance were obtained asf potential and the electrolyte employed. The polycar-tance is due to restricted electron transport, and so isantly inuenced by the electrolyte, while ion transportarbazole is too fast to be resolved from the impedance

e perspective of using polycarbazole as a catalyst sup-ost pertinent results here are its properties at lowor formic acid oxidation the peak voltammetric currentca. 0.2 V vs. SCE [18], and direct formic acid fuel cellsuld generally be operated at anode potentials lowerolycarbazole supported Pt has previously been shown

for formic acid oxidation at potentials as low as -0.1 V vs.gh currents can be severely limited by the conductivityer at low potentials [18]. The observation that changesolyte can inuence the electroactivity of polycarbazoleon could provide a means to increase the activity ofle supported Pt in DFAFC. Since the electrolyte effectsrom cation insertion, polycarbazole with large immo-

or self-doped polycarbazoles could be more effective

gements

k was supported by the Natural Sciences and Engineer-h Council of Canada and Memorial University.

, M. Leclerc, D. Ades, A. Siove, Polycarbazoles: 25 years of progress,lecular Rapid Communications 26 (2005) 761., M. Leclerc, Poly(2,7-carbazole)s. Structure-property relationships,

of Chemical Research 41 (2008) 1110.

(20[5] P.L

Pol[6] S.A

eleAdv

[7] J.L. dev

[8] Y.Pducpolfac

[9] P.AnamforMa

[10] M. me33

[11] W.QforInt

[12] F.Xpolper(20

[13] R. Sbaz

[14] M. Polappnol

[15] M. Impele

[16] M. aleOrg

[17] A. aquche

[18] R.Bsup

[19] R.Bpol

[20] M. vinJou

[21] B. Gpol

[22] R. Nceninc(19

[23] G.Lof P

[24] G. Itrosur

[25] T. Rdetsul

[26] J.F.ins333

[27] E. Bme

[28] W.Jed(19

[29] X. Rcom

[30] H. Mpolby of t

[31] D. Oof hdow786dreault, S. Beaupre, M. Leclerc, Polycarbazoles for plastic electronics,Chemistry 1 (2010) 127., G.A. Sotzing, J.L. Reddinger, J.R. Reynolds, Rapid switching solid stateromic devices based on complementary conducting polymer lms,

Materials 8 (1996) 808..C. Grimsdale, Carbazole-based polymers for organic photovoltaichemical Society Reviews 39 (2010) 2399.namura, J.H. Roberts, J.F. Rubinson, Near-Ohmic behavior for con-olymers: Extension beyond PEDOT on gold-plated platinum to othercounterion/substrate combinations, ACS Applied Materials & Inter-012) 1601.lli, C.T. Sweeney, A.D. Kammerich, B.C.W. Lee, L.H. Rubinson, Y.P. Kayi-. Gale, J.F. Rubinson, Histocompatibility and in vivo signal throughput

T, PEDOP, P3MT, and polycarbazole electrodes, Journal of Biomedical Research Part A 100A (2012) 3455.A review study of (bio)sensor systems based on conducting poly-terials Science & Engineering C-Materials for Biological Applications) 1853.u, J.K. Xu, Y.K. Du, P. Yang, Polycarbazole as an efcient promoterocatalytic oxidation of formic acid on Pt and Pt-Ru nanoparticles,onal Journal of Hydrogen Energy 36 (2011) 1903., F.F. Ren, W.Q. Zhou, Y.K. Du, J.K. Xu, P. Yang, C.Y. Wang, Free-standing(N-vinyl carbazole)]-supported Pt-based catalysts with enhancednce for methanol electro-oxidation in alkaline medium, Fuel 1020.athi, M. Gerard, B.D. Malhotra, Characteristics of aqueous polycar-tteries, Journal of Applied Polymer Science 74 (1999) 145.ac, M. Lira-Cantu, J.O. Sol, I. Baltog, N. Casan-Pastor, P. Gomez-Romero,inyl carbazole) and carbon nanotubes based composites and theirn to rechargeable lithium batteries, Composites Science and Tech-

(2007) 2556.A Comparative Study of Redox Parameters and Electrochemicalce Spectroscopy of Polycarbazole Derivatives on Carbon Fiber Micro-, Fibers and Polymers 11 (2010) 1094.eview study of electrochemical impedance spectroscopy and equiv-trical circuits of conducting polymers on carbon surfaces, Progress inoatings 71 (2011) 1.zyk, M. Skomska, Quantitative studies of silver ions removal fromsolutions by poly(1,8-diaminocarbazole) lms by means of electro-

quartz crystal microbalance, Electrochimica Acta 99 (2013) 62.haddam, P.G. Pickup, Oxidation of formic acid at polycarbazole-d Pt nanoparticle, Electrochimica Acta 97 (2013) 326.haddam, P.G. Pickup, Mechanistic studies of formic acid oxidation atzole supported Pt nanoparticles, Electrochimica Acta 111 (2013) 823.

A.S. Sarac, Capacitive behavior of polycarbazole- and poly(N-azole)-coated carbon ber microelectrodes in various solutions,f Applied Electrochemistry 39 (2009) 2043., A.K. Singh, R. Prakash, Electrolyte effects on various properties ofzole, Thin Solid Films 519 (2010) 1016.i, J. Redepenning, F.C. Anson, Inuence of supporting electrolyte con-n and composition on formal potentials and entropies of redox couplested in Naon coatings on electrodes, Journal of Physical Chemistry 9027.tt, P.G. Pickup, Permselectivity of polypyrrole in acetonitrile, Journalal Chemistry 95 (1991) 9634.

Formation and redox behaviour of polycarbazole prepared by elec-erization of solid carbazole crystals immobilized on an electrodeournal of Solid State Electrochemistry 7 (2003) 503.pp, M. Marandi, T. Tamm, V. Sammelselg, J. Tamm, Study of the factorsing the mobility of ions in the polypyrrole lms doped with aromatic

anions, Electrochimica Acta 53 (2008) 3828.son, Y.P. Kayinamura, Charge transport in conducting polymers:rom impedance spectroscopy, Chemical Society Reviews 38 (2009)

lov, J.R. Macdonald (Eds.), Impedance spectroscopy theory, experi-d applications, 2nd ed, Wiley-Interscience, Hoboken, 2005.ry, C.M. Elliott, A.R. Mount, A transmission line model for modi-rodes and thin layer cells, Journal of Electroanalytical Chemistry 288..G. Pickup, Ion transport in polypyrrole and a polypyrrole/polyanione, Journal of Physical Chemistry 97 (1993) 5356..G. Pickup, In situ measurement of the conductivity of polypyrrole andethyl-3-(pyrrol-1-ylmethyl)pyridinium]+ as a function of potential

ted voltammetry. Redox conduction or electronic conduction? Journalerican Chemical Society 112 (1990) 1776..M. Crooks, M.S. Wrighton, Potential dependence of the conductivity

oxidized polythiophenes, polypyrroles, and polyaniline: Finite win-igh conductivity, Journal of the American Chemical Society 112 (1990)

An electrochemical impedance study of thin polycarbazole films1 Introduction2 Experimental2.1 Chemicals2.2 Working electrode preparation2.3 Instrumentation and impedance fitting

3 Results and discussion3.1 Cyclic voltammetry3.2 Impedance spectroscopy3.3 Modeling of the impedance data3.4 Analysis and discussion of the impedance parameters

4 ConclusionsAcknowledgementsReferences