Carbon Support Effects and Mechanistic Details of ... - ACS Publications

Jan 12, 2017 - Departamento de Química Física, Facultad de Química, Regional Campus of International Excellence “Campus Mare Nostrum”,. Univers...
0 downloads 0 Views 2MB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Carbon Support Effects and Mechanistic Details of the Electrocatalytic Activity of Polyoxometalates Investigated via Square Wave Voltacoulometry Joaquin Gonzalez, Jose Antonio Coca-Clemente, Angela Molina, Eduardo Laborda, José María Gomez-Gil, and Luz Adriana Rincon ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b03392 • Publication Date (Web): 12 Jan 2017 Downloaded from http://pubs.acs.org on January 12, 2017

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.

ACS Catalysis 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 39

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

ACS Catalysis

Carbon Support Effects and Mechanistic Details of the Electrocatalytic Activity of Polyoxometalates Investigated via Square Wave Voltacoulometry

J. Gonzaleza*, J.A. Coca-Clementeb A. Molinaa E. Labordaa, J.M. Gomez-Gila, L.A. Rincona

a

Departamento de Química Física, Facultad de Química, Regional Campus of International Excellence “Campus Mare Nostrum”, Universidad de Murcia, 30100 Murcia, Spain b

Stephenson Institute for Renewable Energy, University of Liverpool, Chadwick Building, Peach Street, L69 7ZF Liverpool, United Kingdom

* Corresponding author: Tel: +34 868 88 7429 Fax: +34 868 88 4148 Email: [email protected]

1 ACS Paragon Plus Environment

ACS Catalysis

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

Abstract The catalytic activity of surface-confined molecular species, as affected by the nature of the support, is investigated by square wave voltacoulometry (SWVC). This technique has proven to be very powerful and advantageous for the study of electroactive and electrocatalytic monolayers. Here, the value of SWVC for the elucidation of the catalytic species and routes when the catalyst can undergo multiple electron transfers is assessed. The redox behaviour in acidic water solution of the immobilized Keggin type polyoxomolybdate [PMo12O40]3- and its catalytic performance towards the reduction of bromate is studied experimentally. Three different supports are considered: boron doped diamond (BDD), bare glassy carbon and graphene oxide-modified glassy carbon. For all of them, the SWVC response enables the accurate identification and analysis of the catalytic pathways and species. Mechanistic details are easily obtained on the basis of the additivity of the contributions associated with the electron transfer and with the catalytic process to the SWVC response. The results obtained reveal the significant influence of the support on the redox properties and on the catalytic activity of the polyoxomolybdate. Among the supports tested, glassy carbon supports show the highest catalytic performance with apparent rate constants that are up to one order of magnitude faster than on BDD.

Keywords Polyoxometalates; Surface-confined electrocatalysts; Square wave voltacoulometry; Support effects; Bromate reduction

2 ACS Paragon Plus Environment

Page 2 of 39

Page 3 of 39

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

ACS Catalysis

1. Introduction Polyoxometalates (POMs)1–4, in particular those of Keggin type, are very attractive compounds in electrocatalysis1 (towards bromate reduction5,6, water oxidation7,8, hydrogen evolution reaction9, nitrite reduction10 and oxidation11, etc.) for their appropriate and tunable redox properties that feature multiple oxidation states and fast stepwise electron transfers. In their use as electrocatalysts, POMs are generally immobilized12,13, frequently on carbon-based supports that show high surface area and pH stability14–16, in order to reduce their degradation in aqueous solutions11 and to facilitate their recovery and reuse. Ideally, the immobilization strategy and the nature of the support must allow for high catalyst coverage and for preserving its activity. Hence, investigating possible support effects on the performance of surfaceconfined electrocalysts is crucial. For the critical and comprehensive evaluation of the electrocatalyst performance and for the improvement of the overall rate of the electrocatalysis, a deep understanding of the redox behaviour of the catalyst and of the reaction mechanism is required. This calls for instrumental techniques to discriminate between different possible pathways and to determine accurately the corresponding kinetic and thermodynamic magnitudes. In this context, electrochemical techniques are very valuable tools since they provide direct access to the above from the analysis of the electrocatalyst’s electrochemical response. Among the techniques available, cyclic voltammetry (CV) is nowadays the reference technique in this type of studies since it enables the simple and rapid qualitative analysis of the system17. Nevertheless, experimental CV curves of surface-bound species can be considerably affected by different drawbacks including the complex influence of non-faradaic contributions18 and deviations from the theoretical response due to the staircase-like (instead of linear) perturbation applied by digital instruments19. Alternatively, potential pulse techniques such as staircase and square wave voltammetries20–24 offer notable advantages: better resolution and higher sensitivity, among others. Moreover, the pulsed nature of the potential perturbation allows for the kinetic discrimination between the different components of the response on the basis of their

3 ACS Paragon Plus Environment

ACS Catalysis

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

intrinsically different temporal dependencies24, a fact that is of great value in the analysis of electrocatalytic processes. In this work, the above questions are investigated by examining the electrochemistry of surface-confined [PMo12O40]3- (POM-Mo) in acidic water solution and, more importantly, its electrocatalytic activity towards the reduction of bromate after immobilization on three different supports: boron doped diamond (BDD), bare glassy carbon (GC) and graphene oxide-modified glassy carbon (GC-Gox). Other immobilization methods involving co-adsorbates, organicinorganic hybrids or support functionalization12 have been avoided to focus on support-only effects. For the experimental studies carried out here, the electrochemical techniques square wave voltacoulometry (SWVC) and square wave voltammetry (SWV) are employed. The SWVC method was designed in our research group25 and it is very powerful in the investigation of electroactive monolayers25–28. Specifically, in the case of redox monolayers with fast electron transfer kinetics (as desired for ideal electrocatalysts) the faradaic charge response can be measured without any problem unlike the current response; also, on the basis of the additivity of the different components in the overall SWVC response26,27, the faradaic contribution can be decoupled from the non-faradaic and catalytic ones. Moreover, the occurrence of the catalytic process is easily detected from the appearance of a peculiar charge-potential sigmoidal response of opposite sign that is highly sensitive to the catalytic rate constant. Finally, the subtractive nature of SWVC leads to responses better defined and less distorted by background signals than CV, which allows for more accurate and simple quantitative analyses. Making use of the SWVC method and of the corresponding analytical theory developed by our team for multiple electron transfers of immobilized redox molecules28,29, the first 6 apparent formal potentials of the surface confined POM-Mo are determined in the three supports above-mentioned, finding an excellent agreement between theoretical and experimental results in all cases. Next, the electrocatalytic performance of POM-Mo towards the reduction of bromate ( BrO3− )6,30 is evaluated. Bromate is considered a potentially carcinogenic and

4 ACS Paragon Plus Environment

Page 4 of 39

Page 5 of 39

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

ACS Catalysis

nephrotoxic, and it has been classified as category I, group B2 carcinogen31,32, with the maximum level in drinking water being set by the European Union and the US Environmental Protection Agency at 10 µg L-1 33. In our study, the catalytic mechanism for the electro-reduction of bromate is not postulated a priori but inferred from the fitting of theoretical and experimental data; this is not an easy task considering that multiple catalytic routes can be envisaged. Thus, the analysis of the experimental SWVC response is tackled assuming a support-dependent catalytic activity and three different catalytic states: H 2 PMo12 O340− , H 4 PMo12 O340− and H6 PMo12O340− . The results offer a general view of the overall mechanism of the electrocatalytic reduction of bromate together with accurate values of the corresponding catalytic rate constants at each support. The latter reveal the degree of suitability of the different conducting supports to maximize the catalytic efficiency.

5 ACS Paragon Plus Environment

ACS Catalysis

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

2. Experimental 2.1. Reagents and instrumentation Na3PMo12O40 (Merck), KBrO3, graphene oxide (Gox) and HClO4 (Sigma-Aldrich) were reagent grade and used as received. The modification of the different electrodes by [PMo12O40]3- was performed by direct immobilisation of the complex from an aqueous 0.1 mM POM-Mo solution, 1.0 M in HClO4 via cyclic voltammetry in the range 1.0 to -0.20 V (vs SCE) at 0.5 V s −1 for 20 minutes. Next, the electrode was thoroughly rinsed with water and transferred to the electrolyte solution. The response of the monolayer thus obtained is stable for several days with no significant loss of signal for all the supports tested. A three-electrode set-up were employed in the experiments with BDD (Windsor Scientific, doping level ca. 0.1%), bare GC (CH Instruments) or modified GC disc electrodes of 0.3 cm diameter acting as the working electrode. The counter electrode was a Pt foil, and the reference electrode was a saturated calomel electrode (SCE, CH Instruments). Solutions were prepared with distilled deionized water (Milli-Q filtering system). Nitrogen gas was passed through solutions for de-aeration for 20 min prior to measurements, with nitrogen atmosphere maintained over the solution during all the experiments.

2.2. Electrode preparation The BDD and GC electrodes was mechanically polished on alumina slurry (1.0 and 0.05 µm, Buehler), washed and cleaned electrochemically by anodic polarization at 1.8 V (vs SCE) for 5 minutes followed by cathodic polarization at -1.0 V for one minute in 1.0 M HClO4 solution. Next, the potential was cycled between 1.5 V and -0.3 V in the same media until a stable voltammogram was obtained. Then, the electrode was rinsed with ethanol and water. The modification of the GC electrode with graphene oxide was performed by drop casting a suspension of Gox (8 microliters) at the electrode surface followed by slow evaporation of the solvent.

6 ACS Paragon Plus Environment

Page 6 of 39

Page 7 of 39

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

ACS Catalysis

2.3. Electrochemical measurements and data analysis SWVC, SWV and CV were performed using a home-made computer-driven potentiostat-galvanostat. The charge record is implemented in the instrument by simply integrating the current obtained during the application of each potential pulse. The net SWVC and SWV response corresponds to the difference of charge or current signals obtained between two successive half-cycles (see below). In both SWVC and SWV, the following potential sequence is applied34,35 (see Figure 1):   m +1  m+1 Em = Einitial −  Int   − 1 ∆Est − ( −1) ESW ; m = 1, 2, …, N 2    

(1) where N is the total number of potential pulses applied, ∆Est is the potential step of the staircase and ESW is the square wave amplitude. Accordingly, three input parameters are necessary to be specified in SWV(C) measurements: ∆Est , ESW and the frequency (f) or the pulse length τ (see Figure 1). < Figure 1> The charge (SWVC) or current (SWV) responses are sampled at the end of each potential pulse and the net response (QSW or ISW) is given by the subtraction of the charge or the current corresponding to a pulse with odd index (forward charge or current) and the signal of the following pulse with even index (reverse charge or current):

QSW,c = Q2c−1 − Q2c = Qf,c − Qr,c ISW,c = I 2c −1 − I 2c = I f,c − I r,c

; c =1, 2, …, (N/2) (2)

where c refers to the c-th pair of pulses or cycle and subscripts f and r to the forward ( 2c − 1 ) and reverse (2c) components of each cycle. The SW charge or current is plotted versus the arithmetic average of the potentials applied:

Eindex,c = Einitial − ( c − 1) ∆Est

(3)

In order to enhance the contribution of the catalytic response to the signal, low frequencies were employed in the SWVC experiments: f = 5 Hz ( τ = 100 ms). Also, constant 7 ACS Paragon Plus Environment

ACS Catalysis

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

stirring of the solution was maintained during the experiments in order to hold pseudo-first order conditions for the catalysis. The baseline correction of the experimental SWVC curves was performed by fitting of the charge values obtained outside the faradaic region to a polynomial of third degree followed by subtraction by means of the ‘Transforms’ routines of SigmaPlot 10.0 for Windows.

8 ACS Paragon Plus Environment

Page 8 of 39

Page 9 of 39

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

ACS Catalysis

3. Results and discussion In this paper, the square wave potential waveform will be applied to examine the electrochemistry of POM-Mo confined at three different conducting supports (BDD, GC and GC-GOx) as well as its electrocatalytic activity towards the reduction of bromate. The study will be based on the quantitative analysis of the corresponding charge-potential ( QSW / E ) and current-potential ( I SW / E ) curves . 3.1. POM-Mo response At the potential interval of the study (0.0 - 1.0 V vs SCE) in strongly acidic solutions where proton transfers are assumed to be not rate-limiting, the surface confined POM-Mo moieties undergo a six-electron six-proton transfer process in agreement with Scheme 113,36,37, PMo12 O340− + e − + H + HPMo12 O340− + e − + H + H 2 PMo12 O340− + e − + H + H3 PMo12 O340− + e − + H + H 4 PMo12 O340− + e − + H + H5 PMo12 O340− + e − + H +

 HPMo12 O340−  Sub-Process 1    H 2 PMo12 O340−    H 3 PMo12 O340−  Sub-Process 2  Overall Process  H 4 PMo12 O340−    H 5 PMo12 O340−    Sub-Process 3  3− H 6 PMo12 O 40  

Scheme 1. Six-electron six-proton electro-reduction of polyoxomolybdate [PMo12O40]3- in acidic aqueous media.

In previous papers28, it has been reported that the SWVC charge-potential response of the overall process in Scheme 1 when the POM-Mo monolayers are formed at BDD electrodes can be modelled as three independent reversible bi-electronic reduction processes coupled with bi-protonations, which will be called sub-processes 1, 2 and 3. This approach is completely valid at BDD since the QSW / E responses (Figure 2) show three well separated peaks (see reference28). For either bare or modified GC supports, the above approach is not strictly valid since there is a partial overlapping of the charge-potential response of sub-processes 1 and 2 (see Figure 2). Nevertheless, when the average apparent formal potentials of the bi-electronic transfers ( E s = ( Es0,′1 + E s0,′2 ) / 2 with s=1, 2 and 3) differ by more than ca. 100mV and 0′

9 ACS Paragon Plus Environment

ACS Catalysis

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 39

∆E s0′ = E s0,′2 − Es0,′1 ≥ − 90 mV, the QSW − E curves corresponding to a six electron-transfer

process are equivalent to that of three independent two-step processes with maximum differences of 7% under the conditions here considered (i.e., ESW = 30 mV and f = 5 Hz) (see Figure S1). This behaviour is directly related to the subtractive nature of the SWVC technique (see Eq. (2)). However, it is worth noting that the individual excesses (or coverages) must be calculated on the basis of a six-electron transfer mechanism (see sections SI1 and SI2) because the assumption of three independent two-electron transfers (three independent EE mechanisms) does not reflect their evolution with the applied potential. According to the above, for the sake of simplicity the analysis of the charge-potential curves of the POM-Mo monolayers at the three supports under study will consider three independent bi-electronic sub-processes: O1,s + e − + H +

O2 ,s

O 2 ,s + e − + H +

O3,s

(E ) (E ) 0′ s ,1

0′ s ,2

    

s =1, 2, 3

Scheme 2. Reaction mechanism considered to model the electro-reduction of [PMo12O40]3-.

where the two electrons are not necessarily transferred simultaneously but according to the corresponding ∆ E s0′ = E s0,′2 − E s0,′1 value (see below). The above approach will also be followed for the analysis of the current-potential curves, such that the expressions for the responses of the overall process in Scheme 1 can be written as Q = Q1EE + Q2EE + Q3EE   I = I1EE + I 2EE + I3EE 

(4)

with Q sEE and I sEE being the charge-potential and current-potential responses of the individual EE vs E ) and SWV bi-electronic transfer sub-process s given by Scheme 2, which in SWVC ( QSW

EE ( ISW vs E ) are given by28,29:

(

) (

)

EE EE, f EE, r f f r r   QSW ,s = QSW ,s − QSW ,s = QF  1 − 2 f O1,s − f O2 ,s − 1 − 2 f O1 ,s − f O2 ,s   

 df f dfOf   dfOr dfOr2 ,s   O1 ,s EE, f EE EE, r 2 ,s 1 ,s      ISW = I − I = Q + − + ,s SW ,s F SW ,s  dt dt   dt dt      

10 ACS Paragon Plus Environment

(5)

Page 11 of 39

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

ACS Catalysis

where QF = FA ΓT (with Γ T being the total excess) and the expressions of the normalized surface excesses (or surface coverages) f Of / r and f Of / r 1, s 2,s

(s=1, 2, 3; ‘f’ and ‘r’ stand for the

forward and reverse pulses of each SW cycle) are defined as

f Of / r = Γ Of / r / Γ T and 1,s 1,s

f Of / r = Γ Of / r / Γ T and their expressions are given in the Supporting Information (see sections 2, s 2,s

SI1 and SI3). Considering EE mechanisms instead of simultaneous two-electron transfers (single two-electron processes) in Scheme 2 is crucial to describe and understand the electrochemical responses obtained experimentally. Note that there are relevant differences in the potential location, peak height and half peak width between the responses of the two situations above-mentioned, as discussed in38. Figure 2 shows the experimental and theoretical QSW / E curves obtained for POM-Mo confined at BDD (Fig. 2a), GC (Fig. 2b) and GC-Gox (Fig. 2c) in 1M HClO4 solutions. The EE responses corresponding to the three hypothetical bi-electronic transfers QSW , s are also

included (dashed blue lines). The experimental QSW / E curves were analysed with the theoretical equations deduced (S12). As can be seen, an excellent agreement between theoretical and experimental curves is achieved for the three supports considered with the values of the six apparent formal potentials indicated in Table 1. At BDD (Fig. 2a), three separated peaks related with sub-processes 1, 2 and 3 in Scheme 1 can be clearly observed. Thus, the value of the peak potential coincides with the 0′



corresponding average apparent formal potential28, E s = ′



E s0,2 + E s0,1

2 ′

(dashed vertical lines) and



the half-peak width enables the determination of ∆ E s0 = E s0,2 − E s0,1 as discussed in28. The total charge QF ( = FAΓ T ) is obtained from the peak heights. Another interesting parameter for each EE process is the "effective electron number", neff , defined as38:

neff ,s = 1 +

% character E2e−

100

=1+

2e

F ∆Es0′ / ( RT )

1 + 2e

F ∆Es0′ / ( RT )

11 ACS Paragon Plus Environment

(6)

ACS Catalysis

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

with (% character E 2e



) / 100

Page 12 of 39

being the probability of the second electron to be transferred in an

apparently simultaneous way with the first one at E = E s0′ . The values of all the above magnitudes are included in Table 1 for the three electrodes. < Figure 2 > In the case of the GC and GC-Gox electrodes (Figs. 2b and 2c), the peak at most negative potentials corresponding to sub-process 3 can be analysed as indicated above to 0′

determine E 3 (from the peak potential), ∆E30′ (from the half-peak width) and QF (from the peak height). On the other hand, the QSW / E responses of sub-processes 1 and 2 are partially overlapped giving rise to a broad peak with a hump at around 0.45-0.5 V. Hence, the fitting of 0′

0′





the whole peak was performed with four adjustable parameters ( E 1 , E 2 , ∆E10 and ∆E20 ) according to two independent EE mechanisms (see also section SI2). The values of the apparent formal potentials corresponding to the best fit of the experimental curves are given in Table 1. 0′

0′

Note that in this case the E 1 - and E 2 -values do not coincide with any particular feature of the

QSW / E curve. Among the data in Table 1 it is worth highlighting that the converted charge values obtained for GC and GC-Gox are quite similar ( QF = 1.09 and 0.90 µ C for GC and GC-Gox, respectively, corresponding to a full monolayer coverage with a total excess of Γ T cm-2), and three times larger than that obtained at BDD ( QF

with Γ T

10−10 mol

0.3 µ C , i.e., a sub-monolayer

10−11 mol cm-2). Since the geometric areas of the BDD and GC electrodes are

identical, the experimental QF -values indicate that GC surfaces show higher roughness and/or that POM-Mo has higher affinity for GC (either naked or modified). Moreover, ∆E s0′ decreases in the order BDD > GC

GC-Gox, especially in the case of process 2 where ∆E20′ varies by

more than -130 mV (from +101mV at BDD to -30mV at GC-Gox). The lower ∆E s0′ -values for GC and GC-Gox may be related to differences in the structure, adsorption and

12 ACS Paragon Plus Environment

Page 13 of 39

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

ACS Catalysis

'microenvironment' of the POM-Mo moiety. These factors affect distinctly the redox properties and the chemical reactivity of the immobilized POM-Mo depending on the nature of the support, as reported with graphene supports11,39. To get further insight into the above question, the evolution of the individual apparent formal potentials of the successive reduction processes ( Ei0′ ) are plotted in Figure 3. The value of E10′ is (statistically) the same in all the supports considered. Then, a quasi-linear decrease of ca. -55 mV per reduction step is observed at both GC and GC-Gox. Clear deviations from this

trend are observed in the second and fourth electro-reductions on BDD with 'inversions' of the ordering of the apparent formal potentials: E20 ≈ E10 and E40 > E30 . This behaviour indicates a particularly strong stabilization of the corresponding reduced states (i.e., H2PMo12O340− and H4PMo12O340− ). If the protonation energy varied linearly with the charge of the POM molecule as

predicted for solution-phase POMs by DFT calculations40, its effect would not break the linear trend of the apparent formal potentials of a given POM isomer. Hence, the deviations observed at BDD may point to the occurrence of reversible POM isomerization and/or re-orientation processes41–43. Regarding the former, POM isomers are known to show different LUMO energies and redox properties41, with β-isomers being more easily reduced than α-isomers and becoming more stable after a few reduction steps41,42. Also, studies of POM on silver surfaces have revealed the re-orientation of the surface-bound molecules as scanning the potential towards more negative values43. Our results would suggest that the extent of these phenomena strongly depends on the support properties since they are only observed at BDD and not at GC and GC-Gox. < Figure 3 > Figure 4 shows the experimental and theoretical ISW / E curves corresponding to POM-Mo monolayers at BDD (Fig. 4a), GC (Fig. 4b) and GC-Gox (Fig. 4c) in 1M HClO4 solutions. The responses corresponding to the three hypothetical individual bi-electronic EE transfers I SW , s are also plotted (dashed blue lines).

13 ACS Paragon Plus Environment

ACS Catalysis

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 39

< Figure 4 > Since fully reversible electron transfer processes are not expected to give any measurable current response in potential pulse or step techniques24,25, then the fact that ISW / E curves were registered indicates a certain degree of non-reversibility. This seems to contrast with the excellent fittings obtained in Figure 2 under the assumption of reversible electron transfers. In order to shed light on this, Figure 5 shows the variation of the normalized I SW / E peak current and potential (Figs. 5a and 5b) and QSW / E peak charge and potential (Figs. 5c

( )

and 5d) of an EE mechanism with log k 0

0 0 (with k = k τ or k 0 = k 0 / ( 2 f ) ) for ∆E 0′

values in the range of those obtained experimentally (Table 1). By comparing the curves in Figs. 5a and 5c, it can be concluded that the peak charge of the QSW / E curves only deviate

( )

significantly (>5%) from the fully reversible behaviour for log k 0 < 0.18 , whereas the rev ISW / E curves show clear deviations from the reversible limit (i.e., from ISW ≅ 0 by more than

( )

5% of the maximum value of the peak current) at larger k 0 -values: log k 0 < 0.41. Therefore,

( )

for log k 0 > 0.18 the QSW / E curves can be treated as "reversible" (Eqs. (S10) of SI) while a

( )

measurable ISW / E response will be obtained if log k 0 < 0.41. It is worth noting that according to Fig. 5b, the peak potential is less sensitive than the peak current to the electron

( )

transfer kinetics. Thus, Epeak retains "reversible features" for log k 0 ≥ −0.5 such that its value coincides with the average apparent formal potential Es0′ (with a maximum difference of 2 mV). < Figure 5 > According to the above discussion, the experimental ISW / E response can be used to study the electron transfer kinetics of the surface-confined POM-Mo. The ISW / E curves were analyzed with the theoretical solutions corresponding to three independent quasi-reversible two-

14 ACS Paragon Plus Environment

Page 15 of 39

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

ACS Catalysis

electron transfers (see Eq. (S17)-(S18)) using the values of the apparent formal potentials in Table 1, αi1 = αi 2 = 0.5 and the standard heterogeneous rate constants ( k1i0 and k2i0 ) as fitting parameters. The ki0 -values corresponding to the best agreement between theoretical and experimental ISW / E curves (Fig. 4) are given in Table 2. They lie within the range 3.8-18.2

( )

s−1 ( − 0.4 < log ki0 < 0.26 ), being of the same order of magnitude for all the supports and slightly slower at GC and GC-Gox than at BDD. In the case of quasi-reversible charge transfer reactions, under the working conditions here employed the assumption of additivity in the current response (Eq. (4)) is justified for the values of the rate constants found for POM-Mo monolayers at BDD since the ISW / E curves are sufficiently separated (see Figure 4a). In the case of GC and GC-Gox, the above does not strictly hold for sub-processes 1 and 2 due to the partial overlapping of the SWV peaks. Therefore, the corresponding values of the rate constants obtained from Eqs. (4), (S17) and (S18) should only be taken as estimative. As discussed above, the SWVC curves of POM-Mo are scarcely affected by the electrode kinetics in such a way that the additivity assumption is fully valid under all the conditions here examined. Therefore, for the following studies only the SWVC responses of POM-Mo will be analyzed.

3.2. Electrocatalysis of the bromate reduction

The electrocatalytic reduction of bromate by POM-Mo confined at the three electrodes considered has been investigated using SWVC. Previous studies in the literature considered the electrochemical activity of POM-Mo attached directly to glassy carbon electrodes5,6 as well as in other contexts including their use as dopants of conducting polymers, combined with nanoparticles and subject to different functionalization procedures6,12,44–49. Regarding the electrocatalytic reduction by POM-Mo of bromate and similar species (iodate, chlorate and oxygen, among others), different reaction mechanisms have been proposed. Most of them

15 ACS Paragon Plus Environment

ACS Catalysis

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 39

assume that H4PMo12O340− is the only active catalytic species, with very few exceptions where the possibility of two (or even three) catalytic routes driven by H2PMo12O340− , H4PMo12O340− and H6PMo12O340− were explicitly considered or at least mentioned44–46,49. Here, we propose that

bromate can be reduced by the two-, four- and six-electron reduced species (i.e., sub-processes 1, 2 and 3 in Scheme 1) giving rise to the following possible catalytic pathways:   Sub-Process 1        H 2 PMo12 O340− + e − + H + H3PMo12 O340−    3− − + 3− H 3PMo12 O 40 + e + H H 4 PMo12 O 40  Sub-Process 2    kcapp ,2 H 4 PMo12 O340− + xBrO3− + ( 6 x − 2 ) H +  → xBr − + H 2 PMo12O340− + 3 xH 2 O      3− − + 3−  H 4 PMo12 O 40 + e + H H5 PMo12 O 40    H5 PMo12O340− + e − + H + H 6 PMo12O340−  Sub-Process 3   kcapp  ,3 H 6 PMo12O340− + xBrO3− + ( 6 x − 2 ) H +  → xBr − + H 4 PMo12O340− + 3 xH 2 O     3− − + 3− HPMo12 O 40 + e + H H 2 PMo12 O 40   kcapp 3− − + − 3− ,1 H 2 PMo12O 40 + xBrO3 + ( 6 x − 2 ) H  → xBr + PMo12 O 40 + 3xH 2O  PMo12O340− + e − + H +

HPMo12 O340−

Scheme 3. Possible electrocatalytic pathways considered for the bromate reduction with POM-Mo as mediator.

and each catalytic sub-process in Scheme 3 can be modelled as: O1,s + e− + H +

O 2 ,s

O 2 ,s + e − + H + k

O3 ,s

app c ,i

O3,s + Ssol  → O1,s + Psol

    

s =1, 2, 3

Scheme 4. Reaction mechanism considered to model the electrocatalytic reduction of bromate by POMMo.

where kcapp ,s is the apparent rate constant of the catalytic step for sub-process s (s=1, 2, 3), also denoted as the 'turnover frequency' (TOF)17. According to Schemes 3 and 4, the occurrence of a "direct" catalytic process is assumed a priori for all the sub-processes. Note that this scheme includes more simple situations where only one or two sub-processes are active, the apparent rate constants of the other routes being negligible or null. Moreover, in the following it will be considered that no concentration gradients of bromate arise due to the constant stirring of the

16 ACS Paragon Plus Environment

Page 17 of 39

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

ACS Catalysis

solution. Hence, the heterogeneous chemical step in each process between the surface-confined POM-Mo and the solution-phase bromate follows a pseudo-first order kinetics (see below). Provided that the electron transfers behave as reversible, the surface excesses are constant over each potential pulse and identical to those calculated in the absence of the catalytic reaction since Eqs. (S5) and (S6) are valid. The charge contributions associated with the electron transfers and to the catalysis are additive such that for the SW potential waveform it can be written that:

QSW = QSW,1 + QSW,2 + QSW,3 = EE r EE app r EE app r = QSW,1 − QF k capp ,1 f O3 + QSW,2 − QF k c ,2 f O5 + QSW,3 − QF k c ,3 f O7

(7)

EE with QSW,s being given by Eq. (5) and k capp the dimensionless apparent catalytic rate constant ,s

* kcapp , s = kc , sτ cBrO−

s = 1, 2, 3

(8)

3

3− and O 3 , O5 and O7 being species H 2 PMo12O340− , H 4 PMo12O340− and H 6 PMo12O40 ,

respectively. Since the QSW / E curves depend on the individual values of the surface excesses of the catalyst states f O 3 ,s as indicated in Scheme 4, they must be calculated considering a six electron-transfer mechanism (Eqs. (S5) and (S6) with the applied potential corresponding to the reverse pulse of the corresponding SW cycle) because the assumption of three independent two electron transfers does not provide appropriate results. In order to quantify the catalytic rate constants for the different routes, SWVC experiments were run in presence of different concentrations of bromate in the range 0.9-5.0 mM in 1M HClO4 solutions. As shown in Fig. 6, the occurrence of the catalytic process affects very apparently the QSW / E curves, mainly at the most cathodic potentials where the sign of the charge changes and a negative charge plateau appears. These negative charges arise as a consequence of the catalytic contributions to the overall response as can be deduced from Eq. (7). Eventually, this can lead to the disappearance of the third peak at large bromate concentrations (see Figs 6b-c). At the three electrodes, negative charge values are only observed in the peaks corresponding to sub-processes 2 and 3 in Scheme 3 whereas the peak of process 1 17 ACS Paragon Plus Environment

ACS Catalysis

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 18 of 39

remains unaffected. The scarce effect on the response related to sub-process 1 indicates that under these conditions kcapp ,1

0.

The experimental QSW / E responses were fitted with Eqs. (7), (8), (S11) and (S12), by using the values of the apparent formal potentials given in Table 1 and the dimensionless app * catalytic rate constants of sub-processes 1 and 2, kc,app s ( = k c, s τ cBrO − with s=1 and 2) as fitting 3

app plateau was directly determined from the charge value of the plateau ( QSW ) parameters. The k c,3

observed at the most negative potentials as follows: kcapp ,3 =

plateau QSW QF

(9)

since in this region the applied potentials are very negative with respect to all the formal EE,1 EE,2 EE,3 potentials such that fOr3 = 0 , fOr5 = 0 and fOr7 = 1 (see Eq. (S11)), QSW = QSW = QSW =0

(see Eqs. (5) and (S11)) and Eq. (9) is immediately deduced from Eq. (7). The results are shown in Figure 6 (symbols) where an excellent agreement between the best-fit theoretical and experimental QSW / E curves is observed for all the bromate concentrations and electrodes tested. < Fig. 6 > Further mechanistic details of the catalysis can be gained by examining the hypothetical s individual contributions of sub-processes 1, 2 and 3 ( QSW ) to the catalysis (red, blue and green

lines, respectively). The theoretical (Eq. (7)) and experimental curves are shown for a given concentration of bromate (4.3 mM, points) in Figure 7 at BDD (Fig. 7a), GC (Fig. 7b) and GCGox (Fig. 7c). In line with the above assumptions, each sub-process has a potential interval of stability where the two-, four- and six-electron reduced species are active. In the case of species H2 PMo12O340− , no effect is observed upon the addition of bromate whereas species H4 PMo12O340−

and H6PMo12O340− give rise to different catalytic contributions to the total response, with the most reduced species being the most active one (see green lines in Fig. 7). Note that species

18 ACS Paragon Plus Environment

Page 19 of 39

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

ACS Catalysis

H4 PMo12O340− is not stable at potentials below 0.15 V such that its contribution to the response

becomes null at the most negative potentials scanned (see blue lines in Fig. 7). < Fig. 7 > By comparing the catalytic activity of the three electrodes, it can be inferred that POMMo monolayers exhibit higher catalytic activity at GC, and specially at GC-Gox, than at BDD. Thus, a 1mM-increase of the BrO3− concentration gives rise to the increase of the plateau charge (relative to the total charge, QF ) by 0.105 µC at BDD, by 0.578 µC at GC and by 0.958 µC at GC-Gox. for the three electrodes have been plotted versus the The values obtained for k capp ,s

BrO3− concentration in Figure 8 for the determination of the true apparent catalytic constants (Table 3). From the values obtained it can be concluded that the catalytic route 1 in k capp ,s Scheme 3 is negligible at the three supports and, therefore, POM-Mo must reach the oxidation state H 4 PMo12O340− to show some catalytic activity in line with the results reported at GC electrodes 47. Hence, routes 2 and 3 are responsible for the catalysis observed. Route 3 shows app app app app higher catalytic efficiency in all cases ( k c,3 ), and k c,2 and k c,3 are one order of > k c,2

magnitude higher at GC-Gox than at BDD. These details are very clearly illustrated by the experimental SWVC curves in Figure 7 where the first peak is unaffected by the presence of bromate whereas the second and third peaks decrease with the concentration of BrO3− , this effect being more apparent at GC-Gox. < Fig. 8 > The combined catalytic activity of species H4PMo12O340− and H6PMo12O340− at GC electrodes makes them adequate supports for the efficient catalysis of the bromate reduction, the modification with graphene oxide improving the catalysis. Note that the poor catalytic performance on BDD can be associated in part with the particular thermodynamic stabilizations

19 ACS Paragon Plus Environment

ACS Catalysis

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

of certain POM-Mo reduced states discussed in Figure 2, which can lead to lower electrocatalytic activity towards reduction reactions.

20 ACS Paragon Plus Environment

Page 20 of 39

Page 21 of 39

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

ACS Catalysis

4. Conclusions The results reported in this work contribute to a more complete mechanistic picture of the electrocatalytic activity of surface-confined polyoxometalates, also pointing out how this can be significantly affected by the nature of the support. Thus, the quantitative electrochemical study of the polyoxomolybdate [PMo12O40]3- (POM-Mo) immobilized on different carbon supports has revealed crucial support effects on both the POM-Mo’s redox behaviour and on the rate of the catalytic reaction. With regard to the former, a quasi-linear dependence of the apparent formal potentials of surface-bound POM-Mo with the reduction state is observed on bare and modified GC electrodes. On the other hand, the reduction potential of POM-Mo adsorbed on BDD shows significant deviations from this trend, with the particular thermodynamic stabilization of certain reduced states. The electrocatalytic activity of POM-Mo towards bromate reduction when confined on BDD also shows considerable differences with respect to GC and GC-Gox, the catalytic rate constants differing by an order of magnitude between them. The analysis of all the experimental SWVC curves in the presence of bromate is consistent with the existence of two different catalytic species: H4PMo12O340− and H6PMo12O340− . Both feature the fastest catalysis when POMMo is attached to graphene oxide-modified GC surfaces and the slowest one with POM-Mo immobilized on BDD. The differences may point out the key role played by the supportdependent stabilization of the different reduced states of POM-Mo in the performance of the electrocatalysis. Quantum chemical calculations as well as infrared spectroscopy can provide further insight into the physicochemical origin of such stabilization. The electrochemical study here performed also demonstrates that square wave voltacoulometry (SWVC), designed by our research group, is a powerful technique for the elucidation of the electrocatalytic mechanism and performance of surface-bound redox catalysts. Important advantages of SWVC versus CV for the study of these systems have been demonstrated theoretically and pointed out experimentally. Mainly, the charge-potential response of reversible mono- and multi-electron transfers in the presence of a coupled catalytic

21 ACS Paragon Plus Environment

ACS Catalysis

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

process can be split into the contributions of the 'intrinsic' electrochemical response of the surface-confined catalyst and that of the catalytic process. Moreover, the occurrence of the catalysis can be easily verified by the appearance of a charge-potential sigmoid of opposite sign, the plateau of which is directly related to the value of the corresponding catalytic rate constant. It is also worth highlighting that the SWVC response of electrochemically quasi-reversible catalysts is less affected by the electrode kinetics than the signal in square wave voltammetry. All these advantages make the elucidation of the electrocatalytic mechanism and its quantitative analysis more accurate and simple.

Supporting Information. - Mathematical derivation of the analytical solution for the SWVC response of reversible multistep electrode processes - Comparison between six-electron transfers, two independent four-electron and two-electron transfers, and three independent two-electron transfers - Mathematical derivation of the analytical solution for the SWV response of non-reversible two-electron electrode processes.

Acknowledgements The authors greatly appreciate the financial support provided by the Fundación Séneca de la Región de Murcia (Projects 19887/GERM/15 and 18968/JLI/13) as well as by the Ministerio de Economía y Competitividad (Projects CTQ-2015-65243-P and CTQ-2015-71955REDT Network of excellence “Sensors and Biosensors”). EL thanks the Ministerio de Economía y Competitividad for the fellowship ‘Juan de la Cierva-Incorporación 2015’. JMGG thanks the Ministerio de Educación, Cultura y Deporte for the fellowship ‘Ayuda de Formación de Profesorado Universitario 2015’.

22 ACS Paragon Plus Environment

Page 22 of 39

Page 23 of 39

2nd cycle

scan 1st cycle

f =

Ιf ∆Est

Potential

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

ACS Catalysis

1 2 tp

τ Eindex,2

ESW E initial = Eindex,1

τ

Ιr

ESW

Time Figure 1. Potential-time waveform in square wave voltammetry (SWV) and voltacoulometry (SWVC). Black dots indicate the time at which the current is measured in SWV (i.e., at the end of each potential pulse). In the case of SWVC, the charge is obtained by integration of the current over the whole pulse length, τ. f denotes the frequency of the SW perturbation. ∆Est

23 ACS Paragon Plus Environment

ACS Catalysis

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

Figure 2. Best-fit theoretical (solid black lines, Eqs. (4) and (5)) and experimental (red points) square wave voltacoulometric curves in 1M HClO4 corresponding to the electroreduction of POM-Mo immobilized on the different supports considered (indicated on the graphs). The responses of the three hypothetical individual bi-electronic transfers have also been included (dashed blue lines) as well as the

24 ACS Paragon Plus Environment

Page 24 of 39

Page 25 of 39

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

ACS Catalysis

corresponding average apparent formal potential (dashed black lines). ∆Est = 5 mV, ESW = 30 mV,

τ = 1 0 0 ms ( f = 5 Hz ).

25 ACS Paragon Plus Environment

ACS Catalysis

0.50

GC BDD GC-Gox

0.45

0.40

0.35

0.30

0'

Ei / mV vs SCE

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 26 of 39

0.25

0.20

1

2

3

4

5

6

Number of reductions, i

Figure 3. Variation of the value of the apparent formal potential in 1M HClO4 with the number of electrons transferred at the different electrodes studied (indicated on the graph). Grey solid lines show the

average decrease of the apparent formal potential with the consecutive electro-reductions at GC and GCGox

26 ACS Paragon Plus Environment

Page 27 of 39

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

ACS Catalysis

Figure 4. Best-fit theoretical (solid black lines, Eqs. (4), (S17) and (S18)) and experimental (red points) square wave voltammetric curves in 1M HClO4 corresponding to the electro-reduction of POM-Mo immobilized on the different supports considered (indicated on the graphs). The responses of the three hypothetical individual bi-electronic transfers have also been included (dashed blue lines) as well as the corresponding average apparent formal potential (dashed black lines). ∆Est = 5 mV, ESW = 30 mV,

( f = 5 Hz ) .

τ = 1 0 0 ms 27 ACS Paragon Plus Environment

ACS Catalysis

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

Page 28 of 39

Figure 5. Variation of the peak height (a and c) and peak potential (b and d) with the electron transfer kinetics in SWV (a and b, Eqs. (S17) and (S18)) and SWVC (c and d, 0′



from Eqs. (S17) and (S18)). α=0.5. The values of ∆E 0 (mV) are: -100, -50, 0, 50 and 100. The curve corresponding to ∆E = −100 mV in Fig b has not been included ′

since the SWV curves obtained for this value show two or more peaks. Dashed lines in Figs. b and d mark the average formal potential E 0 (upper line) and a potential 2 mV below this value (lower line).

28

ACS Paragon Plus Environment

Page 29 of 39

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

ACS Catalysis

Figure 6. Best-fit theoretical (lines, Eq. (7)) and experimental (points) square wave voltammetric curves in 1M HClO4 corresponding to the electro-reduction of immobilized POM-Mo in presence of bromate at different concentrations (indicated on the graphs). ∆Est = 5 mV, ESW = 30 mV, τ = 1 0 0 ms

( f = 5 Hz ) 29 ACS Paragon Plus Environment

ACS Catalysis

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

Figure 7. Theoretical (lines, Eq. (7)) and experimental (points) SWVC curves in 1M HClO4 corresponding to the electro-reduction of POM-Mo immobilized on the different supports considered (indicated on the graphs) in presence of bromate ( c*BrO3− = 4.3 mM). The responses corresponding to the three hypothetical individual sub-processes 1, 2 and 3 have also been included (red, blue and green lines, respectively). Solid black lines correspond to the sum of individual responses for the catalytic response. ∆Est = 5 mV, E SW = 30 mV, τ = 1 0 0 ms ( f = 5 Hz ).

30 ACS Paragon Plus Environment

Page 30 of 39

Page 31 of 39

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

ACS Catalysis

app

app

* Figure 8. Linear regression analysis of the dimensionless catalytic rate constant k c, s = k c, s τ cBrO on the − 3

different supports considered (indicated on the graphs).

31 ACS Paragon Plus Environment

ACS Catalysis

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

Page 32 of 39

Table 1. Total charge ( QF ), average apparent formal potentials ( E s0 ′ ), differences between apparent formal potentials ( ∆Es0′ = E20,s′ − E10,s′ ) and individual ′

apparent formal potentials ( Ei0 ) corresponding to the direct response of [PMo12O40]3- monolayers at BDD, GC and GC-Gox electrodes in 1M HClO4 shown in Figures 2-4. ∆Est = 5 mV, ESW = 30 mV and τ = 1 0 0 ms. All potentials are referred to SCE. BDD

Process

QF / µC

E s0 ′ /mV

∆ E s0′ /mV

1

0.292

475

3

2

362

101

3

193

2

Ei0−′th ET / mV

neff

(1) 474±6 (2) 477±4 (3) 312±9 (4) 413±7 (5) 192±3 (6) 194±6

1.59

(1) 469±8 (2) 380±10 (3) 330±6 (4) 303±9 (5) 220±10 (6) 176±6

1.26

(1) 476±2 (2) 385±3 (3) 339±1 (4) 309±4 (5) 230±10 (6) 184±4

1.26

1.89 1.69

GC 1

1.09

424

-89

2

316

-27

3

198

-44

1.57 1.52

GC-GOx 1

0.900

430

-91

2

324

-30

3

207

-46

32

ACS Paragon Plus Environment

1.53 1.44

Page 33 of 39

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

ACS Catalysis

Table 2. Standard heterogeneous rate constants for the charge transfers steps of the three independent EE processes (s=1, 2 and 3), corresponding to the direct response of [PMo12O40]3monolayers at BDD, GC and GC-Gox electrodes shown in Fig. 4. ∆Est = 5 mV, ESW = 30 mV and τ = 1 0 0 ms. α1 = α2 = 0.5 .

BDD Process

QF / µC

k10i = k20i / s −1

1

0.292

16.5

2

12.0

3

13.2 GC

Process

QF / µC

k1i0 / s−1

k 20i / s−1

1

1.45

12.5

5.0

2

13.0

3

11.5

18.2

GC-Gox Process

QF / µC

k1i0 / s−1

k 20i / s−1

1

1.18

9.5

3.8

2

11.5

3

12.5

33 ACS Paragon Plus Environment

ACS Catalysis

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 34 of 39

Table 3. Catalytic rate constants for the three catalytic routes in Scheme 3 for the three electrodes under study. These data have been obtained from the linear plots shown in Figure 8.

∆Est = 5 mV, ESW = 30 mV and τ = 100 ms. Electrode

−1 −1 kcapp ,1 ( M s )

−1 −1 kcapp ,2 ( M s )

−1 −1 kcapp ,3 ( M s )

BDD

0

784

948

GC

0

603

4903

GC-Gox

0

6525

7435

34 ACS Paragon Plus Environment

Page 35 of 39

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

ACS Catalysis

References (1)

Wang, S.-S.; Yang, G.-Y. Chem. Rev. 2015, 115, 4893–4962.

(2)

Miras, H. N.; Vilà-Nadal, L.; Cronin, L. Chem. Soc. Rev. 2014, 43, 5679–5699.

(3)

Misono, M. In Studies in Surface Science and Catalysis; Elsevier Inc. 2013; Vol. 176, pp 97–155.

(4)

López, X.; Carbó, J. J.; Bo, C.; Poblet, J. M. Chem. Soc. Rev. 2012, 41, 7537–7571.

(5)

Xi, X.; Dong, S. Electrochim. Acta 1995, 40, 2785–2790.

(6)

Papagianni, G. G.; Stergiou, D. V.; Armatas, G. S.; Kanatzidis, M. G.; Prodromidis, M. I. Sensors Actuators B Chem. 2012, 173, 346–353.

(7)

Stracke, J. J.; Finke, R. G. ACS Catal. 2013, 3, 1209–1219.

(8)

Stracke, J. J.; Finke, R. G. ACS Catal. 2014, 4, 79–89.

(9)

Liu, R.; Zhang, G.; Cao, H.; Zhang, S.; Xie, Y.; Haider, A.; Kortz, U.; Chen, B.; Dalal, N.; Zhao, Y. S.; Zhi, L.; Wu, C.-X.; Yan, L.-K.; Su, Z.-M.; Keita, B. Energy Environ. Sci. 2016, 9, 1012–1023.

(10)

Yaqub, M.; Imar, S.; Laffir, F.; Armstrong, G.; McCormac, T. ACS Appl. Mater. Interfaces 2015, 7, 1046–1056.

(11)

Kim, Y.; Shanmugam, S. ACS Appl. Mater. Interfaces 2013, 5, 12197–12204.

(12)

Rafiee, E.; Eavani, S. RSC Adv. 2016, 6, 46433–46466.

(13)

Sadakane, M.; Steckhan, E. Chem. Rev. 1998, 98, 219–238.

(14)

Izumi, Y.; Urabe, K. Chem. Lett. 1981, 10, 663–666.

(15)

Schwegler, M. A.; Vinke, P.; van der Eijk, M.; van Bekkum, H. Appl. Catal. A, Gen. 1992, 80, 41–57.

(16)

Wu, Y.; Ye, X.; Yang, X.; Wang, X. Ind. Eng. Chem. Res. 1996, 35, 2546–2560.

(17)

Rountree, E. S.; McCarthy, B. D.; Eisenhart, T. T.; Dempsey, J. L. Inorg. Chem. 2014, 53, 9983–10002.

(18)

Smith, C. P.; White, H. S. Anal. Chem. 1992, 64, 2398–2405.

(19)

Heering, H. A.; Mondal, M. S.; Armstrong, F. A. Anal. Chem. 1999, 71, 174–182.

(20)

Mirceski, V.; Komorsky-Lovric, S.; Lovric, M. Square-Wave Voltammetry: Theory and 35 ACS Paragon Plus Environment

ACS Catalysis

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

Application; Springer: 2007.

(21)

Mirceski, V.; Laborda, E.; Guziejewski, D.; Compton, R. G. Anal. Chem. 2013, 85, 5586–5594.

(22)

Mirceski, V.; Guziejewski, D.; Lisichkov, K. Electrochim. Acta 2013, 114, 667–673.

(23)

Mirceski, V.; Guziejewski, D.; Bozem, M.; Bogeski, I. Electrochim. Acta 2016, 213, 520–528.

(24)

Molina, Á.; González, J. Pulse Voltammetry in Physical Electrochemistry and Electroanalysis; Monographs in Electrochemistry; Springer International Publishing:

Cham, 2016. (25)

González, J.; Molina, A.; Abenza, N.; Serna, C.; Moreno, M. M. Anal. Chem. 2007, 79, 7580–7587.

(26)

González, J.; Soto, C. M.; Molina, A. J. Electroanal. Chem. 2009, 634, 90–97.

(27)

Gonzalez, J.; Molina, A. J. Solid State Electrochem. 2013, 17, 537–546.

(28)

Gonzalez, J.; Molina, A.; Lopez-Tenes, M.; Karimian, F. Anal. Chem. 2013, 85, 8764– 8772.

(29)

Gonzalez, J.; Lopez-Tenes, M.; Molina, A. J. Phys. Chem. C 2013, 117, 5208–5220.

(30)

Michalski, R.; Łyko, A. Crit. Rev. Anal. Chem. 2013, 43, 100–122.

(31)

Ding, L.; Li, Q.; Cui, H.; Tang, R.; Xu, H.; Xie, X.; Zhai, J. Electrochim. Acta 2010, 55, 8471–8475.

(32)

Sun, C.; Deng, N.; An, H.; Cui, H.; Zhai, J. Chemosphere 2015, 141, 243–249.

(33)

Lan, H.; Mao, R.; Tong, Y.; Liu, Y.; Liu, H.; An, X.; Liu, R. Environ. Sci. Technol. 2016, 50, 11872–11878.

(34)

Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; Wiley: New York, 2000.

(35)

Mann, M. A.; Helfrick, J. C.; Bottomley, L. A. Anal. Chem. 2014, 86, 8183–8191.

(36)

Zhang, T. R.; Feng, W.; Fu, Y. Q.; Lu, R.; Bao, C. Y.; Zhang, X. T.; Zhao, B.; Sun, C. Q.; Li, T. J.; Zhao, Y. Y.; Yao, J. N. J. Mater. Chem. 2002, 12, 1453–1458.

(37)

Xu, M.; Li, Y.; Li, W.; Sun, C.; Wu, L. J. Colloid Interface Sci. 2007, 315, 753–760. 36 ACS Paragon Plus Environment

Page 36 of 39

Page 37 of 39

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

ACS Catalysis

(38)

Lopez-Tenes, M.; Gonzalez, J.; Molina, A. J. Phys. Chem. C 2014, 118, 12312–12324.

(39)

Tessonnier, J.-P.; Goubert-Renaudin, S.; Alia, S.; Yan, Y.; Barteau, M. A. Langmuir 2013, 29, 393–402.

(40)

Fernández, J. A.; López, X.; Poblet, J. M. J. Mol. Catal. A Chem. 2007, 262, 236–242.

(41)

López, X.; Poblet, J. M. Inorg. Chem. 2004, 43, 6863–6865.

(42)

Chiang, M. H.; Dzielawa, J. A.; Dietz, M. L.; Antonio, M. R. J. Electroanal. Chem. 2004, 567, 77–84.

(43)

Teague, C. M.; Li, X.; Biggin, M. E.; Lee, L.; Kim, J.; Gewirth, A. A. J. Phys. Chem. B 2004, 108, 1974–1985.

(44)

Gao, L.-H.; Wang, K.-Z.; Wang, L.-Y. J. Nanosci. Nanotechnol. 2011, 11, 9861–9864.

(45)

Song, W.; Chen, X.; Jiang, Y.; Liu, Y.; Sun, C.; Wang, X. Anal. Chim. Acta 1999, 394, 73–80.

(46)

Hassan, S. S.; Liu, Y.; Sirajuddin; Solangi, A. R.; Bond, A. M.; Zhang, J. Anal. Chim. Acta 2013, 803, 41–46.

(47)

Ding, L.; Liu, Y.; Guo, S.-X.; Zhai, J.; Bond, A. M.; Zhang, J. J. Electroanal. Chem. 2014, 727, 69–77.

(48)

Skunik, M.; Kulesza, P. J. Anal. Chim. Acta 2009, 631, 153–160.

(49)

Li, L.; Li, W.; Sun, C.; Li, L. Electroanalysis 2002, 14, 368–375.

37 ACS Paragon Plus Environment

ACS Catalysis

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

TABLE OF CONTENTS

38 ACS Paragon Plus Environment

Page 38 of 39

Page 39 of 39

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

ACS Catalysis

38x19mm (300 x 300 DPI)

ACS Paragon Plus Environment