Iâ Equilibrium in the Photocatalysis of Oxygen Reduction

Subscriber access provided by UNIV OF TECH SYDNEY

Letter 2

-

On the Role of the I/I Equilibrium in the Photocatalysis of Oxygen Reduction at Ag/AgI. Lukasz Sztaberek, Jenny Kang, William Chen, Francesco F Caruana, Kateryna Huz, Jacob Robinson, John Amato, Christina N Daniels, and John J. McMahon ACS Energy Lett., Just Accepted Manuscript • DOI: 10.1021/acsenergylett.6b00191 • Publication Date (Web): 14 Jul 2016 Downloaded from http://pubs.acs.org on July 20, 2016

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

Page 1 of 23

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

ACS Energy Letters

On the Role of the I2/I- Equilibrium in the Photocatalysis of Oxygen Reduction at Ag/AgI.

Lukasz Sztaberek2, Jenny Kang1, William Chen1, Francesco F. Caruana1, Kateryna Huz1, Jacob Robinson1, John Amato1, Christina N. Daniels1, John J. McMahon1*. 1

Department of Chemistry Fordham University, Bronx NY, 10458, 2Department of

Environmental Control, New York City College of Technology, Brooklyn, NY 11201

Abstract

Photocatalysis of the oxygen reduction reaction at a silver electrode coated with silver iodide is examined with particular attention to the role of the I2/I- equilibrium in the mechanism of catalysis. We demonstrate the dependence of the photovoltage on the standard reduction potentials for the O2/H2O, and the I2/I- equilibria. Iodide, present only via photon-driven charge transfer from silver into the AgI film, oxidizes to iodine in coupling with the oxygen reduction reaction. The photovoltage at the cathode is found tied to that redox couple. The photovoltage is nearly pH-independent owing to a balance between proton consumption during oxygen reduction

 [I ]  and shifting ratio  22  . The catalytic cycle is maintained by regeneration of AgI by the  aI −  thermodynamically allowed reaction of iodine with silver.

1

ACS Paragon Plus Environment

ACS Energy Letters

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

The exchange current density for the oxygen reduction reaction is very small (near 2 x 10-7 A cm-2 at Pt) and limits the power output of fuel cells. As early as 19731 application of the iodine/iodide couple, with its high exchange current density2 near 0.4 A cm-2, was suggested as an intermediate to photocatalysis of the oxygen reduction reaction in a fuel cell. Exploiting the high exchange current advantage of iodine reduction we might expect improved current density from a fuel cell with the iodine/iodide couple controlling the potential at the cathode. However, with the standard reduction potential for iodine in water at +0.536 V vs. NHE compared with the 1.229 V for oxygen reduction the loss of cell voltage would work against overall power output of the hydrogen-iodine-oxygen fuel cell compared with a traditional fuel hydrogen-oxygen fuel cell. In 1975 Schlatter3 et al further precluded the practical application of the hydrogen-iodine-oxygen fuel cell especially in space using calculations that considered weight increases that would accompany use of concentrated iodide solutions and a separate ultraviolet photochemical reactor to oxidize the iodide to iodine.

2


Page 2 of 23

Page 3 of 23

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

ACS Energy Letters

In this report, we reexamine the utility of iodine reduction as a means of catalyzing the oxygen reduction reaction in an aqueous electrolyte that is iodide free. In place of high concentrations of iodide in the supporting electrolyte we use a cathode that is comprised of a silver electrode onto which we deposit a film of silver iodide. The silver iodide film is created either by dipping the silver in an aqueous solution of iodine or by oxidation of the silver in an aqueous solution of iodide. In 2008, we4 showed that the open circuit voltage at an Ag/AgI cathode increased by as much as +400 mV or more upon irradiation of the Ag/AgI electrode in an oxygen-saturated electrolyte. At a fixed applied voltage we observed large increases in the oxygen reduction current upon irradiation. There, we proposed that the light initiates electron transfer from silver into the AgI film making it a source of reducing equivalents for dissolved oxygen diffusing to the surface of the cathode. The absence of photocurrent in nitrogen-saturated electrolyte and the increase in photocurrent with increasing oxygen content of the electrolyte confirms that the observed photocurrent was associated with oxygen reduction. Similar visible light photoreduction action spectrum for different adsorbates suggests that light absorption and subsequent electron transfer likely follows a pathway involving excitation of metallic resonances, perhaps localized surface plasmon resonance LSPR in roughness structures at the surface or silver-adsorbate charge transfer. The action spectrum revealed apparent resonance with both the silver surface plasmon resonance near 400 nm and a localized surface plasmon or charge transfer resonance above 500 nm. Analogous LSPR excitation and charge transfer has been proposed to account for surface enhanced Raman5 intensities and for photocatalytic activity of silver and gold nanoparticles surrounded by Cu2O shells6,7, for example. Visible light electron transfer from silver into the AgI film and subsequent electroreduction of oxygen according to Eq. (1) replaces the ultraviolet oxidation of iodide in the separate photochemical reactors of Schlatter3 and Razumney1. 3


ACS Energy Letters

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

(1)

Page 4 of 23

.

Oxygen reduction is catalyzed by coupling to the oxidation of surface iodide I- to molecular iodine I2. Molecular iodine is subsequently reduced in the dark at the silver surface in the thermodynamically favored regeneration of AgI according to reaction (2) and completing the catalytic cycle. (2)

.

The reductive properties of iodide-modified silver colloids8 and silver nanoparticles9 have been previously recognized where at low temperature the presence of a surface-bound I2 species was identified10 by resonance Raman of iodide-modified nanostructured silver surfaces even in the absence of oxygen. Iodide has also been shown to be catalytic toward oxygen reduction at platinum microparticles on gold, glassy carbon, and indium tin oxide.11 The mixed iodine/iodide/oxygen coupling at the silver cathode presents renewed opportunities for overcoming the ever-present burden posed by the slow kinetics of the oxygen reduction reaction in an economically sound fuel cell. AgI surface film preparation Polycrystalline silver electrodes were polished with progressively smaller aluminum oxide powders (Buehler) to 0.05 micron. The polished electrodes were rinsed with distilled/deionized water and subjected to electrochemical cleaning by polarizing the electrodes to -2.0 V vs. Ag/AgCl in 0.01 M aqueous perchloric acid for ten minutes using a Princeton Applied Research Model 263A Potentiostat/Galvanostat. Cyclic voltammetry on the cleaned electrodes identified a wide electrochemical window bordered anodically by silver oxidation and cathodically by 4


Page 5 of 23

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

ACS Energy Letters

reduction of solvent protons. Films of varying depth of AgI at a polycrystalline silver substrate were realized by moving the clean silver electrode into 0.5 M potassium iodide and polarizing to +0.5 volts vs. Ag/AgI. Anodization continued until desired charge passed (ranging from 0.025 C/cm2 to 1.0 C/cm2) and concomitant film thicknesses were attained at which point the cell circuit was opened to stop the silver oxidation. The electrode surface appeared slightly brown in color following oxidation. Preparation of Ag/AgBr and Ag/AgCl electrodes followed analogous oxidation of silver in potassium bromide and potassium chloride, respectively, at applied potentials above the respective standard reduction potentials for the silver halide. Equivalent preparation of AgI-coated silver, albeit with ultra thin AgI film thickness, was realized by dipping the silver electrode into a saturated aqueous solution of iodine. The AgI-coated silver electrode was then removed and rinsed with distilled/deionized water to remove any unbound iodide or iodine before transfer to a dilute perchloric acid solution for photoresponse measurements. Photoresponse measurements The photovoltage response to light was measured using galvanostatic mode on a PAR Model 263A potentiostat/galvanostat wherein the open circuit voltage was recorded over time while a shutter opened admitting light onto the electrode surface. Equal periods of light off/light on/light off were applied. The time frames measured ranged from 12 seconds overall (4 sec light off/4 sec light on/4 sec light off) to 1800 seconds overall. All voltages are referenced to a leak-free Ag/AgCl reference electrode (Harvard Apparatus) in a dilute perchloric acid (e.g. 0.01 M HClO4 for a pH 2.0 solution). Light excitation was afforded by either a Spectra Physics 2020-05 argon ion laser at 5145 Å or broadband excitation via a 300 W Oriel Model 66013 xenon arc lamp with

5


ACS Energy Letters

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

Model 68811 power source. Oxygen or nitrogen saturation of the supporting electrolyte was afforded by continuous bubbling of the gas through the electrolyte at 1 atm. Photovoltage observed at silver/silver halide cathodes Following Razumney1 we searched for evidence of redox coupling between the AgI surface film and reduction of dissolved oxygen at the electrode surface. We report the photovoltage observed over time during irradiation with a 300 watt xenon arc source for an Ag/AgI electrode and compare with that for Ag/AgBr and Ag/AgCl electrodes. Fig. 1 reports galvanostatic measurements of the open circuit voltage before, during, and after irradiation of each electrode. The open circuit voltage for Ag/AgI jumps from an initial voltage in the dark of -0.014 V to +0.400 V during irradiation for a photovoltage of +0.414 V. Comparatively, Ag/AgBr demonstrates a smaller photovoltage of +0.083 V while Ag/AgCl shows a negative photovoltage of -0.070V.

6


Page 6 of 23

Page 7 of 23

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

ACS Energy Letters

Fig. 1

Open circuit voltage of the Ag/AgI, Ag/AgBr, and Ag/AgCl electrodes in 0.01 M HClO4. 300 Watt unfiltered xenon arc source. The orange zone is light off, the yellow zone is light on, and the green zone is again light off.

The observed photovoltages for the silver/silver halides are found to be proportional to the difference in standard reduction potentials between E°(O 2 /H 2O) = 1.229 V and E°(X2 /X- ) given in Table 1.

7


ACS Energy Letters

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

Page 8 of 23

Table 1. Observed photovoltage for Ag/AgI, Ag/AgBr, and Ag/AgCl and proportionality to the difference between E°(O 2 /H 2O) = 1.229 V and E°(X2 /X- ) where X is I, Br, or Cl. Ag/AgX electrode Ag/AgI

Observed photovoltage +0.414 V

E°(X2 /X- )

E°(O2 /H 2O) - E°(X 2 /X- )

0.536 V

+ 0.693 V

Ag/AgBr

+0.083 V

1.0873 V

+ 0.1417 V

Ag/AgCl

-0.070 V

1.3583 V

- 0.1293 V

A plot of the observed photovoltages as a function of  E°(O 2 /H 2O) - E°(X 2 /X- )  shown in Fig. 2 is linear with a near zero intercept.

Fig. 2

Proportionality of photovoltages observed for Ag/AgI, Ag/AgBr, and Ag/AgCl to the difference between E°(O 2 /H 2O) = 1.229 V and E°(X2 /X- ) . Slope = 0.59 PV / ∆E° .

Open circuit voltage during the post-irradiation period With the intense 300 Watt xenon arc source we found that the Ag/AgI electrode surface darkened during the 15-minute run and that the open circuit voltage did not return to initial 8


Page 9 of 23

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

ACS Energy Letters

values following irradiation. Instead, the voltage settled above +0.270 V vs. Ag/AgCl though slow deterioration over time of the dark open circuit potential was observed as seen in the green section of Fig. 1. In contrast, the open circuit voltages for both Ag/AgBr and Ag/AgCl returned quickly to near the original dark voltage in the green post-irradiation period. When a freshly prepared Ag/AgI surface is irradiated with a lower power 50 mW argon ion laser irradiation at 5145 Å the open circuit potential is observed to more slowly and asymptotically approach the maximum photovoltage attaining only near +0.17 V maximum (compared to + 0.4 V attained in Fig. 1 during intense irradiation) over an elongated 10 minute irradiation period (Fig. 3). The photovoltage relaxes quickly back toward the initial condition during the green dark post-irradiation period.

Fig. 3

Photovoltage at Ag/AgI under 150 mW (red), 125 mW (magenta), 100 mW (blue), and 50 mW (black) argon ion laser irradiation. 0.01 M HClO4, oxygen saturated.

As the laser power is increased the maximum photovoltage also increases. The photovoltage again decays monotonically during the green dark post-irradiation period unless the maximum photovoltage attained during the irradiation period exceeds about +0.235 V vs. Ag/AgCl, or +0.457 V vs. NHE. When the photovoltage attained exceeds +0.235 V, as observed during the

9


ACS Energy Letters

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

150 mW red scan in Fig. 3, then we observe a temporary shoulder in the return to the original open circuit voltage during the green dark post-irradiation period. When we prepared the electrode by briefly (5 minutes) dipping a freshly polished silver electrode into an aqueous solution saturated in iodine, instead of oxidizing in aqueous KI, and irradiated with intense 300 W Xenon arc radiation we observed photovoltage behavior nearly equivalent to the 150 mW results shown in Fig. 1. Shown in red scan of Fig. 4, the maximum voltage again is slowly approached but exceeds +0.235 V and, in the post-irradiation green period, we again see a characteristic temporary delay in the return to the original open circuit voltage. When the 0.01 M HClO4 test solution was contaminated with slight transfer of iodine from the dip solution we observed post-irradiation behavior equivalent to the high intensity experiment shown in Fig. 1. That is, the maximum photovoltage was attained quickly and that voltage decayed but stabilized near 0.275 V in the post-irradiation green period. Additionally, we observe a voltage climb during the pre-irradiation orange period.

Fig. 4

Open circuit voltage of an Ag electrode that was previously dipped in saturated aqueous iodine then transferred to oxygen-saturated 0.01 M HClO4. 300 Watt unfiltered xenon arc source. Red scan: Rinsed of residual iodine from dip solution before transfer to the 0.01 M HClO4 solution. Blue scan: Not rinsed of residual iodine prior to transfer to 0.01 M HClO4, i.e., contaminating the 0.01 M HClO4 test solution with a small amount of iodine. 10


Page 10 of 23

Page 11 of 23

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

ACS Energy Letters

In discussing the above results, we begin by noting that proportionality between the observed photovoltages for Ag/AgI, AgBr, and Ag/AgCl and the difference  E°(O 2 /H 2O) - E°(X 2 /X- )  evidences redox coupling between oxygen reduction and halide oxidation. Both iodine I2 and bromine Br2 are less easily reduced than oxygen while chlorine Cl2 is more easily reduced than oxygen. The observed photovoltages for each electrode reflect this relative ease of reduction. Redox coupling between iodide oxidation and oxygen reduction at Ag/AgI yields a net reaction at the cathode according to

The source of iodide ion I- is photon-assisted electron transfer from silver into the AgI film according to Eq. (4).

(4)

In the dark, the open circuit voltage is defined largely by the Ag/AgI equilibrium, with a small influence from the uncatalyzed reduction of oxygen at the cathode. During irradiation, the voltage at the photocathode can be calculated from the Nernst equation for the net redox reaction given in Eq. (3), i.e.,

Enet = EOo / H O − E Io / I − − 2

2

2

(5)

Enet = 1.229 − 0.536 −

RT  [I 2 ]  ln   2F  aI2− aH2 + 

RT  [I 2 ]  ln   2F  aI2− (.01) 2 

at pH 2.0

11


ACS Energy Letters

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

We confirm this assertion by first reviewing observations made during the post-irradiation period. Molecular iodine generated via Eq. (3) appears to persist, at least temporarily, after the light is turned off. The effect of light intensity on the photoevent (Fig. 3) revealed that a voltage shoulder appeared in the green post-irradiation period only when the voltage during irradiation exceeded +0.235 V vs. Ag/AgCl reference (or +0.457 vs. NHE). During irradiation, when the voltage at the cathode exceeds the equilibrium voltage for iodine reduction , any molecular iodine generated in the redox reaction with oxygen according to Eq. (3) will not be reduced back to iodide electrochemically. Reaction of iodine with Ag(s) to reform AgI(s) according to Eq. (2) remains thermodynamically allowed however and undoubtedly occurs during the radiation period. The photon-driven voltages for Ag/AgBr and Ag/AgCl (yellow irradiation zone of Fig. 1) never exceed the equilibrium voltages for bromine and chlorine reduction. Congruently, no delay in the return to the original open circuit voltage is observed for Ag/AgBr and Ag/AgCl because bromine and chlorine would be electrochemically reduced to halide ions at their respective photovoltages. That remnant iodine persists after the light is turned off, leaving the iodine/iodide equilibrium in control of the cathode voltage for Ag/AgI, is verified when the electrode is prepared by dipping in an aqueous iodine solution (Fig. 4). Again a weak shoulder is observed for the dipped Ag/AgI following a voltage near +0.350 V vs. Ag/AgCl (or +0.572 V vs. NHE) during irradiation, well above the iodine/iodide equilibrium voltage. When the perchlorate electrolyte is contaminated with carry-over from the dip solution the voltage shoulder in the post-irradiation period is extended in time as would be expected to accompany a larger iodine presence. The extended shoulder matches that observed for the thickly coated Ag/AgI electrode in Fig. 1 where again larger remnant iodine that had not yet reacted with silver metal to regenerate AgI might be

12


Page 12 of 23

Page 13 of 23

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

ACS Energy Letters

expected. Observed decay in the voltage during the green post-irradiation period likely accompanies continued reaction of iodine with silver metal depleting the amount of remnant iodine.

Assuming, therefore, that

is in control of the voltage after the light is

 a 2−  turned off we can solve for the ratio  I  in the Nernst equation for I2/I- using Eq. (6).  [I 2 ]  Applying the voltage observed immediately after the light is turned off in Fig. 1, i.e., 0.270 V vs.

 a 2−  Ag/AgCl or 0.492 V vs. NHE . , we find  I  = 30.77 .  [I 2 ] 

(6)

EI / I − = E I / I − o

2

2

2 2 RT  aI −  RT  aI −  − ln  ln   = 0.536 −  = 0.492 2F  [I 2 ]  2F  [I 2 ] 

 a 2−  We can now apply this experimentally determined ratio  I  = 30.77 to calculate the voltage  [I 2 ]  just before the light is turned off by applying the Nernst equation for the redox couple given by

 a 2−  Eq. (5). Using the ratio  I  = 30.77 determined from the green post-irradiation zone we  [I 2 ]  calculate Enet = 0.619 V from Eq. (5). This calculated voltage nearly matches the maximum voltage observed during the irradiation period of +0.400 V vs. Ag/AgCl (Fig. 1) or +0.622 V vs. NHE confirming the redox coupling of oxygen reduction and iodide oxidation . Interestingly, although iodide can only come from photo-assisted electron transfer described by Eq. (4) there appears to be no influence of the Ag/AgI equilibrium voltage on the open circuit

13


ACS Energy Letters

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

Page 14 of 23

voltage observed during irradiation. The overall reaction with light is given by Eq. (1) .

The steepness of the climb of the open circuit voltage upon irradiation of the electrode reflects the kinetics of the approach to the new equilibrium. With light as a reactant in Eq. (1) the more intense the source the quicker the attainment of the new equilibrium voltage (Fig. 3). Equivalently, establishment of equilibrium requires the presence of and back reaction of I2, such that the presence of contaminant I2 from the dip solution also speeds up attainment of the new equilibrium voltage and increases the steepness of the initial response to light (Fig. 4). The pH dependence of the open circuit voltage at irradiated Ag/AgI is shown in Fig. 5. In spite of consuming protons in the overall reaction above, the open circuit voltage at the Ag/AgI photocathode appears largely unchanged with pH compared to the usual paired pH dependence of the oxygen and hydrogen reactions. The absence of pH dependence at the Ag/AgI

 [I ]  photocathode can be understood by considering the  2 2 2  term in Eq. (5). Reduction in aH +  aI − aH +  on moving toward basic pH shifts the equilibrium in Eq. (3) to the left and lowers the Nernstian voltage of the cell. Concomitantly, the equilibrium shift of Eq. (3) to the left with base decreases

 [I ]  the  22  ratio increasing the Nernstian voltage of the cell according to Eq. (5). That is, the two  aI −  shifts counter one another and the open circuit voltage at the photocathode remains unchanged with pH. The resultant cell voltage [E(photocathode)-E(hydrogen anode)] increases with pH because the cathode voltage is unchanged but the voltage at the normal hydrogen anode decreases with pH. That is, while the cell voltage from a normal hydrogen/oxygen fuel cell is pH

14


Page 15 of 23

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

ACS Energy Letters

independent, owing to equivalent pH dependence of the voltages at the cathode and anode, the cell voltage for the photocathode-containing cell increases with pH.

Fig. 5 Measured open circuit voltage vs. NHE for our Ag/AgI photoelectrode () compared with the calculated open circuit voltages for the oxygen electrode ( ) and the hydrogen

electrode ().

We report evidence that the equilibrium established at the Ag/AgI electrode under irradiation involves a redox balance between the oxygen reduction and iodide oxidation to iodine. The open circuit voltage of the photocathode appears pH-independent, which, when combined with the anode voltage yields a cell voltage that increases with pH. Iodine generated in the photoreaction can thermodynamically recombine with metallic silver to rebuild AgI, thereby regenerating the photocatalyst. However, this recombination does not appear to be able to keep up with the iodine generated during the photon-driven redox reaction. Congruently, depletion of AgI layer might be expected to limit the overall current output of a fuel cell constructed with the Ag/AgI photocathode at least at room temperature. 15


ACS Energy Letters

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

While this report is devoted to examination of the open circuit voltage at the Ag/AgI photocathode study of the photocurrent as a function of AgI film thickness might clarify the source of the first reducing equivalent, whether from the bulk silver substrate or from silver nanoparticles that could be present as a result of photographic reduction of the silver halide. Answering this question is not easy however. We have observed in general that higher photocurrents are observed with thicker films of AgI. Where a thicker AgI film might limit light access to the silver substrate and thus reduce the photocurrent if the source of reducing equivalents is the silver substrate, silver nanoparticles may then be flagged as the source of electrons. However, we know now that iodine produced in the photoreaction is not immediately recombined in redox with Ag° to reform the AgI film and cannot keep up with the photoevent generating Ag° and iodine. It should therefore be expected that the photocurrent from a thin film of AgI would be limited because AgI becomes depleted while a thick film would suffer less depletion of available AgI. It therefore becomes difficult to deconvolute the first reducing equivalent as originating from silver nanoparticles vs. silver substrate based upon the film thickness alone. The UV/vis action spectrum reported by McMahon et al.4 where two resonances were observed in the action spectrum, the surface plasmon and localized plasmon resonances, suggests that both the substrate and silver nanoparticles may serve as sources of reducing equivalents.

Acknowledgments J.M. acknowledges financial support from the Fordham University Office of Research. L.S. acknowledges financial support from the Dean’s Office of the School of Technology and Design at the New York City College of Technology.

16


Page 16 of 23

Page 17 of 23

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

ACS Energy Letters

References 1.

Razumney, G.; Wroblowa, H. S. The Photochemical Oxidation of Iodide to Iodine in the Presence of Oxygen. Energy Conversion 1973, 13, 29-37.

2.

Dane, L. M.; Janssen, L. J. J.; Hoogland, J. G. The Iodine/Iodide Redox Couple at a Platinum Electrode. Electrochimica Acta 1968, 13, 507-518.

3.

Schlatter, A.; Plattner, E.; Javet, Ph. Evaluation of the Performances of an HydrogenIodine Fuel-Cell. Energy Conversion 1975, 14, 43-46.

4.

McMahon, J. J., Barry, M., Breen, K. J., Radziwon, A. K., Brooks, L. D. and Blair, M. R. Photocatalysis of the Oxygen Reduction Reaction at Adsorbate-Covered Silver. J. Phys. Chem. C 2008, 112, 1158-1166.

5.

Yamamoto, Y. S.; Itoh, T. Why and How Do the Shapes of Surface-Enhanced Raman Scattering Spectra Change? Recent Progress from Mechanistic Studies. J. Raman Spec. 2016, 47, 78-88.

6.

Li, J.; Cushing, S. K.; Bright, J.; Meng, F.; Senty, T. R.; Zheng, P.; Bristow, A. D.; Wu, N. Ag@Cu2O Core-Shell Nanoparticles as Visible-Light Plasmonic Photocatalysts. ACS Catalysis 2012, 3, 47-51.

7.

Cushing, S. K.; Li, J.; Meng, F.; Senty, T. R.; Suri, S.; Zhi, M.; Li, M.; Bristow, A. D.; Wu, N. Photocatalytic Activity Enhanced by Plasmonic Resonant Energy Transfer from Metal to Semiconductor. J. Am. Chem. Soc. 2012, 134, 15033-15042.

8.

Sibbald, M. S.; Chumanov, G.; Cotton, T. M. Reduction of Cytochrome c by HalideModified, Laser-Ablated Silver Colloids. J. Phys. Chem. 1996, 100, 4672-4678.

9.

Sibbald, M. S.; Chumanov, G.; Cotton, T. M. Reductive Properties of Iodide-Modified Silver Nanoparticles. J. Electroanal. Chem. 1997, 438, 179-185.

17


ACS Energy Letters

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

10.

Chumanov, G.; Sibbald, M. S.; Cotton, T. M. Low-Temperature Resonance Raman Scattering from Iodide Adsorbed on Nanostructured Silver Surfaces. J. Phys. Chem. B 1998, 102, 10836-10841.

11.

Su, B.; Thiagarajan, S.; Chen, S. The Interaction of Iodide Film with Platinum Microparticles on Different Electrode Materials for Various Electrocatalytic Reactions. Electroanalysis 2008, 20, 1987-1995.

18


Page 18 of 23

Page 19 of 23

ACS Energy Letters

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 Paragon Plus Environment

ACS Energy Letters


Page 20 of 23

Page 21 of 23

ACS Energy Letters


ACS Energy Letters


Page 22 of 23

Page 23 of 23

ACS Energy Letters


Iâ Equilibrium in the Photocatalysis of Oxygen Reduction

Recommend Documents

Iâ Equilibrium in the Photocatalysis of Oxygen Reduction