1

Electronic Supplementary Information

Water oxidation catalysis – role of redox and structural dynamics

in biological photosynthesis and inorganic manganese oxides

Ivelina Zaharievaa*, Diego González-Floresa, Baraa Asfaria, Chiara Pasquinia,

Mohammad Reza Mohammadia, Katharina Klingana, Ivo Zizakb, Stefan Loosa, Petko

Cherneva, Holger Daua*

a Freie Universität Berlin, Fachbereich Physik, Arnimallee 14, 14195 Berlin

b Helmholtz-Zentrum-Berlin, Institute for Nanometre Optics and Technology, Albert-Einstein-

Straße 15, 12489 Berlin

*corresponding authors: ivelina.zaharieva@fu-berlin.de, holger.dau@fu-berlin.de

Electronic Supplementary Material (ESI) for Energy & Environmental Science.This journal is © The Royal Society of Chemistry 2016

2

Materials and Methods

Freeze-quench X-ray absorption experiment. Mn oxides deposited on thin glassy carbon

were frozen after 3 min exposure to the selected potential (in 0.1 M phosphate buffer adjusted

to pH 7). The electrolyte-exposed samples were frozen inside of an adapted electrochemical

cell, which subsequently served as a sample holder in the low-temperature XAS measurements;

during the freezing the voltage between the working and counter electrodes was kept constant.

For details on this freeze-quench, quasi-in situ approach 1, see Fig. S1.

EXAFS Fourier-transforms and EXAFS simulations. After transition from an energy to a

wave-vector scale (k-scale) and weighting by k3, the Fourier transforms shown in Fig. 2 in the

main text and Fig. S5b were calculated for k ranging from 2.6 to 12.2 Å-1 (same k-range for

PSII and all oxide films). Cosine windows covering 10% at the low-k and high-k side of the

spectra were applied before calculation of the Fourier transforms. The absolute value of the

Fourier-transform amplitudes is shown.

Simulations of the k3-weighted EXAFS spectra of the catalytically active Mn-only oxide

(MnCat) and the inactive Mn oxide were performed as described in Refs.2, 3 using the k-range

from 2 to 14 Å-1. In order to minimize the number of free parameters and to improve the

significance of the simulation results, a joint fit approach was used where the interatomic

distances were kept the same for all spectra of one data set (the same oxide type at various

potentials). In addition, the Debye-Waller parameters for all oxygen shells were kept the same

for all potentials. The Debye-Waller parameters for all manganese shells were fixed to 0.063

Å. Also, the sum of the EXAFS coordination numbers in the two oxygen shells was fixed to 6.

The EXAFS coordination numbers for the Mn-Mn vectors were varied freely. For the

catalytically active oxide, the short Mn-Mn distance (~ 2.9 Å) was simulated with two separate

shells. For the inactive material, only one shell was used because splitting of this distance into

two shells did not result in significant improvement of the fit quality, but in an increased

number of fit parameters and thus increased uncertainty in the parameter values. This

observation suggests that the short Mn-Mn distances are more uniform in the inactive Mn

oxide, as expected from the better ordered structure. The simulation results are presented in

Table S1. The used amplitude reduction factor (S02) was 0.7. For calculation of the Fourier-

filtered error (described in Ref.4), the range from 1 to 3.5 Å on the reduced distance scale was

used.

Time-resolved in situ X-ray absorption measurements. For an overview scheme of the time-

resolved in situ experiment, see Fig. S9. The measurements were performed at 20 °C at the

BESSY II synchrotron radiation source at beamline KMC-3. A silicon (111) double-crystal

monochromator was used for selecting a fixed X-ray excitation energy (6553.3 eV).

Manganese oxide deposited on thin glassy carbon was attached as a window to the wall of a

home-made Teflon cell filled with 0.1 M phosphate buffer (adjusted to pH 7) and placed in the

path of the X-ray beam. The size of the area illuminated by the X-ray beam was approximately

11 x 3 mm. All X-ray absorption signals were collected in fluorescence mode. The excited Mn

X-ray fluorescence passed through the glassy carbon and then through a Cr foil (10 μm)

shielding against scattered X-rays. The fluorescence was monitored perpendicular to the

3

incident beam by a scintillation detector (19.6 cm2 active area, 51BMI/2E1-YAP-Neg, Scionix;

shielded by 2 μm Al foil against visible light). The detector consisted of a scintillating crystal

(YAP) converting X-ray photons into visible light (~50% efficiency) detected by a fast

photomultiplier operated at 1.1 kV. The current signal from the photomultiplier passed through

a 1 MΩ resistor for current-to-voltage conversion and was fed into a low-noise amplifier

(Stanford Research Systems, model SR560; 10 Hz low-pass filtering, 6 dB/oct; amplification

factor of 20). The amplified signal was finally recorded with a time resolution of 10 ms by the

potentiostat (Biologic SP-300) that also operated the electrochemical cell and recorded the

signal from an ionization chamber monitoring the intensity of the incoming X-ray beam. Short-

range Mn K-edge absorption spectra of the oxide film on the glassy carbon electrode were

recorded immediately before and after each time-resolved experiment, in order to facilitate

normalization of time-resolved data and to ascertain that there was no significant film

dissolution. In separate X-ray scans, absorption spectra of KMnO4 powder were measured for

precise energy calibration, setting the maximum of the pre-edge absorption peak of the KMnO4

to 6543.3 eV.

4

Figure S1. Freeze-quench (quasi-in situ) approach for investigation of Mn(Ca) oxide film by

low-temperature X-ray absorption spectroscopy (XAS). (A) Sample cell made of transparent

PVC (33 mm x 19 mm x 2 mm) with a window (14 mm x 14 mm) and two holes drilled for

insertion of a platinum wire, which can serve as a counter electrode (CE). The window is closed

by the working electrode (WE) consisting of a glassy carbon sheet (thickness of 100 µm)

supported by Kapton® tape. Copper tape was connected to the glassy carbon, facilitating

electric contact by using crocodile clamps (not shown). (B) The Mn(Ca) oxide were

electrodeposited on the working electrode as described in the Materials and Methods section

of the main article. After electrodeposition of the Mn(Ca) oxide films, ~250 μl of Mn-free

electrolyte (0.1 M potassium phosphate, pH 7.0) were filled in the cell and a mercury sulfate

REF

WE: Glassy carbon

PVC frame

copper tapeconnected

to WE

CE: Pt wire

WE: Glassy carbon

A

CE: Pt wire

A B

Electrolytecovering oxidefilm and fillingcavity(Pi, pH 7)

Oxide film deposited on WE

V

XAS sample holder before EC Operation as 3-electrode electrochemical cell

C D

Oxide film and electrolyte1st frozen in liquid N2 (l)2nd WE/CE disconnected by cutting

under liquid nitrogen3rd XAS measurement at 20 K

A

V

WE: ITO on PET

CE: Pt wire

Electrolyte(Pi, pH 7)Mn oxide film

deposited on WE

Operation as 2-electrode cell during freezing Cell for XAS measurements

5

reference electrode (REF, Hg/Hg2SO4/K2SO4 +650 mV vs. NHE) was immersed into the

electrolyte as shown in the scheme. Then the desired potential was applied for 2 minutes and

the voltage, VWE/CE, between the working electrode and the counter electrode was measured.

(C) The reference electrode was removed and VWE/CE was applied between the working

electrode and counter electrode (meaning operation in a two-electrode configuration). (D) After

1 minute the whole sample cell was quickly frozen by spilling liquid nitrogen over the cell

assembly. Only after the whole cell had been frozen, the electrode clamps were detached and

the copper tape was cut off (under liquid nitrogen). Finally, the platinum wire was removed

and the samples were stored in liquid nitrogen until use in the XAS measurements.

6

Figure S2. Cyclic voltammograms (CVs) of catalytically active Mn oxide (MnCat, red line)

and inactive Mn oxide (Inactive, blue line). Scan rate 20 mV/s; the second CV of a series is

shown.

Figure S3. Calibration line relating Mn K-edge energy position and the formal Mn oxidation

state. Edge energies of the commercial Mn compounds MnIIO, -MnIII2O3 and -MnIVO2 were

used for calibration. Using additional Mn oxide references results in a very similar calibration

curve (see ref. 2). The linear regression line results in the following calibration equation for

estimating the mean Mn oxidation state: zox = (Eedge – 6538.4 eV)/3.8. All edge energies were

determined using an ‘integral method’, which is clearly less sensitive to edge-shape changes

than other methods.4

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

0.9 1 1.1 1.2 1.3 1.4

Cu

rren

t d

ensi

ty [

mA

/cm

2]

Potential vs. NHE [V]

ITO-coated glass electrode

MnCat

Inactive

6545

6546

6547

6548

6549

6550

6551

6552

6553

6554

6555

1.5 2.0 2.5 3.0 3.5 4.0 4.5

Ed

ge

en

erg

y [

eV

]

Mn oxidation state

7

Figure S4. k3-weighted EXAFS spectra (top) and their Fourier transforms (bottom) of the

catalytically active Mn oxide (MnCat, left) and the inactive oxide (right) equilibrated at

different potentials (vs. NHE, pH 7). Black lines represent experimental data, red lines –

simulated spectra. The simulation parameters are listed in Table S1.

-70

-60

-50

-40

-30

-20

-10

0

10

20

2 3 4 5 6 7 8 9 10 11 12 13 14

Inactive

0.45 V

0.70 V

0.85 V

1.15 V

1.35 V

1.45 V

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

10

20

2 3 4 5 6 7 8 9 10 11 12 13 14

0.45 V

0.85 V

0.95 V

1.15 V

1.20 V

1.35 V

1.45 V

MnCat

0.70 V

k [Å-1] k [Å-1]

k3(k

)[Å

-3]

k3(k

)[Å

-3]

-35

-30

-25

-20

-15

-10

-5

0

5

10

0 1 2 3 4 5 6 7

MnCat

0.45 V

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.35 V

1.45 V-25

-20

-15

-10

-5

0

5

10

0 1 2 3 4 5 6 7

Inactive Mn oxide

0.45 V

0.70 V

0.85 V

1.15 V

1.35 V

1.45 V

Reduced distance [Å]Reduced distance [Å]

FT

of

EX

AF

S

FT

of

EX

AF

S

8

Table S1. Simulation parameters for the EXAFS spectra of the catalytically active Mn-only

oxide (MnCat) and of the inactive Mn oxide. The interatomic distances, R, for each shell are

shown in bold (in Å) and were kept the same at all potentials. The rest of the values show the

coordination numbers at the respective potential. The errors representing 68% confidence

intervals of the respective fit parameters are shown in parentheses. For the MnCat, the Debye-

Waller parameter () for the two oxygen shells was 0.065 0.001 Å (at all potentials) and the

filtered R-factor was 10.9%. For the inactive Mn oxide, the Debye-Waller parameter () for

the two oxygen shells was 0.059 0.001 Å and the filtered R-factor 11.4%. The asterisks

indicate that the sum of the EXAFS coordination numbers of the two oxygen shells was forced

to be equal to six; the error is indicated only for one of the two coordination numbers. The

Debye-Waller parameters for all manganese shells were fixed to 0.063 Å.

Mn-O Mn-O Mn-Mn Mn-Mn Mn-Mn Mn-Mn Mn-Mn

catalytically active MnCat

all 1.90 (0.001) 2.28 (0.01) 2.85 (0.01) 2.99

(0.01) 3.16 (0.01) 3.42 (0.01) 3.81 (0.02)

0.45 V 4.7 (0.2)* 1.3* 2.6 (0.4) 1.7 (0.6) 1.1 (0.5) 0.7 (0.4) 0.4 (0.5)

0.70 V 5.0 (0.2)* 1.0* 3.0 (0.4) 2.8 (0.7) 2.0 (0.6) 1.3 (0.4) 0.2 (0.5)

0.85 V 5.3 (0.2)* 0.7* 3.6 (0.4) 3.3 (0.7) 2.8 (0.7) 2.0 (0.5) 0.4 (0.5)

0.95 V 5.4 (0.2)* 0.6* 3.6 (0.4) 3.0 (0.7) 2.6 (0.6) 1.7 (0.4) 0.5 (0.5)

1.15 V 5.4 (0.2)* 0.6* 3.5 (0.4) 2.9 (0.7) 2.3 (0.7) 2.0 (0.4) 0.9 (0.5)

1.20 V 5.5 (0.2)* 0.5* 3.7 (0.4) 3.1 (0.7) 2.7 (0.7) 2.3 (0.5) 0.4 (0.5)

1.35 V 5.4 (0.2)* 0.6* 3.7 (0.4) 3.0 (0.7) 2.3 (0.7) 1.9 (0.4) 0.5 (0.5)

1.45 V 5.6 (0.2)* 0.4* 4.1 (0.4) 3.1 (0.7) 2.2 (0.7) 1.8 (0.4) 0.6 (0.5)

inactive Mn oxide

all 1.90 (0.001) 2.28 (0.01) 2.87 (0.002) 3.14 (0.01) 3.49 (0.01) 3.78 (0.01)

0.45 V 5.0 (0.2)* 1.0* 4.6 (0.2) 0.7 (0.3) 1.7 (0.4) 0.4 (0.5)

0.70 V 5.3 (0.2)* 0.7* 4.8 (0.2) 0.9 (0.3) 1.7 (0.4) 0.7 (0.5)

0.85 V 5.5 (0.2)* 0.5* 4.9 (0.2) 0.9 (0.3) 1.3 (0.4) 0.1 (0.5)

1.15 V 5.7 (0.2)* 0.3* 4.9 (0.3) 1.5 (0.3) 1.5 (0.4) 1.3 (0.5)

1.35 V 5.7 (0.2)* 0.3* 4.6 (0.2) 0.8 (0.3) 1.6 (0.4) 0.9 (0.5)

1.45 V 5.8 (0.2)* 0.2* 4.8 (0.2) 0.7 (0.3) 1.4 (0.4) 1.1 (0.5)

9

Figure S5. Averaged MnO distances calculated from the simulation results presented in Table

S1. Mn oxidation state calculated according to the bond-valence sum rules5 is +3.48 for the

MnCat and +3.61 for the inactive Mn oxide at 0.45 V and +3.86 for the MnCat and +3.95 for

the inactive Mn oxide at 1.45 V.

1.90

1.92

1.94

1.96

1.98

2.00

0.4 0.6 0.8 1 1.2 1.4

Mn

O

bo

nd

le

ng

th [

Å]

Potential vs. NHE [V]

MnCat

Inactive Mn oxide

10

Figure S6. Fourier-isolated EXAFS oscillations corresponding to the MnMn peak (Fourier

isolation from 2.1 Å to 2.9 Å on the reduced distance scale) of the data for the MnCat shown

in Fig. S4. For clarity, only the positive amplitudes of the EXAFS oscillations are shown. In

panel A, the experimental data is presented. In panel B, the simulated oscillations with fixed

Debye-Waller parameter (using the simulation approach presented in Fig. S4 and Table S1) are

shown. Panel C shows the results from simulations in which (i) the coordination numbers for

the two short Mn-Mn distances (2.85 Å and 2.99 Å) were fixed to the values obtained at 0.45 V

and (ii) the Debye-Waller factors for these two shells were freely varied. Only four potentials

are shown for clarity. The experimental data is better modeled with fixed Debye-Waller

parameter (in B), indicating that the observed potential-induced increase in the amplitude of

the second peak is a result rather from an increase in the coordination number than from a

decreased inhomogeneity in the length of the shortest Mn-Mn distances.

0

5

10

15

20

25

30

35

40

2 3 4 5 6 7 8 9 10 11 12 13 14

EX

AF

S a

mp

litu

de

[a

.u.]

Wevenumber [Å-1]

Simulation: fixed coordination number (N)C

0

5

10

15

20

25

30

35

40

2 3 4 5 6 7 8 9 10 11 12 13 14

EX

AF

S a

mp

litu

de

[a

.u.]

Wevenumber [Å-1]

Simulation: fixed Debye-Waller factor (σ)B

0

5

10

15

20

25

30

35

40

2 3 4 5 6 7 8 9 10 11 12 13 14

EX

AF

S a

mp

litu

de [

a.u

.]

Wevenumber [Å-1]

Experimental0.45 V

0.70 V

0.95 V

1.35 V

A

11

Figure S7. XANES spectra (a) and Fourier transformed EXAFS spectra (b) of the MnCa

oxides equilibrated at the indicated potentials (vs. NHE, pH 7). The inset in (a) show the

extracted pre-edge features. For each FT peak, the corresponding structural motif is

schematically shown (Mn, magenta; O, red). In the MnCa oxide, the third FT-peak also

contains contributions from Mn-Ca distances.

0

1

6535 6540 6545 6550 6555 6560 6565

No

rma

lize

d f

luo

res

ce

nc

e

X-ray energy [eV]

0.45 V

0.70 V

0.85 V

0.95 V

1.20 V

1.35 V

1.45 V

0

0.05

6535 6540 6545

3.3 3.8

a

0

2

4

6

8

10

12

0 1 2 3 4 5 6 7

FT

of

EX

AF

S

Reduced distance [Å]

0.45 V0.70 V0.85 V0.95 V1.20 V1.35 V1.45 V

1

2

3

45 6

b

12

Figure S8. Absorption spectra of (a) the catalytically active Mn-only oxide (MnCat), (b) of

the inactive Mn oxide and (c) the absorption 370 nm of the catalytically active Mn-only oxide

(MnCat, red circles) and the inactive film (blue circles) as a function of the applied potential

(vs. NHE at pH 7). The catalyst films were deposited on optically transparent ITO-covered

glass slides and equilibrated at the indicated potentials (vs. NHE, pH 7) for 2 min before

measuring the UV-vis spectra.

0.1

0.2

0.4 0.6 0.8 1.0 1.2 1.4

Ab

so

rban

ce [

OD

]

Potential vs NHE [V]

MnCat

Inactive film

0

0.1

0.2

0.3

300 400 500 600 700 800 900

Ab

so

rba

nc

e [

OD

]

Wavelength [nm]

0.45V

0.70V

0.95V

1.10V

1.20V

1.35V

1.40V

1.45V

0

0.1

0.2

0.3

300 400 500 600 700 800 900

Ab

so

rba

nc

e [

OD

]

Wavelength [nm]

0.45 V

0.70 V

0.95 V

1.10 V

1.20 V

1.35 V

1.40 V

1.45 V

a

b

c

13

Figure S9. Estimation of the initial rate of Mn reduction from the UV-vis time courses of Fig. 4

in (a) active oxide (MnCat) and (b) inactive oxide. To obtain a rough estimate, the time period

for an absorption change corresponding to I/I = 0.5 10-3 was determined graphically. Values

of about 20 ms and 3000 ms are obtained for active and inactive oxide, respectively. (In this

estimation, it is assumed that equal I/I values correspond to roughly similar changes in the

average Mn oxidation state, as suggested by comparison of Fig. S8c with Fig. 3.)

Tra

ns

mit

ed

lig

ht

ins

en

tity

, I(

t)

a

2,347

2,348

2,349

2,350

2,382

2,383

2,384

2,385

2,386

9,9 10 10,1 10,2

Tra

ns

mis

sio

n [

a.u

.]

Time [s]

MnCat

Inactive

1.4

V

OC

P

100 ms

I/I = 0.5 10-3

~20 m

s

I/I = 0.5 10-3

~3000 ms

b

14

Figure S10. Time course of current and open-circuit potential (OCP) measured in parallel to

the UV-vis experiment of Figure S9. Top panels: Time courses of the current for stepping the

electrode working conditions from OCP top application of 1.4 V vs NHE (pH 7). Bottom

panels: Time courses of the electrode potential for switching between application of 1.4 V (vs.

NHE, pH 7 and OCP) conditions. The right panels show the respective time course in form of

a double-logarithmic plot.

0,001

0,01

0,1

1

0,01 0,1 1 10

Cu

rre

nt

de

ns

ity [

mA

/cm

2]

Time [s]

MnCat

Inactive

at 1.4 V

1,2

1,25

1,3

1,35

1,4

0,01 0,1 1 10

Po

ten

tia

l v

s N

HE

[V

]

Time [s]

MnCat

Inactive

at OCP conditions

15

Figure S11. Time-resolved in situ X-ray absorption experiment. (a) Illustration of the

experimental concept. Panel-a shows XANES spectra collected in freeze-quench experiments

for the MnCat equilibrated at different electrode potentials. The X-ray absorption at 6553.3 eV

(arrow) decreases (see inset) because the edge position shifts with increasingly positive

potentials to higher energies, as a result of Mn oxidation. This absorption change at 6553.3 eV

was traced in the time-resolved X-ray experiments. (b) Scheme of the experimental setup for

time-resolved X-ray absorption measurements. Monochromatic synchrotron radiation (6553.3

eV) passes through an ionization chamber (IC, for recording the variations in the incoming

beam intensity) and a photoshutter (PS) before hitting the Mn oxide film (Mn). The Mn oxide

film deposited on a thin, X-ray-transparent glassy carbon (GC) electrode serves as the working

electrode in a 3-electrode electrochemical setup, with reference electrode (RE) and a platinum

mesh serving as counter electrode (CE). The cell is filled with 0.1 M phosphate buffer (pH 7).

The excited Mn X-ray fluorescence passes through the glassy carbon and then through a 10 μm

chromium foil (Cr), which absorbs a major fraction of the scattered X-ray photons. The

fluorescence is monitored perpendicular to the incident beam by a scintillation detector

consisting of a scintillating crystal and a fast photomultiplier operated at 1.1 kV. The signal

from the photomultiplier passes through a 1 MΩ resistor for current-to-voltage conversion and

is fed into an amplifier. The amplified signal is finally recorded by the same potentiostat device

that also operates the electrochemical cell and records the IC signal. (c) Mn oxidation-state

changes of catalytically active Mn oxide (red) and inactive oxide (blue) induced by stepping

the electrode potential between 1.4 V and 0.6 V and tracked by recording the K/ X-ray

fluorescence intensity for excitation at 6553.3 eV (same data as in Fig. 5a, but with extended

time window shown and estimated Mn oxidation states indicated). The Mn oxidation state

estimates were derived from rapid XAS scans before and after the voltage-jump experiments;

the corresponding numbers are less precise than in Figure 3.

0

1

6535 6540 6545 6550 6555 6560 6565

No

rma

lize

d f

luo

res

ce

nc

e

X-ray energy [eV]

0.45 V

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.35 V

1.45 V

0.5

0.8

0.4 0.9 1.4Potential vs NHE [V]

Synchrotron

radiation

6553.3 eV

MnGC

RE

CE

PS

Scintillationdetector

with photo-multiplier

Cr

Amplifier

IC

Potentiostat

Electrochemical

cell

3.4

3.5

3.6

3.7

3.8

3.9

4.0

0.55

0.6

0.65

0.7

0.75

0.8

8 10 12 14 16 18 20 22 24 26 28 30

No

rmalize

d X

-ray f

luo

rescen

ce

Time [s]

MnCat

Inactive

0.6 V 1.4 V 0.6 V

oxidation reduction

Mn

ox

ida

tion

sta

te

a b

c

16

Figure S12. Bode plots for electrochemical impedance spectra obtained at different potentials

(vs. NHE, pH 7) for the catalytically active Mn-only oxide (MnCat, in a and b) and the inactive

Mn oxide (c, d). The experimental data are presented by points, the simulation results according

to the model in Fig. S14 are indicated by solid lines.

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

0.1 1 10 100 1000 10000

Ph

ase []

Frequency [Hz]

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

10

100

1000

10000

0.1 1 10 100 1000 10000

Z [

]

Frequency [Hz]

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

0.1 1 10 100 1000 10000

Ph

ase []

Frequency [Hz]

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

10

100

1000

10000

100000

0.1 1 10 100 1000 10000

Z [

]

Frequency [Hz]

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

a

b

c

d

17

Figure S13. Nyquist plots for electrochemical impedance spectra obtained at different

potentials for the catalytically active Mn-only oxide (MnCat, in a, b, and c) and the inactive

Mn oxide (d, e f). In the different plots, different scales are used. The experimental data are

presented by points; the simulations according to the model in Fig. S14, by solid lines. The

simulations parameters are shown in Fig. S14.

-2000

-1500

-1000

-500

0

0 500 1000 1500 2000

Im(Z

) [

]

Re(Z) []

-30000

-25000

-20000

-15000

-10000

-5000

0

0 5000 10000 15000 20000 25000 30000

Im(Z

) [

]

Re(Z) []

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

-500

-400

-300

-200

-100

0

0 100 200 300 400 500

Im(Z

) [

]

Re(Z) []

-2000

-1500

-1000

-500

0

0 500 1000 1500 2000

Im(Z

) [

]

Re(Z) []

-30000

-25000

-20000

-15000

-10000

-5000

0

0 5000 10000 15000 20000 25000 30000

Im(Z

) [

]

Re(Z) []

0.70 V

0.85 V

0.95 V

1.15 V

1.20 V

1.25 V

1.35 V

1.40 V

-500

-400

-300

-200

-100

0

0 100 200 300 400 500

Im(Z

) [

]

Re(Z) []

a

b

c

d

e

f

18

Figure S14. Parameters for the simulation of the impedance spectra according the equivalent

circuit shown bottom right. ROhm – summed Ohmic resistance of electrolyte and ITO electrode;

Cdl – double-layer capacitance, modelled as a constant-phase element; Cox – capacitance

describing oxidation state changes of the catalyst film (in other contexts also denoted as

pseudo-capacitance), modelled as a constant phase element; Rox – resistance describing the

oxidation state changes of the catalyst film; Rcat – catalytic resistance of the oxide (in other

contexts also denoted as charge-transfer resistance). The values for the Ohmic resistance were

fixed during the simulations (46.5 for the MnCat and 53 for the inactive Mn oxide). The

double layer capacitance (Cdl) and the oxidation capacitance (Cox) were calculated according

to the given equation (following the Brug conversion rules6), where α is the phase parameter

of the constant phase element (with values between 0.75 and 1 in all simulations) and Q is the

capacitance parameter of the constant phase element in units of Fs(α-1). The formal uncertainty

ranges for the fit parameters were below 4%.

ROhm Rox Rcat

Cdl

Cox

1

10

100

1000

10000

100000

0.6 0.8 1.0 1.2 1.4

Ro

x[

]

Potential vs. NHE [V]

RoxMnCat

Inactive Mn oxide

10

100

1000

10000

100000

1000000

0.6 0.8 1.0 1.2 1.4

Rcat[

]

Potential vs. NHE [V]

Rcat

0.001

0.01

0.1

1

10

0.6 0.8 1.0 1.2 1.4

cd

l, c

ox

[mF

]

Potential vs. NHE [V]

Cox

Cdl

19

References

1. M. Risch, F. Ringleb, M. Kohlhoff, P. Bogdanoff, P. Chernev, I. Zaharieva and H. Dau,

Energy Environ. Sci., 2015, 8, 661-674.

2. I. Zaharieva, P. Chernev, M. Risch, K. Klingan, M. Kohlhoff, A. Fischer and H. Dau,

Energy Environ. Sci., 2012, 5, 7081-7089.

3. M. Wiechen, I. Zaharieva, H. Dau and P. Kurz, Chem. Sci., 2012, 3, 2330-2339.

4. H. Dau, P. Liebisch and M. Haumann, Anal. Bioanal. Chem., 2003, 376, 562-583.

5. I. D. Brown and D. Altermatt, Acta Crystallographica Section B, 1985, 41, 244-247.

6. G. J. Brug, A. L. G. van den Eeden, M. Sluyters-Rehbach and J. H. Sluyters, J.

Elelectroanal. Chem., 1984, 176, 275-295.