S1

Supporting Information

Electrocatalytic Water Oxidation at Quinone-on-Carbon: A Model

System Study

Yangming Lin,[a] Kuang-Hsu Wu,[b] Qing Lu,[a] Qingqing Gu,[a, c] Liyun Zhang,[c] Bingsen Zhang,[c]

Dangsheng Su,[c] Milivoj Plodinec [d] Robert Schlögl [a, d] and Saskia Heumann, [a]*

a Max Planck Institute for Chemical Energy Conversion, Stiftstrasse 34–36, Mülheim an der Ruhr,

45470, Germany

b School of Chemical Engineering, The University of New South Wales, Sydney, Kensington, NSW

2052, Australia

c Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy

of Sciences, Shenyang, 110016, P. R. China

d Department of Inorganic Chemistry, Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6,

Berlin, 14195, Germany

S2

Materials

9, 10-phenanthrenequinone (PQ, 99%), 9-phenanthrenol (PE), 9-phenanthrenecarboxylic acid (PCA),

xanthene (X, 99%), anthraquinone (AQ), anthrone (AT), 5, 12-naphthacenequinone (5, 12-NQ) and

6,13-pentacenequinone (6, 13-PQ, 99%) and isopropanol (99.9%) were obtained from Aldrich. All

chemicals were used as received without further purification. Cyclohexane was purchased from Merck

Co. (Germany). Purified ultra-dispersed nanodiamond (UDD) was bought from Beijing Grish Hitech Co.

(China), produced by detonation and followed by acid washing. The average particle size is about 5 nm.

The surface area of OLC is about 460 m2/g. MWCNTs supplied by Shandong Dazhan Co. (China) was

purified by water-diluted HCl (volume ratio is 1:1) for 24 h before using. The product was collected by

filtration, washed with H2O until the pH of the filtrate reaches 7 and then dried at 393 K. The surface

area of MWCNTs is about 230 m2/g.

Synthesis of onion-like carbon (OLC)

OLC with graphite-like shell structures was prepared by annealing purified nanodiamond at 1500 °C

for 30 min in argon atmosphere. The yield of OLC was about 75%. The surface area of OLC is about

460 m2/g.

Synthesis of modified onion-like carbon (OLC) and MWCNTs

In a typical procedure, 100 mg OLC and 20 mg of model catalysts, such as

9,10-phenanthrenequinone (PQ), 9-phenanthrenol (PE), 9-phenanthrenecarboxylic acid (PCA), xanthene

(X), anthraquinone (AQ), anthrone (AT), 5, 12-naphthacenequinone (5, 12-NQ) and

6,13-pentacenequinone (6, 13-PQ), were dissolved in 70 mL of cyclohexane to form a mixture which

was then sonicated for 1 h under the room temperature. The obtained mixture was then transferred into a

Teflon-lined stainless steel autoclave with a capacity of 200 mL and 400 rpm rotation rate and

maintained at about 110 °C for 24 h. The final products were filtered and washed with a large amount of

ethanol to remove the excess of model small molecules and cyclohexane and then dried them at 70 °C

for 12 hours. The as-prepared modified OLC were labeled as PQ+OLC, PE+OLC, X+OLC, PCA+OLC,

S3

AT+OLC, AQ+OLC, 5, 12-NQ+OLC and 6, 13-PQ+OLC, respectively. Pure OLC was treated by the

same process without the introduction of model catalysts and acted as reference materials (labeled as

T-OLC). The T-MWCNTs and PQ-modified MWCNTs samples were prepared by using the same

procedure.

Synthesis of IrO2/OLC

The synthesis of IrO2/OLC was performed by using a slightly modified method based on the previous

report. (see Y. Zhao, et al. Nat. Commun. 2013, 4, 2390 and T. Y. Ma, et al. Angew. Chem. Int. Ed.,

2016, 55, 1138). Briefly, 0.05 g of K2IrCl6 was added to 50 mL of aqueous solution containing 0.835 g

of disodium hydrogen citrate sesquihydrate. The pH of the red-brown solution was adjusted to 7.5 using

0.25 M NaOH solution. The resulting mixture was heated to 95 °C for 30 min under constant stirring,

and cooled down to room temperature. Afterwards, the solution was transferred to a round-bottom flask

with a reflux condenser followed by dispersion of 0.092 g of OLC. The mixture was maintained at 95

°C for 2 h bubbled with O2 through the solution. After vacuum-drying at 60 °C, the sample was calcined

at 300 °C for 40 min to remove the organic compounds, and washed with deionized water. Finally, the

sample collected by filtration and dried at 60 °C was denoted as IrO2/OLC, which contained ~20 wt.%

of IrO2.

Materials Characterization

High-resolution transmission electron microscopy (HRTEM) images were achieved on a FEI Tecnai G2

F20 microscope. The X-ray photoelectron spectroscopy (XPS) spectra were quickly carried out on an

ESCALAB 250 XPS system with a monochromatized Al Ka X-ray source to avoid the evaporation of

molecules from supports under the irradiation. The impurities of samples were determined by

inductively coupled plasma mass spectrometry (ICP-MS) and elemental analysis (Vario EL

CHNOS-Analyzer). Thermogravimetric (TG) analysis was performed on a NETZSCH STA449 F3

thermal analyzer. The atmosphere was argon, and the heating rate was 10 K/min starting from 40 °C to

900 °C. FTIR (Fourier transform infrared) spectra were collected with a Thermo Scientific Nicolet iS50

S4

in attenuated total reflection (ATR) mode with a resolution of 4 cm-1 at room temperature. High

performance liquid chromatography (HPLC) spectra were recorded by using a setup from Agilent with

an O-18 column. Static contact angle measurements were carried out on an OCA 15plus instrument

(Data Physics Instruments GmbH, Filderstadt, Germany) with Milli-Q water drawn from a Millipore

Direct Q8 system (Millipore, Schwalbach, with Millimark Express 40 filter, Merck, Germany) with a

resistivity of 18.0 MΩ cm. The powder samples were pressed into the pellets (the used pressure is 105 N)

prior to measurement.

Electrochemical measurements were performed using a potentiostat/galvanostat (BioLogic VSP,

France) with a conventional three-electrode electrochemical cell. Pt wire as a counter electrode and a

reversible hydrogen electrode (RHE, HydroFlex, Gaskatel GmbH) as a reference electrode were used.

The glassy carbon electrode (GCE, 5 mm diameter, 0.196 cm2) with rotating disk electrode (RDE) as a

working electrode was polished by 0.1 µm and 0.05 µm alumina powder and rinsed with deionized

water, followed by sonication in ethanol and deionized water. The modified samples were transferred

onto the GCE according to the following procedures: (1) 5 mg catalysts were added into the mixed

solution of 960 µL H2O, 4 mL isopropanol and 40 µL nafion, and then ultrasonicated for 15 min. (2) 10

µL of the ink was dropped on the GCE surface and dried at 70 °C for few minutes in air. The final

loading of catalyst on the electrode is about 0.051 mg/cm2. Cyclic voltammograms (CVs) were obtained

by scanning between 0 V to 1.0 V with a scan rate of 100 mV/s in argon-saturated solution. OER

polarization curves were obtained by linear sweep voltammetry scanning from 1.0 V to 1.8 V/1.9V with

a scan rate of 5 mV/s in argon-saturated 0.1 M KOH with RDE (rotation rate of 2250 rpm). Prior to the

impedance measurements, the electrodes were held at a constant desired potential for 10 min to ensure a

stable current. Subsequently measurements were performed by applying an AC voltage with 5 mV

amplitude in a frequency range from 100 kHz to 100 mHz. The detailed process can be found in Scheme

S1. All the experiments were carried out at room temperature. The calculation of electrochemically

active surface area (ECSA) is based on the measured double layer capacitance of the pure OLC and

PQ-modified OLC on glassy carbon RDE in 0.1M KOH according to a previous published report 1.

Briefly, a potential range where no apparent Faradaic process happened was determined firstly using the

S5

static CV. This range is typically a 0.1 V potential window centered at the open-circuit potential (OCP)

of the system. All measured current in this non-Faradaic potential region is assumed to be due to

double-layer charging. The charging current, ic, is then measured from CVs at different scan rates. The

working electrode was held constant at each potential vertex for 10 s before beginning the next sweep.

The relation between ic, the scan rate (ν) and the double layer capacitance (Cdl) was given in equation 1.

Cdl = ic/ν (1)

Therefore, the value of Cdl can be obtained from the slope of ic as a function of ν.

For the calculation of ECSA, a specific capacitance (Cs) value Cs ≈ 0.019 mF cm-2 (estimated value)

in 0.1 M NaOH is adopted from previous reports.2 The relative equation is:

ECSA =Cdl/Cs (2)

Reference

[1] McCrory C C L, Jung S, Peters J C, Jaramillo T F. Benchmarking heterogeneous electrocatalysts for

the oxygen evolution reaction. Journal of the American Chemical Society 2013, 135, 16977-16987.

[2] Badawy W A, Gad-Allah A G, Abd El-Rahman H A, Abouromia M M, Kinetics of the passivation

of molybdenum in acids and alkali solutions as inferred from impedance and potential measurements.

Surface and Coatings Technology 1986, 27, 187-196.

S6

Scheme S1. Activity and stability measurements for OER catalysts evaluation.

Calculation Methods

1. The possible amounts of molecules on the surface of individual OLC nanoparticle.

Where surface area(OLC) is 460 m2/g, Micropore area(OLC) is 71 m2/g, M(molecules) represents molar

mass of each molecules, Geometric area is geometric area of each molecules and NA is 6.02*1023.

S7

By using the above equation, the theoretical saturated mass fraction is:

14.04% for PQ 12.68% for PE 15.65% for X 11.92% for PCA

By using TG, the real mass fraction is:

2.1% for PQ 1.9% for PE 1.6% for X 2.0% for PCA

By using the above equation, the possible coverage rate for each molecules on OLC is:

14.95% for PQ 14.98% for PE 10.22% for X 16.77% for PCA

Where Spherical area(OLC) is about 78.5 nm2 (If the particle size of every OLC is 5 nm and the

morphology of every OLC is spherical).

By using this equation, the theoretical amounts of molecules on the surface of individual OLC

nanoparticle is: ~12 for PQ ~12 for PE ~11 for X ~11 for PCA

2. Calculation of Mass activity, theoretical TOF and Faradaic efficiency.

S8

Details concerning the calculation of mass activity and turnover frequency (TOF) are shown below:

The values of mass activity (A g-1) were calculated from the catalyst loading m (0.051 mg/cm2) and the

measured current density j (mA cm-2 ) at = 0.430 V:

Mass activity = j/m

The theoretical values of TOF were calculated by assuming that every oxygen group is involved in

the catalysis.

TOF = j×S/4×F×n

Where j is the measured current density at = 0.340 or 0.430 V and S is the surface area of glass

carbon disk electrode (0.196 cm2). F represents Faraday constant (96485.3 C mol-1), and the number 4

originates from the four electron transfer. It means that 4 electrons/mol per produced O2 are required.

The moles of the pure model catalysts that are deposited onto the glass carbon disk are represented by n.

Prior to the calculation of TOF, the mass relevance fraction of pure model catalysts is obtained from

TG.

The Tafel slope was calculated according to the Tafel equation as follows:

= b·log(j/j0)

where denotes the overpotential, b denotes the Tafel slope, j denotes the current density, and j0

denotes the exchange current density.

To investigate the reaction mechanism for OER, the rotating ring-disk electrode (RRDE)

voltammograms were conducted on a RRDE configuration consisting of a GCE and a Pt ring electrode.

A scan rate of 10 mV s-1 and a rotation rate of 2250 rpm were applied for RRDE tests. Specifically, in

order to determine the reaction pathway (electron transfer number) for OER by detecting the HO2-

formation, the ring potential was held constantly at 1.465 V vs. RHE for oxidizing HO2- intermediate in

argon-saturated 0.1 M KOH. The HO2- intermediate production percentages (% HO2

-) were determined

as follows:

S9

%HO2- = 200Ir/(IdN + Ir)

where Id is the disk current, Ir is the ring current and N is the current collection efficiency.

The Faradaic efficiency of the system was measured to ensure that the oxidation current derived from

oxygen evolution rather than other side reactions. Here, the ring potential was held constantly at 0.465

V vs. RHE to reduce the O2 formed from the catalyst on the disk electrode in argon-saturated 0.1 M

KOH solution. During Faradaic efficiency measurements, the disc potential was held at 1.8 V vs. RHE

to allow the OER to proceed for 2 h. The Faradaic efficiency (FE) determined as follows:

FE= Ir/IdN

S10

Scheme S2. The synthetic process of aromatic molecules-modified onion-like carbon (OLC) and

multi-walled carbon nanotubes (MWCNTs). The surface of OLC and MWCNTs should be rather

depicted as irregular sp2 carbon fragments, nevertheless multiple perfect graphitic shells were displayed

for simplicity.

S11

Table S1. Summary of XPS and TG data of T-OLC and modified OLC samples.

XPS (Ox/C, at%)Subtractting the XPS of T-OLC

(Ox/C, at%)

Samples

Total O C=OC-O-C/

O=C-O

C-O/

C-OHTotal O C=O

C-O-C/

O=C-O

C-O/

C-OH

Total desired O

(Proportion in

total O)

ΔTG (wt%)a

T-OLC 1.1 ---- 0.6 0.5 0 0 0 0 ---- 0

PQ+OLC 4.2 2.8 1.0 0.4 3.1 2.8 0.4 ----2.8

(90.3%)2.1

PE+OLC 3.6 ---- 1.2 2.4 2.5 ---- 0.6 1.91.9

(76.0%)1.9

X+OLC 3.4 ---- 2.8 0.60 2.3 ---- 2.2 0.12.2

(95.6%)1.6

PCA+OLC 4.0 ---- 3.0 1.0 3.3 ---- 2.4 0.52.4

(72.7%)2.0

aTG was calculated by mass of modified OLC minus mass of T-OLC.

S12

Figure S1. (a) XPS full spectra and (b) XPS N 1s spectra of unmodified T-OLC, PQ-, PE-, X- and

PCA-modified OLC samples.

As displayed in Figure S1b, there is a trace inherent nitrogen (~0.5 at%) present in OLC. The

introduction of aromatic molecules (PQ, PE, X and PCA) does not change the content and chemical

state of original nitrogen species.

S13

Figure S2. TG spectra of unmodified T-OLC, PQ-, PE-, X- and PCA-modified OLC samples.

S14

Figure S3. Attenuated total reflectance (ATR) infrared spectra of pure PQ powder (red), pure solvent

(cyclohexane) (black), dissolved PQ (cyclohexane as solvent) at 100 °C for 24 h (blue) and peeled-off

PQ (turquoise). The peeled-off PQ was obtained by removing the PQ molecules from as-prepared

PQ+OLC sample by sonication for 8 h in cyclohexane.

As shown in Figure S3, the main characteristic peak of pure PQ powder located at 1672 cm-1 is

designated as stretching vibration of the C=O group. The peaks located at 1509 and 1589 cm-1 are

ascribed as vibrations of aromatic rings. When pure PQ powder is dispersed in cyclohexane

(PQ+solvent) similar characteristic peaks are observed in the solution. It is clearly seen that the

exfoliated PQ molecules from the as-prepared PQ+OLC sample using sonication method also show the

main characteristic peaks of aromatic rings and the C=O group. The slight shift of peaks can be

attributed to the change of π-π interaction between PQ molecules in solvent. The measurements confirm

that the aromatic structure of PQ is not destroyed during the preparation process and that the structure is

also present while the organic molecule is adsorbed on the OLC surface.

S15

Figure S4. High performance liquid chromatography (HPLC) spectra of dissolved PQ (cyclohexane as

solvent) at 100 °C for 24 h (blue,bottom) and peeled-off PQ (turquoise, top). The peeled-off PQ was

obtained by removing the PQ molecules from as-prepared PQ+OLC sample by sonication for 8 h in

cyclohexane.

Two peaks with the similar retention time can be assigned to PQ. Confirming that the PQ structure

is very stable during the preparation process.

S16

Figure S5. ATR spectra of pure PE powder, pure solvent (cyclohexane), dissolved PE (cyclohexane as

solvent) at 100 °C for 24 h and peeled-off PE. The peeled-off PE was obtained by removing the PE

molecules from as-prepared PE+OLC sample by sonication for 8 h in cyclohexane.

As shown in Figure S5, the main characteristic peaks of pure PE powder located at 3250 and 1318

cm-1 are designated as stretching and bending vibrations of the -OH group, respectively. The peaks

located at 1602 and 1629 cm-1 are ascribed as vibrations of aromatic rings. The peak located at 1221

cm-1 is ascribed as vibration of C-O group. When pure PE powder is dispersed in cyclohexane

(PE+solvent) at 100 °C for 24 h, similar characteristic peaks are observed in the solution. It is clearly

seen that the exfoliated PE molecules from the as-prepared PE+OLC sample using the sonication

method also show the main characteristic peaks of aromatic rings and the -OH group. The slight shift of

peaks can be attributed to the change of π-π interaction between PE molecules in solvent. The

measurements confirm that the aromatic structure of PE is not destroyed during the preparation process

and that the structure is also present while the organic molecule is adsorbed on the OLC surface.

S17

Figure S6. ATR spectra of pure X powder, pure solvent (cyclohexane), dissolved X (cyclohexane as

solvent) at 100 °C for 24 h and peeled-off X. The peeled-off X was obtained by removing the X

molecules from as-prepared X+OLC sample by sonication for 8 h in cyclohexane.

As shown in Figure S6, the main characteristic peak of pure X powder located at 1297 cm-1 is

designated as stretching vibration of the C-O group. The peaks located at 1479 and 1576 cm-1 are

ascribed as vibrations of aromatic rings. When pure PE powder is dispersed in cyclohexane

(PE+solvent) at 100 °C for 24 h, similar characteristic peaks are observed in the solution. It is clearly

seen that the exfoliated X molecules from the as-prepared X+OLC sample using the sonication method

also show the main characteristic peaks of aromatic rings and the C-O group. The slight shift of peaks

can be attributed to the change of π-π interaction between X molecules in solvent. The measurements

confirm that the aromatic structure of X is not destroyed during the preparation process and that the

structure is also present while the organic molecule is adsorbed on the OLC surface.

S18

Figure S7. ATR spectra of pure PCA powder, pure solvent (cyclohexane), dissolved PCA (cyclohexane

as solvent) at 100 °C for 24 h and peeled-off PCA. The peeled-off PCA was obtained by removing the

PCA molecules from as-prepared PCA+OLC sample by sonication for 5 h in cyclohexane.

As shown in Figure S7, the main characteristic peaks of pure PCA powder located at 1293 and

1679 cm-1 are designated as stretching vibrations of the C-O and C=O group from the carboxylic

-COOH group. The peaks located at 1496 and 1529 cm-1 are ascribed as vibrations of the aromatic rings.

When pure PCA powder is dispersed in cyclohexane (PCA+solvent) at 100 °C for 24 h, similar

characteristic peaks are observed in the solution. It is clearly seen that the exfoliated X molecules from

the as-prepared PCA+OLC sample using the sonication method also show the main characteristic peaks

of aromatic rings and the -COOH group. The slight shift of peaks can be attributed to the change of π-π

interaction between PCA molecules in solvent. The measurements confirm that the aromatic structure of

PCA is not destroyed during the preparation process and that the structure is also present while the

organic molecule is adsorbed on the OLC surface.

S19

Figure S8. TG spectra of representative pure PQ, pure X, PQ- and X-modified OLC .

As shown in Figure S8, the onset temperature of modified OLC is significantly increased compared

with pure molecules, suggesting that there is a strong π-π interaction between molecules and OLC.

S20

Figure S9. Structural and compositional characterization of T-OLC and PQ-modified OLC.

HAADF-STEM and EDX mapping images of (a-c) T-OLC and of (e-g) PQ modified OLC. Colors in

elemental mapping images: red for C; and green for O. The values in the spectra (c, g) show the atomic

percentages for Oxygen from EDX. HRTEM images of (d) T-OLC and (h) modified OLC.

S21

Figure S10. (a) A plausible reversible proton-coupled electron transfer (PCET) between phenolic and

quinone groups and the possible intermediate radicals during electrochemical process. (b)Cyclic

voltammetry curves of PQ-modified OLC and pure PQ samples in argon-saturated 0.1M KOH measured

at a scan rate of 100 mV s-1. The loadings of all samples on electrode are 0.051 mg/cm2.

It has been reported that there is a good proton-coupled electron transfer process (PCET) between

quinone and phenolic groups, which is a reversible redox process (Figure S10a). In this process,

intermediate phenoxy radicals (e.g., PQ radical, PQH radical) with lone pair, originated from the PQ or

PQH2 redox reaction, have been highlighted. The intermediates can be stabilized due to the good

electron and proton transfer between support and pure PQ. Furthermore, the good electron transfer can

be experimentally confirmed by comparing the CV curves of pure PQ and PQ-modified OLC (Figure

S10b). The potential difference between the redox peaks of pure PQ (42 mV) and PQ-modified OLC

(20 mV) is shortened by 22 mV. The reduction and oxidation potentials of modified OLC shift to more

positive (+ 15 mV) and more negative (- of 7 mV) values, respectively. This means, that PQ molecules

on the surface of OLC are easier reduced or oxidized. All these findings point out the direct existence of

fast charge transfer between organic molecules and OLC support. This behavior provides great potential

for electrochemical applications.

S22

Figure S11. Electrochemical measurements of various pure aromatic organic molecules measured in

argon-saturated 0.1M KOH with a scan rate of 5 mV s-1 with iR correction. Here, the blank experiment

means that pristine working electrode without other catalysts is directly tested in OER.

S23

Figure S12. Nyquist plots of the pure PQ, T-OLC and PQ-modified OLC catalysts recorded at 1.6 V.

S24

Figure S13. Mass activities of representative PQ- and PE-modified OLC catalysts at =430 mV.

S25

Figure S14. Detection of H2O2 (purple dots) and O2 evolution (orange dots) from (a) PQ-modified OLC

and (b) T-OLC using rotating ring disk electrode (RRDE) measurement. The Pt ring was biased at 1.465

V versus RHE to collect H2O2 and was fixed at 0.465 V versus RHE to collect O2 in argon saturated 0.1

M KOH without iR compensation.

S26

Figure S15. (a-b) CV obtained with (a) pure T-OLC and (b) PQ-modified OLC loaded glassy carbon

electrodes in a capacitance current range between 1.02 V and 1.12V vs. RHE with scan rates of 5, 10,

25, 50, 100, 200 and 400 mV s-1, respectively. (c-d) The cathodic (red triangle) and anodic (blue circle)

capacitance currents measured at 1.07 V vs. RHE plotted as a function of the scan rate. The double-layer

capacitance (Cdl ) is determined from slopes of the anodic and cathodic linear fits.

The Cdl values of pure OLC and PQ-modified OLC measured by the scan rate dependent CVs are

0.112 mF and 0.216 mF, respectively. By using equation 2, the ECSA values of pure OLC and

PQ-modified OLC are about 5.89 cm2 and 11.37 cm2, respectively.

S27

Figure S16. Contact angles of T-OLC and various aromatic molecules-modified OLC. The data are

collected directly after the water drop was dropped onto the surface of samples.

S28

Table S2. The metal impurities of T-OLC and modified OLC samples measured by ICP-MS and

elemental analysis.

CatalystsFe a (ppm)

Cr a (ppm)

Mgb (ppm)

Al a (ppm)

Cub (ppm)

Ti a (ppm)

Ca a

(ppm)

T-OLC 26 18 227 84 117 8 49

PQ+OLC 28 17 336 81 118 7 52

PE+OLC 29 25 200 92 77 13 58

X+OLC 31 36 127 99 105 12 48

PCA+OLC 36 21 296 74 124 18 61

a obtained by ICP-MS. b obtained by elemental analysis

S29

Figure S17. CV curves before (red) and after the reaction at 10 mA/cm2 for 2 h (green) or 9 h (black) of

(a) pure PQ, (b) T-OLC and (c) PQ-modified OLC measured in argon-saturated 0.1 M KOH. The

loading of all samples on the glassy carbon electrode was 0.051 mg/cm2.

S30

Figure S18. TG spectra of AQ- and AT-modified OLC samples.

S31

Figure S19. The net contents of 5, 12-NQ and 6, 13-PQ molecules on modified OLC catalyst.

S32

Figure S20. ATR spectra of pure AQ powder, pure solvent (cyclohexane), dissolved AQ (cyclohexane

as solvent) at 100 °C for 24 h and peeled-off AQ. The peeled-off AQ was obtained by removing the AQ

molecules from the as-prepared AQ+OLC sample by sonication for 8 h in cyclohexane.

As shown in Figure S20, the main characteristic peak of pure AQ powder located at 1671 cm-1 is

designated as stretching vibration of the C=O group. The peaks located at 1572 and 1588 cm-1 are

ascribed as vibrations of aromatic rings. When pure AQ powder is dispersed in cyclohexane

(AQ+solvent) at 100 °C for 24 h, similar characteristic peaks are observed in the solution. It is clearly

seen that the exfoliated AQ molecules from the as-prepared AQ+OLC sample using the sonication

method also show the main characteristic peaks of aromatic rings and the C=O group. The slight shift of

peaks can be attributed to the change of π-π interaction between AQ molecules in solvent. The

measurements confirm that the aromatic structure of AQ is not destroyed during the preparation process

and that the structure is also present while the organic molecule is adsorbed on the OLC surface.

S33

Figure S21. ATR spectra of pure 5, 12-NQ powder, pure solvent (cyclohexane), dissolved 5, 12-NQ

(cyclohexane as solvent) at 100 °C for 24 h and peeled-off 5, 12-NQ. The peeled-off 5, 12-NQ was

obtained by removing the 5, 12-NQ molecules from the as-prepared 5, 12-NQ+OLC sample by

sonication for 8 h in cyclohexane.

As shown in Figure S21, the main characteristic peak of pure 5, 12-NQ powder located at 1676

cm-1 is designated as stretching vibration of the C=O group. The peaks located at 1579 and 1616 cm-1

are ascribed as vibration of the aromatic rings. When pure 5, 12-NQ powder is dispersed in cyclohexane

(5, 12-NQ+solvent) at 100 °C for 24 h, similar characteristic peaks are observed in the solution. It is

clearly seen that the exfoliated 5, 12-NQ molecules from the as-prepared 5, 12-NQ+OLC sample using

sonication method also show the main characteristic peaks of aromatic rings and the C=O group. The

slight shift of peaks can be attributed to the change of π-π interaction between 5, 12-NQ molecules in

solvent. The measurements confirm that the aromatic structure of 5,12-NQ is not destroyed during the

preparation process and that the structure is also present while the organic molecule is adsorbed on the

OLC surface.

S34

Figure S22. ATR spectra of pure 6,13-PQ powder, pure solvent (cyclohexane), dissolved 6,13-PQ

(cyclohexane as solvent) at 100 °C for 24 h and peeled-off 6,13-PQ. Here, the peeled-off 6,13-PQ was

obtained by removing the 6,13-PQ molecules from the as-prepared 6,13-PQ+OLC sample by sonication

for 8 h in cyclohexane.

As shown in Figure S22, the main characteristic peak of pure 6,13-PQ powder located at 1670 cm-1

is designated as stretching vibration of the C=O group. The peaks located at 1571 and 1611 cm-1 are

ascribed as vibrations of the aromatic rings. When pure 6,13-PQ powder is dispersed in cyclohexane

(6,13-PQ+solvent) at 100 °C for 24 h, similar characteristic peaks are observed in the solvent. It is

clearly seen that the exfoliated 6,13-PQ molecules from the as-prepared 6,13-PQ+OLC sample using the

sonication method also show the main characteristic peaks of aromatic rings and the C=O group. The

slight shift of peaks can be attributed to the change of π-π interaction between 6,13-PQ molecules in

solvent. The measurements confirm that the aromatic structure of 6,13-PQ is not destroyed during the

preparation process and that the structure is also present while the organic molecule is adsorbed on the

OLC surface.

S35

Figure S23. Cyclic voltammetry curves of pure AQ (red), T-OLC (black) and AQ-modified OLC (purple).

S36

Figure S24. Tafel slopes of (a) various quinone-modified catalysts and (b) AT- and AQ-modified OLC.

S37

Figure S25. Dependence of loading of 6, 13-PQ-modified OLC catalyst on the catalytic performance.

S38

Figure S26. Long-term stability tests of PQ-modified OLC, pure PQ and pure 6, 13-PQ catalysts at a

constant potential of 1.66 V.

S39

Table S3. Comparison of the electrocatalytic OER activity of the highly active transition-metal and

carbon-based catalysts. = overpotential determined at 10 mA/cm2

Reported Catalysts (mV)

@10mA/cm2Electrolyte

Tafel slope

(mVdec-1)Literature

N-doped graphene nanoribbon

()360 1M KOH 47 Sci. Adv., 2016, 2, e1501122

Oxidized CNTs

()360 1M KOH 48 ACS Energy Lett., 2017, 2, 294–300

C3N4/CNT

(Δ)370 0.1M KOH 83

Angew. Chem. Int. Ed., 2014, 53, 7281

–7285

N-doped porous carbon cloth

()360 1M KOH 98

Energy Environ. Sci., 2016, 9,

3411--3416

PrBa0.5Sr0.5Co1.5Fe0.5Oδ

(Δ)358 0.1M KOH 55 Nat. Commun., 2017, 8, 14586.

N-doped CNTs

(Δ)370 0.1 M KOH 93 Nat. Energy, 2016, 1, 15006.

N,S-codoped carbon materials

()360 1M KOH 56 Adv. Energy Mater., 2017, 1602068

Oxidized CNTs

(Δ)450 0.1 M KOH 72

J. Am. Chem. Soc., 2015, 137, 2901–

2907

NiCoP/C

()330 1 M KOH 96 Angew. Chem. Int. Ed., 2017, 56,1 – 5

S40

S,N-Fe/N/C-CNT

(Δ)370 0.1 M KOH 82

Angew. Chem. Int. Ed., 2017, 56,

610-614.

Co-C3N4/CNT

()380 1 M KOH 69

J. Am. Chem. Soc., 2017, 139, 3336–

3339

N-doped carbon microtube sponge

(Δ)290 0.1 M KOH 246

Energy Environ. Sci., 2016, 9,

3079-3084

S-doped CNTs/Graphene Nanolobes

()350 1 M KOH 95 Adv. Energy Mater., 2016, 6, 1501966

6,13-PQ-modified OLC

(﹡)350 0.1 M KOH 71 This work

S41

Figure S27. Isotopic electrochemical studies of representative AQ-modified OLC catalyst recorded in

0.1 M KOH dissolved in H2O (purple) and D2O (blue) (99.9%).

S42

Figure S28. Tafel slopes of PQ-modified OLC in KOH-H2O and KOH-D2O.

S43

Figure S29. The net content of PQ molecule on PQ-modified MWCNT catalyst.