Hydrothermally Driven Transformation of Oxygen Functional Groups at

Subscriber access provided by BOSTON UNIV

Article

Hydrothermally Driven Transformation of Oxygen Functional Groups at Multiwall Carbon Nanotubes for Improved Electrocatalytic Applications Bryan H.R. Suryanto, Sheng Chen, Jingjing Duan, and Chuan Zhao ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b14090 • Publication Date (Web): 02 Dec 2016 Downloaded from http://pubs.acs.org on December 6, 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 Applied Materials & Interfaces 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 30

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 Applied Materials & Interfaces

Hydrothermally Driven Transformation of Oxygen Functional Groups at Multiwall Carbon Nanotubes for Improved Electrocatalytic Applications Bryan H. R. Suryanto, Sheng Chen, Jingjing Duan and Chuan Zhao* School of Chemistry, The University of New South Wales, Sydney, NSW 2052, Australia Abstract The role of carbon nanotubes in the advancement of energy conversion and storage technologies is undeniable. In particular, carbon nanotubes have attracted significant applications for electrocatalysis. However, one central issue related to the use of carbon nanotubes is the required oxidative pre-treatment that often leads to significant damage of graphitic structures which deteriorates their electrochemical properties. Traditionally, the oxidized carbon nanomaterials are treated at high temperature under inert atmosphere to repair the oxidation induced defect sites, which simultaneously removes significant number of oxygen functional groups. Nevertheless, recent studies have shown that oxygen functional groups on the surface of MWCNT are the essential active centres for a number of important electrocatalytic reactions such as hydrogen evolution reaction (HER), oxygen evolution reaction (OER) and oxygen reduction reaction (ORR). Herein we firstly show that, hydrothermal treatment as a mild method to improve the electrochemical properties and activities of surface oxidized MWCNT for OER, HER and ORR without significantly altering the oxygen content. The results indicate that hydrothermal treatment could potentially impair the defects without significantly reducing the pre-existing oxygen content, which has never been achieved before with conventional high-temperature thermal annealing treatment. KEYWORDS carbon, carbon nanotubes, electrochemistry, hydrogen evolution reaction, hydrothermal, oxygen evolution reaction, oxygen reduction reaction, water splitting. 1

ACS Paragon Plus Environment


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 2 of 30

1. Introduction Carbon nanotube (CNT) is one of the most investigated carbon materials for various applications because of their enthralling properties such as a high electrical conductivity up to 5000 S cm-1, large surface area of 1315 m2 g-1 and impressive thermal and mechanical stabilities.1-2 In electrocatalysis, CNTs have garnered significant interests for their application as catalyst support and even as a stand-alone catalyst for electrochemically important reactions such as oxygen evolution reaction (OER), oxygen reduction reaction (ORR) and hydrogen evolution reaction (HER) for energy conversion and storage.3-7Consequently, significant amounts of study have followed and demonstrated the enhancing effect of CNT for OER, ORR and HER catalysts.4,

8-9

Moreover

the

increasing

evidence

regarding

the

tailorable

electrocatalytic property of functionalised/doped CNTs for these reactions proving the solid position of CNT in the development of energy conversion and storage technologies. Particularly multi-walled carbon nanotubes (MWCNT) are known to be more active compared to the single-walled, as their inner multi-layers of graphitic tubes which are important for fast electron transport are protected from the required damaging and harsh oxidative treatments of outer layers.10 One issue with utilization of MWCNT for chemical and electrochemical process in water is their high hydrophobicity, similar to that of graphite-water interface.11 Therefore, surface oxidation in strong acids is often required to functionalize the surface of MWCNT with oxygen functional groups (such as hydroxyl, epoxy, and carboxyl) to better disperse them in water as well as to enable further modification with other catalysts.4, 12

However this approach inevitably leads to graphitic damage and shortening that can

substantially reduce their electrochemical properties.13-16

2


Page 3 of 30

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


Hitherto, the common procedure to improve the graphitization of CNTs is hightemperature (T > 1000 oC) thermal annealing in inert atmosphere or in vacuo.17-18 These methods are known to be impractical for large scale application and require expensive instrumentations. Moreover, the high-temperature treatment always lead to the removal of oxygen functional groups rendering MWCNT to return to its initial hydrophobic state.19 On the other hand, recent studies have demonstrated the important role of oxygen functional group, in particular ketonic as active sites in electrocatalysis for OER along with evidence of catalytically active oxygen functionalities for ORR and HER.20-22 Therefore the real dilemma is to choose between electrochemical properties and oxygen functionality. Consequently, it is crucial to develop methods to improve the electrochemical properties of surface oxidized MWCNT without significantly sacrificing the oxygen functionalities. And it is important to note that recently, the development multifunctional electrocatalyst in particular for OER, HER and ORR based on earth abundant element has been gaining impetus.23-24 In this work, we developed a simple and chemical free hydrothermal method to improve the electrochemical properties of surface oxidized MWCNT without substantially reducing its initial oxygen content (as shown in Figure 1). Early investigations indicates that hydrothermal treatment could improve electronic properties of graphene oxide and MWCNT, however in depth investigation and mechanism remain unavailable and not clear.20, 25-26 Besides, the hydrothermal method is observed to improve the hydrophilicity (wettability), induces transformation of oxygen functional groups as well as reduces the amount of defects created during the chemical oxidation step. Potential applications of this approach have also been demonstrated for using MWCNT as electrocatalysts for OER, HER, and ORR.

3



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 4 of 30

2. Experimental 2.1 Synthesis of surface oxidized MWCNT In a typical synthesis, 150 mg of MWCNT (Nanotech Port, China) was transferred into a 100 ml round bottomed flask, followed by the addition of 75 ml of concentrated H2SO4 and the mixture was sonicated for 5 minutes until homogenous dispersion was obtained. Into the mixture, 25 ml of 30% of H2O2 solution (Univar, Ajax APS) was gradually added to prevent overheating. The final mixture was stirred continuously at 500 r.p.m. with magnetic stirrer and refluxed for 5 hours at room temperature. The reaction was quenched using 100 ml of Milli-Q water and washed with Milli-Q water until neutral pH was achieved. The piranha treated MWCNT (p-MWCNT) was then re-suspended using centrifugation and dried in an oven at 40oC. 2.2 Hydrothermal treatment Hydrothermal treatments were conducted in a 50 ml Teflon lined stainless steel autoclave. Generally, 50 mg of MWCNT were mixed with 35 ml of Milli-Q water. The mixture was then sonicated for 5 minutes to achieve homogenous dispersion then transferred into the autoclave. The autoclave was then kept in an oven for 18 h and the treatment temperature was adjusted accordingly. To recover the treated MWCNT centrifugation was performed and no further material purification was performed. 2.3 Electrochemical analysis For all comparison purposes, the SCE potential used in this study were calibrated against reversible hydrogen electrode (RHE) potential values according to the following formula: E(RHE) = E(SCE) + 0.241+(0.059*pH) 4


Page 5 of 30

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


Oxygen evolution reaction (OER). Electrochemical analysis was performed on CHI 760 Electrochemical Workstation (CH Instrument, Texas, USA), with a standard three electrode in a cell arrangement with a graphite rod as the counter electrode and a saturated calomel electrode (SCE, with saturated KCl) reference electrode. All analyses were performed in 0.1 M KOH, pH = 13 with a sweeping rate of 5 mV s-1. All of the working electrodes were made of glassy carbon and they were thoroughly polished using an electrode-polishing pad with 0.05 µm alumina slurry (Buehler, USA). All inks in this study were prepared by mixing a correct amount of MWCNT with 50% EtOH and 20 µl Nafion solution (5 wt.%, Sigma Aldrich), mixture was sonicated for 20 min to obtain a homogenous 5 mg ml-1 ink. The ink was then dropcasted to the surface of the glassy carbon electrode to achieve of catalyst loading of 0.2 mg cm-2 and dried overnight in ambient condition. For electrochemical impedance spectroscopy, a Solartron SI 1287 – Electrochemical interface coupled with SI 1260 – Impedance/gain phase analyzer was used. Hydrogen evolution reaction (HER). Electrochemical analysis for hydrogen evolution reaction were performed in 0.5 M H2SO4 solution (pH = 0.3) or otherwise stated. The reference electrode used in the HER analysis was a SCE (with saturated KCl) electrode and a graphite rod electrode was used as the counter electrode. Oxygen reduction reaction (ORR). All oxygen reduction reaction measurements were performed in 0.1 M KOH solution. All electrodes were modified with MWCNT with catalyst loading amount of 0.2 mg cm-2. All voltammograms showing the oxygen reduction peaks, the electrolytes were purged with pure O2 gas for 15 minutes. The steady state voltammograms were collected using RDE electrodes under continuous O2 gas purging. The reference electrode used in ORR study was SCE (saturated KCl) electrode.

5



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 6 of 30

2.4 Physical analysis X-ray photoelectron spectroscopy (XPS). All of XPS spectra were collected by ESCALAB 250 Xi (Thermo Scientific, U.K.) instrument. The X-ray source employed for measurements was a monochromated Al K-α (1486.6 eV). The pass energy used to generate survey scan was 100 eV, while high-resolution region scan was generated at 20 eV. The generated spectra was fitted using an Avantage XPS spectra fitting software. The reported binding energy was calibrated with adventitious C 1s peak at 284.8 eV. Peak fitting of the spectra was conducted following a Shirley background subtraction. Raman spectroscopy. Multiwall carbon nanotube samples were dispersed in dimethylformamide (DMF) by ultrasonication for 2 minutes and drop casted to the surface of cleaned glass slides. The samples were then dried in the vacuum oven at 50oC. Raman spectra were then obtained using Renishaw Raman Microscope (514 nm), (Renishaw Plc., U.K.). The ratio of IG/ID was calculated by determining the area under the peaks. ICP-MS. Inductively coupled mass spectrometry samples were prepared by digesting a known amount of MWCNT sample in 1 ml of concentrated HNO3 using a microwave reactor (Discover SP system, CEM Corporation). Digestion was performed for 1.5 hours and power was set at 30 W. Transmission electron microscopy (TEM). TEM measurements were performed with Phillips CM200. The TEM samples were prepared by homogenously dispersing 1 mg of MWCNT in absolute ethanol solutions by ultra-sonication. The ethanol-MWCNT mixture was drop casted onto formvar coated Cu-grid, and then dried overnight at room temperatures.

6


Page 7 of 30

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


Contact angle measurements. Contact angle measurements were performed with Rame-Hart 100-00 gonio-metry. The MWCNT films were fabricated by drop casting a total of 500 µl of 5 mg ml-1 MWCNT inks (without nafion) onto glass surface (area = 1 cm2) which was dried at ambient condition, contact angle measurements were performed at three different spots.

3. Results and discussion 3.1 Physical characterisation

Figure 1. The effect of hydrothermal treatment towards physical properties transformation the subsequent enhancement of electroacatalytic performance (values in the graph are for demonstration purpose only, notes: *ketonic group has been demonstrated to catalyse a number of electrochemically important reaction such as OER and HER, **the ORR performance of r-MWCNT were found to be higher due to the presence of surface bound metal impurities).

7



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 30

The modification of untreated MWCNT (r-MWCNT, r stands for raw) by oxygen functional groups was achieved via chemical oxidation in piranha solution (3:1, conc. H2SO4 : 30% H2O2). From the survey XPS scans in Fig S1, it is confirmed that in both MWCNT and r-MWCNT, only peaks correspond to C and O were observed. Further elemental composition analysis revealed that p-MWCNT (p stands for piranha treatment) contains 4.94 at.% of O, higher than r-MWCNT (1.5 at.%) which validates the successful oxidation reaction. Following the chemical oxidation, the dried pMWCNT samples were hydrothermally treated in Milli-Q water with autoclaves. Initially, three hydrothermal treatment temperatures of 100oC, 140oC and 180oC were selected to investigate

8


Page 9 of 30

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. High resolution XPS for O 1s peak from (A) r-MWCNT, (B) p-MWCNT, (C) H-100, (D) H-140 and (E) H-180. The peak shaded in blue (~531.2 eV) corresponds to ketonic/carboxylates (C=O), the peak with red (~533.1 eV) shading corresponds C-O as well as adsorbed H2O; (F) The plot of I(C-O)/I(C=O) from the obtained high-resolution O 1s XPS region scans.

their effect on MWCNT electrochemical performance (samples denoted as H-T, T is the corresponding hydrothermal treatment temperatures, e.g. H-180). Initially, the impact of hydrothermal treatments on the amount of oxygen content were investigated by XPS and summarized in Table S1. Generally hydrothermal treatment induces slight 9



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 30

removal of oxygen content up to 1.0 at.% from the initial O content of p-MWCNT (~4.94 at.%). The high resolution O 1s XPS spectra (Fig 2) reveal that the oxygen peaks can be resolved into two major peaks, the peak at ~531.2 eV (shaded in blue) corresponds to ketonic or carboxylates groups (C=O), while the peak at ~533.1 eV corresponds to C-O as well as surface adsorbed H2O.27-28 The amount of oxygen content was

Figure 3. C 1s high resolution XPS region scans of (A) p-MWCNT, (B) H-100, (C) H-140 and (D) H180.

compared using the calculated ratio between the two peaks (I(C-O)/I(C=O)) as shown in Fig 2F. As shown by Fig 3A, it was found that the fraction of peaks correspond to C-O and adsorbed H2O were increased (~533.1 eV). The increment is consistent with the increased I(C-O)/I(C=O) as a direct result from elevating the hydrothermal treatment

10


Page 11 of 30

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


temperatures. While the blue line suggests that the total oxygen contents in all hydrothermally treated samples were relatively constant. To exclude the XPS signal contribution from H2O, the effect of hydrothermal treatment on the surface-bound oxygen functionalities was investigated by studying the high resolution XPS of the C 1s peaks. As shown by Fig 3 (A, B, C and D), the C 1s peak could be resolved into five main peaks, C=C (284.5 eV), C-C (~285.0 eV), CO (epoxy/esters, ~286.0 eV), C=O (ketonic, ~287.0 eV) and O-C=O (carboxylate, ~288.0 eV) 29-30.

Formatted: Font: (Default) Times New Roman, Character scale: 108%

11



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 12 of 30

Figure 4. (A) FT-IR spectroscopy analysis of all MWCNT samples (from top to bottom, H180, H-100, p-MWCNT and r-MWCNT); (B) Raman spectroscopy of p-MWCNT, H-100 and H-180. The red-shaded area indicate the D-band (1360 cm-1), blue is G-band (1572 cm-2) and green is D’-band (1608 cm-1). (C) Schematic showing thermally driven transformation from

12


Page 13 of 30

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


edge confined ketonic and hydroxyl groups to carboxyl group (carbon, hydrogen and oxygen atoms are represented in the diagram as grey, white and red colored atoms).

From Table S2, I(C-O)/I(C=O) values were found to decrease with the increase of temperatures, indicating that oxygen functionalities undergo a thermally driven transformation process. The trend opposes the result from O1s analysis. Which rather confirms the origin of XPS signal contribution from non C-O bonds such as adsorbed H2O. Hence, it can be deduced that higher hydrothermal treatment temperature improves the wettability of MWCNT resulting in more adsorbed H2O. To validate the observed transformation, the oxygen functional group species were further investigated with FT-IR. Shown in Fig 4A, the FT-IR spectra of all MWCNT samples were compared. It is clear that oxygen functionalities such as C-O (1050 cm-1) and CO-C (1065 cm-1) were observed in all of the tested samples. The C=O peaks (1752 cm1

) was observed the strongest in H-180 and also observed in all other MWCNT

samples except for r-MWCNT. The increased wettability could be attributed to the increased amount of ketones and carboxylic acid functionalities which could facilitates the formation of H-bond with water, as shown by Table S2 and Fig 3. Additionally, the surface transformation of CO bonds (hydroxyl) into carboxyl/ketonic group has previously been validated in graphene oxide (GO) systems.31 Density functional theory (DFT) calculation points out that at the edge of GO, the transformation from hydroxyl and ketonic group carboxyl groups is a thermally driven process (Fig 4C). The carboxylates acid functionalities are negatively charged, which is imperative in preventing MWCNT from aggregation in polar solution, due to the electrostatic repulsion effect. To validate the effect of hydrothermal treatment on the wettability of and dispersion stability,

13



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 30

water contact angle measurements of the hydrothermally treated MWCNT were carried out. Shown in Fig S2 and Table S3, increased spread of water droplets as well as decreased contact angles of 147.1o,131.1o, 125.7o and 120.2o were observed for rMWCNT, p-MWCNT, H-100 and H-180, respectively. Dispersion stability of MWCNT were tested according to reported method,32 the dispersion stabilities of the hydrothermal treated MWCNT were also improved as shown in Fig S3. It was demonstrated that in Milli-Q water, p-MWCNT start to destabilize only after 1 week and mostly precipitated only after being kept for 3 weeks. Raman spectroscopy was also employed to monitor the effect of hydrothermal treatment on the amount of defect in MWCNT. In the spectrum shown by Fig 4B, the G-band at 1572 cm-1 reflects the crystallinity of the graphitic lattices which arises from tangential stretching of C-C bonds. In contrast, the D-band and D’-band at 1360 and 1608 cm-1 are induced by lattice disorder such as defects, vacancies, kinks, molecules intercalation, heptagon-pentagon pairs as well as presence of heteroatoms, (e.g. oxygen in this case), as well as intercalating molecule.29 The ratio shown in Table S4 was obtained by fitting the G-band and D-band peak using a mixture between Lorentzian and Gaussian peak fitting method. As shown in Fig 4B, the untreated pMWCNT shows the lowest IG/ID of 0.95. With hydrothermal treatments the IG/ID of the treated samples are higher.33 The IG/ID was also found to increase with hydrothermal treatment temperatures implying higher hydrothermal temperature is favorable for the improvement of the graphitic lattice ordering. On the other hand, the D’-band increased following the hydrothermal treatments, presumably owing to the intercalated water molecules into the layers of MWCNT as a result of increased pressure and temperature during the hydrothermal treatment.

14


Page 15 of 30

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


Lastly, TEM investigations were performed on p-MWCNT and H-180, respectively. Fig S4 shows that there are no distinguishable physical features between p-MWCNT and H-180 with core-confined metal impurities observed in both samples. Realizing the potential contribution from metal impurities to catalytic activity, their concentration levels were investigated using ICP-MS. Fig S5 shows that the piranha treatment results in significant decrease of metallic impurities such as Co, Fe and Ni which often contribute to the electrocatalytic activity of carbon materials.34-35 It was also found that the hydrothermal treatment can further reduce the metal impurities in MWCNT, possibly attributed to the dissolution at high temperatures by the residual acids from the initial chemical oxidation stage.36 The removal of surface bound metal impurities was also confirmed using XPS (Fig S6). The spectra of Ni 2p, Co 2p and Fe 2p for the p-MWCNT and hydrothermal treated MWCNT all show no detectable impurities, confirming that the surface of MWCNT is free from metal impurities.

3.2 Electrochemical performance of hydrothermally treated, surface oxidized MWCNT The effect of hydrothermal treatment on electrochemical performances of MWCNT was initially investigated by comparing their electrochemically active surface area (ECAS), according to a reported method.37 Double layer capacitance values were determined by collecting cyclic voltammetry at different scan rates (v) of 0.005, 0.01, 0.025, 0.05, 0.1, 0.2 and 0.4 V s-1. The cyclic voltammetries are shown (CV) in Fig 5A for H-180 and Fig S7 for the p-MWCNT and H-100. The potential range that we employed to obtain the CV is typically 0.1 V, centred at the open circuit potential (OCP) of the electrochemical cell. The current (i) values obtained at OCP were plotted against the corresponding v as shown in Fig 5B. The double layer capacitance (CDL) was calculated from i-v plot according the following formula:

15



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

I = vCDL

Page 16 of 30

(1)

Figure 5. (A) Double layer capacitance measurement obtained by cyclic voltammetry with

scanning range of 0.14 V to 0.24 V. The difference colors indicate different scan rates (v) of 5, 10, 25, 50, and 400 mV s-1 (smallest to largest capacitance). (B) Plot of (i) at 0.19 V versus scan rates.

Hence, the slopes from i-v plot are the CDL values. The CDL values for p-MWCNT, H100 and H-180 were determined to be 0.12, 0.25 and 0.26 mF, respectively. From the obtained CDL, the ECAS was then determined with the following formula: ECAS = CDL/Cs

(2)

16


Page 17 of 30

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


Where, CS = 0.04 mFcm-2 is the specific capacitance of MWCNT, for MWCNT

20

.

Thus the ECAS for p-MWCNT, H-100 and H-180 are 3.00, 6.25 and 6.50 cm2, respectively. The

effect

of

hydrothermal

treatment

in

improving

electrochemical

performance of MWCNT is further demonstrated by using these MWCNT for three key reactions in electrochemical energy conversion and storage device, namely OER, ORR and HER.

3.2.1 Oxygen evolution reaction Assessment on OER activity of hydrothermal treated MWCNT was performed using linear sweep voltammetry (LSV) in 0.1 M KOH with a typical scan rate of 5 mV s-1. Initially, all of the MWCNT were electrochemically activated by cycling between two potential values of 1.0 V to 1.76 V vs RHE until stable current is achieved (Fig. S8). The LSVs in Fig 6A demonstrate the OER activity of the MWCNT samples. The OER performances of the MWCNT samples were compared in terms of onset potential and current density (j) and the results are summarized in Table S5. It is observed that the OER performances of the hydrothermally treated samples are higher than p-MWCNT. As shown by Table S5, the measured OER j at 1.7 V (RHE) for H180 is 3.4 mA cm-2, which is more than 2 times higher than that obtained with pMWCNT. Moreover, as revealed by the ICP-MS previously, it is worth noting that as p-MWCNT contains much higher amount of metal impurities compared to H-180, the above results also provide evidence that the enhanced OER catalytic activity of MWCNT after hydrothermal treatment is not originated from the metal impurities, rather the oxygen functional groups on MWCNT in catalyzing OER process.

17



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 30

Figure 6. (A) LSVs showing the dependency of hydrothermal treatment temperatures towards

OER activity of MWCNT, (B) Tafel plot derived from LSV in (A) and (C) Electrochemical impedance spectroscopy (Nyquist plot) data of hydrothermal treated MWCNT obtained at 1.60 V (RHE). (All data was obtained in 0.1 M KOH electrolyte).

For all hydrothermally treated samples in Fig 6A, the OER onset potentials were observed to occur at 1.57 V vs RHE accompanied with gas bubble evolution on the working electrode. 18


Page 19 of 30

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


A significant decrease of Tafel slope from 118 to 85 mV dec-1 following the chemical oxidation of r-MWCNT (Fig 6B), suggests the change in OER kinetic as a result of formation of oxygen groups on MWCNT.38 Furthermore, the increase in OER activity also echoes with the XPS results, where the values of I(C-O)/I(C=O) were found to decrease with the increase of OER activity. Collectively, the results suggest that oxygen functional groups, especially the C=O functionalities play an important role in catalysing OER, presumably due to its electron withdrawing properties.20, 39 Finally, the Nyquist plot in Fig 6C revealed that H-180 exhibited the smallest charge transfer resistance compared to H-100 as indicated by smaller semi-circle radius. This result also suggests that hydrothermal treatments could enhance of the electrical conductivity of the acid oxidized MWCNT samples. 3.2.2 Hydrogen evolution reaction The electrochemical activity of oxidized MWCNT for hydrogen evolution reaction has been reported by several studies.22, 40 Herein, we show that hydrothermal treatment of MWCNT is also effective for improving their HER activity. The HER activities of MWCNT were evaluated by LSV from 0 V to -0.8 V (vs. RHE) with a typical scan rate of 5 mV s-1 in 0.5 M H2SO4. The LSV in Fig 7A shows that the H180 shows an improved HER performance, better than H-100, indicated by the lower HER onset potential. In comparison to r-MWCNT and p-MWCNT, H-180 exhibits a significant improvement in HER activity (Fig. 7B). Using H-180, a j of 10 mA cm-2 can be achieved at a potential of -0.68 V vs RHE. In Fig 7C, a Tafel slope of 71.35 mV dec-1 was obtained for the H-180 which is comparable to that reported in literature 22

. The HER onset for the hydrothermal treated MWCNT is 50 mV which is

comparable to other reported metal-free and metal based HER

19



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 20 of 30

Figure 7. (A) Linear sweep voltammogram shows the polarisation curves for HER of the hydrothermal treated MWCNT. (B) Polarisation curves comparing the HER performance of preactivated MWCNT (dotted line) to activated samples (solid line) as well as 30% Pt/C, (C) Tafel plots comparison of catalysts. All experiments were performed in 0.5 M H2SO4 oxygen reduction reaction.

20


Page 21 of 30

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


catalysts such as cathodic activated MWCNT (100 mV),

22

C60(OH)8(210 mV),41

C3N4(240 mV),42 and the well-developed MoS2 based catalyst (150 mV).43 It is noted that the HER activity of hydrothermally treated MWCNT can be further enhanced by electrochemical activation. The electrochemical activation procedure was done according to previously reported cathodic pre-treatment method reported by Cui et al.,22 where the cathodic pre-treatment was performed at -2.0 V vs RHE for 400 s. According to Fig S9, the pre-treatment leads to a shift of HER onset potential towards more positive values for about 350 mV. The XPS spectra in Fig S9 shows that during the pre-treatment the amount of oxygenated functional groups were increased from 4.39 at.% to 11.90 at.%, possibly due to the carbon etching that takes place during cathodic activation in acids.22,

44

The high

resolution XPS shows that during the cathodic pre-treatment at -2.0 V vs SCE, the amount of –OH (533.41 eV) and –C=O (532.05 eV) groups were enhanced significantly.45 Therefore, the increased HER activity can also be correlated to the increased amount of surface bound oxygen functional groups.41 The increased conductivity after hydrothermal treatment can also add up to the very high HER activity observed. Extending the cathodic pre-treatment duration into 1 hour (denoted as H-180 (3600 s) can further enhance enhance the HER activity, approaching the benchmark HER catalyst, in this case 30% Pt/C. 3.2.3 Oxygen reduction reaction All ORR tests were carried out using a glassy carbon rotating disk electrode (RDE) in 0.1 M KOH. Fig 8A shows the cyclic voltammetry curves for the treated MWCNT. It is demonstrated that r-MWCNT exhibits the most positive 1st ORR peak of 0.68 V vs RHE. Following acid treatment, the p-MWCNT exhibits a lower 1st ORR peak at 0.62 V indicating a lower performance. The decrease of ORR performance is somewhat expected due to the removal of surface bound ORR active metal impurities (Fig S4).

21



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 22 of 30

Figure 8. (A) CV of MWCNT for ORR (without rotation); (B) Comparison of electron transfer number (n) obtained from various MWCNTs samples across the potential range of 0.6 V to -0.2 V, in 0.1 M KOH, for CV the scan rate = 50 mV s-1

All ORR tests were carried out using a glassy carbon rotating disk electrode (RDE) in 0.1 M KOH. Fig 8A shows the cyclic voltammetry curves for the treated MWCNT. It is demonstrated that r-MWCNT exhibits the most positive 1st ORR peak of 0.68 V vs RHE. Following acid treatment, the p-MWCNT exhibits a lower 1st ORR peak at 0.62 V indicating a lower performance. The decrease of ORR performance is somewhat expected due to the removal of surface bound ORR active metal impurities (Fig S4). 22


Page 23 of 30

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 1. Electrochemical parameter summary of the MWCNTs for ORR Sample

Eonset1

E1/22

|Jl|3

n4

r-MWCNT

0.75

0.67

4.5

3.45

p-MWCNT

0.70

0.63

4.6

3.75

H-100

0.72

0.64

4.9

3.90

H-180

0.74

0.66

4.9

4.00

1

Onset potential; 2Half-wave potential; 3Limiting current density (mA cm-2); 4electron transfer number calculated at -0.2 V (vs RHE). The data were obtained from LSVs collected in 0.1 M KOH, collected with scan rate of 5 mV s-1 and rotation speed of 1600 rpm.

Meanwhile, the CVs of the hydrothermally treated p- MWCNT (H-100 and H-180) display a progressive improvement as shown positive shift of both 1st and 2nd ORR peaks. The presence of the two ORR peaks reveals that the ORR proceeds via a 2-step reduction process. Steady-state measurement using RDE (Fig S10 and Table 1) also suggests the ORR performance of p-MWCNT can be gradually recovered by hydrothermal treatment. The ORR onset potential for H-180 was at 0.74 V which is comparable to the recently developed nitrogen doped CNT-graphene hybrid.7 The background current for H-180 obtained in N2saturated 0.1 M KOH is shown in Fig S11. The ORR mechanisms at the four MWCNT samples were further assessed using the Koutecky-Levich equation.46 The electron transfer numbers (n) are determined which shows significant effect of hydrothermal treatment on ORR at MWCNT. Shown in Fig 8B and Table 1, the comparison of ORR electron transfer numbers for treated and r-MWCNT. The H-180 exhibits the highest among the MWCNT samples at all potentials lower than 0.6 V. Importantly, at the potential of -0.2 V, the n = 4 only can be achieved with H-180, followed by H-100 and p-MWCNT. The least ORR activity was observed for r-MWCNT (n

Hydrothermally Driven Transformation of Oxygen Functional Groups at

Recommend Documents