Synthesis and Characterization of Modified Silicas and ...modification de la surface du carbone pour améliorer les densités d’énergie et de puissance. Pickup, P.G., Rowe, A.,

Defence R&D Canada – Atlantic

DEFENCE DÉFENSE&

Synthesis and Characterization of ModifiedSilicas and Carbons for Use as Electrodes inElectrochemical SupercapacitorsFirst Annual Report

Peter G. Pickup, Aaron Rowe, Xiaorong Liu and Derrick DesRochesMemorial University of Newfoundland

Memorial University of NewfoundlandDepartment of ChemistrySt. John’s, Newfoundland A1B 3X7

Contract Manager: Colin G. Cameron, 902-427-1367Contract Number: W7707-063350Contract Scientific Authority: Colin G. Cameron, 902-427-1367

The scientific or technical validity of this Contract Report is entirely the responsibility of the contractorand the contents do not necessarily have the approval or endorsement of Defence R&D Canada.

Contract Report

DRDC Atlantic CR 2007-120

August 2007

Copy No. _____

Defence Research andDevelopment Canada

Recherche et développementpour la défense Canada

This page intentionally left blank.

Synthesis and Characterization of Modified Silicas and Carbons for Use as Electrodes in Electrochemical Supercapacitors First Annual Report

Peter G. Pickup Aaron Rowe Xiaorong Liu Derrick DesRoches Memorial University

Prepared by: Memorial University of Newfoundland Department of Chemistry St. John’s, NL A1B 3X7

Contract Manager and Scientific Authority: Colin G. Cameron 902-427-1367 Contract Number: W7707-063350

Defence R&D Canada – Atlantic Contract Report DRDC Atlantic CR 2007-120 August 2007

The scientific or technical validity of this Contract Report is entirely the responsibility of the contractor and the contents do not necessarily have the approval or endorsement of Defence R&D Canada

Author

Peter G. Pickup

Approved by

Colin G. Cameron

Scientific Authority

Approved for release by

James L. Kennedy

DRP Chair

© Her Majesty the Queen in Right of Canada, as represented by the Minister of National Defence, 2008

© Sa Majesté la Reine (en droit du Canada), telle que représentée par le ministre de la Défense nationale, 2008

Original signed by Colin G. Cameron

Original signed by James L. Kennedy

DRDC Atlantic CR 2007-120 i

Abstract

A series of commercial high surface area carbons, ruthenium (Ru) oxide prepared in-house, and a series of carbon supported Ru oxide materials prepared in-house have been evaluated as capacitive materials for supercapacitors. A proton conducting organically modified silica gel has been evaluated as an electrode binder and shown to increase capacitances and decrease ionic resistances in carbon electrodes. Prototype supercapacitors have been built with Ru oxide and carbon electrodes. A device capacitance of 14 F has been demonstrated. The effective series resistance has been decreased to 0.24 Ω cm2. Average power densities during full discharge exceed 25 kW/kg (peak power > 100 kW/kg), and energy densities exceed 30 Wh/kg.

Résumé

On a évalué des carbones à surface efficace élevée d’origine commerciale, de l’oxyde de ruthénium (Ru) préparé sur place, ainsi qu’une série de matériaux de type oxyde de Ru sur support de carbone, en tant que matières capacitives pour des supercondensateurs. On a évalué un gel de silice modifié organiquement, conducteur de protons, en tant que liant pour électrode, et on a montré qu’il augmentait la capacité et diminuait la résistance ionique des électrodes en carbone. Des prototypes de supercondensateurs ont été construits avec de l’oxyde de Ru et des électrodes en carbone. On a pu mettre en évidence ue capacité de 14 F pour un dispositif. La résistance en série effective a été réduite à 0,24 Ω cm2. La densité de puissance moyenne lors d’une décharge complète était supérieure à 25 kW/kg (puissance pic > 100 kW/kg), et la densité d’énergie était supérieure à 30 Wh/kg.

ii DRDC Atlantic CR 2007-120


DRDC Atlantic CR 2007-120 iii

Executive summary

Introduction

The focus of the project in year one has been to evaluate the performances of silica-based sol-gels as ion conductors in supercapacitor electrodes. Two types of capacitive material commonly used in supercapacitors, Ru oxide and high surface area carbon, were selected for this work. The Ru oxide was synthesized by us, by adapting literature methods, while various commercial carbons were used.

Results

A series of commercial high surface area carbons, Ru oxide prepared in-house, and a series of carbon supported Ru oxide materials prepared in-house have been evaluated as capacitive materials for supercapacitors. Black Pearls 2000 (Cabot Corp; BP2000) was selected as the best carbon powder for use in the project. It has a specific capacitance of ca. 170 F g-1 in 1 M H2SO4(aq), but its low density limits the amount that can be loaded onto an electrode, and so limits the capacitance that can be obtained per cm2 of electrode to below 0.5 F. A proton containing ormosil was shown to improve the performance of BP2000 in supercapacitors.

A high surface area carbon fabric (Spectracarb 2225) was found to be very convenient for use in high capacitance electrodes, with a maximum capacitance of 2.6 F being achieved for a 1cm2 electrode in 1 M H2SO4(aq) (190 F g-1). Supercapacitors could be charged to 2 V in acetonitrile. These electrodes had low ionic resistances, which were not improved by use of the ormosil. However, capacitances in acetonitrile were improved by the ormosil.

Ru oxide with a specific capacitance of 680±70 F g-1 was prepared by modification of a literature method and used to prepare supercapacitors with loadings as high as 100 mg and capacitances as high as 14 F (max electrode capacitance of 35 F). The effective series resistance was decreased to 0.24 Ω cm2. Average power densities during full discharge exceeded 25 kW/kg (peak power > 100 kW/kg), and energy densities exceeded 30 Wh/kg.

Ru has been deposited onto high surface area carbon (BP2000) and allowed to oxidize to Ru oxide to give composite materials with specific capacitance as high as 570 F g-1. The Ru + oxide component has a specific capacitance of 700 F g-1.

Significance

The Ru oxide that we have prepared has yielded very impressive power and energy densities compared with literature results. We have shown it to be far superior to carbon as an electrode material; commercially available double-layer capacitors from Montena produce around 4 kW/kg with an energy density of 4 Wh/kg (including device packaging). For military applications, the weight advantage over carbon will likely be a critical factor. The ormosil that we have developed shows promise for use in carbon supercapacitors, particularly if corrosive acid electrolytes are to be avoided.

iv DRDC Atlantic CR 2007-120

Future plans

Work in year 2 will focus on:

Combining Ru oxide with carbon and other oxides to decrease the amount (cost) of Ru oxide required.

Use of sulphonated sol-gels to improve capacitances and lower resistances

Surface modification of carbon to improve energy and power densities.

Pickup, P.G., Rowe, A., Liu, X, DesRoches, D. 2007 Synthesis and Characterization of Modified Silicas and Carbons for Use as Electrodes in Electrochemical Supercapacitors: First Annual Report. DRDC Atlantic CR 2007-120. DRDC Atlantic.

DRDC Atlantic CR 2007-120 v

Sommaire

Introduction

L’objectif pour la première année du présent projet a été d’évaluer les performances de matériaux de type sol/gel à base de silice comme conducteurs ioniques pour électrodes de surpercondensateurs. Deux types de matériaux communément utilisés dans les supercondensateurs, l’oxyde de Ru et le carbone à surface efficace élevée, ont été retenus pour le projet. Nous avons nous-mêmes synthétisé l’oxyde de Ru, en adaptant des méthodes trouvées dans la littérature. Les carbones utilisés étaient quant à eux d’origine commerciale.

Résultats

On a évalué des carbones à surface efficace élevée d’origine commerciale, de l’oxyde de ruthénium (Ru) préparé sur place, ainsi qu’une série de matériaux de type oxyde de Ru sur support de carbone, en tant que matières capacitives pour des supercondensateurs. La poudre de carbone Black Pearls 2000 (Cabot Corp; BP2000) s’est avérée la meilleure poudre a utilisé pour le projet. Elle a une capacité spécifique d’environ 170 F g-1 dans du H2SO4(aq) 1 M. Toutefois, sa faible densité limite la quantité pouvant être chargée sur une électrode et limite donc la capacité pouvant être obtenue par cm2 d’électrode jusqu’à 0,5 F. On a montré qu’un ormosil contenant des protons permet d’améliorer la performance du BP2000 dans les supercondensateurs.

On a montré qu’un tissu en carbone à surface efficace élevée (Spectracarb 2225) était très utile dans des électrodes de capacité élevée, avec une capacité maximale de 2,6 F pouvant être obtenue avec une électrode de 1 cm2 dans du H2SO4(aq) 1 M (190 F g-1). Les supercondensateurs pourraient être chargés jusqu’à 2 V dans l’acétonitrile. Ces électrodes ont des résistances ioniques faibles, qui n’ont pas pu être améliorées en utilisant l’ormosil. Toutefois, les capacités dans l’acétonitrile étaient améliorées par l’ormosil.

On a préparé de l’oxyde de Ru ayant une capacité spécifique de 680 ± 70 F g-1 en modifiant une méthode trouvée dans la littérature, Cet oxyde a servi à préparer des supercondensateurs ayant des charges allant jusqu’à 100 mg et des capacités allant jusqu’à 14 F (capacité maximale de l’électrode de 35 F). La résistance en série effective était résuite à 0,24 Ω cm2. Les densités de puissance moyennes lors d’une décharge complète étaient supérieures à 25 kW/kg (puissance pic > 100 kW/kg). Les densités d’énergie étaient supérieures à 30 Wh/kg.

On a déposé du Ru sur du carbone à surface efficace élevée (BP2000), puis on l’a laissé s’oxyder afin d’obtenir des matériaux composites ayant des capacités spécifiques allant jusqu’à 570 F g-1. Le composant Ru + oxyde a une capacité de 700 F g-1.

vi DRDC Atlantic CR 2007-120

Importance des résultats

L’oxyde de Ru que nous avons préparé a permis d’obtenir des densités de puissance et d’énergie très impressionantes en comparaison de celles publiées dans la littérature. Nous avons montré qu’il était bien supérieur au carbone comme matériau pour electrode; des dispositifs à couche double disponibles de Montena produisent environ 4 kW/kg et une densité d’énergie de Wh/kg (y compris l’embellage). Pour des applications militaires, son avantage de poids par rapport au carbone sera probablement un élément critique. L’ormosil que nous avons développé s’est avéré prometteur pour une utilisation dans des supercondensateurs au carbone, en particulier si le recours à des électrolytes acides corrosifs doit être évité.

Plans d’avenir

Le travail au cours de la deuxième année portera sur :

combinaison de l’oxyde de Ru avec du carbone et d’autres oxydes pour réduire la quantité (coût) d’oxyde de Ru requise.

utilisation de sol/gels sulfonés pour améliorer les capacités et réduire les résistances

modification de la surface du carbone pour améliorer les densités d’énergie et de puissance.

Pickup, P.G., Rowe, A., Liu, X, DesRoches, D. 2007 Synthesis and Characterization of Modified Silicas and Carbons for Use as Electrodes in Electrochemical Supercapacitors: First Annual Report. DRDC Atlantic CR 2007-120. DRDC Atlantic.

DRDC Atlantic CR 2007-120 vii

Table of contents

Abstract........................................................................................................................................ i

Executive summary ................................................................................................................... iii

Sommaire.................................................................................................................................... v

Table of contents ...................................................................................................................... vii

List of figures ............................................................................................................................ ix

List of tables ............................................................................................................................... x

Acknowledgements ................................................................................................................... xi

1. Introduction ................................................................................................................... 1

2. Some theoretical considerations .................................................................................... 3

3. Synthesis, Characterization and Capacitance of Ruthenium Oxide .............................. 4 3.1 Preparation of hydrous ruthenium oxide powder ............................................. 4 3.2 The preparation of electrodes ........................................................................... 4 3.3 Electrochemical characterization...................................................................... 4

4. Synthesis, Characterization and Capacitance of Carbon Supported Ru Oxide ............. 8 4.1 Synthesis, characterization and electrode preparation...................................... 8 4.2 Summary of results........................................................................................... 8 4.3 Conclusions .................................................................................................... 12

5. Evaluation of Commercial Carbons ............................................................................ 13

6. Ruthenium Oxide Supercapacitors .............................................................................. 15 6.1 Cyclic Voltammetry ....................................................................................... 15 6.2 Impedance Spectroscopy ................................................................................ 15

6.2.1 Representative Capacitance Plots...................................................... 16 6.2.2 Impedance and capacitance vs. Ruthenium oxide loading ................ 18

6.3 Constant current discharge ............................................................................. 19

viii DRDC Atlantic CR 2007-120

6.4 Conclusions .................................................................................................... 22

7. Carbon Supercapacitors............................................................................................... 24 7.1 Experimental .................................................................................................. 24 7.2 Black Pearls 2000........................................................................................... 24 7.3 Spectracarb 2225 carbon fabric ...................................................................... 26 7.4 Conclusions .................................................................................................... 31

8. References ................................................................................................................... 33

List of symbols/abbreviations/acronyms/initialisms ................................................................ 34

Distribution list ......................................................................................................................... 35

DRDC Atlantic CR 2007-120 ix

List of figures

Figure 3.1. Cyclic voltammograms (20 mV/s) of 1.4 mg Ru oxide in 1 M H2SO4. ................... 5

Figure 3.2. Specific capacitance of Ru oxide ............................................................................. 6

Figure 3.3. Anodic to cathodic charge ratio for Ru oxide .......................................................... 7

Figure 4.1. Cyclic voltammagrams of 30 wt % Ru (not fully aged) on Vulcan composite showing(a) current and (b) specific capacitance as a function of potential at different scan speeds in 0.5 M H2SO4. ....................................................................................................... 9

Figure 4.2. Cyclic voltammagrams of (a) 30 wt % Ru on Vulcan and (b) 43 wt % Ru on Black Pearl composites aged for different time intervals in air. Blank carbon CMEs are shown for the respective carbon supports as references. .............................................................. 10

Figure 6.1. Cyclic voltammogram (20 mV/s) of Ru oxide supercapacitors with different amounts of Nafion. Ru oxide loadings were a) 9.51, b) 10.35, c) 10.34 and d) 10.14 mg. ...................................................................................................................... 16

Figure 6.2. Capacitance plots for Ru oxide supercapacitors. The total loading of Ru oxide is specified ............................................................................................................................ 17

Figure 6.3. Capacitance plots for Ru oxide (10 mg) supercapacitors....................................... 18

Figure 6.4. Interfacial capacitance as a function of hydrous ruthenium oxide loading ............ 19

Figure 6.5. Constant current discharge plots for a supercapacitor with a total ruthenium oxide loading of 10.3 mg ............................................................................................................ 20

Figure 6.6. Ragone plots for supercapacitors with different amounts of Nafion binder. Ruthenium oxide loadings were 9.51, 10.14, 10.34 and 10.14 mg for 0%, 2.5%, 5% and 10% Nafion, respectively .................................................................................................. 21

Figure 6.7. Ragone plots obtained by different measurement method. The ruthenium oxide loading was 10.34 mg, with 2.5 % Nafion ........................................................................ 22

Figure 7.1. ESR corrected capacitance vs resistance plot for 1 cm2 BP2000 electrodes......... 25

Figure 7.2. Cyclic voltammograms of Spectracarb supercapacitors ........................................ 27

Figure 7.3. Nyquist plots for Spectrocarb Supercapacitors in 1 M H2SO4(aq) ....................... 28

Figure 7.4. ESR corrected capacitance plots for Spectrocarb supercapacitors in 1 M H2SO4(aq) ........................................................................................................................ 29

Figure 7.5. Capacitance plots for Spectrocarb supercapacitors in acetonitrile ......................... 31

x DRDC Atlantic CR 2007-120

List of tables

Table 4.1. Table of average specific capacitance values for various % Ru on carbon............. 11

Table 5.1. Specific capacitances and resistances measured for commercial carbon blacks ..... 13

Table 7.1. Selected data from impedance on BP2000 electrodes ............................................. 26

Table 7.2. Paremeters from the impedance of Spectracarb supercapacitors............................. 29

DRDC Atlantic CR 2007-120 xi

Acknowledgements

This work was supported in part by Memorial University and NSERC.

xii DRDC Atlantic CR 2007-120


DRDC Atlantic CR 2007-120 1

1. Introduction

The focus of the project in year one has been to evaluate the performances of silica-based sol-gels as ion conductors in supercapacitor electrodes. Two types of capacitive material commonly used in supercapacitors, Ru oxide and high surface area carbon, were selected for this work. The Ru oxide was synthesized by us, by adapting literature methods, while various commercial carbons were used. It became apparent during the course of the project that we were able to obtain much better performances from these materials than expected based on literature reports. Resistances measured by impedance spectroscopy were very low, and so thick layers were required in order to obtain meaningful results. This necessitated the development of supercapacitor cells much earlier in the project than originally anticipated (Device fabrication and testing was originally scheduled for Year 2). Consequently the development of other sol-gels has been delayed.

High surface area carbon is used as the capacitive material in most commercial supercapacitors [1, 2]. The main attractions of carbon are its low cost, and the high cell voltages (up to ca. 3.2 V [1]) that it can provide with non-aqueous electrolytes. Because of the voltage advantage, non-aqueous electrolytes are typically used in carbon supercapacitors; acetonitrile containing Et4NBF4 now being the preferred electrolyte. Specific capacitances as high as 250 F/g has been reported [2]. Ref. [3] provides a good description of the construction and performance of a state-of-the-art carbon supercapacitor.

Ru oxide has been widely studied as a supercapacitive material because of its high specific capacitance. However, its use in real capacitors has been limited by its cost (ca. $10/g) and the limited cell voltages (ca. 1 V) that it can sustain because it can only be used with aqueous electrolytes. Nonetheless, its very high specific capacitance and a high conductivity aqueous acid electrolyte combine to provide the highest power densities available. There have been many recent papers on improving the synthesis of Ru oxide to achieve higher specific capacitances, and on depositing layers of Ru oxide with high capacitance per cm2. The best synthesis methods give materials with specific capacitances of over 700 F/g, although such values have only been reported for thin layers of oxide (< 2 mg cm-2) [4, 5]. The sharp drop in specific capacitance at higher loadings appears to be due to mechanical instability of thick films [4], or the formation of denser films [5].

In order to improve the useable capacitance of Ru oxide, and to decrease the cost of electrodes, composites with carbon have been widely investigated. Specific capacitances for the Ru oxide component as high as 1580 F/g have recently been reported [6], although the accuracy is this number is doubtful.

This report details our work on the synthesis of highly capacitive Ru oxide (Section 3) and carbon supported Ru oxide (Section 4), the evaluation of various commercial carbons (Section 5), and the construction and evaluation of Ru oxide (Section 6) and carbon (Section 7) supercapacitors. Both aqueous acid and acetonitrile electrolytes have been used in the carbon supercapacitors. Nafion and a sulphonated sol-gel have been investigated as binders. Nafion and Celgard electrode separators have been used. Nafion has been found to be very

2 DRDC Atlantic CR 2007-120

effective as both the binder and separator in Ru oxide supercapacitors, producing impressive power and energy densities.


2. Some theoretical considerations

From consideration of the basic equations that determine the power of a supercapacitor [7], it was concluded that the maximum power (and hence power density) occurs with the minimum amount of active material. The maximum power is:

Pmax = ΔV2/4ESR

Where ΔV is the initial voltage across the device, and ESR is its effective series resistance, which is dominated by the electrolyte resistance. For a fixed current collector area, increasing the amount of capacitive material cannot increase the power, but will eventually increase R and so decrease the power. The power density is therefore inversely proportional to the mass of material, or worse. The benefit of using more material is that the energy density of the device increases, but this is only true if the full mass of the device is used (the energy density decreases with increased loading when only the mass of the capacitive material, mmaterial, is used), and the device can deliver useful power for a longer time. We estimate the device mass (mdevice), excluding mmaterial, to be ca. 100 mg cm-2. The loading of capacitive material needs to be similar to this mass to achieve the targeted energy density of 40 J/g.

Since the loading of active material will be chosen based on the optimum compromise between power and energy densities, and this will change with the application, we are focusing on minimizing the resistivity of the capacitive material, since this is the fundamental parameter that, for a certain capacitive material, determines the energy density at a chosen operating power (or power available at a specified energy density).


3. Synthesis, Characterization and Capacitance of Ruthenium Oxide

3.1 Preparation of hydrous ruthenium oxide powder

The hydrous ruthenium oxide power was prepared by a modified version of the sol-gel method described by Zheng and coworkers [8]. 3 M Na2CO3 was slowly added to a 0.05 M aqueous solution of RuCl3.xH2O (Precious Metals Online Pty Ltd) with stirring at room temperature. A ruthenium oxide sol-gel was formed when the pH reached ca. 7, and this was heated to 75 oC. After ca. 5 min, the precipitate completely or partly deposited on the bottom of the container was collected by filtration (Whatman 4) and washed many times with deionized water until the filtrate became cloudy (due to peptization). Finally the Ru oxide sample was dried for a period of 3 hours at 110 oC in air.

3.2 The preparation of electrodes

Ru oxide (2-60 mg), an equal mass of 5 % Nafion solution (DuPont) and a few drops of water were ground together to form a paste. This was painted with a brush onto a 1 cm2 disc of dried and wieghed carbon fibre paper (CFP; TorayTM Carbon Paper, TGP-H-090). The electrodes were dried in air for 10 minutes at 110 oC, and then re-weighed. The mass loading of Ru oxide was calculated by equation (1) to account for the 5% of Nafion binder.

2 2.0.95( )RuO xH O CFP paste CFPm m m+= − (1)

Where 2 2.RuO xH O

m , CFP pastem + and CFPm are the mass of hydrous ruthenium oxide, electrode and blank electrode, respectively.

3.3 Electrochemical characterization

The electrochemical properties of the ruthenium oxide electrodes were characterized by cyclic voltammetry using a three-electrode setup (Potentiostat and Galvanostat, EG&G 273A) with a Pt wire counter electrode and Ag/AgCl reference electrode. The electrolyte was 1 M H2SO2(aq) unless stated otherwise.

The specific capacitance (Csp(cv)) was calculated by the following equation (2)

2 2

( ).2

a csp cv

RuO xH O

Q QCm V

+⏐ ⏐=

Δ (2)

Where aQ , cQ , 2 2.RuO xH Om and VΔ are the anodic charge, cathodic charge, the mass of ruthenium oxide and potential widow, respectively.


Fig. 3.1-3.3 show cyclic voltammograms for a typical electrode together with plots of specific capacitance and anodic to cathode charge ratio recorded over a period of 50 cycles. After a few cycles, the electrode exhibited stable capacitive behaviour with good reversibility (Qa/Qc ~ 1). The capacitance was stable (except for one anomalously high result) at 650 F/g. Similar results for four other electrodes gave an average specific capacitance of 678±69 F/g.

-30

-20

-10

0

10

20

30

0 200 400 600 800 1000

E(mV)

I(mA

)

1cycle10cycle25cycle26cycle50cycle

Figure 3.1. Cyclic voltammograms (20 mV/s) of 1.4 mg Ru oxide in 1 M H2SO4.


598.9

652.6 651.8

712.3

651.3

580

600

620

640

660

680

700

720

0 10 20 30 40 50 60

cycle number

C(s

peci

fic c

apac

itanc

e) F

/g

Figure 3.2. Specific capacitance of Ru oxide


0.893

0.999 1.0011.014 1.01

0.88

0.9

0.92

0.94

0.96

0.98

1

1.02

1.04

0 10 20 30 40 50 60

cycle number

Qa/

Qc

Figure 3.3. Anodic to cathodic charge ratio for Ru oxide


4. Synthesis, Characterization and Capacitance of Carbon Supported Ru Oxide

4.1 Synthesis, characterization and electrode preparation

The ruthenium-carbon composites were prepared as follows: x milligrams of either Vulcan Carbon XC-72 or Black Pearl 2000 were combined with 30 mL of distilled deionized water (DDW). Next, (100 – x) mg of the RuCl3·xH20 precursor was combined with 30 mL of DDW. The ruthenium precursor solution was then added dropwise to the carbon black suspension with stirring and sonication. Next, an aqueous solution containing a 2 times excess of NaBH4 was added to the ruthenium precursor / carbon black solution with stirring and sonication. The final product was collected via suction filtration and rinsed with copious amounts of DDW. The composites were air-dried for at least 24 hours prior to electrode preparation. The Ru nano-particles (3.1-4.3 nm by XRD) are (in theory) electrochemically converted to hydrous RuO2. XRD and TEM were used to measure particle sizes and XPS was used to probe the Ru oxidation state. Specific capacitances were determined by CV.

Chemically modified electrodes (CMEs) incorporating the carbon-supported ruthenium composites were prepared as follows: carbon fibre paper (CFP) strips were cut with dimensions 3.0 cm (l) x 0.4 cm (w) x 0.025 cm (d). To make the CMEs, both inks and pastes of the composites were prepared using small amounts of isopropanol (0.1 to 0.5 mL) and/or Nafion binder (1% w:w). The pastes were spread onto the CFP electrode surfaces using a spatula, while the inks were used to impregnate the CFP using a fine brush. All CMEs were air-dried for at least 12 hours prior to electrochemical characterization.

CVs were measured from -0.2 V to +1.0 at 10 mV/s in 1.0 M H2SO4. For each electrode, the anodic and cathodic currents from 0 V to +1.0 V were used to calculate the specific capacitance; the absolute average was then obtained. The values reported represent the average specific capacitance of at least 3 electrodes prepared with the same percentage of ruthenium. However, the mass loadings of the Ru/C on the CFP were variable (0.50 mg to 2.0 mg), and did appear to have an influence on the specific capacitance. The lower mass loadings tended to yield higher specific capacitances than higher loadings.

4.2 Summary of results

Fig. 4.1 shows the scan rate dependence of the CV of a composite, and the conversion of the CVs to plots of specific capacitance vs. potential (cyclic capacitance plots) by division of the current by the scan speed and mass. All of the composites exhibited high specific capacitances, with little loss of capacitance with increasing scan speed. Capacitances were found to increase with aging (storage in air), as shown in Fig. 4.2, indicating that the Ru was slowly converted to Ru oxide. This conclusion is supported by XRD and XPS data that will be provided in Aaron’s thesis.


-15

-10

-5

0

5

10

15

-0.2 0 0.2 0.4 0.6 0.8 1 1.2

E / V vs. SCE

I / m

A

10 mV/s50 mV/s100 mV/s

-150

-100

-50

0

50

100

150

-0.2 0 0.2 0.4 0.6 0.8 1 1.2

E / V vs. SCE

Spec

ific

Capa

cita

nce

/ F

g-1

10 mV/s50 mV/s100 mV/s

Figure 4.1. Cyclic voltammagrams of 30 wt % Ru (not fully aged) on Vulcan composite showing(a) current and (b) specific capacitance as a function of potential at different scan speeds in 0.5 M H2SO4.


-300

-200

-100

0

100

200

300

400

-0.2 0 0.2 0.4 0.6 0.8 1 1.2

E / V vs. SCE

Spec

ific

Capa

cita

nce

/ F g

-1

Vulcan Carbon blankAged 1 weekAged 20 weeks

-600

-400

-200

0

200

400

600

800

-0.2 0 0.2 0.4 0.6 0.8 1 1.2

E / V vs. SCE

Spec

ific

Cap

acita

nce

/ F g-

1

Aged 8 weeks

Black Pearl Blank

Figure 4.2. Cyclic voltammagrams (10 mV/s) of (a) 30 wt % Ru on Vulcan and (b) 43 wt % Ru on Black Pearl composites aged for different time intervals in air. Blank carbon CMEs are shown for the respective

carbon supports as references.


A summary of specific capacitances (i.e. based on the sum of the C + Ru oxide masses) of the composite materials (aged for at least 8 weeks) is given in Table 4.1, together with specific capacitances calculated for the Ru + oxide component. The % Ru + oxide indicated, and used to determine the latter specific capacitance is based on commercial carbon analyses (% Ru + oxide = 100 % - %C). It includes the masses of Ruo, anhydrous and/or hydrous oxides of unknown composition, oxygen groups on the carbon, and other impurities. Ru analyses are planned for year 2 to better quantify the compositions of selected materials. The high apparent Ru + oxide loadings for the black pearl samples, relative to the targeted values, are due in part to the high oxygen content of BP2000, which is only 93.43 % C by elemental analysis. There is probably also a higher % conversion of Ru to Ru oxide, as indicated by the higher specific capacitances of the Ru + oxide component.

Table 4.1. Table of average specific capacitance values for various % Ru on carbon

CARBON TYPE % Ru + OXIDE (TARGETED % Ru)

SPECIFIC CAPACITANCE (F/g)

SPECIFIC CAPACITANCE OF Ru

+ OXIDE (F/g)

9.6 (4.6) 56 460

23.5 (16) 106 409

29.9 (30) 182 577

53.4 (50) 250 456

Vulcan XC72

81.1 (79) 300 367

19.0 (4.6) 162 234

30.6 (16) 253 498

42.9 (30) 384 701

59.5 (50) 482 711

Black Pearls 2000

-a (79) 574 -

a. not determined due to lack of material

The specific capacitances of both types of composite (Vulcan and BP) increase approximately linearly with Ru loading, as expected. For the Vulcan carbon, specific capacitances of the Ru + oxide component do not show a significant dependence on Ru loading, with an average of 454±70 F/g. There is an apparent increase with loading for BP2000, but this is probably due to errors in the masses of Ru oxide due to the ca. 6.5 % oxygen functionality on the carbon. This would cause a much larger error at low Ru loadings. The effect is seen in the discrepancies between the targeted Ru loadings and measured Ru + oxide loadings. As such, the ca. 700 F/g measured at 43 and 60 % Ru + oxide is probably the most accurate measure of the Ru oxide specific capacitance. Even higher values would be obtained if the mass of the oxide on the carbon was discounted.


4.3 Conclusions

The maximum specific capacitances for the Ru oxide component in our composites (ca. 700 F/g) are similar to most values in the literature for C + Ru oxide composites, and similar to the values that we have obtained with pure Ru oxide (Section 3). There does not therefore appear to be any particular advantage to either our method of preparing the composites, or the use of a carbon support. However, it should be noted that ca. 50 % of the Ru in our aged composites is still in the metallic form (based on XRD and XPS results). If this can be completely converted to Ru oxide, we would approach the highest specific capacitances reported for Ru oxide in composites, and have very attractive materials. Further investigation of electrochemical methods and heat treatment to produce the oxide, or non-reductive methods to deposit Ru oxide on BP2000 are warranted.


5. Evaluation of Commercial Carbons

Four commercial carbon blacks were obtained and evaluated in 1 M H2SO4(aq). Electrochemical measurements on the power samples (Black Pearls 2000, Norit SX Ultra and Ketjunblack ED600JD) were performed with electrodes prepared by physically impregnating carbon blacks into carbon fibre paper electrodes (CFP; 75 % porosity), using a spatula or paintbrush. Strips of CFP ca. 3 cm long and 0.5 cm wide were used, and the carbon black was coated on the lower 1 cm. A small amount of Nafion (as a 5 % solution) was added to the carbon paste/ink (with propane-2-ol) as a binder. The optimum Nafion loading was determined to be 1 mass %. Nafion does not significantly influence the specific capacitance or resistance of the electrodes at loadings up to 5 % by mass, but does make electrode preparation easier and more reproducible, and allows higher carbon loadings to be used. Results for loadings between 1.0 and 3.2 mg cm-2 are reported in Table 5.1. The CFP (0.28 mm thick) can accommodate ca. 1.5 mg of carbon black (density ca. 0.1 g cm-2) per cm2, so at higher loadings some must be spread on the surface. Cyclic voltammetry (CV) at 100 mV s-1 and impedance (IS) experiments were conducted in a conventional glass cell. The ionic resistance of each electrode was determined from the low and high frequency resistances as RI = 3*(Rlow – Rhigh). RIA (A = area of CPF supporting carbon black) was found to be independent of carbon loading per cm2, indicating that the effective thickness of the electrode was fixed at the CFP thickness.

Table 5.1. Specific capacitances and resistances measured for commercial carbon blacks

Carbon type Reported Specific area (m2 g-1)

Specific Capacitance from CV (F g-1)

Specific Capacitance from IS (F g-1)

RIA from IS

(Ω cm2)

Black Pearls 2000 (Cabot)

2100 169±42 173±50 0.55±0.15

Norit SX Ultra 1320 145±71 132±73 0.77±0.44

Ketjunblack ED600JD (Akzo Nobel)

1400 102±11 97±22 0.38±0.14

Spectacarb 2225 (Engineered Fibers Technology)

2500 197a 154±37 0.52±0.16

a single (best) result


One sample, Spectracarb 2225, was obtained as a fabric (ca. 14 mg cm-2) and evaluated in a sandwich cell (see Section 7.3). Results are included here for comparison with the powder samples.

It is clear from the data in Table 5.1 that Black Pearls 2000 offers the best specific capacitance of the carbon black powders, and that its contribution to the electrodes’ resistance is not significantly higher than for the others. It was therefore used in all further work with carbon black powders. Several electrodes prepared with Norit SX Ultra gave anomalously large specific capacitances (250 and 290 F g-1) and this may be worth further examination at some point.


6. Ruthenium Oxide Supercapacitors

In light of the excellent capacitive behaviour described in Section 3 for our Ru oxide, and in order to measure the resistances of Ru oxide layers as a function of thickness, further experiments were conducted in a prototype supercapacitor (sandwich cell). In this cell an electrolyte separator (Nafion or Celgard 3400 impregnated with H2SO4 (aq)) is sandwiched between two similar electrodes consisting of the capacitive material with a binder spread on carbon fibre paper. Ti plate current collectors are used, and the whole cell is immersed in a H2SO4 (aq) solution containing a reference electrode. CV and impedance can be obtained for each electrode separately (3 electrode mode, with the other electrode acting as the counter electrode), or for the two electrodes simultaneously (i.e. acting as a supercapacitor in 2-electrode mode). The sandwich cell allows us to use thicker layers of capacitive material and minimize the electrolyte resistance. In supercapacitor mode it allows us to determine energy and power densities as a function of discharge rate, and therefore construct Ragone plots. Electrodes were prepared as described in Section 3.2.

6.1 Cyclic Voltammetry

Figure 6.1 shows representative cyclic voltammograms in supercapacitor mode for electrodes prepared with various amounts of Nafion binder, and with a Nafion 115 separator. The Ru oxide loading on each electrode was kept constant at ca. 10 mg. All of the cells showed high quality capacitive behaviour with excellent reversibility. The Nafion binder enhances reversibility. The optimum Nafion loading is below 5%.

6.2 Impedance Spectroscopy

The measurements of impedance spectroscopy were as follows. The frequency range was from 10 kHz to 5 mHz or from 10 kHz to 1 mHz, with an amplitude of 10 mV and a DC bias potential of 1 V. The specific capacitance is calculated by equation (3) for the three-electrode setup or equation (4) for the two-electrode setup.

2 2

( ).

EISsp EIS

RuO xH O

CCm

= or 2 2

( ).

EISsp EIS

RuO xH O

CCS

= (3)

( )( )EIS a c

sp EISa c

C m mCm m× +

=×

(4)

Where ( )sp EISC , EISC , 2 2.RuO xH OS , 2 2.RuO xH Om , am and cm are specific capacitance, measured capacitance, interfacial surface of cathode, the mass of hydrous ruthenium oxide on the working electrode, anode electrode and cathode, respectively. The capacitance of each cell generally increased somewhat (typically ca. 10 %) with time/use, while the resistance decreased. Reported data were obtained after stabilization of these trends.


-50

-40

-30

-20

-10

0

10

20

30

40

50

-1000 -500 0 500 1000

cell voltage (mV)

curr

ent (

mA

)

a: no Nafion

b: 2.5 % Nafion

c: 5 % Nafion

d: 10 % Nafion

b

b

a

a

Figure 6.1. Cyclic voltammogram (20 mV/s) of Ru oxide supercapacitors with different amounts of Nafion. Ru oxide loadings were a) 9.51, b) 10.35, c) 10.34 and d) 10.14 mg.

6.2.1 Representative Capacitance Plots

Fig. 6.2 shows plots of the series capacitance (Cs = 1/ωZ″) vs real impedance (resistance) for supercapacitors (2 electrode mode) with high Ru oxide loadings (the total loading on both electrodes is specified). High frequency resistances (ESR) were ca. 0.5 Ω and maximum capacitances were approached (i.e. 90 %) at ca. 0.65-1.0 Ω. These are remarkably low resistances for such high loadings, particularly for the 49 mg device. The lack of a clear correlation with loading, and the significant charge transfer resistances seen before the capacitance rises, suggest that the resistances could be improved further. The capacitance increased with loading as expected, and this is discussed further in section 6.2.2.


0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

0.0 0.5 1.0 1.5 2.0

Resistance (Ohm)

Cap

acita

nce

(F)

27 mg49 mg103 mg

Figure 6.2. Capacitance plots for Ru oxide supercapacitors. The total loading of Ru oxide is specified

Fig. 6.3 shows results from optimization of the supercapacitor’s resistance at a fixed (10 mg) Ru oxide loading. A thinner Nafion membrane (Nafion 112 vs 115) and thinner CFP support both improved the resistances at both high and low frequencies, and capacitance. Use of a Celgard membrane decreased the resistance relative to Nafion 115, but the capacitance was lower. The lowest ESP of ca. 0.24 Ω would give a peak power density for the 10 mg loading of over 100 kW/kg, and even higher power densities for lower loadings.


0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0.0 0.2 0.4 0.6 0.8 1.0

Resistance (Ohm)

Cap

acita

nce

(F)

Celgard; 0.3 mm CFP0.13 mm Nafion; 0.3 mm CFP0.05 mm Nafion; 0.1 mm CFP)

Figure 6.3. Capacitance plots for Ru oxide (10 mg) supercapacitors

6.2.2 Impedance and capacitance vs. Ruthenium oxide loading

Figure 6.4 shows that the interfacial capacitance of ruthenium oxide electrodes increased linearly with increasing ruthenium oxide loading. The results are consistent with those of Fang et al [9], however, we get much higher interfacial capacitance than their results. They reported a maximum interfacial capacitance of 4 F/cm2 obtained with a 10 mg/cm2 Ru oxide loading. Compared to their results, we obtained 4.18 F/cm2 (694 F/g) with a 6 mg/cm2 loading. When the ruthenium oxide loading was increased to 51.2 mg, we obtained a remarkable interfacial capacitance of 34.9 F/cm2 (682 F/g). Jang et al [4, 10] reported that specific capacitance gradually decreased with increasing ruthenium oxide loading, and the authors think the poor mechanical stability of electrode layers contributes to lower specific capacitance.


0 5 10 15 20 25 30 35 40 45 50 55 6002468

10121416182022242628303234363840

R= 0.99971

Y = 0.29764 + 0.68115X

Inte

rfaci

al C

apac

itanc

e(F/

cm2 )

Loading Mass of RuO2(mg/cm2)

experimental results results of linear fit

Figure 6.4. Interfacial capacitance as a function of hydrous ruthenium oxide loading

Although the thickness increases with increasing ruthenium oxide loading, the whole ruthenium oxide layer can still effectively be utilized. The proton-conductivity of the layer plays a crucial role in the charging and discharging processes. The Nafion binder provides a proton transfer path between particles, so the ruthenium oxide particles located deeper within the layer can participate in the process of proton transfer, thus enhancing the efficiency of charging and discharging.

6.3 Constant current discharge

The results of a series of constant current discharge experiments are shown in Fig. 6.5. The potential dropped sharply at the beginning of each experiment due to the internal resistance (ESR) and then decreased approximately linearly as expected for a capacitor. ESR values determined from the initial potential drop were typically ca. 0.4 Ω, which is consistent with values determined by impedance. Values as low as 0.25 Ω cm2 have been reported in the literature [8], and we have achieved lower values than this in our most recent work.


0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.00

100

200

300

400

500

600

700

800

900

1000

1100

1mA 10mA 0.1A 0.2A 0.3A 0.4A 0.5A 0.6A 0.7A 0.8A 0.9A 1A

Pot

entia

l(vol

t)

time/second

Figure 6.5. Constant current discharge plots for a supercapacitor with a total ruthenium oxide loading of 10.3 mg

Usable energy density and average power density – The usable energy density and power density of supercapacitors are obtained by a series of mathematic calculations from experimental data. The relevant equations (6)-(9) are as follows

t tP I V= × (6)

0 0

0 0

V Vt t

t tt t

E Pdt I V dt= =

= =

= = ×∫ ∫ (7)

spa c

EEm m

=+

(8)

0

0

Vt

tsp t

sp

I V dtE

Pt t

=

=

×= =

Δ Δ

∫ (9)


Where Pt , I, Vt, E, ma, mc, Esp, Psp, tΔ are transient power at discharging time t, constant current, potential at discharging time t, energy, the mass of ruthenium oxide on the anode, the mass of ruthenium oxide on the cathode, usable energy density, average power density and total discharging time.

Ragone plots derived from results of cells with different amounts of Nafion binder are shown in Fig. 6.6, and results for 2.5 % binder from constant current, CV, and impedance experiments are compared in Fig. 6.7 as log Ragone plots. Results from the former two techniques agree well, but further work is needed to improve the accuracy of energy and power results derived from impedance.

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 3202468

1012141618202224262830

0% Nafion Binder 2.5% Nafion Binder 5% Nafion Binder 10% Nafion Binder

Ave

rage

pow

er d

ensi

ty (K

W/K

g)

Usable energy density(Wh/Kg)

Figure 6.6. Ragone plots for supercapacitors with different amounts of Nafion binder. Ruthenium oxide loadings were 9.51, 10.14, 10.34 and 10.14 mg for 0%, 2.5%, 5% and 10% Nafion, respectively


0.25 0.5 1 2 4 8 16 32

0.0625

0.125

0.25

0.5

1

2

4

8

16

32

64

results of cyclic voltammetry results of constant current discharging results of impedance spectrometry

Pow

er d

ensi

ty (K

W/K

g)

Energy density(Wh/Kg)

Figure 6.7. Ragone plots obtained by different measurement methods. The ruthenium oxide loading was 10.34 mg, with 2.5 % Nafion

6.4 Conclusions

The Ru oxide that we have prepared performs extremely well in supercapacitors with a 1 M H2SO4(aq) electrolyte. Specific capacitances of nearly 700 F/g are maintained at electrode loadings as high as 50 mg cm-2, and the effective series resistance (ESR) has been decreased to 100 kW/kg), and energy densities exceed 30 Wh/kg (108 J/g). This compares favourably to conventional electrolytic and ceramic capacitors (typically around 100 kW/kg and 10 mWh/kg, and 10 MW/kg and < 1mWh/kg, respectively).

We believe that these outstanding performances are due primarily to our synthesis method and the use of CFP (4 mg cm-2 for the thinnest sample) as a support for the Ru oxide. The CFP decreases the interfacial resistance between the Ru oxide and the Ti current collector, and may also stabilize the structure of the Ru oxide layer. The use of Nafion as a binder and a Nafion electrode separator are also beneficial.

The impedance results have shown that the ionic resistance of the Ru oxide layer is small, even at high loadings (e.g.


using higher concentrations of H2SO4, and so the possible benefits of adding a silica sol-gel are minimal. Several attempts were made to incorporate a sulphonated silane into the Ru oxide electrodes, but resistances were high and capacitances were low. It is not thought that further work with the silicas and Ru oxide is warranted.


7. Carbon Supercapacitors

The primary objective of the work described in this Section was to investigate the use of sulphonated silica gels as binders and ion conducting agents for carbon black supercapacitors. Nafion was also used for comparative purposes.

A single composition of gel prepared from a 8.3:1 molar ratio of tetramethyl orthosilicate (TMOS) and a sulphonated silane was used [11].

7.1 Experimental

A mixture of 0.6 mL of 2(4-chlorosulfonylphenyl)ethyl-trichlorosilane (50 % in CH2Cl2 from United Chem. Technol., Inc., Petrarch Silanes, Silicones and Bonded Silicas) and 2.4 mL of methanol was added to 1.5 mL of tetramethyl orthosilicate (TMOS; 98% from Aldrich), 1.3 mL deionized water and 0.3 mL 0.1 M HCl in a round bottom flask and heated at reflux for 1.5 hours. Following partial gellation at ambient temperature overnight, the gel was stored in a fridge or freezer (for longer time storage with less gellation) until needed. Thicker gels were mixed with propan-2-ol for electrode preparation, while more liquid samples were used neat.

The cell described at the beginning of Section 6 was used with a Nafion 115 electrode separator and immersed in 1 M H2SO4(aq) unless otherwise stated.

7.2 Black Pearls 2000

Electrodes were prepared by mixing Black Pearls 2000 (BP2000) with propan-2-ol and either Nafion solution or an ormosil gel as a binder, and spread on CFP (0.28 mm thick) with a spatula. 1 cm2 discs were punched out, weighed, and put into the sandwich cell. The two electrodes were matched as closely as possible to prevent damaging potential excursions. We have found it very difficult to prepare electrodes with > 2 mg cm-2 loadings because of the very low density of the carbon black. This means that ionic resistances of the carbon black layers measured by impedance are small and have large uncertainties.

All measurements with BP2000 in the sandwich cell were conducted in 3-electrode mode, with a Ag/AgCl reference electrode in the electrolyte solution. This was later found to give somewhat inaccurate impedance characteristics, and so this should be kept in mind when assessing the following results.

Fig. 7.1 shows capacitance plots for electrodes with Nafion or ormosil binders. Additional data are provided in Table 7.1. Within experimental error, the specific capacitances (CSP) of the BP2000 and the ESR were the same for all electrodes. The average CSP of 161±37 F g-1 agrees well with the value reported in Table 5.1. The average ESR of 0.16±0.04 Ω cm2 corresponds to a full-cell ESR of 0.32 Ω cm2, similar to typical values obtained for our Ru oxide capacitors with Nafion 115 (Section 6.2.1).


0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.00 0.05 0.10 0.15 0.20 0.25 0.30Real Impedance - ESR (Ohm)

-1/w

Z"

(F)

0.82 mg + 10 % ormosil 0.76 mg + 10 % ormosil1.14 mg + 40% ormosil1.08 mg + 40 % ormosil0.33 mg + 1 % Nafion0.66 mg + 1% Nafion1.1 mg + 3 % Nafion

low frequency limiting capacitance

slope proportional to ion conductivity

Figure 7.1. ESR corrected capacitance vs resistance plot for 1 cm2 BP2000 electrodes

In the capacitance plots the slope of the medium frequency region of the curve is proportional to the ionic conductivity of the carbon black layer, and the limiting capacitance at low frequency is the total capacitance of the layer. It can be seen that use of 40 % gel + solvents by mass gives the best conductivity and specific capacitance. Use of 3 % Nafion (i.e. 40 % Nafion + solvents) also appears to give a high conductivity, but the data are offset along the real impedance axis by a charge transfer resistance, and the specific capacitance is slightly inferior.

The masses given in the legend of Fig. 7.1 are the mass of BP2000. The ormosil gel contained ca. 80-90 % water and 2-propanol by mass, and so the mass loading of dry ormosil in the 40% electrodes was ca. 5-10 %. Use of Nafion at such loading levels leads to a large loss of specific capacitance.


Table 7.1. Selected data from impedance on BP2000 electrodes

BINDER LOADING OF BP2000 (mg cm-2)

SPECIFIC CAPACITANCE (F g-1)

ESR (Ω)

10 % ormosil 0.82 138 0.19

10 % ormosil 0.76 158 0.23

40 % ormosil 1.14 147 0.15

40 % ormosil 1.08 173 0.14

1 % Nafion 0.33 234 0.10

1 % Nafion 0.66 174 0.19

3 % Nafion 1.10 106 0.15

CSP does not included the mass of the binder

7.3 Spectracarb 2225 carbon fabric

Electrodes were prepared by simply punching 1 cm2 discs from the fabric. The average mass was 14.1±0.4 mg. Initially, the two Spectracarb discs were placed in direct contact with the Ti current collector plates, with a Nafion 115 membrane between them. However, impedance revealed the presence of large interfacial resistances that were traced to the Ti/Spectracarb interafaces. These resistances were eliminated by placing a CFP disc between each Spectracarb electrode and its Ti current collector. Celgard 3400 was used as the electrode separator for experiments with acetonitrile electrolytes. Ormosil containing electrodes were prepared by adding one or two drops of sol to a Spectracarb disc and allowing it to gel overnight in a sealed vial. The discs were generally then allowed to dry briefly in air and weighed.

Initially, experiments with Spectracarb electrodes were run in 3-electrode (half-cell) mode with a reference electrode in the electrolyte solution. However, the two nominally identical electrodes in each cell always gave significantly different impedance responses. The cause was found to be uneven current densities at the edges of the electrodes, and the problem was alleviated, but not fully corrected, by drilling holes through the cell body into the back of the electrodes and inserting a luggin capillary. In light of this unavoidable uncertainty in the 3-electrode measurements, it was decided to use 2-electrode measurements with matched electrodes for subsequent work. Thus all data reported in this Section are for 2-electrode (full-cell) experiments.

Fig. 7.2 shows cyclic voltammograms for spectrcarb supercapacitors in 1 M H2SO4(aq) and acetonitrile (AC) with two different electrolytes. For both solvent systems, data for electrodes with and without ormosil are shown. The –1 to +1 V scans are for the H2SO4(aq) electrolyte, and it is clear that the capacitor can only be reversibly charged to ca. 0.7 V (+ve


or –ve). The broad peak at ca. 0 V is due to quinone groups on the surface of the carbon. Addition of ormosil to the carbon fabric (ca. 15 mg to each electrode), before cell assembly, did not significantly change the voltammetric behaviour. The best specific capacitance measured was 197 F g-1, similar to values reported by Conway and coworkers for similar conditions [12].

-0.03

-0.02

-0.01

0.01

0.02

0.03

-2.00 -1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 2.00

cell voltage (V)

curr

ent (

A)

1M H2SO4

30 mg ormosil + 1M H2SO4

AC + 1M Et4NBF4

AC + 1M LiBF4

Ormosil + AC + 1 M LiBF4

10 mV/s

Figure 7.2. Cyclic voltammograms of Spectracarb supercapacitors

In AC, the supercapacitors could be charged to 2 V with reasonable reversibility. Capacitances (proportional to the currents in Fig. 7.2) were somewhat lower than in H2SO4(aq), mainly due to a smaller contribution from the quinone sites. The differences between the three results in AC are probably due to differences in the water content of the electrolyte, which was not well controlled. The ormosil would certainly be expected to trap water in the electrode structure, and this would account for the higher capacitance that it produces. A disadvantage of the ormosil however, is the higher irreversible currents at high cell voltages, which are again presumably due to water in the electrodes.

Fig. 7.3 shows complex plane impedance (Nyquist) plots for Spectracarb supercapacitors with and without ormosil in H2SO4(aq). All four supercapacitors gave similar Nyquist plots


exhibiting almost ideal porous electrode behaviour (45o then 90o regions). ESR values ranged between 0.33 and 0.42 Ω, without any clear trend. The differences are best attributed to differences in contact resistances. The electrodes with ormosil exhibited higher total resistances than those without, but this may not be significant (see below).

-1.0

-0.8

-0.6

-0.4

-0.2

0.00.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Real Impedance (Ohm)

Imag

inar

y im

peda

nce

(Ohm

)

no ormosilno ormosil (16 h)30 mg ormosil40 mg ormosil

Figure 7.3. Nyquist plots for Spectrocarb Supercapacitors in 1 M H2SO4(aq)

Differences between the electrodes can be more clearly seen in the ESR corrected capacitance plots in Fig. 7.4. Now it can be seen that ionic conductivity (proportional to slope) and the limiting capacitance were both significantly higher for one of the cells without ormosil, while the differences in the other three electrodes were probably not significant. The anomalously good cell (no ormosil (16 h)) had been assembled following saturation of the electrodes with H2SO4(aq), and was not immersed in H2SO4(aq) until the following day. Its outstanding performance may have been due to more effective displacement of air bubbles from the carbon fabric, or possibly activation of the carbon surface. In future work, the effects of entrapment of air will be investigated and minimized.


0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

0.0 0.1 0.2 0.3 0.4

Resistance - ESR (Ohm)

Cap

acita

nce

(F)

no ormosilno ormosil (16 h)30 mg ormosil40 mg ormosil

Figure 7.4. ESR corrected capacitance plots for Spectrocarb supercapacitors in 1 M H2SO4(aq)

It is encouraging that one of the ormosil containing electrodes exhibited better conductivity and capacitance than the other ormosil-free electrode, although better reproducibility is required to establish whether the ormosil is of significant value in Spectracarb electrodes. It should be noted that although the ormosil adds to the mass of the electrode, it will decrease the mass of H2SO4(aq) required and so its net mass will be small – possibly negative.

ESR, specific capacitance and electrode ion resistance values from the impedance data in Figs. 7.3 and 7.4 are presented in Table 7.2.

Table 7.2. Paremeters from the impedance of Spectracarb supercapacitors

ELECTROLYTE ESR (Ω) CS (F/g) RION (Ω)

1 M H2SO4(aq) 0.33 117 0.64


1 M H2SO4(aq) (16 h) 0.38 190 0.36

30 mg ormosil + 1 M H2SO4(aq)

0.33 129 0.74

40 mg ormosil + 1 M H2SO4(aq)

0.42 153 0.58

AC + 1 M Et4NBF4 0.99 95 6.6

AC + 1 M Et4NBF4 (Nafion separator)

0.80 87 9.0

Ormosil +AC + 1 M Et4NBF4

1.1 120 10.2

AC + 1 M LiBF4 1.0 78 9.3

Fig. 7.5 shows capacitance plots for Spectracarb supercapacitors in AC electrolytes. Pertinent parameters at listed in Table 7.2. The ESR in AC was only 2-3 times higher than in H2SO4(aq), indicating the existence of significant contact and current collector/cable resistances in the measurements. There were not significant differences between Celgard and Nafion as the separator (Celgard is preferred because Nafion is so hygroscopic). Et4NBF4 and LiBF4 gave similar results. The specific capacitance was lower in AC vs. H2SO4(aq) by at least 30 %, and RION was higher by a factor of >10, as expected. Use of the ormosil improved the specific capacitance to close to the H2SO4(aq) value, as also indicated by cyclic voltammetry, but surprisingly did not improve RION.


0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

Resistance (Ohm)

Cap

acita

nce

(F)

1M Et4NBF41 M Et4NBF4 (Nafion separator)22 mg ormosil + 1 M Et4NBF41 M LiBF4

Figure 7.5. Capacitance plots for Spectrocarb supercapacitors in acetonitrile

7.4 Conclusions

Spectracarb 2225 carbon fabric provides excellent supercapacitor performance in both H2SO4(aq) and AC electrolytes. It is very easy to use, provides a convenient and reproducible loading of 14 mg cm-2 and exhibits resistances as low as those for much lower loadings of BP2000 (Table 5.1). With more effective removal of air and reduction of ESR by the methods described in Section 6, even better performances can be achieved.

The ormosil has yielded some encouraging results with both BP2000 and Spectracarb 2225, although better control of experimental errors is required to verify that the ormosil can improve the specific capacitance and/or ionic resistance. The effects are clearly quite small, but could be significant with optimization. However, it should be noted that use of higher H2SO4(aq) concentrations is probably a better way to improve performances, and the effect of the ormosil may then be even lower. The most useful role of the ormosil may be to


improve capacitances in AC, if this effect does not diminish with better drying (needed to decrease background currents).


8. References 1. R. Kotz and M. Carlen, Electrochim. Acta 45, 2483 (2000).

2. A.G. Pandolfo and A.F. Hollenkamp, J. Power Sources 157, 11 (2006).

3. P.L. Taberna, C. Portet, and P. Simon, Applied Physics A-Materials Science & Processing 82, 639 (2006).

4. J.H. Jang, A. Kato, K. Machida, and K. Naoi, J. Electrochem. Soc. 153, A321 (2006).

5. B.O. Park, C.D. Lokhande, H.S. Park, K.D. Jung, and O.S. Joo, J. Power Sources 134, 148 (2004).

6. C.C. Hu and W.C. Chen, Electrochim. Acta 49, 3469 (2004).

7. W.G. Pell and B.E. Conway, J. Power Sources 63, 255 (1996).

8. J.P. Zheng, P.J. Cygan, and T.R. Jow, J. Electrochem. Soc. 142, 2699 (1995).

9. Q.L. Fang, D.A. Evans, S.L. Roberson, and J.P. Zheng, J. Electrochem. Soc. 148, A833 (2001).

10. J.H. Jang, K. Machida, Y. Kim, and K. Naoi, Electrochim. Acta 52, 1733 (2006).

11. W.M. Aylward and P.G. Pickup, Journal of Solid State Electrochemistry 8, 742 (2004).

12. J.J. Niu, W.G. Pell, and B.E. Conway, J. Power Sources 156, 725 (2006).


List of symbols/abbreviations/acronyms/initialisms

DND Department of National Defence

Et4NBF4 Tetraethylammonium tetrafluoroborate

LiBF4 Lithium tetrafluoroborate


Distribution list LIST PART 1: CONTROLLED BY DRDC Atlantic LIBRARY 5 DRDC Atlantic Library (4 CDs + 1 hardcopy) 2 Colin Cameron (1 CD + 1 hardcopy) 1 Jeff Szabo (GL Tailored Polymers) 1 Calvin Hyatt (Head, Emerging Materials) 1 Trisha Huber 10 TOTAL PART 1 LIST PART 2: DISTRIBUTED BY DRDKIM 1 NDHQ/DRDC/DRDKIM (PDF copy) 1 TOTAL PART 2 LIST PART 3: EXTERNAL DISTRIBUTION 1 Prof. Michael Freund (hardcopy) Department of Chemistry University of Manitoba Winnipeg, MB R3T 2N2 1 Prof. Alex Adronov (hardcopy) Department of Chemistry McMaster University 1280 Main St. W Hamilton, ON L8S 4M1 1 Prof. Daniel Belanger (hardcopy) Dept. de chimie Universite de Quebec a Montreal CP 888 Succ. Centre Ville Montreal, QC H3C 3P8 2 Prof. P.G. Pickup (1 CD + 1 hardcopy) Department of Chemistry Memorial University of Newfoundland St. John’s, NL A1B 3X7 5 TOTAL PART 3 16 TOTAL COPIES

DRDC Atlantic mod. May 02

DOCUMENT CONTROL DATA (Security classification of title, body of abstract and indexing annotation must be entered when the overall document is classified)

1. ORIGINATOR (the name and address of the organization preparing the document. Organizations for whom the document was prepared, e.g. Centre sponsoring a contractor's report, or tasking agency, are entered in section 8.)

Memorial University of Newfoundland Department of Chemistry St. John's, NL A1B 3X7

2. SECURITY CLASSIFICATION (overall security classification of the document

including special warning terms if applicable).

UNCLASSIFIED

3. TITLE (the complete document title as indicated on the title page. Its classification should be indicated by the appropriate abbreviation (S,C,R or U) in parentheses after the title).

Synthesis and Characterization of Modified Silicas and Carbons for Use as Electrodes in Electrochemical Supercapacitors: First Annual Report

4. AUTHORS (Last name, first name, middle initial. If military, show rank, e.g. Doe, Maj. John E.)

Pickup, Peter G.; Rowe, Aaron; Liu, Xiaorong; DesRoches, Derrick

5. DATE OF PUBLICATION (month and year of publication of document)

August 2007

6a. NO. OF PAGES (total containing information Include Annexes, Appendices, etc).

52

6b. NO. OF REFS (total cited in document)

12

7. DESCRIPTIVE NOTES (the category of the document, e.g. technical report, technical note or memorandum. If appropriate, enter the type of

report, e.g. interim, progress, summary, annual or final. Give the inclusive dates when a specific reporting period is covered).

CONTRACT REPORT

8. SPONSORING ACTIVITY (the name of the department project office or laboratory sponsoring the research and development. Include address).

Defence R&D Canada – Atlantic PO Box 1012 Dartmouth, NS, Canada B2Y 3Z7

9a. PROJECT OR GRANT NO. (if appropriate, the applicable research

and development project or grant number under which the document was written. Please specify whether project or grant).

12sz07

9b. CONTRACT NO. (if appropriate, the applicable number under which the document was written).

W7707-063350

10a ORIGINATOR'S DOCUMENT NUMBER (the official document

number by which the document is identified by the originating activity. This number must be unique to this document.)

DRDC Atlantic CR 2007-120

10b OTHER DOCUMENT NOs. (Any other numbers which may be assigned this document either by the originator or by the sponsor.)

11. DOCUMENT AVAILABILITY (any limitations on further dissemination of the document, other than those

imposed by security classification) ( x ) Unlimited distribution ( ) Defence departments and defence contractors; further distribution only as approved ( ) Defence departments and Canadian defence contractors; further distribution only as approved ( ) Government departments and agencies; further distribution only as approved ( ) Defence departments; further distribution only as approved ( ) Other (please specify):

12. DOCUMENT ANNOUNCEMENT (any limitation to the bibliographic announcement of this document. This will normally correspond to the Document

Availability (11). However, where further distribution (beyond the audience specified in (11) is possible, a wider announcement audience may be selected).

DRDC Atlantic mod. May 02

13. ABSTRACT (a brief and factual summary of the document. It may also appear elsewhere in the body of the document itself. It is highly desirable that the abstract of classified documents be unclassified. Each paragraph of the abstract shall begin with an indication of the security classification of the information in the paragraph (unless the document itself is unclassified) represented as (S), (C), (R), or (U). It is not necessary to include here abstracts in both official languages unless the text is bilingual).

A series of commercial high surface area carbons, ruthenium (Ru) oxide prepared in-house, and a series of carbon supported Ru oxide materials prepared in-house have been evaluated as capacitive materials for supercapacitors. A proton conducting organically modified silica gel has been evaluated as an electrode binder and shown to increase capacitances and decrease ionic resistances in carbon electrodes. Prototype supercapacitors have been built with Ru oxide and carbon electrodes. A device capacitance of 14 F has been demonstrated. The effective series resistance has been decreased to 0.24 cm2. Average power densities during full discharge exceed 25 kW/kg (peak power > 100 kW/kg), and energy densities exceed 30 Wh/kg.

14. KEYWORDS, DESCRIPTORS or IDENTIFIERS (technically meaningful terms or short phrases that characterize a

document and could be helpful in cataloguing the document. They should be selected so that no security classification is required. Identifiers, such as equipment model designation, trade name, military project code name, geographic location may also be included. If possible keywords should be selected from a published thesaurus. e.g. Thesaurus of Engineering and Scientific Terms (TEST) and that thesaurus-identified. If it not possible to select indexing terms which are Unclassified, the classification of each should be indicated as with the title).

Electrochemical Supercapacitors, modified carbons, high surface area carbons

Documents

Synthesis and Characterization of Modified Silicas and ...modification de la surface du carbone pour améliorer les densités d’énergie et de puissance. Pickup, P.G., Rowe, A.,