WO2002011227A1

WO2002011227A1 - Gelled electrolyte vanadium battery

Info

Publication number: WO2002011227A1
Application number: PCT/AU2001/000923
Authority: WO
Inventors: Maria Skyllas Kazacos; Amedeo Banzato
Original assignee: Unisearch Limited
Priority date: 2000-07-28
Filing date: 2001-07-27
Publication date: 2002-02-07
Also published as: AUPQ908700A0

Abstract

The present invention relates generally to gelling of vanadium electrolytes by the addition of an inorganic gelling agent in the form of fumed silica. The gelled vanadium electrolyte battery has relatively high concentrations of vanadium and thus storage capacity can be included in the gelled electrolyte. The gelled electrolyte increases the charge/discharge cycles of the battery.

Description

GELLED ELECTROLYTE VANADIUM BATTERY

FIELD OF THE INVENTION

The present invention relates generally to a gelled electrolyte vanadium battery.

BACKGROUND TO THE INVENTION

A conventional vanadium redox flow battery uses two solutions of vanadium which are stored in external tanks. The vanadium solutions are pumped through a cell stack where electron transfer reactions occur, producing energy. A V(II)/V(III) redox couple electrolyte is used for the negative half cell and a V(IV)/V(V) redox solution for the positive. The capacity of the system is a function of the vanadium ion concentration and electrolyte volume, so the battery is ideally suited to applications requiring 2 or more hours of storage. With a vanadium ion concentration of 2-3 moles per litre, the energy density of 25-35 Wh/kg is adequate for a wide range of stationary applications and some specialised mobile applications. For example, electric buses that require continuous operation could take advantage of the battery' s unique feature which allows conventional electrical recharging in addition to mechanical refuelling by exchanging solutions. While the vanadium redox flow battery has been successfully tested in an electric golf car, wider acceptance for other mobile applications requires further improvements in energy density to provide the vehicle driving ranges stipulated by vehicle manufacturers. On "the other hand, several car manufacturers have recently been promoting the hybrid vehicle concept which would operate in electric mode during stop-star heavy traffic driving, changing over to conventional gasoline propulsion under highway driving conditions.

The specification of International patent application No. PCT/AU96/00268 relates generally to a vanadium redox cell. The application designates Skyllaz-Kazacos and Kazacos as the co-inventors and is entitled "High Energy Density Vanadium Solutions, Methods of Preparation Thereof and All Vanadium Redox Cells and Batteries Containing High Energy Density Vanadium Electrolyte Solutions" . The speci ication describes a vanadium electrolyte including a wide range of organic immobilising agents such as Xanthum gum, starch, gelatin, methyl cellulose and other gums and gels. These immobilising agents are susceptible to oxidation and decomposition by reaction with strongly oxidising V (V) in the charged positive half-cell of the vanadium redox battery.

SUMMARY OF THE INVENTION According to one aspect of the present invention there is provided a method of preparing a gelled electrolyte of a vanadium battery, said method involving the addition of an inorganic gelling agent to a vanadium electrolyte to form the gelled electrolyte.

According to another aspect of the invention there is provided a gelled electrolyte of a vanadium battery, said gelled electrolyte prepared by the addition of an inorganic gelling agent to a vanadium electrolyte.

According to a further aspect of the invention there is provided a half cell of a vanadium battery, said half cell including a gelled electrolyte prepared by the addition of an inorganic gelling agent to a vanadium electrolyte.

Preferably the inorganic gelling agent includes silica and in particular fumed silica. More preferably the concentration of the fumed silica in the vanadium electrolyte is at least 3% by weight. Alternatively the inorganic gelling agent includes titanium dioxide or alumina/silica mixtures.

Typically the vanadium electrolyte has a vanadium ion concentration of at least 3 Molar (M) . More typically the vanadium ion concentration is about 4M.

Preferably the vanadium electrolyte is prepared in a sulphate solution such as sulphuric acid. More preferably the sulphuric acid concentration is about 6M.

BRIEF DESCRIPTION OF THE DRAWINGS In order to achieve a better understanding of the nature of the present invention a preferred embodiment of a method of preparing a gelled electrolyte of a vanadium battery together with other aspects of the invention will now be described, by way of example only, with reference to the accompanying representations in which:

Figure 1 is a schematic diagram of a half cell of a vanadium test battery in an assembled and exploded condition;

Figures 2 to 5 are plots of conductivity with respect to time for a gelled vanadium electrolyte in various oxidation states in sulphate solutions;

Figures 6 and 7 are cyclic voltammograms of ungelled and gelled vanadium electrolyte sulfate solutions; Figure 8 is typical charge/discharge curves for a gelled vanadium electrolyte sulfate solution;

Figures 9 and 10 are plots of capacity for gelled and ungelled vanadium electrolytes with respect to first and second cycles, respectively;

Figures 11 and 12 are plots of cell efficiency calculations for both gelled and ungelled vanadium electrolytes against cycle number;

Figures 13 to 20 are photographs of both gelled and ungelled vanadium electrolytes in various oxidation states after specified _.times .

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

In a preferred embodiment of the invention gelling of vanadium electrolytes is achieved by the addition of an inorganic gelling agent in the form of fumed silica. The viability and characteristics of a gelled electrolyte vanadium battery were investigated using the fumed silica inorganic gelling agent. The following experiments were conducted to examine the effects of gelling on the electrochemical activity, cell performance characteristics and precipitation behaviours of a supersaturated vanadium electrolyte .

Preparation of solutions

For the tests on the effect of the inorganic gelling agent on properties of the vanadium electrolyte, a stock solution of 2 M V0S0₄ in 3 M sulfuric acid was prepared. To prepare the vanadium sulfate electrolyte for the cell cycling tests, the V(IV) solution was partially reduced to the V(3.5+) oxidation state (i.e. 50% V³ + 50% V(IV) ), by electrolysis in a two compartment cell employing lead electrodes and sulfuric acid as the anolyte, the anolyte and catholyte being separated by a Selemion AMV anion exchange membrane (Asahi Glass Co., Japan) . Other solutions of the various oxidation states of vanadium in varying sulfate concentrations were also prepared by electrolysis of the corresponding V0S0₄ solution.

Preliminary Gelling Tests

Preliminary tests with the gelling agents were carried out to determine the minimum fumed silica concentration required to produce as stable gel. The commercially available inorganic silica gellings agents used in this embodiment were Aerosil 200, Aerosil 300, and Aerosil Cok 84 supplied by Degussa. To determine if the Aerosil products could actually gel the vanadium electrolyte, the gelling agents were mixed with the prepared vanadium electrolytes using the following steps: (i) 2M V (IV) in 5M total sulfates was prepared as described above and 50 mL of the electrolyte was placed in each of six sample tubes with lids; (ii) 2.5% by weight of the Aerosil 200 was added to the V(IV) solution and mixed with a hand-held mixer at a high-speed setting for five minutes and the sample was then allowed to stand at room temperature for observation; and (iii) step (ii) was repeated using 3 wt% Aerosil 200, 2.5 and 3 wt% Aerosil 300, 2.5 and 3 wt% Aerosil 200.

All subsequent experiments involving the gelled vanadium electrolyte were performed using 3 wt% Aerosil 200.

Cell Cycling Tests

The membrane used for all cell cycling trials was a Nafion 112 cation exchange membrane. This membrane has a low resistance, low vanadium ion permeability and good chemical stability. Toyoba carbon felt was used an electrode material being particularly hydropholic. In an alternative trial, the carbon felt electrode ^'was replaced by graphite fibres to eliminate any possibility of incomplete soaking of the felt electrode sample. An exemplary method for cell preparation is described as follows :

(i) after mixing a vanadium electrolyte and an Aerosil 200 gelling agent, the resultant mixture was placed onto carbon felt electrodes; (ii) the electrolyte was forced into each of the felt electrodes using a spatula with continuous motion being applied to the electrolyte to delay gelation; and

(iii) the electrolyte held within the felt electrode was allowed to completely gel (approximately one hour) and placed onto a glassy carbon electrode substrate in each half cell.

The final cell assembly formed from two of the half cells is shown in Figure 1.

The left hand representation is the cell in its assembled form whilst on the right hand the cell is shown in an exploded configuration. The various components of the cell

10 are designated as follows:

Cation exchange membrane 12

Carbon felt electrode 14 Frame member 16

Spacer element 18

Glassy carbon sheet 20

Copper current collector 22 Rubber gasket 24

PVC end block 26.

The cell was connected to a power supply and battery controller for charge-discharge cycling. The voltage efficiency, coulombic efficiency, energy efficiency, capacity, and cycle life are determined from the resulting voltage versus time plots.

Optical Examination of Effects of Gelling on Crystallisation

Optical microscopy was used to visually examine the effect of gelling on the crystallisation behaviour of the vanadium solutions. The following procedure was used: (i) solutions of 2M V(III) and V(IV) in 6.6M total sulfates and 2M V (V) in 3M total sulfates were prepared, these concentrations being chosen to increase supersaturation for each oxidation state; (ii) half of each solution was gelled as described above and a drop of each solution was applied to a glass microscope slide and covered with another glass cover to prevent air oxidation or drying out of the gels;

(iii) the solutions on each of the slides were then forced to precipitate by placing the slides containing the V(V) solution in an oven at 70°C, and those containing the V(III) and V(IV) solutions in a refrigerator at 5°C;

(iv) inspection of the slides was carried out using an Optical Microscope and photographs were taken of relevant slides using a Camera. Preliminary Tests

The preliminary results with the gelling agents are presented in the following Table.

Table: Concentration Comparison of Gelling Agents to Achieve Gelation of 2M V(IV) in 3M Total Sulfates

The Aerosil 200 and Aerosil 300 succeeded in gelling a sample of 2M V93.5+) in 3M total sulfates. It was found that a minimum of 3 wt% fumed silica content was required to achieve complete gelation. Agitation of the gelled electrolyte was effective in restoring a gelled sample to a solution.

Conductivity

Figure 2 shows the reduction of conductivity during the gelling of 2M V(V) in 3M total sulfates. The plot displays an exponential decay in conductivity followed by a levelling off to a constant value. This indicates that as the viscosity increases during the onset period, hydrogen bonding is increasing throughout the solution, until a "solid" gel is achieved. At this stage, the hydrogen bridging has approached its limiting value, and hence the conductivity of the gelled solution reaches a limiting value. A gelled solution of 2 M V (V) in 3M total sulfates is seen to have a conductivity of approximately 480 mS/cm, compared with a value of 515mS/cm for the ungelled solution. This represents a drop of 7% in conductivity which is not expected to have a significant impact on cell resistance. The gelled electrolyte demonstrates a linear relationship between conductivity and temperature as shown in Figure 5. The results from the conductivity measurements were influenced by ambient temperature fluctuations, particularly the measurements taken during the onset of gelling (Figures 2 to 4) . The small fluctuation in conductivity seen in these plots is associated with minor temperature variations, leading to small errors in the measurements. This problem was eliminated by the use of a controlled temperature water bath for subsequent measurements.

Cyclic Voltammetry Figure 6 shows the typical cyclic voltammogram obtained for an ungelled 2M V(IV) solution in 5M total sulfates, while Figure 7 is the cyclic voltammogram obtained for the corresponding gelled solution of 2M V(IV) in 5M total sulfates .

Clearly all the relevant peaks associated with the reduction and oxidation of the vanadium species are present. This implies that the vanadium ions in the gelled vanadium electrolyte can undergo oxidation and reduction to all of the oxidations states involved in the charge- discharge processes of the vanadium battery. The redox peak potentials are also very similar in both the gelled and ungelled electrolytes, showing that the gelled state has little or no effect on the electrochemical reversibility of the vanadium redox couples. Cell Cycling Tests

Increasing the vanadium ion concentration to over 3 M in a 5M sulphuric acid solution was expected to cause precipitation in a conventional redox flow cell. This precipitation of vanadium salts in electrolyte channels would lead to cell blockages in the conventional flow cell. On the other hand, in the field of immobilised vanadium batteries, cell blockages are not a problem since the electrolyte is not pumped through a cell stack. Therefore, it was possible to trail higher vanadium concentrations for the gelled electrolyte, provided that severe precipitation did not lead to electrode or membrane passivation. A 4 M vanadium electrolyte in 6M total sulfates was chosen for use in cell cycling tests, although less concentrated electrolytes were also tested. It was of particular interest to establish whether the gelled supersaturated 4 M electrolyte exhibited greater stability against precipitation compared with an ungelled solution of the same concentration. Figure 8 shows typical charge/discharge curves obtained for the gelled 4M vanadium electrolyte in 6M total sulfates.

Cell Capacity

Electrolyte losses during cell assembly made it difficult to accurately estimate the total solution volume for a theoretical capacity calculation. The theoretical capacity of the cell was thus calculated from the volume of each half-cell cavity assuming that the graphite felt volume was negligible. Figures 9 and 10 illustrate the variation in the capacity of the cell employing a 4 M vanadium electrolyte in 6 M total sulfates with respect to cycle number. In Figure 9 each cycle was normalised against the first cycle which was set at 100% theoretical basis, while

Figure 10 uses the second cycle. This eliminates any problems with initial leakage and errors in capacity calculations arising from inaccurate volume estimates. It is clear that capacity losses are much greater in the case of the ungelled electrolyte and that the gelled electrolyte displays almost double the cycle life. In Figure 10 it is seen that after 100 charge/discharge cycles, the gelled electrolyte has retained 65% of its original capacity, compared to only 20% for the ungelled. This suggests that the gel is reducing the rate or degree of precipitation of vanadium in the supersaturated electrolyte, so that it is available for the charge- discharge processes.

Coulombic, Voltage and Energy Efficiencies

Results of cell efficiency calculations are shown in Figures 11 and 12 for the ungelled and gelled 4 M vanadium electrolyte in 6M total sulfates, respectively. High coulombic efficiencies close to 100% can be observed. The voltage efficiencies compared favourably for the gelled electrolyte. The ungelled cell displayed a decreasing trend toward a value of about 50% after some 400 cycles, whereas the gelled electrolyte remained steady at a value around 65% efficiency, even up to 900 cycles. Surprisingly, the gelled electrolyte cell did not exhibit a lower voltage efficiency than the ungelled cell, despite the higher conductivity of the latter electrolyte. The higher voltage efficiency with increasing cycle number in the case of the gelled electrolyte, is understood to be due to a decrease in amount of crystallisation on the carbon felt electrode and membrane surfaces which would reduce the active surface area and increase polarisiation losses. Optical Microscopy Study of Effects of Gelling on Crystallisation

The results of the cell cycling tests suggested that the gelled supersaturated vanadium electrolyte was more stable against precipitation of vanadium from the supersaturated solution, this leading to a decrease in the observed capacity losses with increasing cycle number. To confirm this hypothesis and to determine the effect of the gelling agent on the crystal habit and growth of the various vanadium species, an optical microscopy study was undertaken on both the gelled and ungelled solutions. The effect of the habit modification and crystal size is discussed below for each oxidation state investigated. The sulfate concentrations are chosen to maximise the supersaturation of the solutions to promote crystal growth and to reduce the induction time, for purposes of accelerated testing.

Vanadium (III) Figures 13 to 16 are photographs of both the ungelled and gelled forms of the 2M V(III) solutions at a concentration of 6.6M total sulfates. The solutions were held at a temperature of 5°C and the photographs taken at a magnification of 66X. The photograph of: (i) Figure 13 is of the ungelled V(III) solution after 1 day; (ii) Figure 14 is one portion of the ungelled V(III) solution after four days; (iii) Figure 15 is another portion of the ungelled V(III) solution after 4 days.

The gelled 2M V(III) in 6.6 total sulfates did not precipitate in the first day at 5°C whilst the ungelled V(III) did show some very small precipitation. Following four days at 5°C, both the gelled and ungelled V(III) displayed visual crystallisation. The ungelled and the gelled V(III) precipitated after four days at 5°C, but with quite different results. The ungelled solutions included relatively large and irregular crystals whereas the crystals for the gelled V(III) solution were about ten times smaller than those in the ungelled V(III) . This suggests that the gel environment is reducing either the crystal growth rate or the ultimate crystal size. The habit of the crystals grown in the gel environment is slightly different than that in the ungelled solution. The gel environment seems to allow more perfect crystal growth due to nucleation suppression and by the uniform force exerted by the gel .

Vanadium (IV)

Figures 17 to 20 are photographs of both the ungelled and gelled forms of the 2M V(IV) solutions at a concentration of 6.6M total sulfates. The solutions were held at a temperature of 5°C and the photographs taken at a magnification of 66X. The photograph of:

(i) Figure 17 is of the ungelled V(IV) solution after 1 day; (ii) Figure 18 is of the gelled V(IV) solution after 1 day; (iii) Figure 19 is of the ungelled V(IV) solution after 4 days ; (iv) Figure 20 is of the gelled V(IV) solution after 4 days .

Both the ungelled and gelled 2M V(IV) in 6.6M total sulfates precipitated after one day at 5°C. It was surprising that for V(IV) after one day, in terms of overall amount of precipitation, the gelled and ungelled V(IV) solutions were similar. The increase of nucleation in the gel may be due to silica particles which have not contributed to gelling, and have acted as nucleation centres for crystal growth. Comparing the gelled and ungelled V(IV) slides after 4 days immediately suggests that the gel is both changing habit, and the rate of crystal growth. The star-shaped crystals grown in the ungelled solution are now more than ten times larger than those which have grown in the gel environment. Also, it maybe be observed again that the habit is more perfect in the gel environment .

Vanadium (V)

Due to the thermal precipitation of vanadium (V) , these samples were placed in an oven at 70°C to induce crystal growth. After three hours in these conditions, the slides were found to contain an amorphous precipitate of V₂0₅. There was no obvious distinction between the gelled and ungelled 2M V(V) in the 3M -total sulfate solution. The amorphous precipitate observed in the gelled solution was exactly the same for the ungelled. It should be noted, however that these were extreme conditions for the V(V) electrolytes and that higher acid concentrations and lower temperatures would be expected to show up greater differences in the precipitation behaviour of the gelled and ungelled V (V) solutions.

The experimental results pertaining to these embodiments of the invention have shown that gelling the vanadium electrolyte essentially has no effect on the electrochemical behaviour of the vanadium redox couples so that the short-term charge/discharge behaviour of a gelled vanadium redox is similar to that of an ungelled electrolyte cell. Gelling the vanadium electrolyte, with in this example fumed silica, led to a slight decrease in conductivity and vanadium ion diffusivity due to increased viscosity. However, in a highly supersaturated 4M vanadium solution, gelling significantly enhanced the stability of the vanadium solutions against precipitation during charge/discharge cycling of the vanadium redox cell. In contrast to the gelled vanadium solutions, severe capacity losses were observed for the ungelled cell of the supersaturated vanadium solutions after extended cycling. In the experiments conducted after 100 charge/discharge cycles, the gelled electrolyte retained 65% of its original capacity compared to only 20% for the ungelled electrolyte. Therefore, it is understood that the gel reduces the rate or degree of precipitation of vanadium in the supersaturated electrolyte so that it is available for the charge/discharge process.

Now that a preferred embodiment of the present invention has been described in some detail it will be apparent to those skilled in the relevant art that the gelled vanadium electrolyte battery has at least the following advantages: (i) relatively high concentrations of vanadium and thus storage capacity can be included in the gelled electrolyte; (ii) the gelled electrolyte is particularly suited to an "immobilised" electrolyte cell; and (iϋ) the gelled electrolyte increases the effective charge/discharge cycles of the battery. Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. For example, the inorganic gelling agent is not restricted to silica but may include other inorganic gelling agents such as titanium dioxide or alumina/silica mixtures. Furthermore, it should be appreciated that the concentrations of silica and sulfate described are for experimental and illustrative purposes only and are not to restrict the scope of the invention. For example, the vanadium ion concentration is not limited to at least 3M.

All such variations and modifications are to be considered within the scope of the present invention the nature of which to be determined from the foregoing description.

Claims

CLAIMS :

1. A method of preparing a gelled electrolyte of a vanadium battery, said method involving the addition of an inorganic gelling agent to a vanadium electrolyte to form the gelled electrolyte.

2. A method of preparing a gelled electrolyte as defined in Claim 1 wherein the inorganic gelling agent includes silica.

3. A method of preparing a gelled electrolyte as defined in Claim 2 wherein the silica is fumed silica.

4. A method of preparing a gelled electrolyte as defined in Claim 3 wherein the concentration of the fumed silica in the vanadium electrolyte is at least 3% by weight .

5. A method of preparing a gelled electrolyte as defined in Claim 1 wherein the inorganic gelling agent includes titanium dioxide or alumina/silica mixtures.

6. A method of preparing a gelled electrolyte as defined in any one of the preceding claims wherein the vanadium electrolyte has a vanadium ion concentration of at least 3 Molar (M) .

7. A method of preparing a gelled electrolyte as defined in any one of Claims 1 to 5 wherein the vanadium ion concentration of the vanadium electrolyte is about 4M.

8. A method of preparing a gelled electrolyte as defined in any one of the preceding claims wherein the vanadium electrolyte is prepared in a sulfate solution.

9. A method of preparing a gelled electrolyte as defined in Claim 8 wherein the sulphate solution is sulphuric acid at a concentration of about 6M.

10. A gelled electrolyte of a vanadium battery, said gelled electrolyte prepared by the addition of an inorganic gelling agent to a vanadium electrolyte.

11. A half cell of a vanadium battery, said half cell including a gelled electrolyte prepared by the addition of an inorganic gelling agent to a vanadium electrolyte.

12. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 10 or 11 respectively wherein the inorganic gelling agent includes silica.

13. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 12 wherein the silica is fumed silica.

14. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 13 wherein the concentration of the fumed silica in the vanadium electrolyte is at least 3% by weight.

15. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 10 or 11 respectively wherein the inorganic gelling agent includes titanium dioxide or alumina/silica mixtures.

16. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 10 or 11 respectively wherein the vanadium electrolyte has a vanadium ion concentration of at least 3 Molar (M) .

17. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 10 or 11 respectively wherein the vanadium ion concentration of the vanadium electrolyte is about 4M.

18. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 10 or 11 respectively wherein the vanadium electrolyte is prepared in a sulfate solution.

19. A gelled electrolyte or a half cell of a vanadium battery as defined in Claim 18 wherein the sulphate solution is sulphuric acid at a concentration of about 6M .