The present invention is generally directed to materials that can be used for the counter electrode of dye sensitized solar cells. While the present invention is described with respect to dye sensitized solar cells, other applications are also envisaged.

BACKGROUND OF THE INVENTION

Dye sensitized solar cells (DSC) have been developed as a relatively low cost alternative to conventional inorganic semiconductor solar cells. A DSC conventionally consists of a conducting electrode (anode) coated with a photovoltaically active layer in the form of a nanoporous (for example titanium dioxide) semiconductor electrode layer, and an opposing counter electrode (cathode) to form a sandwich structure. A thin coating of photosensitive dye covers the surfaces of the nanoporous electrode layer, and an electrolyte interpenetrates that nanoporous electrode layer. The counter electrode of the DSC is typically made from platinum (Pt). The high cost of Pt and the limited supplies of that material however warrants consideration of alternative materials to use for the counter electrode.

PEDOT:PSS (Poly (3, 4-ethylenedioxythiophene) poly(styrenesulfonate)), a conductive polymer mixture, has been considered as an alternative material for platinum. PEDOT:PSS shows good catalytic effects towards the l ₃ 7l ^" redox electrolyte system which makes this material a feasible option for replacing the Pt. However, the relatively high resistance of pure PEDOT:PSS electrodes has limited its application in DSCs. Furthermore a pure PEDOT:PSS film usually has a low surface area limiting its catalytic reaction capacity.

BRIEF DESCRIPTION OF THE INVENTION

It would therefore be advantageous to be able to use an alternative material in place of the Pt in the counter electrode of a DSC.

It would also preferably be advantageous that the material has a comparable performance as Pt when used in a DSC. According to one aspect of the present invention, there is provided a dye sensitized solar cell having a counter electrode formed from a nano composite material of metal carbide, graphene and PEDOT:PSS.

The metal carbide may be in the form of nano particles, and the graphene may coat the outer surface of the metal carbide nano particles. The coating of graphene, which itself displays superior conductivity, helps to greatly enhance the conductivity of the metal carbide nano particles.

The metal carbide may be Titanium Carbide which has excellent mechanical and functional properties such as high hardness and wear resistance, low thermal expansion, high electrical and thermal conductivity and great chemical inertness under non-oxidizing conditions. The use of other metal carbides including, but not limited to, transitional metal carbides such as Zirconium Carbide (ZrC) and Hafnium Carbide (HfC), as well as calcium and aluminium carbides is also envisaged.

The metal carbide may be supported on a conductive or a non-conductive material substrate. When used for flexible dye sensitized solar cells, the substrate may be a flexible polymer.

According to another aspect of the present invention, there is provided a method of producing metal carbide, graphene and PEDOT:PSS nano composite material for a counter electrode of a dye sensitized solar cell, including mixing graphene coated metal carbide nano particles with PEDOT:PSS to form a slurry, coating the slurry on a surface of conductive material, and drying the slurry to form the counter electrode.

The metal carbide nano particles and PEDOT:PSS may be mixed together with water or ethanol to form a mixture that is milled to form the slurry. The mixture may include 5%-40% (wt%) of graphene coated metal carbide nano particles, and 0.2%-10% (wt%) of PEDOT:PSS.

According to a further aspect of the present invention, there is provided a metal carbide, graphene and PEDOT:PSS nano composite material produced according to the method as described above.

According to yet another aspect of the present invention, there is provided a method of synthesizing a mesoporous carbon-bonded metal carbide nano composite including synthesizing a Metal-O-C precursor, and subjecting the precursor to carbothermal reduction.

The precursor may be synthesized using a sol-gel process, and the carbothermal reaction may occur at 1 100 - 1450 °C.

According to another aspect of the present invention, there is provided a mesoporous carbon-bonded metal carbide nano composite synthesized according to the method as described above.

The metal carbide of the above-described methods may preferably be titanium carbide.

Electrochemical and photovoltaic experiments using this nano composite material have shown excellent catalytic and photon to electrical energy conversion properties. The nano composite material has demonstrated comparable performance as Pt, and can therefore be used as an alternative counter electrode material for DSCs on glass and flexible substrates such as metals and polymers.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be described with reference to the accompanying drawings which illustrate a preferred example of the present invention. Other examples are also envisaged, and consequently, the particularity of the accompanying drawings is not to be understood as superseding the generality of the preceding description of the invention.

In the drawings:

Figure 1 shows the XRD patterns of gel precursor TSC-G, M-O-C composites fired at 550 ^Q C in nitrogen TSC-550, and high temperature 1450 ^Q C (in Ar) treated sample TSC-1450;

Figure 2 shows the Nitrogen adsorption isothermals of TSC-550 and TSC- 1450 (The pore size distribution (PSD), inserted at the left corner, were calculated with the DFT plus software (Micromeritics), applying the Barrett-Joyner-Halenda (BJH) model with cylindrical geometry of the pores.);

Figure 3(a) shows the HRTEM image of mesoporous carbon-bonded titanium-carbide composite TSC-1450, showing nanocrystalline titanium carbide grains coated with graphene layers; and Figure 3(b) shows a schematic of three kinds of carbon coating which could provide bonding between carbide grains (The arrows indicate the <1 1 1 > direction of titanium carbide grains. The denoted 1 , 2 and 3 mechanisms were observed and marked in Figure 3(a));

Figure 4 shows the Raman spectra of M-O-C composites (TSC-550) and its carbonized sample (TSC-1450);

Figure 5 shows the Nyquist impedance result of DSCs with the same working electrode but different counter electrodes;

Figure 6 shows the Cyclic voltammograms for I7I ₃ ^" redox couple with TiC/Graphene PEDOT SS film, Pt film and PEDOT:PSS film.

Figure 7 shows the l-V curve and l-V performances for DSCs with the same working electrodes and various counter electrodes; and

Figure 8 shows the SEM of films: pure PEDOT:PSS (x5000 magnification) and TiC graphene PEDOT:PSS (X5000 magnification) respectively.

DETAILED DESCRIPTION OF THE INVENTION

The various procedures involved in synthesizing the nano structured metal carbide with graphene coating and the final nano composite material according to the present invention will initially be described. The performance of the nano composite material within a DSC will also be reviewed. While the experiments were in relation to Titaniun Carbide (TiC), it should be appreciated that other metal carbides can also be used according to the present invention.

1. Synthesis of graphene coated TiC nano particles

In experiments conducted by the applicant, mesoporous carbon-bonded TiC bulk composites with a high specific surface area (221 m ² /g), ultra fine grains (10-50 nm) and high crystallinity were synthesized by direct carbothermal reduction of sol-gel prepared monolithic M-O-C precursors.

1.1. Experimental details:

Chemicals: To synthesize the mesoporous carbon-bonded TiC ceramic, sol-gel process were carried out as follows. Typically, 4.64 g of P123 (EO20PO70EO20, MW5800, Sigma-Aldrich) was dissolved in 6.9 g ethanol (anhydrous, Aldrich) to form a solution under continuous stirring, to which 1 1 .368 g titanium tetraisopropoxide (TTIP, Ti(OCH(CH ₃ ) ₂ ) ₄ ; Sigma-Aldrich) was added. Separately, 2g 10 M hydochloric acid (HCI, Merck) was mixed with 2.3 g ethanol using vigorous magnetic stirring. The solution containing TTIP and surfactant was added dropwise into the HCI solution under vigorous stirring. After 30 minutes, 5.886g furfuryl alcohol (FA C ₅ H ₆ 0 ₂ ; 99%, Aldrich) was added into the TTIP under stirring. The solution turned brown immediately, indicating the start of FA polymerization. After continuous stirring for 6 hours, the sol was dark brown in color. All bottles with solution were capped immediately after mixing. The solution was aged for 3 days at the ambient temperature of 25 ^Q C. The aged solution was poured into a petri dish and dried in oven at 80 ^Q C.

1 .2. Results of Experiments:

Phase of C-TiC

The gel material was amorphous, as indicated by the corresponding X-ray diffraction (XRD) pattern (Fig. 1 TSC-G). Sample TSC-G was then calcined at 550 ^Q C under nitrogen for 5 hours to obtain M-O-C composites (named TSC-550). A shrinkage of -23% (in diameter) was found for the porous bulk material after heat-treatment at 550 ^Q C. Anatase was the only crystalline phase detected after calcination at 550 ^Q C (Fig. 1 ). The size of anatase nanocrystals was 3-5 nm, estimated from the full width at half intensity of the (101 ) peak using the Scherrer equation. TSC-550 samples were further heat treated at 1450 ^Q C under argon for 5 hours (samples were named TSC-1450) and TiC is the only detected crystalline phase by XRD. It has been found that TiC can be synthesized where carbothermal reduction occurs anywhere between 1 100 to 1450 °C.

Mesostructure of M-O-C Precursor and C-TiC

The mesoporosity of the metal-oxide/carbon precursor and the final C-TiC material was verified by N ₂ -sorption measurements, shown in Fig. 2. TSC-550 and TSC-1450 samples demonstrated typical type IV curves with H3 and H1 hysteresis loops, respectively (Fig. 2), as defined by lUPAC, characteristic of mesoporous materials. The capillary condensation step for TSC-550 was between P/P ₀ = 0.4-0.6, and appeared at a higher relative pressure (P/P ₀ = 0.7- 0.9) for TSC-1450, suggesting a larger average pore size and wider pore-size distribution (PSD) for the carbonized sample. As confirmed in PSD curves (Fig. 2), TSC-550 had a narrow pore size distribution with a peak pore size at 3.5 nm, while TSC-1450 had a majority of pores smaller than 15 nm and displayed a bimodal pore distribution at 3.5 nm and 7.6 nm. The small pores (~3 nm), for both samples, were generated by removing of the surfactant (P123) during gel calcinations, and they partially remained in the residue carbon for TSC-1450. The TiC particle size is around 40nm. Large pores (-10 nm) formed by the packing of nanoparticles. The Brunauer-Emmett-Teller (BET) surface area of the carbide composite TSC-1450 was 221 m ² /g, which was slightly increased comparing to TSC-550 (213 m ² /g), indicating the mesoporosity was maintained during carbonization.

Microstructures of C-TiC

Of interest is carbon atoms arranged as ordered layers on (1 1 1 ) faces (indicated by the carbide lattice fringe spacing, -0.25 nm) of carbide grains, forming 2-3 layers of graphene (identified by the interplanar spacing, -0.34 nm, which is close to the accepted value for graphite (0.335 nm)), shown in Fig. 3a. The single-layer graphene, also referred to as a monolayer graphite, can epitaxitally grow on a TiC (1 1 1 ) surface under chemical vapor deposition using ethylene as carbon source. This could be a convenient method to synthesize graphene-related materials, although the actual growth mechanism still needs to be explored. The first-layer graphene is bonded with TiC (1 1 1 ) plane by its orbital hybridization which weakens the C-C sp ² bond and enlarges the graphite lattice to match the hexagonal matrix of TiC (1 1 1 ) plane. While no such weakening effects are observed on other plane, suggesting no direct bonding between graphene and carbide, which explains the less ordered graphene observed when it attached to other crystal facets, such as (1 10) and (100) (see Fig. 3a).

The HRTEM microstructures (Fig. 3a) suggest the following possible bonding mechanisms from excess carbon (Fig. 3b): 1 ) direct connection by graphene sheets which stretch out and bridge several grains. The bonding derives from the sp ² -bonded carbon atoms in single layer graphite network. It is quite difficult to find ideal bonding as illustrated, because it requires grains having small gap and similar (1 1 1 ) orientation; 2) carbide grains can also be mechanically "entangled" by distorted graphene which can coat separate particles. Van der Waals interaction between graphitic layers and the energy required to stretch curved graphene provides the source of bonding; 3) another possible bonding comes from amorphous carbon regions which are located between carbide grain boundaries. These highly amorphous carbon atoms can act as adhesive and "glue" carbide grains together. Carbon Network Evolution

The changes of carbon structure induced by temperature were investigated by Raman spectroscopy (Fig. 4). Two peaks, at -1 350 and -1 600 cm ^"1 , referred as to D and G peaks, were observed for TSC-550. Both the D and G peaks are related to vibrations of sp ² -hybridized carbon bonds. The D peak is caused by breathing motion of sp ² bond of carbon rings, activated by structural defects, which is essential for the structure study of carbon. The G peak is related to stretching of sp ² bond in both carbon rings and chains, corresponding to the Brillouin zone center E _2g vibration mode, which is always observed in all carbon based materials. Their positions, shape and intensity ratio of I _D /IG provide the information about the degree of order in carbon network.

After carbonization, the broad D peak became sharp while G peak showed a maximum intensity at -1 605 cm ^"1 . The 2D peak, the overtone of D peak, is normally used to identify the monolayer graphene when it is sharp and high, appearing broadly at -2670 cm ^"1 . The combination mode (D+D') shows at -2930 cm ^"1 , which is also defect activated. These observed changes in Raman spectra and the increased ratio of I _D /IG (from 0.79 for TSC-550 to 1 .52 for TSC-1450) indicated the phase transition of carbon network from amorphous to nanocrystalline as the heat treatment temperature increased. This result is in a good agreement with the HRTEM observation that most carbon is arranged as distorted graphenes in a nano scale order.

2. Formation of the TiC-graphene-PEDOT:PSS nano composite material

In producing the nano composite material, graphene-coated TiC nano particles ranging from 5%-40 %( wt %) with PEDOT SS range from 0.2% to 2% (wt%) were mixed together in a water or ethanol solution by ball milling method. The mixture was milled in a planetary ball milling machine with a rotating speed from 50 rpm to 250 rpm and rotating time from 0.5h to 24h. After the mixture was fully milled, this slurry was coated on the surface of conductive materials (plastic, ceramics, metals and alloy) via spin coating or doctor blading method. During the milling process, the PEDOT:PSS can be fully stuck on the surface of TiC- graphene particles. After the film is dried in the air or on a hot plate being heated to be range from 50 ^Q C to 1 50 ^Q C or in a vacuum furnace being heated to below 1 50 ^Q C, the drying film is used as counter electrodes in DSC. The following Figures 5 to 8 show some characteristics of the graphene-coated TiC PEDOT SS counter electrodes based DSCs.

In general, nano sized metal carbide particles are used as a template to create a porous structure for PEDOT:PSS by simple ball milling and the graphene-coated TiC nano particles can offer extra conductivity to the whole system. Furthermore, PEDOT:PSS works as a binder to enable the TiC- graphene PEDOT:PSS water slurry becoming printable. This technique is also suitable for producing counter electrodes on flexible polymer substrates as the counter electrode does not require high temperatures for its formation.

3. Tests showing the suitability of the nano composite material as a replacement for Pt

Various tests were conducted to compare the performance of TiC/Graphene/PEDOT:PSS with Pt and pure PEDOT:PSS within DSC devices.

Figure 5 shows the Nyquist impedance result of DSCs with the same working electrode, but with different counter electrodes. According to this figure, the overall impedance of DSCs using Pt or TiC-graphene PEDOT:PSS is nearly the same but much lower than the one with pure PEDOT:PSS. The overall impedance of DSCs can be obtained in this figure by recording the start point of semicircles in the figure. Moreover, the first semicircle represents the resistance of the electrons transporting from electrolyte redox (Ι ₃ 7 redox based electrolyte system is used in this work) to counter electrodes. Smaller radius of the first semicircle reflects smaller transporting resistance. According to this figure, the DSC with a TiC-graphene PEDOT:PSS counter electrode has the smallest electron transporting resistance, which is very close to the one with Pt, but much smaller than the one with pure PEDOT:PSS.

Figure 6 shows the cyclic voltammograms of TiC/Graphene PEDOT:PSS film, Pt film and PEDOT:PSS film towards the I7I ₃ ^" redox couple. In this figure, the peak at around 0.4V and 0.6V represent the catalytic possibility of films toward the I7I ₃ ^" redox couple, likewise, their heights (current I) and the distance between the two peaks represent films catalytic ability. Therefore, obviously, the TiC- graphene PEDOT:PSS based films has the best catalytic ability toward the l ^" /l ₃ ^" redox couple , followed by Pt and pure PEDOT:PSS is the worst. Figure 7 shows photovoltaic performance of DSCs with various counter electrodes. The light to photon conversion efficiency of the DSC with Pt is nearly the same with that with TiC/Graphene PEDOT:PSS but both are much higher than that with pure PEDOT:PSS. The short circuit current of the cell with TiC- graphene PEDOT:PSS is higher than the one with Pt and phenomenally higher than that with pure PEDOT:PSS.

Figures 8 show the microstructure of films made of pure PEDOT:PSS and TiC graphene PEDOT:PSS. According to this figure, a nearly dense layer can be found for the pure PEDOT:PSS film, but an obviously porous structure can be observed in the TiC graphene PEDOT:PSS film.

These tests show that the TiC/Graphene PEDOT:PSS nano composite has a comparable performance as Pt and thus can be used as an alternative counter electrode for DSCs.

While the graphene coated metal carbide PEDOT:PSS nano composite material has been described with reference to its use in DSCs, this nano composite material may also be applicable for use in other electrochemical devices that involve catalytic reactions.

Modifications and variations as would be deemed obvious to the person skilled in the art are included within the ambit of the present invention as claimed in the appended claims.