surface modification of a-SiC films for PEC

Slide 1: 

www.MVSystemsInc.com SPIE Optics+Photonics, San Diego, CA, 2010 Surface modification of a-SiC photoelectrodesfor photocurrent enhancement Ilvydas Matulionisa, Jian Hua, Feng Zhua, Josh Gallona, Nicolas Gaillardc, Todd Deutschb, Eric Millerc, and Arun Madana a. MVSystems, Inc., Golden, CO, USA b. National Renewable Energy Laboratory, Golden, CO, USA. c: Hawaii Natural Energy Institute, University of Hawaii at Manoa, HI, USA Supported by the U.S. Department of Energy Contract # DE-FC36-07GO17105

Slide 2: 

Outline PEC – photoelectrochemical PECVD – plasma enhanced chemical vapor deposition a-SiC (or a-SiC:H) – hydrogenated amorphous silicon carbide Hybrid Device – a-SiC electrode integrated with a-Si/a-Si tandem solar cell Introduction and device fabrication Motivation for a-SiC surface modification PECVD-based modification methods Surface modification by use of nanoparticles Discussion and energy band diagram of PEC devices Conclusions

Slide 3: 

Why hydrogen production by photoelectrochemical (PEC) technique? Advantages of the a-SiC photoelectrode: Bandgap of about 2 eV captures most of the solar spectrum so the photoelectrode can produce current in excess of 10 mA/cm2 The a-SiC bandgap can be tuned from 1.9 eV to 2.3 eV by adjustment of PECVD process gas ratio Hybrid devices (tandem a-Si solar cell with a-SiC electrode) can be made in the same PECVD multi-chamber system (cluster tool) The a-SiC material resists corrosion in electrolytes Solar-to-Hydrogen (STH) efficiency is related to current density via water molecule splitting energy of 1.23 eV, i.e. 10 mA/cm2 converts to about 12% STH.

Slide 4: 

Deposition system – PECVD/Sputter cluster tool RF power: 10-20 W Excitation frequency: 13.56 MHz Pressure: 300-550 mTorrr SiH4 flow rate: 20 sccm CH4 flow rate: 0-20 sccm H2 flow rate 0-100 sccm p-doping: B2H6 n-doping: PH3 Substrate temperature 200°C Sputtering chamber ZnO, ITO, and metals PECVD chambers Load Lock Main deposition parameters: All a-SiC:H films, photoelectrodes, solar cells, and PEC hybrid devices were fabricated in the cluster tool PECVD/Sputtering system, designed and manufactured by MVSystems, Inc. www.MVSystemsInc.com

Slide 5: 

Current extraction from a PV/PEC hybrid device a-SiC (i), 100nm SnO2 p a-Si p-i-n (top cell) a-Si p-i-n (bottom cell) p n p n a-Si (i), 150nm a-SiC p-i a-Si (i), 500nm a-SiC (i), 100nm a-Si (i), 150nm a-Si (i), 500nm Transparent-conductive ITO contact for current measurement Exposed to electrolyte light light Glass Glass HYBRID PV/PEC DEVICE ANALOGOUS DEVICE FINISHED WITH AN ITO CONTACT STH efficiency of hybrid PEC cell should be > 6% based on solid state version (right) Low current in hybrid PEC cell (left) Charge carrier extraction problem at the a-SiC/electrolyte interface SiOx etched with HF ITO

Slide 6: 

Surface modification by PECVD methods Surface modification can be accomplished by alteration of process gases at the end of the a-SiC deposition or an application of a thin film in another chamber in the same cluster tool Surface modification film should prevent oxidation or change surface band alignment to better match H2/H2O and O2/H2O levels in electrolyte Simplified structure (a “photoelectrode”) has been used for rapid PECVD film evaluation as shown in the figure

Slide 7: 

Application of a carbon-rich film Carbon-rich (C-rich) film was created by turning of SiH4 flow at the end of a deposition (only CH4 remained). Optical Emission Spectroscopy (OES) spectra helped determine Si residence time (about 40 sec). The C-rich was intended to eliminate formation of silicon oxide (SiOx), however apparently it created a barrier for charge carriers. Conclusion: This method has low flexibility (only timing can be changed) and does not appear promising from the existing data.

Slide 8: 

Application of silicon nitride (a-SiNx) layer The bandgap of SiNx can be tuned by alteration of SiH4 and NH3 ratio. Charge built-in in the film should alter surface band alignment for a better match with H2/H2O and O2/H2O levels in the electrolyte. SiNx thickness was about 2 nm SiNx film with bandgap of 3 eV shows improved performance. Adjustment of deposition parameters to create slightly (+) or slightly (–) film as well as p-type or n-type film may create proper band alignment.

Slide 9: 

Surface modification by a-SiF film a-SiF films were deposited by employing SiH4 and SiF4 as process gases. Infrared spectra of films are shown below. 830 cm2 F is a very electronegative element, i.e. it has a tendency to capture an electron. F-containing film should dramatically change surface band alignment and possibly disrupt the Helmholtz solid-liquid layer, which may hinder charge transport. Conclusion: PECVD surface modification shows some promise and there is a vast parameter space and other materials to be explored. Rapid evaluation of samples would be very helpful.

Slide 10: 

Surface modification with PdAu nanoparticles PdAu nano-particles were deposited using a compact sputter system at HNEI Before nano-particle deposition, a-SiC was etched in 5% HF to remove silicon oxide Nano-particles were about 2 – 4 nm in size All at 160 mT  30 sec 60 sec 90 sec Conclusion: PdAu nanoparticles work well for other materials, however their high work function (> 5 eV) creates a barrier on the a-SiC surface

Slide 11: 

Coating of a-SiC with Pt/Au nanoparticles Pt/Au nanoparticles have been sputtered on hybrid devices in Dr. Yanfa Yan’s lab at NREL. Estimated nanoparticle size was a few nanometers. Conclusion: No improvement observed in this set of data. Lower work function nanoparticles must be used.

Slide 12: 

Reduction of barrier via use of lower work function metal Barrier created by a high work function metal on the surface Titanium nanoparticles show promise (next slide)

Slide 13: 

Surface modification by use of Ti nanoparticles Sputter deposition of titanium nanoparticles has been accomplished at MVSystems. SEM of the nanoparticles on a textured SnO2  Conclusion: Optimized Ti nanoparticle coating combined with an SiOx etch by HF should eliminate the barrier and significantly increase the current at zero external voltage.

Slide 14: 

Conclusions Thanks to Ed Valentich (MVS) for help with sample preparation and equipment maintenance. PECVD-based surface modification methods show some promise – there is a lot more to be explored Titanium or other low work function nanoparticles combined with SiOx etch should lead to much higher currents Rapid evaluation methods are being designed, for example metal/n/a-SiC/metal Schottky barrier structure Other materials that can be fabricated in the same cluster tool should be evaluated, e.g. nc-SiC, a-SiCN, nc-SiNx, a-SiON