Growth of III-V materials on Si could enable dramatic cost reduction for III-V PV by eliminating the need for expensive III-V substrates and enabling high-efficiency tandem solar cells. The direct heteroepitaxy of III-V materials on Si for high-efficiency photovoltaics has progressed in recent years, but most studies have focused on off-cut wafers polished using an expensive chemical-mechanical planarization process (which is not used for commercial solar cells). Alternative growth approaches are needed that can enable the integration of high material quality III-Vs with PV-grade Si materials.
III-V photovoltaics have demonstrated the highest device efficiencies, but one of the factors limiting widespread deployment is the availability of low-cost substrates. There have been decades of work on growth of III-Vs on Si, either to enable low-cost III-Vs alone or to build tandem device structures with an active Si bottom cell. However, such approaches have focused on offcut, polished wafers. GaP grown directly on Si has potential to either be a nucleation layer for a graded buffer layer to a lattice-mismatched material such as GaAsP or can serve as a direct heteroemitter for crystalline Si solar cells. To produce high quality solar cells from III-V/Si heteroepitaxy, there are several challenges such as lattice mismatch, thermal expansion mismatch, and the formation of antiphase boundaries (APBs) due to the growth of polar materials on a non-polar substrate. Recent advances in GaP/Si heteroepitaxy have shown that careful control of Si (001) surfaces (with either deliberate or nominal offcuts) prior to GaP nucleation can enable APB-free GaP films by preferentially forming double atomic steps. GaP nucleation on Si (111) has also been investigated, and the polarity of the III-V material can be controlled by varying pre-nucleation conditions.
Current approaches for 1-sun III-V/Si cell designs utilize precisely polished surfaces that can add significant cost to the manufacturing process. Demonstrated epitaxial or wafer-bonded tandems required polished Si interfaces (<0.5 nm RMS roughness for bonding, and 2-6 offcut for growth), while mechanically stacked approaches require reuse of the growth substrate to be cost effective. Adding chemomechanical polishing (CMP) can contribute at least $1/Wp,DC to the cost of the cell, which is too expensive for flat panel PV applications. For photovoltaic applications it is compelling to investigate ways to integrate III-Vs with unpolished solar-grade wafers (eliminating the need for CMP).
In an aspect disclosed herein is a method for patterning and etching an unpolished silicon substrate that is capable of III-V epitaxial growth. In an embodiment, the method further comprises treating the silicon substrate with AsH3. In another embodiment he method further comprises patterning v-grooves on the silicon substrate. In yet another embodiment, the method, the silicon substrate is oriented in the (001) direction and the v-grooves are patterned in the {110} direction before etching. In an embodiment, the v-grooves on the silicon substrate comprise (001) facets. In an embodiment, v-grooves on the silicon substrate comprise (001) facets and have not been etched to pyramids. In yet another embodiment, the layer of an oxide or SiNX is on the (001) facets of the v-grooves.
In an aspect, disclosed herein is a method for epitaxial growth of a III-V material on an unpolished silicon substrate comprising patterning v-grooves on the silicon substrate; and treating the silicon substrate with AsH3; and epitaxially growing a III-V material on {001} and {111} surfaces of the silicon substrate. In an embodiment, the III-V material comprises a molar ratio of a group V element to a group III element of from 10 to 5000. In an embodiment, the epitaxial growth of the III-V material is at between about 600° C. and about 800° C. In an embodiment, the III-V epitaxial growth comprises nucleation of the III-V material with registry between pairs of intersecting {111} surfaces of the silicon substrate. In another embodiment, the the morphology of the nucleation of the epitaxially grown III-V material is controlled by varying growth temperatures and V/III ratios.
In an aspect, disclosed is a method for making solar cells comprising epitaxial growth of a III-V material on an unpolished silicon substrate comprising patterning v-grooves and etching. In an embodiment, the silicon substrate is oriented in the (001) direction and the v-grooves are patterned in the {110} direction. In an embodiment, the v-grooves on the silicon substrate have (001) facets. In an embodiment, the v-grooves on the silicon substrate have (001) facets and have not been etched to pyramids. In an embodiment, the silicon substrate is treated with AsH3. In an embodiment, a layer of an oxide or SiNX is on the (001) facet of the v-grooves. In an embodiment, the III-V material comprises a molar ratio of a group V element to a group III element of from 10 to 5000. In another embodiment, the epitaxial growth of the III-V material is at between about 600° C. and about 800° C. In an embodiment, the solar cells lack antiphase boundary defects.
In an aspect, a method is disclosed for patterning and etching to create templates for selective area epitaxy on PV-grade Si substrates comprising the nucleation of III-V materials on the PV-grade Si substrates. In an embodiment, the method further involves exposing crystalline facets for epitaxy by patterning and selective wet chemical etching. In another embodiment, a solar cell is made by using the method.
In another aspect, a method of making substrates for III-V heteroepitaxy that uses planarization and patterning v-grooves on cut silicon solar-grade wafers is disclosed. In an embodiment, the method further uses AsH3 pretreatments to remove contaminants and improve the surface structure for III-V epitaxy on {001} and {111} Si surfaces. In another embodiment, a solar cell is made by using the method.
Exemplary embodiments are illustrated in referenced figures of the drawings. It is intended that the embodiments and figures disclosed herein are to be considered illustrative rather than limiting.
a,
1
b, and 1c depict an optical image of PV-grade Czochralski Si.
Disclosed herein is the use of cost-effective patterning and etching approaches to create templates for selective area epitaxy on PV-grade Si substrates and the nucleation of III-V materials on these substrates.
Direct epitaxy of III-Vs on non-CMP polished surfaces is an important area of research for the integration of III-V materials with Si for photovoltaic applications. In certain embodiments disclosed herein, patterning and selective wet chemical etching were used to expose the desired crystalline facets for epitaxy, demonstrating that these surfaces are suitable for high quality heteroepitaxy of GaP.
Commercial Si solar cells are fabricated from nominally (001) oriented wafers that are textured using base etching to selectively expose Si {111} facets to create a random pyramid structures that enhance light trapping. Combining nanoscale patterning with the same facet selective etch chemistries can provide a pathway to control the facets that are exposed for epitaxy, without the need for CMP. If Si (001) substrates are patterned with lines in the {110} direction before etching, a v-grooved pattern is formed, rather than random pyramids. Epitaxy on v-grooved substrates can enable high quality nucleation of III-Vs by forcing registry between pairs of intersecting {111} surfaces to prevent defect formation. There have been several reports of using v-grooved surfaces for the metamorphic growth of high-quality lattice mismatched III-Vs such as GaAs or InP for integrating photonic or logic devices onto Si substrates, but few reports of applying controlled patterning to non-CMP Si wafers. In an embodiment, disclosed herein are approaches to low-cost planarization and patterning v-grooves on diamond-cut silicon solar-grade wafers, as well as the applicability of these substrates for III-V heteroepitaxy.
Direct patterning of as-sawn Si wafers is challenging due to the surface damage and large peak-to-valley heights left from the sawing process. Different chemical planarization techniques were investigated to minimize the peak-to-valley height difference, and surface roughness was measured across multiple sample areas using a 50× objective. As sawn PV-grade Czockralski wafers had an average RMS roughness of 716 nm (
Prior work has demonstrated that AsH3 pretreatments can remove contaminants and improve the surface structure for III-V epitaxy on both {001} and {111} Si surfaces. Before nucleating GaP, the effectiveness of the AsH3 pre-treatment on v-grooved Si was investigated using low-energy electron diffraction (LEED) and Auger electron spectroscopy (AES). A partially etched sample on CMP polished Si {001} was annealed under standard AsH3 conditions, and then transferred to an ultra-high vacuum (UHV) surface analysis chamber.
When v-grooves were not fully etched to pyramids and (001) facets remained along the ridge-tops (as depicted in
As depicted in
As depicted in
Patterning was investigated on Czochralski (CZ) solargrade Si wafers cut with a diamond saw as well as CMP polished CZ wafers. For solar-grade wafers, aqueous KOH and HF:HNO3:CH3COOH (HNA) etch chemistries, as well as reactive ion etching were investigated as planarizing etches. Patterns were transferred into SiNx or SiOx dielectric layers using nanoimprint lithography (NIL) and laser interference lithography (LIL). Reactive ion etching and dilute (1%) HF were used to selectively remove the dielectric masks, then v-grooves were formed by etching the samples in 10% TMAH.
III-V growth was carried out in a low pressure MOCVD reactor using triethylgallium and PH3 as precursors. In an embodiment, trimethylgallium may be used. Prior to growth, samples were annealed in dilute AsH3, which has previously been shown to remove C and O contaminants and for Si (001) substrates, to create a single domain arsenic terminated surface. For surface analysis, samples were transferred to a linked ultrahigh-vacuum (UHV) surface analysis chamber equipped with low-energy electron diffraction (LEED) and Auger electron spectroscopy (AES). The surface roughness of Si wafers was measured using a Keyence VK-X250 3D scanning laser microscope with 0.5 nm resolution in the z-direction.
GaP growth was carried out in an MOCVD reactor, using PH3 as a group V precursor and either triethyl gallium or trimethyl gallium as a group III precursor. Wet chemical etching and an in-situ surface clean was done to prepare the surface for growth. In an embodiment, V/III ratios ranged from 10 to 5000. In an embodiment, GaP nucleation and growth was carried out between 600° C. and 800° C.
This application claims priority under 35 U.S.C. 119(e) to U.S. Provisional Application No. 62/682,535 filed 8 Jun. 2018, the contents of which are incorporated herein by reference.
The United States Government has rights in this invention under Contract No. DE-AC36-08GO28308 between the United States Department of Energy and Alliance for Sustainable Energy, LLC, the Manager and Operator of the National Renewable Energy Laboratory.
Number | Date | Country | |
---|---|---|---|
62682535 | Jun 2018 | US |