Designing TaC Virtual Substrates for Vertical AlxGa1−x<mml:mrow…

Power electronics are critical for a sustainable energy future, playing a key role in electrification and integration of renewable energy sources into the grid. Advances in ultrawide band gap materials are needed to handle higher powers in smaller form factors while reducing electrical and thermal losses. High <a:math xmlns:a="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><a:mi>Al</a:mi></a:math> content <d:math xmlns:d="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><d:msub><d:mi>Al</d:mi><d:mi>x</d:mi></d:msub><d:msub><d:mi>Ga</d:mi><d:mrow><d:mn>1</d:mn><d:mo>−</d:mo><d:mi>x</d:mi></d:mrow></d:msub><d:mrow><d:mrow><d:mi mathvariant="normal">N</d:mi></d:mrow></d:mrow></d:math> is theoretically capable of meeting these demands, but its impact in power electronics has been severely restricted by a lack of substrates that can satisfy conductivity, lattice matching, and/or thermal expansion requirements. We demonstrate that electrically conductive <h:math xmlns:h="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><h:mrow><h:mi>Ta</h:mi><h:mi mathvariant="normal">C</h:mi></h:mrow></h:math> can be used as a virtual substrate for <l:math xmlns:l="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><l:msub><l:mi>Al</l:mi><l:mi>x</l:mi></l:msub><l:msub><l:mi>Ga</l:mi><l:mrow><l:mn>1</l:mn><l:mo>−</l:mo><l:mi>x</l:mi></l:mrow></l:msub><l:mrow><l:mrow><l:mi mathvariant="normal">N</l:mi></l:mrow></l:mrow></l:math> heteroepitaxy. Scaleably sputtered <p:math xmlns:p="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><p:mrow><p:mi>Ta</p:mi><p:mi mathvariant="normal">C</p:mi></p:mrow></p:math> grown on <t:math xmlns:t="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><t:msub><t:mi>Al</t:mi><t:mn>2</t:mn></t:msub><t:msub><t:mrow><t:mrow><t:mi mathvariant="normal">O</t:mi></t:mrow></t:mrow><t:mn>3</t:mn></t:msub></t:math>, followed by high-temperature face-to-face annealing, produces a thin film <x:math xmlns:x="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><x:mrow><x:mi>Ta</x:mi><x:mi mathvariant="normal">C</x:mi></x:mrow></x:math> template with an effective hexagonal lattice constant matched to <bb:math xmlns:bb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><bb:msub><bb:mi>Al</bb:mi><bb:mrow><bb:mn>0.70</bb:mn></bb:mrow></bb:msub><bb:msub><bb:mi>Ga</bb:mi><bb:mrow><bb:mn>0.30</bb:mn></bb:mrow></bb:msub><bb:mrow><bb:mrow><bb:mi mathvariant="normal">N</bb:mi></bb:mrow></bb:mrow></bb:math>. Annealing of the <fb:math xmlns:fb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><fb:mrow><fb:mi>Ta</fb:mi><fb:mi mathvariant="normal">C</fb:mi></fb:mrow></fb:math> promotes recrystallization, significantly improving crystallinity and reducing crystalline defects from as-deposited columnar grains to a step-and-terrace surface morphology, enabling the subsequent growth of high-quality <jb:math xmlns:jb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><jb:msub><jb:mi>Al</jb:mi><jb:mrow><jb:mn>0.70</jb:mn></jb:mrow></jb:msub><jb:msub><jb:mi>Ga</jb:mi><jb:mrow><jb:mn>0.30</jb:mn></jb:mrow></jb:msub><jb:mrow><jb:mrow><jb:mi mathvariant="normal">N</jb:mi></jb:mrow></jb:mrow></jb:math> by molecular beam epitaxy. X-ray diffraction and scanning transmission electron microscopy confirm that the <nb:math xmlns:nb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><nb:msub><nb:mi>Al</nb:mi><nb:mi>x</nb:mi></nb:msub><nb:msub><nb:mi>Ga</nb:mi><nb:mrow><nb:mn>1</nb:mn><nb:mo>−</nb:mo><nb:mi>x</nb:mi></nb:mrow></nb:msub><nb:mrow><nb:mrow><nb:mi mathvariant="normal">N</nb:mi></nb:mrow></nb:mrow></nb:math> layer is heteroepitaxially aligned, strain-free, and lattice-matched, transitioning abruptly from <rb:math xmlns:rb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><rb:mrow><rb:mi>Ta</rb:mi><rb:mi mathvariant="normal">C</rb:mi></rb:mrow></rb:math> to <vb:math xmlns:vb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><vb:msub><vb:mi>Al</vb:mi><vb:mi>x</vb:mi></vb:msub><vb:msub><vb:mi>Ga</vb:mi><vb:mrow><vb:mn>1</vb:mn><vb:mo>−</vb:mo><vb:mi>x</vb:mi></vb:mrow></vb:msub><vb:mrow><vb:mrow><vb:mi mathvariant="normal">N</vb:mi></vb:mrow></vb:mrow></vb:math> without intermediate phases. These results demonstrate <zb:math xmlns:zb="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><zb:mrow><zb:mi>Ta</zb:mi><zb:mi mathvariant="normal">C</zb:mi></zb:mrow></zb:math> virtual substrates as electrically conductive, lattice-matched, and thermally compatible templates for vertical <dc:math xmlns:dc="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"><dc:msub><dc:mi>Al</dc:mi><dc:mi>x</dc:mi></dc:msub><dc:msub><dc:mi>Ga</dc:mi><dc:mrow><dc:mn>1</dc:mn><dc:mo>−</dc:mo><dc:mi>x</dc:mi></dc:mrow></dc:msub><dc:mrow><dc:mrow><dc:mi mathvariant="normal">N</dc:mi></dc:mrow></dc:mrow></dc:math> devices that can meet the growing power needs of a sustainable energy future.