金属有机物化学气相沉积同质外延GaN薄膜表面形貌的改善
|
摘要
为了获得高质量的GaN薄膜材料,研究了金属有机物气相沉积系统中GaN插入层对GaN衬底同质外延层表面宏观缺陷和晶体质量的影响.研究发现,插入层生长温度是影响GaN同质外延膜表面形貌和晶体质量的关键因素.由于生长模式与插入层生长温度相关,随着插入层生长温度的降低,外延膜生长模式由准台阶流模式转变为层状模式,GaN同质外延膜表面丘壑状宏观缺陷逐渐减少,但微观位错密度逐渐增大.通过对插入层温度和厚度的优化,进一步调控外延层的生长模式,最终有效降低了外延层表面的宏观缺陷,获得了表面原子级光滑平整、位错密度极低的GaN同质外延膜,其X射线衍射摇摆曲线(002),(102)晶面半峰宽分别为125arcsec和85arcsec,表面粗糙度均方根大小为0.23nm.
Free-standing GaN is generally regarded as an ideal substrate for GaN-based devices due to its advantage of low threading dislocation density(TDD) and good thermal conductivity. However, new surface features such as hillocks and ridges appear on the Ga N homoepitaxy films. In this paper, the influences of the intermediate GaN(IM-GaN) layer on the surface defects and crystal quality of Ga N homoepitaxy films grown on c-plane GaN substrates by metalorganic chemical vapor deposition are investigated. It is found that hexagonal hillocks and ridges on the surface can be avoided by inserting an IM-GaN layer grown at an intermediate temperature(650–850?C), prior to the growth of GaN at 1050?C. The results based on X-ray diffraction(XRD) measurements and differential interference contrast microscopy images demonstrate that the growth temperature of the IM-GaN layer has a significant influence on GaN homoepitaxy layer,which is one of the most critical parameters determining the surface morphology and crystal quality. As the IM-GaN growth temperature decreases from 1050?C to 650?C, thed densities of hillocks and ridges on the surface reduce gradually. While, the XRD full width at half maximum(FWHM) values of(002) and(102) peaks for the homoepitaxy films are increased rapidly, indicating the adding of the TDD in the films. The atomic force microscopy(AFM) images show that the quasi-step growth mode change into layer-layer growth mode with the growth temperature decreasing from 1050?C to 650?C during the IM-GaN layer growing. It is speculated that the growth mode is determined by the diffusion length of adatom on the growing surface, which is proportional to the growth temperature. In the case of IM-GaN grown at low temperature, the formation of hillocks can be suppressed by reducing the adatom diffusion length. Finally, High crystal quality GaN homoepitaxy films(2 μm) without hillocks is achieved by optimizing the growth parameters of IM-GaN layer, which is about 150 nm in thickness and grown at 850?C. The crystal quality of GaN homoepitaxy film is assessed by XRD rocking curve measured with double-crystal optics. The FWHMs of the(002)and(102) peaks are 125 arcsec and 85 arcsec respectively, indicating that rather low TDD is formed in the film. And well defined steps are observed on the image of AFM test, the root-mean square roughness value of the which is only about0.23 nm for 5 μm × 5 μm scan area.
Free-standing GaN is generally regarded as an ideal substrate for GaN-based devices due to its advantage of low threading dislocation density(TDD) and good thermal conductivity. However, new surface features such as hillocks and ridges appear on the Ga N homoepitaxy films. In this paper, the influences of the intermediate GaN(IM-GaN) layer on the surface defects and crystal quality of Ga N homoepitaxy films grown on c-plane GaN substrates by metalorganic chemical vapor deposition are investigated. It is found that hexagonal hillocks and ridges on the surface can be avoided by inserting an IM-GaN layer grown at an intermediate temperature(650–850?C), prior to the growth of GaN at 1050?C. The results based on X-ray diffraction(XRD) measurements and differential interference contrast microscopy images demonstrate that the growth temperature of the IM-GaN layer has a significant influence on GaN homoepitaxy layer,which is one of the most critical parameters determining the surface morphology and crystal quality. As the IM-GaN growth temperature decreases from 1050?C to 650?C, thed densities of hillocks and ridges on the surface reduce gradually. While, the XRD full width at half maximum(FWHM) values of(002) and(102) peaks for the homoepitaxy films are increased rapidly, indicating the adding of the TDD in the films. The atomic force microscopy(AFM) images show that the quasi-step growth mode change into layer-layer growth mode with the growth temperature decreasing from 1050?C to 650?C during the IM-GaN layer growing. It is speculated that the growth mode is determined by the diffusion length of adatom on the growing surface, which is proportional to the growth temperature. In the case of IM-GaN grown at low temperature, the formation of hillocks can be suppressed by reducing the adatom diffusion length. Finally, High crystal quality GaN homoepitaxy films(2 μm) without hillocks is achieved by optimizing the growth parameters of IM-GaN layer, which is about 150 nm in thickness and grown at 850?C. The crystal quality of GaN homoepitaxy film is assessed by XRD rocking curve measured with double-crystal optics. The FWHMs of the(002)and(102) peaks are 125 arcsec and 85 arcsec respectively, indicating that rather low TDD is formed in the film. And well defined steps are observed on the image of AFM test, the root-mean square roughness value of the which is only about0.23 nm for 5 μm × 5 μm scan area.
引文
[1]Palacios T,Chakraborty A,Rajan S,Rajan S,Poblenz C,Keller S,Den Baars S P,Speck J S,Mishra U K 2005IEEE Elec.Dev.Lett.26 781
[2]Webb J B,Tang H,Rolfe S,Bardwell J A 1999 Appl.Phys.Lett.75 953
[3]Limb J B,Xing H,Moran B,Mc Carthy L,Den Baars S P,Mishra U K 2000 Appl.Phys.Lett.76 2457
[4]Qin P,Song W D,Hu W X,Zhang Y W,Zhang C Z,Wang R P,Zhao L L,Xia C,Yuan S Y,Yin Y A,Li S T,Su S C 2016 Chin.Phys.B 25 088505
[5]Liu Y L,Jin P,Liu G,Wang WY,Qi Z Q,Chen C Q,Wang Z G 2016 Chin.Phys.B 25 087801
[6]Kikkawa T 2005 Jpn.J.Appl.Phys.44 4896
[7]Zhang J Q,Wang L,Li L A,Wang Q P,Jiang Y,Zhu H C,Ao J P 2016 Chin.Phys.B 25
[8]Duan X L,Zhang J C,Xiao M,Zhao Y,Ning J,Hao Y2016 Chin.Phys.B 25 087304
[9]Killat N,Montes M,Paskova T,Evans K R,Leach J,Li X,?zgürü,Morko?H,Chabak K D,Crespo A,Gillespie J K,Fitch R,Kossler M,Walker D E,Trejo M,Via G D,Blevins J D,Kuball M 2013 Appl.Phys.Lett.103193507
[10]Oehlern F,Zhu T,Kappers M J,Kappers M J,Humphreys C J,Oliver R A 2013 J.Cryst.Growth 38312
[11]Zhou K,Liu J,Zhang S M,Li Z C,Feng M X,Li D Y,Zhang L Q,Wang F,Zhu J J,Yang H 2013 J.Cryst.Growth 371 7
[12]Kizilyalli I C,Buiquang P,Disney D,Bhatia H,Aktas O 2015 Microelectron.Reliab.55 1654
[13]Kubo S,Nanba Y,Okazaki T,Manabe S,Kurai S,Taguchi T 2002 J.Cryst.Growth 236 66
[14]Leszczynskia M,Beaumont B,Frayssinet E,Knap W,Prystawko P,Suski T,Grzegory T,Porowski S 1999Appl.Phys.Lett.75 1276
[15]Okada S,Miyake H,Hiramatsu K,Miyagawa R,Eryu O,Hashizume T 2016 Jpn.J.Appl.Phys.55 01AC08
[16]Cho Y,Ha J S,Jung M,Lee H J,Park S,Park J,Fujii K,Toba R,Yi S,Kil G S,Chang J,Yao T 2010 J.Cryst.Growth 312 1693
[17]Tian W,Yan W Y,Dai J N,Li S L,Tian Y,Hui X,Zhang J B,Fang Y Y,Wu Z H,Chen C Q 2013 J.Phys.D:Appl.Phys.46 065303
[18]Heying B,Wu X H,Keller S,Li Y,Kapolnek D,Keller B P,Den Baars S P,Speck J S 1996 Appl.Phys.Lett.68 643
[19]Heinke H,Kirchner V,Einfeldt S,Hommel D 2000 Appl.Phys.Lett.77 2145
[20]Scheel H J 2001 J.Cryst.Growth 211 1
[21]Tanabe S,Watanabe N,Uchida N,Matsuzaki H 2016Phys.Status Solidi A 213 1236
[22]Corrion A L,Wu F,Speck J S 2012 J.Appl.Phys.112054903
[23]Perret E,Highland M J,Stephenson G B,Streiffer S K,Zapol P,Fuoss P H,Munkholm A,Thompson C 2014Appl.Phys.Lett.105 051602
[2]Webb J B,Tang H,Rolfe S,Bardwell J A 1999 Appl.Phys.Lett.75 953
[3]Limb J B,Xing H,Moran B,Mc Carthy L,Den Baars S P,Mishra U K 2000 Appl.Phys.Lett.76 2457
[4]Qin P,Song W D,Hu W X,Zhang Y W,Zhang C Z,Wang R P,Zhao L L,Xia C,Yuan S Y,Yin Y A,Li S T,Su S C 2016 Chin.Phys.B 25 088505
[5]Liu Y L,Jin P,Liu G,Wang WY,Qi Z Q,Chen C Q,Wang Z G 2016 Chin.Phys.B 25 087801
[6]Kikkawa T 2005 Jpn.J.Appl.Phys.44 4896
[7]Zhang J Q,Wang L,Li L A,Wang Q P,Jiang Y,Zhu H C,Ao J P 2016 Chin.Phys.B 25
[8]Duan X L,Zhang J C,Xiao M,Zhao Y,Ning J,Hao Y2016 Chin.Phys.B 25 087304
[9]Killat N,Montes M,Paskova T,Evans K R,Leach J,Li X,?zgürü,Morko?H,Chabak K D,Crespo A,Gillespie J K,Fitch R,Kossler M,Walker D E,Trejo M,Via G D,Blevins J D,Kuball M 2013 Appl.Phys.Lett.103193507
[10]Oehlern F,Zhu T,Kappers M J,Kappers M J,Humphreys C J,Oliver R A 2013 J.Cryst.Growth 38312
[11]Zhou K,Liu J,Zhang S M,Li Z C,Feng M X,Li D Y,Zhang L Q,Wang F,Zhu J J,Yang H 2013 J.Cryst.Growth 371 7
[12]Kizilyalli I C,Buiquang P,Disney D,Bhatia H,Aktas O 2015 Microelectron.Reliab.55 1654
[13]Kubo S,Nanba Y,Okazaki T,Manabe S,Kurai S,Taguchi T 2002 J.Cryst.Growth 236 66
[14]Leszczynskia M,Beaumont B,Frayssinet E,Knap W,Prystawko P,Suski T,Grzegory T,Porowski S 1999Appl.Phys.Lett.75 1276
[15]Okada S,Miyake H,Hiramatsu K,Miyagawa R,Eryu O,Hashizume T 2016 Jpn.J.Appl.Phys.55 01AC08
[16]Cho Y,Ha J S,Jung M,Lee H J,Park S,Park J,Fujii K,Toba R,Yi S,Kil G S,Chang J,Yao T 2010 J.Cryst.Growth 312 1693
[17]Tian W,Yan W Y,Dai J N,Li S L,Tian Y,Hui X,Zhang J B,Fang Y Y,Wu Z H,Chen C Q 2013 J.Phys.D:Appl.Phys.46 065303
[18]Heying B,Wu X H,Keller S,Li Y,Kapolnek D,Keller B P,Den Baars S P,Speck J S 1996 Appl.Phys.Lett.68 643
[19]Heinke H,Kirchner V,Einfeldt S,Hommel D 2000 Appl.Phys.Lett.77 2145
[20]Scheel H J 2001 J.Cryst.Growth 211 1
[21]Tanabe S,Watanabe N,Uchida N,Matsuzaki H 2016Phys.Status Solidi A 213 1236
[22]Corrion A L,Wu F,Speck J S 2012 J.Appl.Phys.112054903
[23]Perret E,Highland M J,Stephenson G B,Streiffer S K,Zapol P,Fuoss P H,Munkholm A,Thompson C 2014Appl.Phys.Lett.105 051602