The wing tilt was completely eliminated by partly-contacting with seed layer. The coalesced front at the boundary of two nucleation strips has the average linear TDD of 6.4 x 103 cm-1. The monolayer microsphere mask was proved to meet these conditions. Realization of PC-ELO demands three key conditions.
![overgrowth free download 2016 overgrowth free download 2016](https://dotawallpapers.com/_ph/1/191481903.jpg)
In conclusion, we demonstrate the PC-ELO method and achieve it without lithography by the monolayer microspheres mask. Such obvious reduction of TDs was not only due to the absence of the TDs threading at nucleation windows, which will be discussed in other paper, but also due to the depression of the nucleation tilt. AFM micrograph of more than one pattern period, showing the coalescence fronts mark by dashed line ellipses.īy coalescence boundary length per area.
![overgrowth free download 2016 overgrowth free download 2016](http://cdn.akamai.steamstatic.com/steam/apps/326380/ss_17b3e65413443ba6d7c80de5e5e69ed3c46db21d.1920x1080.jpg)
And the average linear TDD at coalescence boundary is approximately 6.4 x 103 cm-1 calculated by dividing area TDDįigure 7. 6 shows the distribution of TDs with low linear TDD, of which the worst coalescence boundary has the maximum linear TDD of 2 x 104 cm-1. The screw or mixed TDD was estimated to be 1.8 x 107 cm-2 by the tips and step-edge terminations in AFM image. It is much lower than that (4 x 108 cm-2) reported by Li et al.15 using silica microsphere of 1000 nm in diameter as growth mask on GaN template. The TDD reduces from ~2 x 109 cm-2 to 6.2 x 107 cm-2, which is about 30 magnitudes lower than that of seed layer GaN. Due to the low misorientation of the overgrowth regions (wing), surface atom steps show rare discontinuities as they cross the coalescence front6.Īn assessment of the TDs reduction can be directly made by counting the number of TDs in CL mapping26,27. In vicinity of the B- and C-type coalescence boundary, there is existence of pure edge TDs and absence of screw-component TDs. In other words, in the vicinity of A-type coalescence boundary, there is existence of screw-component TDs and absence of pure edge TDs. These TDs exists screw-component, since the tips and step-edge terminations in AFM image are the screw or mixed TDs24,25. 6 due to the coincident arrangement and density. They corresponded with the TDs at the worst coalescence boundary of the CL mapping in Fig. The small CL mapping at the upper-right corner shows the distribution of TDs at local, and marks the worst coalescence front by an arrow. Large area CL mapping illustrates the low TDD of overgrowth layer. The small cross-section SEM image in (b) shows the uncoalesced nucleation stripe.įigure 6. (b) Double crystal x-ray rocking curves of the GaN (0002) peak for coalesced and uncoalesced stripes, measured with the scattering plane perpendicular to the and directions, respectively. (a) Double crystal x-ray rocking curves of the GaN (0002) peak for uncoalesced stripes and two representive splitting peak rocking curves due to wing tilt with mask and without mask, measured with the scattering plane perpendicular to the direction. The contact interfaces are marked in figure (b-d) by the white arrows.įigure 5. (d) g = gives a clear contrast for the stacking faults. (c) g = shows the presence of edge dislocations and mixed dislocations. (b) g = highlights the presence of screw dislocations and partial dislocations that laterally terminate the stacking faults.
![overgrowth free download 2016 overgrowth free download 2016](https://steamcdn-a.akamaihd.net/steam/apps/25000/ss_29c31940d30633b0023aeca38eb9e6c9f6e36a9d.1920x1080.jpg)
(b-d) correspond to the white rectangular zone in figure (a) with g =, and, respectively.
![overgrowth free download 2016 overgrowth free download 2016](https://desktopography.net/wp-content/uploads/2016/09/original-46-1.jpg)
(a) Dark field TEM cross-section image of the GaN epitaxial layer when g =. The contact interface quality analysis of overgrowth layer and seed layer by dark field TEM cross-section image with zone axis. From the results of scanning electron microscopy, cathodoluminescence, x-ray diffraction (XRD), transmission electron microscopy, and atomic force microscope (AFM), overgrowth layer shows no tilt angle relative to the seed layer and high quality coalescence front (with average linear dislocation density ^įigure 4. Moreover, we report an improved monolayer microsphere mask method without lithography of PC-ELO method, which was used to grow GaN. What is remarkable in PC-ELO method is that the tilt angle of overgrowth stripes could be eliminated by contacting with seed layer. The passage also illustrates special mask structures with and without lithography and provides three essential conditions to achieve the PC-ELO method. We have discussed a new crystal epitaxial lateral overgrowth (ELO) method, partly-contacted ELO (PC-ELO) method, of which the overgrowth layer partly-contacts with underlying seed layer. Ming Xiao, Jincheng Zhang, Xiaoling Duan, Hengsheng Shan, Ting Yu, Jing Ning & Yue Hao Received: 24 December 2015 Accepted: 15 March 2016 Published: 01 April 2016 OPEN A partly-contacted epitaxial lateral overgrowth method applied to GaN