Electron-beam-induced strain within InGaN quantum wells: False indium “cluster” detection in the transmission electron microscope

Abstract

InGaN quantum wells have been found to be extremely sensitive to exposure to the electron beam in the transmission electron microscope (TEM). High-resolution TEM images acquired immediately after first irradiating a region of quantum well indicates no gross fluctuations of indium content in the InGaN alloy. During only a brief period of irradiation, inhomogeneous strain is introduced in the material due to electron beam damage. This strain is very similar to that expected from genuine nanometer-scale indium composition fluctuations which suggests there is the possibility of falsely detecting indium-rich “clusters” in a homogeneous quantum well.

Does In form In-rich clusters in InGaN quantum wells?

“…In the light of the Smeeton et al [16,17] papers, which suggested that the gross indium-rich clusters in InGaN quantum wells reported by many researchers might be due to electron beam damage, the Gerthsen group revised their earlier conclusions [11,12]. They observed that the indium concentration in the clusters increased with increasing irradiation time in the electron microscope.…”

“…At low electron dose, the lattice fringe image of the quantum well and the lattice parameter map are both reasonably uniform ( figure 1a, 1c), [16,17]. We have studied the effect of 200, 300, and 400 keV incident electrons.…”

“…Figure 1a was recorded within 20 seconds of first exposing this part of the quantum well to the electron beam, figure 1b is the same area after a few minutes of exposure. We have analysed these images to produce lattice parameter maps [16,17] using a process similar to the DALI technique [11]. After only a few minutes exposure to the electron beam we found nanometre-size indium clusters formed which caused local strains of up to 10%, corresponding to an indium fraction x of 60%.…”

“…Insert Fig.1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 Since publishing the Smeeton et al [16,17] papers it has been suggested to us that our results may apply only to InGaN grown on MOVPE equipment at Cambridge, or may be related to our TEM specimen preparation procedures, rather than being a general effect. We have therefore purchased a very bright blue commercial LED and examined the InGaN quantum wells it contains by TEM.…”

Comment on “Three-dimensional atom probe studies of an InxGa1−xN∕GaN multiple quantum well structure: Assessment of possible indium clustering” [Appl. Phys. Lett. 90, 061903 (2007)]

“…4 The paper cites work by Smeeton et al, 5,6 which also relates the time evolution of dotlike structures of In x Ga 1−x N samples to heavy electron beam damage. These publications do not address the ion beam damage during sample preparation, which is equally relevant 7,8 and it is well known by now that high indium concentration regions with x Ͼ 0.7 are electron and ion beam induced artifacts.…”