Diamond & Related Materials 58 (2015) 221–225
Contents lists available at ScienceDirect
Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond
HPHT synthesis and crystalline quality of large high-quality (001) and (111) diamond crystals Hitoshi Sumiya a,⁎, Katsuko Harano a, Kenji Tamasaku b a b
Advanced Materials R&D Laboratories, Sumitomo Electric Industries, Ltd., Itami, Hyogo 664-0016, Japan XFEL Research and Development Division, RIKEN SPring-8 Center, Sayo, Hyogo 679-5148, Japan
a r t i c l e
i n f o
Article history: Received 7 June 2015 Received in revised form 15 August 2015 Accepted 17 August 2015 Available online 20 August 2015 Keywords: Synthetic diamond High pressure and high temperature Crystal growth Crystalline quality Crystal defect
a b s t r a c t High-quality type IIa diamond crystals measuring up to 12 mm in diameter (8 to 10 carats) were successfully grown by the temperature gradient method at high pressure and high temperature (HPHT) on defect-free (001)-oriented and (111)-oriented seed crystals. The crystalline quality of the grown crystals was qualitatively evaluated by synchrotron based X-ray topograph and rocking curve (RC) measurements. The results revealed that the crystals grown on (001)-oriented seeds have extremely high quality in the (001) growth sector, which extends straight upward from the seed surface. This sector contains very few dislocations and stacking faults in the upper part in particular, where the RC almost agrees with the theoretical curve. For the crystals grown on (111)-oriented seeds, it was confirmed that crystal defects are fewer in the outer (100), (010), and (001) growth sectors than in the central (111) growth sector extending straight upward from the seed surface. The RCs in selected regions of these {100} growth sectors are close to the theoretical curve, indicating high crystalline quality. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Diamond crystals synthesized by the temperature gradient method at high pressure and high temperature (HPHT) may be far superior in crystalline quality and consistency among individual crystals compared to natural diamonds. In particular, type IIa diamond crystals synthesized by advanced control of impurities and crystal defects have extremely high crystalline quality [1–4]. They are, therefore, very useful as monochromators or other optical elements of synchrotron X-ray radiation beams [5–10]. The next-generation of highly brilliant beams (free-electron X-ray laser, XFEL) requires diamonds of much higher crystalline quality, and so there are high expectations for these large synthetic type IIa diamond crystals. They are also expected to be introduced to other applications such as various high-sensitivity detectors (sensors), or as substrates for electronic devices in the future. In the 1990s we demonstrated that high-quality type IIa diamond crystals of 5–6 mm (1–2 carats) in diameter could be produced by eliminating chemical impurities below 0.1 ppm [1] using low-defect density (001)-oriented seeds [2]. Recently, we succeeded in producing much larger high-quality type IIa diamond crystals measuring up to 12 mm in diameter by improving the synthesis technology. The (001) growth sectors of the large crystals grown on (001)-oriented seeds were found to be almost defect-free [11]. In addition, we successfully synthesized large type IIa crystals of 12 mm in diameter with (111)-oriented seeds. In this study, we investigate the distribution of dislocations and ⁎ Corresponding author. E-mail address:
[email protected] (H. Sumiya).
http://dx.doi.org/10.1016/j.diamond.2015.08.006 0925-9635/© 2015 Elsevier B.V. All rights reserved.
stacking faults in these large crystals grown on (001)-oriented and (111)-oriented seeds and the dependency of crystalline quality on the major grown sectors through X-ray topograph and rocking curve (RC) measurement using synchrotron X-ray beams. We also indicate the differences between the crystals grown from these two different orientations of seeds. 2. Experimental methods High-quality type IIa diamond crystals measuring up to 12 mm in diameter (8 to 10 carats) were synthesized by the temperature gradient method under HPHT conditions of 5.5 GPa and 1340–1350 °C using Tiloaded high-purity Fe–Co solvents, high-purity carbon sources, and defect-free seed crystals [1,2]. Seed crystals measuring 0.5 × 0.5 mm cut from high crystalline quality type IIa crystals synthesized by the temperature gradient method were used. Two types of seed crystal, (001)-oriented and (111)-oriented, were prepared. The diamond crystals grown on the (001)-oriented and (111)-oriented seed surfaces were sliced by laser cutting as shown in Fig. 1 to prepare diamond plates of 0.8 to 1.2 mm in thickness. The (001) surfaces were polished using metal-bonded diamond wheel. The (111) surfaces were not polished in this study because the (111) plane is too hard and mechanical damage is easily introduced while polishing. UV-excited luminescence images of these diamond plates were taken using the DiamondView™ instrument (Diamond Trading Company, [12]) in order to distinguish growth sectors in the diamond crystals. The crystalline quality of these diamond plates was qualitatively evaluated by synchrotron X-ray RC measurement and topography as
222
H. Sumiya et al. / Diamond & Related Materials 58 (2015) 221–225
Fig. 1. Large (001) and (111) high-quality type IIa diamond crystal plates. (a) (001)/(001) crystal: (001) plate cut from the crystal grown on the (001)-oriented seed, (b) (111)/(001) crystal: (111) plate cut from crystal grown on (001)-oriented seed, (c) (111)/(111) crystal: (111) plate cut from crystal grown on (111)-oriented seed.
described next. At the 1-km beamline of SPring-8 (BL29XUL) [13], the topographs were acquired under quasi-plane wave incidence using a Si collimator [14]. Table 1 shows the experimental conditions (reflection plane, collimator, photon energy). A CCD-based X-ray camera with a pixel size of 12 μm was used to capture the topographs. The topographs were acquired at the peak and low-angle side tail region of the RC. The RCs were measured with footprints over the entire surface of the crystals and in a narrow area (250 × 250 μm) selected by a slit.
The UV-excited luminescence images of the (001)/(001) crystal plate and the (111)/(001) crystal plate captured with the DiamondView™ instrument are shown in Fig. 2. The {100} growth
3. Results and discussion 3.1. Synthesis of high-quality large type IIa diamond crystals Crystal growth experiments were performed for more than 200 h under very precise temperature control within a narrow high-pressure and high-temperature region where good-quality crystals can be obtained [4,11]. Examples of the resulting large good-quality crystals are shown in Fig. 1. Large diamond plates of both (001) and (111) orientations measuring up to 12 mm in diameter were produced. Fig. 1 (a) shows a (001) crystal plate cut out from the central region of the crystal grown on (001)-oriented seed by cutting in the horizontal direction, parallel to the (001) surface. Fig. 1(b) shows a (111) crystal plate cut in an oblique direction, parallel to the oblique (111) surface from the crystal grown on (001)-oriented seed. Fig. 1(c) shows a (111) crystal plate prepared from the crystal grown on (111)-oriented seed by cutting in the horizontal direction, parallel to the (111) seed surface. Here, the diamond crystal plates cut as shown in Fig. 1(a), (b), and (c) are referred to as “(001)/(001) crystal", “(111)/(001) crystal”, and “(111)/(111) crystal”, respectively. The amount of chemical impurities such as nitrogen and boron in each of these large crystals was less than 0.1 ppm (below the limit of analysis by optical absorption). Table 1 Experiment conditions. Reflection plane of diamond
Collimator (asymmetry)
Photon energy (keV)
400 Bragg geometry 111 Bragg geometry 220 Laue geometry
Si(531) Si(220) Si(331)
19.744 9.439 14.547
Fig. 2. UV-excited luminescence images. (a) (001) crystal plate cut out from the diamond crystal grown on a (001)-oriented seed, (b) (111) crystal plate cut out from the diamond crystal grown on a (111)-oriented seed. Several images were taken and patched because the diamond crystal plates were larger than the field of view.
H. Sumiya et al. / Diamond & Related Materials 58 (2015) 221–225
sectors of these plates exhibit little fluorescence, which indicates that the purity is very high. The blue light emitted from the {111} growth sectors is attributed to trace amounts of nitrogen and boron impurities [15].
3.2. Crystalline quality a. (001)/(001) crystal Fig. 3 shows a (001) crystal plate cut from the middle section of the crystal grown on a (001)-oriented seed as shown in Fig. 1(a). Fig. 3 (a) and (b) shows the X-ray topographs of the 220 transmission Laue geometry taken from the tail and the peak of the RC, respectively, and panel (c) shows the X-ray topograph of the 400 reflection Bragg geometry taken from the tail of the RC. The interior of the (001) growth sector in the central region exhibits few dislocation defects, no stacking faults and no inclusions, indicating extremely high crystalline quality. It can be confirmed through observation of UV-excited luminescence image that the interior of the (001) growth sector is very pure (Fig. 2(a), [11]). Fig. 3(d) shows the RC of this crystal plate. In the (001) growth sector (the central region) of the crystal grown on a (001)-oriented seed, the RC was almost equivalent to the theoretical value. The small bump on the low-angle side is attributable to the reflection on the rear face. On
223
the other hand, stacking faults are found inside the (111) growth sector. The tail of the entire RC of this crystal (“whole surface” in Fig. 3(d)) is slightly wider because of the effect of the stacking faults inside the {111} growth sectors. In general, dislocation defects of synthetic diamonds originate from the seed or inclusions, regardless of the growth sector. Especially, dislocation defects induced mainly by the seed crystal are noticeable. Such dislocation defects can be significantly reduced by increasing the crystalline quality of seed crystals [2]. On the other hand, many stacking faults exist inside the surrounding {111} growth sectors. No such stacking faults are observed inside the (001) growth sectors. The stacking faults inside the {111} growth sectors are formed radially, or in a fanlike pattern, from the seed toward the four surfaces (111), (1-11), (-111), and (-111), when the seed is (001)-oriented [11].
b. (111)/(001) crystal Fig. 4 shows the experimental results for the (111) crystal plate cut parallel to the oblique (111) surface from the crystal grown on (001)oriented seed, as shown in Fig. 1(b). Fig. 4(a) shows the X-ray topograph image taken from the tail of the RC with the 111 reflection Bragg geometry. In the (001) growth sector in the central region (“top-side” in Fig. 4), few crystal dislocations are found. Fig. 4(b) is the
Fig. 3. X-ray topograph images (a, b, c) and rocking curves (d) of the (001)/(001) crystal: (001) plate cut from the crystal grown on the (001)-oriented seed. (a) X-ray topograph image taken from the tail of the RC with a 220 transmission Laue geometry, (b) X-ray topograph image taken from the peak of the RC with a 220 transmission Laue geometry, (c) X-ray topograph image taken from the tail of the RC with a 400 reflection Bragg geometry, (d) Rocking curves measured with a beam footprint over the whole surface (red) and the central selected region (blue) with a 400 reflection Bragg geometry. The black line indicates the theoretical curve. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
224
H. Sumiya et al. / Diamond & Related Materials 58 (2015) 221–225
Fig. 4. X-ray topograph image (a) and rocking curves (b) of the (111)/(001) crystal: (111) plate cut from the crystal grown on the (001)-oriented seed. (a) X-ray topograph image taken from the tail of the RC with a 111 reflection Bragg geometry, (b) Rocking curves measured on the top-side selected region (red) and the seed-side selected region (blue) with a 111 reflection Bragg geometry. The black line indicates the theoretical curve. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
RC of this plate crystal. The tail of the RC is slightly wider on the side close to the seed (“seed-side” in Fig. 4). The results indicate that the distribution and density of crystal defects depend on the position and that the upper part of the grown crystal has high crystalline quality. In
contrast, the lower region close to the seed crystal sustains a little internal strain and a few crystal defects attributable to unstable growth at an early stage of crystal growth, leading to slightly lower crystalline quality. The probable reason why the RC in the high-quality region is a
Fig. 5. X-ray topograph images (a, b, c) and rocking curves (d) of the (111)/(111) crystal: (111) plate cut from the crystal grown on the (111)-oriented seed. (a) X-ray topograph image taken from the tail of the RC with a 220 transmission Laue geometry, (b) X-ray topograph image taken from the peak of the RC with a 220 transmission Laue geometry, (c) X-ray topograph image taken from the tail of the RC with a 111 reflection Bragg geometry, (d) Rocking curves measured on the outer selected region (red) and the central selected region (blue) with a 111 reflection Bragg geometry. The black line indicates the theoretical curve. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
H. Sumiya et al. / Diamond & Related Materials 58 (2015) 221–225
little wider than the theoretical curve is slight surface roughness produced by laser cutting. c. (111)/(111) crystal Fig. 5 shows X-ray topographs and RC data of a (111) crystal plate cut parallel to the (111) surface by laser from the central region (middle layer) of the crystal grown on a (111)-oriented seed, as shown in Fig. 1 (c). Fig. 5(a) and (b) shows the X-ray topographs of the 220 transmission Laue geometry taken from the tail and the peak of the RC respectively, and Fig. 5(c) shows the X-ray topograph of the 111 reflection Bragg geometry taken from the tail of the RC. Although some stacking faults are observed, there are few dislocations. The defect distribution of this crystal reveals that, unlike the crystal grown on (001)-oriented seed, there are fewer crystal defects in the region surrounding the crystal plate (“outer region” in Fig. 5) than in the growth sector in the upper region of the seed crystal (“central region” in Fig. 5). Few dislocations are observed inside the surrounding (100), (010) and (001) growth sectors, in particular. Fig. 5(d) shows the RC (111 reflection Bragg geometry) of this crystal plate. The RC (red line) of one of the surrounding {100} growth sectors is narrower and has a lower tail than the RC (blue line) of the central region with some stacking faults. However, both of the RCs are almost the same as the theoretical curves, indicating that the crystalline quality of this (111) crystal plate is very high. The lower slopes of the RCs are slightly wider than the theoretical curve due to surface roughness formed by laser cutting. 4. Conclusions (1) Large type IIa diamond crystals measuring 12 mm in diameter grown on (001)-oriented and (111)-oriented seed crystals were produced, and the crystalline quality was qualitatively evaluated by X-ray topograph and RC measurement using synchrotron X-ray beams. (2) The quality of the interior of the (001) growth sector of the crystal grown on the (001)-oriented seed crystal, which extended straight upward from the seed surface, is extremely high, and the upper region of the sector in particular has very few dislocation and stacking faults, with an RC almost equivalent to the theoretical curve. (3) With regard to the defect distribution of the crystal grown on the (111)-oriented seed crystal, there are fewer crystal defects in the (100), (010) and (001) growth sectors around the crystal than in the (111) growth sector just above the seed surface. The RCs of these {100} growth sectors are close to the theoretical curve, indicating extremely high crystalline quality.
225
Prime novelty statement Very large high-quality type IIa diamond crystals up to 12 mm in diameter were grown by the temperature gradient method at HPHT on (001)-oriented and (111)-oriented seed crystals, and the crystalline quality of the grown crystals was evaluated by synchrotron X-ray topography and rocking curve (RC) measurement. It was revealed that both large crystals grown on (001)-oriented and (111)-oriented seeds have very few crystal defects in the {100} growth sectors and that the RCs in these {100} growth sectors are close to the theoretical curve, indicating high crystalline quality.
References [1] H. Sumiya, S. Satoh, High-pressure synthesis of high-purity diamond crystal, Diam. Relat. Mater. 5 (1996) 1359–1365. [2] H. Sumiya, N. Toda, Y. Nishibayashi, S. Satoh, Crystalline perfection of high purity synthetic diamond crystal, J. Cryst. Growth 178 (1997) 485–494. [3] R.C. Burns, J.O. Hansen, R.A. Spits, M. Sibanda, C.M. Welbourn, D.L. Welch, Growth of high purity large synthetic diamond crystals, Diam. Relat. Mater. 8 (1999) 1433–1437. [4] H. Sumiya, N. Toda, S. Satoh, Growth rate of high-quality large diamond crystals, J. Cryst. Growth 237-239 (2002) 1281–1285. [5] G. Grübel, D. Abernathy, G. Vignaud, M. Sanchez del Rio, A. Freund, A diamond double-crystal transmission monochromator for the TROIKA II station at ESRF, Rev. Sci. Instrum. 67 (1996) 3349. [6] M. Suzuki, N. Kawamura, M. Mizumaki, A. Urata, H. Maruyama, S. Goto, T. Ishikawa, Polarization–modulation technique with diamond phase retarder to improve the accuracy of XMCD measurements, J. Synchrotron Radiat. 6 (1999) 190–192. [7] A.K. Freund, J. Hoszowska, J.P.F. Sellschop, R.C. Burns, M. Rebak, Recent developments of high quality synthetic diamond single crystals for synchrotron X-ray monochromators, Nucl. Instrum. Methods A 467-468 (2001) 384–387. [8] J.P.F. Sellschop, A. Freund, J. Hoszowska, S.H. Connell, M. Rebak, R.C. Burns, The quest for diamond crystal perfection to meet the needs of physics: unrealistic dream or attainable target? Phys. Status Solidi 193 (2002) 415–422. [9] M. Yabashi, S. Goto, Y. Shimizu, K. Tamasaku, H. Yamazaki, Y. Yoda, M. Suzuki, Y. Ohishi, M. Yamamoto, T. Ishikawa, Diamond Double-Crystal Monochromator for SPring-8 Undulator Beamlines, Synchrotron Radiation Instrumentation: Ninth International Conference, CP879, 2007 922–925. [10] R.C. Burns, A. Chumakov, S.H. Connell, D. Dube, H.P. Godfried, J.O. Hansen, J. Härtwig, J. Hoszowska, F. Masiello, L. Mkhonza, M. Rebak, A. Rommevaux, R. Setshedi, P. Van Vaerenbergh, HPHT growth and x-ray characterization of high-quality type IIa diamond, J. Phys. Condens. Matter 21 (2009) 364224. [11] H. Sumiya, K. Tamasaku, Large defect-free synthetic type IIa diamond crystals synthesized via high pressure and high temperature, Jpn. J. Appl. Phys. 51 (2012) 090102. [12] C.M. Welbourn, M. Cooper, P.M. Spear, The De Beers natural versus synthetic diamond verification instruments, Gems & Gemology 32 (1996) 156–169. [13] K. Tamasaku, Y. Tanaka, M. Yabashi, H. Yamazaki, N. Kawamura, S. Suzuki, T. Ishikawa, SPring-8 RIKEN Beamline III for Coherent X-Ray Optics, Nucl. Instrum. Methods Phys. Res. Sect. A (467-468) (2001) 686–689. [14] K. Tamasaku, T. Ueda, D. Miwa, T. Ishikawa, Goniometric and topographic characterization of synthetic IIa diamonds, J. Phys. D. Appl. Phys. 38 (2005) A61–A66. [15] K. Watanabe, S.C. Lawson, J. Isoya, H. Kanda, Y. Sato, Phosphorescence in highpressure synthetic diamond, Diam. Relat. Mater. 6 (1997) 99–106.