Effect of Al evaporation temperature on the properties of Al films grown on sapphire substrates by molecular beam epitaxy

Abstract

High-quality Al films with an in-plane epitaxial relationship of Al[1−10]//sapphire[1−100] have been epitaxially grown on sapphire substrates by molecular beam epitaxy. The as-grown and ∼200 nm thick Al films prepared at an Al evaporation temperature of 1100 °C were highly crystalline, with a full-width at half-maximum of 180 arcseconds, and had a very smooth surface, with a root mean square roughness of 0.6 nm. There was no interfacial layer between the Al and sapphire. Furthermore, the effect of the Al evaporation temperature on the properties of the as-grown ∼200 nm thick Al films has been studied in detail. This work of achieving high-quality Al films is of great importance for the fabrication of high-performance Al-based devices.

---

1. Introduction

Due to its excellent properties, there is increasing interest in the epitaxial growth of Al film^1–4 for its applications in metal-oxide semiconductor microelectronic devices and subwavelength surface plasmonic devices.^5–7 Recent studies have found that the performance of these Al-based devices depends on the surface morphologies, crystallinity, and interfacial properties of Al films.^8–10 On one hand, the flat Al surface will reduce the resistance of devices, and eventually assist in the development of high-efficiency devices. On the other hand, the high crystallinity of Al films means that few dislocations are formed in the as-grown Al films, which helps to reduce the leakage current in devices. Furthermore, the excellent interfacial properties of Al/oxide hetero-interfaces are beneficial to the migration of electrons in devices. In this regard, the growth of high-quality Al films is of paramount importance for the fabrication of high-performance Al-based devices.^11,12 So far, there have been some reports about growing Al films. H. Liu et al. and H. Uchida et al. grew Al films on Si substrates by molecular beam epitaxy (MBE), and studied the Al surface morphologies and the corresponding epitaxial relationships.^6,7 However, due to the large lattice mismatch between Al and Si, it is hard to obtain high-quality Al films on Si substrates. D. Medlin et al. used sapphire as a substrate for the growth of Al films by the evaporation of Al from an effusion cell.⁵ Thanks to the small lattice mismatch between Al and sapphire, the quality of Al films has been improved to some extent. However, the Al/sapphire hetero-interfaces are still not abrupt, and a ∼1 nm thick interfacial layer exists between Al films and sapphire substrates.⁵ It may be that this is because the suitable conditions for the growth of Al films on sapphire substrates have not yet been deduced.

In this work, we have epitaxially grown Al films on sapphire substrates by MBE with a suitable Al evaporation temperature. On one hand, the sapphire substrate shows a very small lattice mismatch of 3.9% with the Al films, which is good for the nucleation of Al films during the initial growth. On the other hand, the suitable Al evaporation temperature in this work is beneficial to the migration of Al precursors. Employing both of these aspects would lead to the growth of high-quality Al films.

Herein, we report on the growth of high-quality Al films with sharp and abrupt hetero-interfaces on sapphire substrates by MBE. The effect of the Al evaporation temperature on the surface morphologies, crystallinity and interfacial properties of the as-grown and ∼200 nm thick Al films was also studied in detail using various techniques, such as in situ reflection high energy electron diffraction (RHEED), white-light interferometry, atomic force microscopy (AFM), high-resolution X-ray diffraction (HRXRD), and high-resolution transmission electron microscopy (HRTEM). It was found that the as-grown and ∼200 nm thick Al films prepared at an Al evaporation temperature of 1100 °C had very smooth surfaces, were highly crystalline, and exhibited sharp and abrupt hetero-interfaces. This work of growing high-quality Al films is of paramount importance for the application of Al-based devices.

2. Experimental

The as-received 2-inch sapphire substrates underwent a degassing treatment in an ultra-high vacuum (UHV) load-lock chamber with a background pressure of 1.0 × 10⁻⁸ Torr for 30 min, and were then transferred into an UHV MBE growth chamber with a background pressure of 2.0 × 10⁻¹⁰ Torr. Afterwards, the as-transferred sapphire substrates were annealed at 850 °C for 60 min to remove the surface contamination. During the epitaxial growth, high-purity (99.9999%) aluminum slugs with a 3.175 mm diameter and 3.175 mm length bought from Alfa Aesar were used as the precursors of Al. Meanwhile, the Al evaporation temperature was set in the range from 1000 to 1150 °C in a nitrogen atmosphere with an optimized nitrogen flow rate of 1 sccm. The rotation rate of the sapphire substrates was kept at 5 rotations per min to guarantee the growth of Al films with a homogeneous thickness at 750 °C. As for the calculation of the growth rate, the growth of Al films for 30 min at various temperatures ranging from 1000 to 1150 °C was carried out. After the epitaxial growth had been completed, the surface morphologies, crystallinity, and interfacial properties of the as-grown ∼200 nm thick Al films were characterized using in situ RHEED, white-light interferometry (Y-Wafer GS4-GaN-R-405), AFM (Bruker Dimension Edge, American), HRXRD (Bruker D8 X-ray diffractometer with Cu Kα1 X-ray source λ = 1.5406 Å), and HRTEM (JEOL 3000F, field emission gun TEM working at a voltage of 300 kV, which gives a point to point resolution of 0.17 nm).

3. Results and discussion

During the epitaxial growth, in situ RHEED measurements were taken to monitor the growth process. Fig. 1a shows the sharp and streaky RHEED patterns for sapphire substrates after the annealing process, which confirms that the sapphire surface was very smooth, which is advantageous for the subsequent growth process.^13,14 Fig. 1b reveals the clear and streaky RHEED patterns for the as-grown and ∼200 nm thick Al films grown at 750 °C with an Al evaporation temperature of 1100 °C, which proves that single-crystalline Al films with smooth surfaces have been obtained. Meanwhile, we also found that the single-crystalline Al films on sapphire substrates can be obtained with various Al evaporation temperatures ranging from 1000 to 1150 °C, based on the RHEED observations. After carefully studying the RHEED patterns, it was found that an in-plane epitaxial relationship of Al[1−10]//sapphire[1−100] was obtained.^15–17 It is known that the a for sapphire is 0.4765 nm, and the corresponding length of a_sapphire/3[1−100] is 0.2751 nm; while the a for Al is 0.4040 nm, and the corresponding length of a_Al/2[1−10] is 0.2857 nm, as shown in Fig. 1c. Therefore, the lattice mismatch between sapphire[1−100] and Al[1−10] was calculated to be 3.9%.^18,19

Fig. 1 RHEED patterns for as-annealed (a) sapphire substrates and (b) the as-grown ∼200 nm thick Al films grown with an evaporation temperature of 1100 °C. (c) The schematic diagram of the lattice arrangement of Al films grown on sapphire substrates.

To calculate the growth rate of Al films grown with various Al evaporation temperatures, white-light interferometry was employed. Fig. 2a shows a thickness distribution image of Al films grown on 2-inch sapphire substrates for 30 min with an Al evaporation temperature of 1100 °C, and the thickness of the as-grown Al films was measured to be about 200 nm. Using the same method, the thicknesses of the Al films grown for 30 min with Al evaporation temperatures of 1000, 1050, and 1150 °C were measured to be about 159, 178, and 212 nm, respectively. Therefore, the average growth rates for Al films grown with the Al evaporation temperatures of 1000, 1050, 1100, and 1150 °C were about 318, 356, 400, and 424 nm h⁻¹, respectively (Fig. 2b). In this case, we find that the growth rate for Al films grown on sapphire substrates is gradually increased as the Al evaporation temperature increases from 1000 to 1150 °C.

Fig. 2 (a) The thickness distribution of the Al films grown at 1100 °C on a 2-inch sapphire substrate. (b) The growth rate of Al films grown with various Al evaporation temperatures ranging from 1000 to 1150 °C.

The influence of the Al evaporation temperature on the surface morphologies of the as-grown ∼200 nm thick Al films was studied using AFM. It can be clearly seen that the ∼200 nm thick Al films grown with an Al evaporation temperature of 1000 °C show a very rough Al surface, with a root mean square (RMS) surface roughness of 2.8 nm in Fig. 3a. If the Al evaporation temperature increases from 1000 to 1100 °C, the surface morphologies are gradually improved. Actually, the ∼200 nm thick Al films grown with an Al evaporation temperature of 1100 °C show a very smooth surface, with a RMS surface roughness of 0.6 nm, as shown in Fig. 3b. However, if the Al evaporation temperature is increased further, the surface morphologies for the ∼200 nm thick Al films become poorer, with a RMS surface roughness of 2.1 nm, as shown in Fig. 3c. This can possibly be attributed to the suitable Al evaporation temperature. If the Al evaporation temperature is too low, the flow rate for Al atoms will be low as well, which results in the low growth rate of Al films. During the growth of the films, the low growth rate results in a poorly formed surface, due to the fact that the surface kinetics are not sufficient for the migration of the adatoms to their lowest energy crystal positions. Furthermore, these adatoms are superimposed by additional adatoms, and many dislocations are formed under these conditions.^20,21 If the Al evaporation temperature is too high, the flow rate for Al atoms will be high as well, which would lead to the high growth rate of Al films. The high growth rate of Al films results in a rough surface, due to the fact that the surface kinetics do not allow sufficient time for adatoms to move into their lowest energy crystal positions before being superimposed by additional adatoms.^22,23 In fact, many dislocations are also formed in this case.^22,23 Both of these cases lead to the formation of poor-quality Al films. In this regard, the Al evaporation temperature of 1100 °C seems to be the optimal temperature for the growth of high-quality Al films on sapphire substrates.

Fig. 3 AFM images for the as-grown ∼200 nm thick Al films grown on sapphire substrates with the different Al evaporation temperatures of (a) 1000, (b) 1100, and (c) 1150 °C.

XRD was employed to further study the structural properties of the as-grown ∼200 nm thick Al films. Fig. 4a shows a typical XRD 2θ–ω scan for the ∼200 nm thick Al films grown with various Al evaporation temperatures. It can clearly be seen that when the Al evaporation temperature increases from 1000 to 1100 °C, the peaks for Al(111) and Al(222) both become sharp and narrow, which reveals the increase in crystallinity.¹¹ However, when the temperature is further increased to 1150 °C, the peaks for Al(111) and Al(222) broaden, which confirms the decrease in crystallinity. Furthermore, the out-of-plane epitaxial growth relationship of Al(111)//sapphire(0001) can be deduced from Fig. 4a. Fig. 4b shows the typical φ scans for Al(1−13) and sapphire(1−102), where the six-fold rotational peaks for Al(1−13) and three-fold rotational peaks for sapphire(1−102) can be identified. This result proves the in-plane epitaxial relationship of Al[1−10]//sapphire[1−100] between Al and sapphire,²⁴ which is in good agreement with the RHEED findings. Based on these results, another in-plane epitaxial relationship of Al[11−2]//sapphire[11−20] between Al and sapphire can be obtained, as shown in Fig. 4c.

Fig. 4 XRD 2θ–ω scan for the as-grown ∼200 nm thick Al(111) films grown with various Al evaporation temperatures, and (b) φ scans for Al(1−13) and sapphire(1−102). (c) The epitaxial relationship between Al films and sapphire substrates for Al films grown on sapphire substrates.

The crystallinity of the as-grown ∼200 nm thick Al films was evaluated using X-ray rocking curves (XRCs). It is known that the full-width at half-maximum (FWHM) of the XRC for Al films is related to the dislocation density in the films, and therefore the FWHM is widely used to evaluate the crystallinity of as-grown films.^25,26 Fig. 5a shows an XRC for the ∼200 nm thick Al films grown with the Al evaporation temperature of 1100 °C. The FWHM for Al(111) is as small as 180 arcseconds, which stands in stark contrast to that of Al films grown by the Czochralski method, with a FWHM of 1800 arcseconds. We attribute this result to the very small lattice mismatch between Al and sapphire, which is good for the nucleation of Al films on the sapphire substrates during the initial growth and is thereby ultimately beneficial to the growth of high-quality Al films.^5,11 Additionally, the influence of the Al evaporation temperature on the FWHM for Al(111) films was also studied (Fig. 5b). One can observe that, when the Al evaporation temperature increases from 1000 to 1100 °C, the FWHM for Al(111) films is greatly reduced from 1080 to 180 arcseconds, which shows that the crystallinity of the Al films is improved significantly. However, when the Al evaporation temperature is further increased to 1150 °C, the FWHM for Al(111) films is broadened to 288 arcseconds, revealing the decline in crystallinity.

Fig. 5 (a) XRC for the ∼200 nm thick Al(111) films grown with the Al evaporation temperature of 1100 °C and (b) the influence of the Al evaporation temperature on the FWHM for Al(111).

The interfacial properties of the as-grown ∼200 nm thick Al films grown on sapphire substrates were studied using cross-sectional TEM. Fig. 6a is a low-magnification cross-sectional TEM image for 201 nm thick Al films grown with an Al evaporation temperature of 1100 °C. Fig. 6b shows a high-magnification HRTEM image for the Al/sapphire hetero-interfaces grown with the Al evaporation temperature of 1100 °C. There is no interfacial layer between the Al and sapphire.^27,28 This result confirms the excellent interfacial properties of the as-grown Al films grown with the Al evaporation temperature of 1100 °C. After careful study, another in-plane epitaxial relationship of Al[11−2]//sapphire[11−20] can be obtained.^29–31 Additionally, the influence of the Al evaporation temperature on the interfacial layer thickness of the Al/sapphire hetero-interfaces was deduced, as shown in Fig. 6c. One can find that when the Al evaporation temperature is increased from 1000 to 1100 °C, the interfacial layer thickness of the Al/sapphire hetero-interfaces is decreased from 2.0 to 0 nm. However, when the Al evaporation temperature is further raised, the interfacial layer thickness is increased to 1.0 nm. We attributed this to the suitable Al evaporation temperature, which is beneficial to the migration of Al precursors on the sapphire substrates during initial growth and eventually leads to the formation of sharp and abrupt hetero-interfaces. Additionally, the trend in the interfacial layer thickness is good agreement with that for the surface morphologies of Al films grown on sapphire substrates with various Al evaporation temperatures. In this regard, the Al evaporation temperature of 1100 °C is the optimal temperature.

Fig. 6 (a) Low- and (b) high-magnification cross-sectional TEM images for Al/sapphire hetero-interfaces grown with an Al evaporation temperature of 1100 °C. (c) The influence of the Al evaporation temperature on the interfacial layer thickness.

4. Conclusions

In summary, ∼200 nm thick Al films have been grown on sapphire substrates with various Al evaporation temperatures. The influence of the Al evaporation temperature on the crystallinity, surface morphologies, and interfacial properties of the as-grown ∼200 nm thick Al films grown on the sapphire substrates were carefully studied. It was found that the crystallinity, surface morphologies, and interfacial properties of the as-grown ∼200 nm thick Al films were first improved and then worsened as the Al evaporation temperature increased from 1000 to 1150 °C, and that the optimal Al evaporation temperature was 1100 °C. We ascribe these results to two aspects of the films’ growth. One is the small lattice mismatch between Al and sapphire, and the other is the suitable Al evaporation temperature. The former is advantageous for the nucleation of Al films on sapphire substrates, and is beneficial for the growth of high-quality Al films. The latter is good for the coherence of the Al precursors. If the Al evaporation temperature is too low, the flow rate for the Al atoms will be low as well, which results in the low growth rate of Al films. During the growth of the films, the low growth rate results in a poorly formed surface, due to the fact that the surface kinetics are not sufficient for the migration of the adatoms to their lowest energy crystal positions. Furthermore, these adatoms are superimposed by additional adatoms, and many dislocations are formed under these conditions. If the Al evaporation temperature is too high, the flow rate for the Al atoms will be high as well, which would lead to the high growth rate of Al films. The high growth rate of Al films results in a rough surface, due to the fact that the surface kinetics do not allow sufficient time for adatoms to move into their lowest energy crystal positions before being superimposed by additional adatoms. In fact, many dislocations are also formed in this case. Both of these cases lead to the formation of poor-quality Al films. This work on obtaining high-quality Al films is of great importance for the fabrication of high-performance Al-based devices.

Acknowledgements

This work is supported by National Science Fund for Excellent Young Scholars of China (no. 51422203), National Natural Science Foundation of China (no. 51372001), Outstanding Youth Foundation of Guangdong Scientific Committee (no. S2013050013882), Key Project in Science and Technology of Guangdong Province (no. 2011A080801018), and Strategic Special Funds for LEDs of Guangdong Province (nos 2011A081301010, 2011A081301012, and 2012A080302002).

References

1. F. Zhang, X. Zou, F. Sun, H. Ren, Y. Jiang and G. Zhu, CrystEngComm, 2012, 14, 5487–5492.
2. F. Fei, J. Zhuang, W. Wu, M. Song, D. Zhang, S. Li, W. Su and Z. Cui, RSC Adv., 2015, 5, 608–611.
3. E. Hosono, S. Fujihara and T. Kimura, J. Mater. Chem., 2004, 14, 881–886.
4. X. Zhou, R. E. Jones, P. E. Hopkinsb and T. E. Beechem, Phys. Chem. Chem. Phys., 2014, 16, 9403–9410.
5. D. Medlin, K. McCarty, R. Hwang, S. Guthrie and M. Baskes, Thin Solid Films, 1997, 299, 110–114.
6. H. Liu, Y. Zhang, D. Wang, M. Pan, J. Jia and Q. Xue, Surf. Sci., 2004, 571, 5–11.
7. H. Uchida, T. Kuroda, F. Mohamad, J. Kim, K. Nishimura and M. Inoue, Surf. Sci., 2004, 566–568, 197–202.
8. Y. Lim, M. Chaudhri and Y. Enomoto, J. Mater. Res., 1999, 14(6), 2314–2327.
9. S. W. Whangbo, Y. K. Choi, K. B. Chung, Y. D. Chung, W. S. Koh, H. K. Jang, H. W. Yeom, K. Jeoung, S. K. Kang, D.-H. Ko and C. N. Whang, J. Mater. Chem., 2002, 12, 2559–2562.
10. Z. Tang, J. Teng, Y. Jiang, J. Jia, J. Guo and K. Wu, J. Appl. Phys., 2007, 102, 053504.
11. W. Wang, W. Yang, Z. Liu, Y. Lin, S. Zhou, H. Qian, H. Wang, Z. Lin and G. Li, CrystEngComm, 2014, 16, 7626–7632.
12. B. Cho, T. Schwarz-Selinger, K. Ohmori and D. Cahill, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 66, 195407.
13. W. Wang, Z. Liu, W. Yang, Y. Lin, S. Zhou, H. Qian, H. Wang, Z. Lin and G. Li, RSC Adv., 2014, 4, 39651–39656.
14. W. Wang, W. Yang, Z. Liu, Y. Lin, S. Zhou, H. Qian, H. Wang, Z. Lin and G. Li, RSC Adv., 2014, 4, 27399–27403.
15. A. Ichimiya and P. I. Cohen, Reflection high-energy electron diffraction, Cambridge University press, 2004.
16. C. Gaire, F. Tang and G. C. Wang, Thin Solid Films, 2009, 517, 4509–4514.
17. W. Wang, W. Yang, Z. Liu, Y. Lin, S. Zhou, H. Qian, F. Gao and G. Li, CrystEngComm, 2014, 16, 4100–4107.
18. L. Liu and J. H. Edgar, Mater. Sci. Eng., R, 2002, 37, 61–127.
19. W. Wang, W. Yang, H. Wang and G. Li, J. Mater. Chem. C, 2014, 2, 9342–9358.
20. B. Cho, T. Schwarz-Selinger, K. Ohmori and D. Cahill, Phys. Rev. B: Condens. Matter Mater. Phys., 2002, 66, 195407.
21. T. Krzyzewski and T. Jones, J. Appl. Phys., 2004, 96, 668–673.
22. G. Capellini, M. De Seta and F. Evangelisti, J. Appl. Phys., 2003, 93, 291–295.
23. A. Kurtz, S. Kulin and B. Averbach, J. Appl. Phys., 1956, 27, 1287–1290.
24. W. Wang, S. Zhou, Z. Liu, W. Yang, Y. Lin, H. Qian, F. Gao and G. Li, Mater. Res. Express, 2014, 1, 025903.
25. W. Wang, H. Yang and G. Li, J. Mater. Chem. C, 2013, 1, 4070–4077.
26. M. A. Moram and M. E. Vicker, Rep. Prog. Phys., 2009, 72, 036502.
27. W. Wang, Z. Liu, W. Yang, Y. Lin, S. Zhou, H. Qian, H. Wang, Z. Lin and G. Li, Mater. Lett., 2014, 129, 39–42.
28. H. Yang, W. Wang, Z. Liu and G. Li, CrystEngComm, 2013, 15, 7171–7176.
29. Y. H. Ra, R. Navamathavan, J. H. Park and C. R. Lee, CrystEngComm, 2013, 15, 1874–1881.
30. Y. H. Ra, R. Navamathavan and C. R. Lee, CrystEngComm, 2012, 14, 8208–8214.
31. W. Wang, W. Yang, Z. Liu, Y. Lin, S. Zhou, H. Qian, F. Gao, H. Yang and G. Li, Appl. Surf. Sci., 2014, 294, 1–8.

---