- Email: [email protected]
Effect of nitrogen flow rate on structural, morphological and optical properties of In-rich InxAl1−xN thin films grown by plasma-assisted dual source reactive evaporation
Applied Surface Science 378 (2016) 150–156
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Effect of nitrogen flow rate on structural, morphological and optical properties of In-rich Inx Al1−x N thin films grown by plasma-assisted dual source reactive evaporation M. Alizadeh a,∗ , V. Ganesh a , B.T. Goh a , C.F. Dee b , A.R. Mohmad b , S.A. Rahman a,∗ a b
Low Dimensional Materials Research Centre (LDMRC), Department of Physics, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia Institute of Microengineering and Nanoelectronics (IMEN), Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 24 December 2015 Received in revised form 26 February 2016 Accepted 24 March 2016 Keywords: Inx Al1−x N Plasma-assisted deposition Raman spectra Band gap
a b s t r a c t In-rich Inx Al1−x N thin films were deposited on quartz substrate at various nitrogen flow rates by plasmaassisted dual source reactive evaporation technique. The elemental composition, surface morphology, structural and optical properties of the films were investigated by X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FESEM), Raman spectroscopy, X-ray diffraction (XRD), UV–vis spectrophotometer and photoluminescence (PL) measurements. XPS results revealed that the indium composition (x) of the Inx Al1−x N films increases from 0.90 to 0.97 as the nitrogen flow rate is increased from 40 to 100 sccm, respectively. FESEM images of the surface and cross-sectional microstructure of the Inx Al1−x N films showed that by increasing the N2 flow rate, the grown particles are highly agglomerated. Raman and XRD results indicated that by increasing nitrogen flow rate the In-rich Inx Al1−x N films tend to turn into amorphous state. It was found that band gap energy of the films are in the range of 0.90–1.17 eV which is desirable for the application of full spectra solar cells. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Inx Al1−x N semiconductor has attracted so much attention for various applications such as distributed Bragg reflector, field effect transistors, facilitation layers in pseudomorphic epitaxy and solar cells [1–4]. The increasing interest is due to its tunable band gap varying remarkably from 0.7–6.2 eV depending on the amount of indium (or aluminum) constituent [5]. Therefore, varying the composition of Inx Al1−x N film makes it possible to customize the electronic structures of the alloy films for specific applications. For example, In-rich Inx Al1−x N thin films with tunable band gap of 0.7–2.4 eV can be used for multi-junction solar cell devices since this band gap (energy) range almost covers solar spectrum [6]. To date, there have been many reports on the growth of Inx Al1−x N alloys with different band gaps (Eg ) by metal organic chemical vapor deposition (MOCVD) [7,8], molecular beam epitaxy (MBE) [9–11] and sputtering-based deposition [12–14]. However, less works have been devoted to the effect of growth parameters on structural, morphological and optical properties of both In-rich
∗ Corresponding authors. E-mail addresses: alizadeh [email protected] (M. Alizadeh), [email protected] (S.A. Rahman). http://dx.doi.org/10.1016/j.apsusc.2016.03.174 0169-4332/© 2016 Elsevier B.V. All rights reserved.
and Al-rich Inx Al1−x N films. Investigation of Inx Al1−x N properties under different growth conditions is essential for understanding the defect mechanism and application of this trinary compound in the above-mentioned devices. This enables one to control the properties of the Inx Al1−x N films for desire purposes. On the other hand, there have been remarkable disagreements among the bang gap energies measured for Inx Al1−x N films grown using above-mentioned methods. The recent revision of InN bandgap energy from 1.9 eV to a much smaller value of 0.7 eV [5] is a possible reason for discrepancies among the reported results. Larger Eg values for InN (thus for In-rich Inx Al1−x N films) are usually obtained by sputtering [15–19] although the sputtered films are highly crystalline. Monteagudo-Lerma et al. [15] and ValduezaFelip et al. [16] obtained the Eg value of 1.7 eV from rf-sputtered InN films on AlN- and GaN-buffered sapphire substrate, respectively. Sasaoka et al. [17] deposited InN thin films by rf magnetron sputtering and the measured Eg value was approximately 1.4 eV. He et al. [18] reported the Eg value of 1.75 and 1.50 eV from highly crystalline rf-magnetron-sputtered Al0.18 In0.82 N and Al0.06 In0.94 N thin films, respectively. The optical measurement of the sputtered InN thin films reported by Cai et al. [19] showed an absorption edge at around 850 nm which may result in a larger band gap energy than 0.7 eV. These obtained large band gap energies limit the sputtered InN and In-rich Inx Al1−x N films for solar cell applications as
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
151
Fig. 1. In3d5/2 core-level photoelectron spectra of the In-rich Inx Al1−x N films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d). The inset of (a–d) are the corresponding Al2p core-level photoelectron spectra.
the films with narrow band gaps are highly desirable as the base materials in the solar cell devices. Plasma-assisted reactive evaporation method is employed here to deposit In-rich Inx Al1−x N films. Hot filaments are used for evaporation of aluminum and indium wires. In our previous work [20] it was shown that the rates of Al and In evaporation can be effectively controlled by varying an applied AC voltages to the filaments resulting in different Al(In) mole fractions in the obtained film structure. Moreover, the energized neutral species [21,22] along with electric-field-directed ion fluxes from the plasma ambient [23,24] contribute in local heating of top-most layers of the substrate, thus mobilizing the adsorbed adatoms and their incorporations in the structure of a growing film. Therefore, a controlled combination of reactive evaporation and plasma-based processing holds promise for deposition of Inx Al1−x N films. In this work, In-rich Inx Al1−x N thin films were successfully deposited by plasmaassisted dual source reactive evaporation technique and the effect of N2 flow rate on structural, optical and morphological properties of the grown films were investigated and discussed. 2. Experimental details In-rich Inx Al1−x N films were deposited on quartz substrate using a home-built plasma-assisted dual source reactive evaporation system. A schematic of the system can be seen in our previous work [20]. Prior to the deposition, the quartz substrate was ultrasonically degreased in decon 90 soap, rinsed in de-ionized water, ethanol and acetone and subsequently dried by nitrogen gas. Two separate tungsten wires were coiled and used as the hot filaments to evaporate aluminum and indium wires (with 99.999% purity). The filaments were clamped between the grounded substrate holder and a radio-frequency (RF) powered stainless steel electrode. In order to activate of the surface bonds (prior to the deposition), the
substrates were treated in H2 -plasma for 10 min at a fixed hydrogen flow rate of 100 sccm. The deposition process was conducted in three steps, as follows: (1) N2 plasma was generated in the chamber and stabilized for 5 min, (2) filament holding the Al wire (FAl ) was heated to the temperature required for evaporation of Al atoms (1650 ◦ C) and, then cooled down immediately to 1450 ◦ C to avoid high evaporation rates, (3) Immediately, filament holding the In wire (FIn ) was heated to around 1100 ◦ C and the Inx Al1−x N film growth was conducted at the evaporation of Al and In wires under N2 plasma (RF power: 250 W) for 5 min. The quartz substrates were initially heated to 200 ◦ C and during the deposition process the final (substrate holder) temperature was around 360 ◦ C. After the film deposition, the FIn was switched off, FAl was cooled down to 1300 ◦ C and the obtained samples were in situ annealed under N2 plasma (RF power: 50 W and N2 flow rate: 40 sccm) for 30 min to improve the quality of the films [18]. The N2 flow rate during the growth process was controlled as the growth parameter in this study. The deposition was carried out with different N2 flow rates of 40, 60, 80 and 100 sccm. The structural properties of the grown Inx Al1−x N films were investigated by Raman spectroscopy (Renishaw inVia Raman Microscope, with a laser excitation wavelength of 514 nm) and Xray diffraction (SIEMENS D5000 X-ray diffractometer, Cu K␣ X-ray ˚ The chemical compositions and bonding radiation = 1.54060 A). configuration of the samples were analyzed by X-ray photoelectron spectroscopy (XPS, PHI Quantera II). Field emission scanning electron microscopy (FESEM, JEOL JSM-7600F microscope) was employed to observe surface morphologies and cross-sectional microstructures of the as-gown films. The thickness of the films was measured using surface profiler (KLA-Tencor). The optical transmittance and reflectance for the grown Inx Al1−x N thin films was measured using an UV–vis spectrophotometer (Lambda 750, PerkinElmer). A 532-nm-line CW laser was used as an excitation
152
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
source, and photoluminescence (PL) signals were detected using an InGaAs photodiode array.
Table 1 The calculated compositional and optical values of the In-rich Inx Al1−x N films deposited at various nitrogen flow rates. N2 flow rate (sccm) XPS
3. Results and discussion XPS analysis was performed to investigate the Al (In) mole fraction in the as-prepared Inx Al1−x N films. Fig. 1(a)–(d) displays the In3d5/2 core-level photoelectron spectra of the films deposited at various N2 flow rates. The core-level photoelectron spectra of the Inx Al1−x N films were deconvoluted into three main sub-peaks centered at 445.4 ± 0.1, 444.3 and 443.6 eV which are attributed to In O, In N and In In bonds, respectively [25,26].The corresponding Al2p core-level photoelectron spectra of the films are presented in the inset of Fig. 1. The spectra were deconvoluted into two main components at 75.2 ± 0.2 and 74.1 eV which are assigned to Al O and Al N bonds, respectively [27,28]. In addition, Al2p core-level photoelectron spectra of the Inx Al1−x N samples deposited at N2 flow rate of 60 and 100 sccm exhibit a minor bump at the lower bonding energy side (73 eV) corresponding to Al Al bonding within the films [28]. The peak area of In N and Al N bonds as well as the estimated In composition (In mole fractions) in the Inx Al1−x N films (x = In N/(Al N + In N) [29]) are presented in Table 1. The analysis on the chemical composition of the Inx Al1−x N films showed that the x value increases from 0.90 to 0.97 as the nitrogen flow rate is increased from 40 to 100 sccm, respectively. The high values of x indicate that the Inx Al1−x N films are compositionally In-rich. The slight increase in indium composition of the Inx Al1−x N films may be attributed to gas phase reactions in the plasma and vicinity of the hot filament as well as to surface reactions at the growth sites on the substrates. By increasing the N2 flow rate, more excited atoms are generated in the plasma and in the vicinity of the hot filament. The increase in the N2 flow rate also results in the presence of more unreacted hot N2 molecules in the vicinity of the hot filament. More
40 60 80 100
Bandgap value (eV)
In-N peak area
Al-N peak area
In composition (x)
Optical absorption
PL
20607 27034 43817 50924
2347 2016 2168 1705
0.90 0.93 0.95 0.97
1.08 0.95 0.93 0.90
1.17 0.99 0.95 0.93
gas phase reactions occur due to collisions between the excited atoms and molecules. These reactions decrease the kinetic energy of ad-atoms/ions impinging on the growth surface of the film. As a result, lower mobility is imparted to the ad-atoms/ions at the growing surface of the thin films. The bonding energy of Al N bonds is much larger than that of In N [30]. Therefore, under a higher nitrogen flow rate ad-atoms/ions with lower energy are energetically favorable for the formation of In N bonds, rather than Al N resulting in higher In mole fraction of Inx Al1−x N alloys. Fig. 2(a)–(d) shows the N1s core-level photoelectron spectra of the Inx Al1−x N films grown at various N2 flow rates. The N1s core-level photoelectron spectra of all films were deconvoluted into two main sub-peaks centered at 396 ± 0.1 and 396.7 ± 0.1 eV which are assigned to N In [31] and N Al [27] bonds, respectively. The presence of N In and N Al components further confirms growth of Inx Al1−x N films. The spectra also exhibit a broad bump at the higher bonding energy side (398.5 ± 0.2 eV) corresponding to N Al O bond within the film [32]. This indicates that the unsatisfied bonds on AlN have higher affinity for O atoms compared to those on InN.
Fig. 2. N1s core-level photoelectron spectra of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
153
Fig. 3. FESEM images of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
Fig. 4. Cross-sectional images of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
Fig. 3 displays the FESEM images of the surface of the In-rich Inx Al1−x N films grown at different N2 flow rates. It is evident that surface morphology of the films strongly depends on the gas flow rate. The surface morphology of the Inx Al1−x N films grown at N2 = 40 sccm (Fig. 3(a)) exhibits clusters of near-spherical-shaped nanoparticles over the surface. The FESEM image of the Inx Al1−x N films deposited at N2 = 60 sccm (fig. 3(b)) shows that the nanoparticles tend to agglomerate with increase in N2 flow rate. Further increase in nitrogen flow rate enhances agglomeration of the nanoparticles and a few excessively large grains with the size of ∼0.5–1 m are formed on the surface of the Inx Al1−x N films grown at N2 = 80 sccm (fig. 3(c)). The FESEM image of the Inx Al1−x N sample prepared at nitrogen flow rate of 100 sccm (fig. 3(d)) shows cauliflower-shaped features with appearance of dark pits in between. To better understanding the growth mechanism of the Inx Al1−x N films at various N2 flow rates, the cross-sectional microstructure of the films are displayed in Fig. 4. It can be seen
that the thickness of the films is around 1.5–1.7 m. Cross-sectional image of the Inx Al1−x N films grown at the N2 flow rate of 40 sccm (fig. 4(a)) shows columnar growth of nanoparticles at the early stages of the deposition process. These nanoparticles are clustered at the topmost layers resulting in formation of grains of dimension of 200–300 nm. A dense and compact microstructure is clearly observed from the cross-sectional image of the films deposited at N2 = 60 sccm. The boundaries of coalesced grains (shown by arrow) appear on the microstructure indicating agglomeration tendency of the particles at the higher N2 flow rate. Columnar growth of islands of micrometers scale, perpendicular to the substrate, is obviously seen from the cross-sectional image of the Inx Al1−x N films grown at N2 = 80 sccm (fig. 4(c)). Fig. 4(d) revealed that combining of these micro-islands resulted in formation of cauliflower-shaped features and appearance of wide pits (shown by arrow) in the micro structure of the films deposited at the N2 flow rate of 100 sccm.
154
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
spectrum of Inx Al1−x N films deposited at N2 = 60 sccm, the A1 (TO) phonon mode is highly broadened whereas the A1 (LO) mode experiences a blue shift which is possibly due to the induced residual compressive stress as a result of increased nitrogen flow rate [35]. However, the position of this peak is also close to that of the phonon associated with the bcc-structured In2 O3 [36]. The E2 (high) and A1 (LO) phonon modes of the thin films deposited at N2 = 80 and 100 sccm are very broad showing that the structure of the In-rich Inx Al1−x N thin films tends to turn into an amorphous state at high nitrogen flow rate. This is probably due to congelation of disorder state of ad-atoms/ions under high N2 flow rates, as explained earlier. The XRD patterns of the In-rich Inx Al1−x N thin films are shown in Fig. 5(b). The XRD spectra of the Inx Al1−x N thin films grown at nitrogen flow rates of 40 and 60 sccm show peaks at 2 = 31.30◦ and 31.25◦ , respectively, which are assigned to (002) plane of wurtzite In0.89 Al0.11 N and In0.91 Al0.09 N based on Vegard’s law: c Inx Al1−x N = xcInN + (1 − x)c AlN
Fig. 5. Micro Raman spectra (a) and XRD patterns (b) of the In-rich Inx Al1−x N thin films.
The dependence of surface morphology of the Inx Al1−x N thin films on the nitrogen flow rate can be also explained in terms of plasma dynamics. As mentioned earlier, at a high N2 flow rate lower kinetic energy (and hence lower mobility) is imparted to the ad-atoms/ions on the growing surface of the thin film. It is well known that during growth process the crystalline particle must move within a certain distance on the substrate surface to find an optimal site which matches its crystal structure. At high N2 flow rate conditions, before these atoms or ions have the opportunity to get the optimal energy site, the congelation of their disorder state has a higher probability of occurring due to the low mobility of the adatoms on the surface. As a result, enhanced agglomeration of particles and cauliflower–like morphology are observed on the surface of the Inx Al1−x N thin films grown at N2 = 80 and 100, respectively. The structural results of the Inx Al1−x N thin films are shown in Fig. 5(a) and (b). Fig. 5(a) displays the Raman spectra of the samples deposited under various nitrogen flow rates. For the films deposited at nitrogen flow rate of 40 sccm, two major peaks centered at 502 and 582 cm−1 are observed which are assigned to E2 (high) and A1 (LO) phonon modes of the In0.90 Al0.10 N films, respectively [33]. Also, one broad peak at around 457 cm−1 is seen in the Raman spectra of the films. Since the Al content in the sample is very low it can be easily attributed to A1 (TO) mode of the sample originating from that mode of InN (at 445 cm−1 [34]). In the Raman
(1)
where cInN = 5.793 A˚ and cAlN = 5.012 A˚ are the c lattice constant of crystalline InN and AlN, respectively [37]. The In incorporation (x) values calculated from XPS analysis are very close to those obtained from Vegard’s law. The spectra of the films grown at N2 = 60 and 80 sccm show no diffraction peaks assigned to crystalline Inx Al1−x N which is in good agreement with the Raman results (fig. 5(b)). Phases of Al and In crystalline grains are observed from XRD patterns of the films grown at the nitrogen flow rates of 40, 60 and 100 sccm. The presence of these metallic phases was also deduced from the XPS spectra of the films and is probably due to higher concentration of reactive In and Al atoms reaching the growth sites compared to the reactive N atoms. The particle size of the Inx Al1−x N thin films grown at N2 = 40 and 60 sccm was measured using the well-known Scherrer formula [38] which are 12 and 33 nm, respectively. The calculated values are close to the average particle size estimated from FESEM results (fig. 3(a) and (b)). Optical properties of the In-rich Inx Al1−x N films deposited at various nitrogen flow rates were studied by obtaining transmittance and reflectance spectra. Fig. 6(a) displays the transmittance curves of the In-rich Inx Al1−x N films grown at various nitrogen flow rates. It can be clearly observed that the absorption edge of the all samples is in near infrared (NIR) region which confirms that the Inx Al1−x N films are compositionally In-rich. The wavelength dependence of optical reflectance spectra of the Inx Al1−x N films is shown in Fig. 6(b). The reflectance spectra of the thin films deposited at N2 = 40 and 60 sccm show a minimum at NIR region. It is well-known that minimum value of the reflection occurs in the vicinity of the wavelength related to the optical bandgap energy. Hence, the band gap value of the samples is expected to be in NIR region. Insignificant reflectance (e. g., R ∼ 0%) is seen for the thin films grown at N2 = 80 and 100 sccm which is mainly due to very rough surface of the films as shown in Fig. 3(c) and (d). The absorption coefficient (␣) was also calculated for each film with the measured transmittance T, reflection R and film thickness d from equation [39]:
˛=
ln (1 − R)2 /T d
(2)
It is known that the squared absorption coefficient of a semiconductor with a direct band gap is a function of photon energy:
˛2 = A h − Eg (h)
2
(3)
where A and Eg are proportional constant and band gap energy, respectively [40]. Direct band gap of the Inx Al1−x N films at various growth condition was estimated by drawing the tangential line for the (˛h)2 versus (h) as shown in (Fig. 6(c)). This method has been
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
155
Fig. 7. PL spectra at room temperature for the In-rich Inx Al1−x N thin films grown at different N2 flow rates.
PL spectra of the Inx Al1−x N films (fig. 7). The exact peak positions of the PL spectra are also shown in Table 1. The PL spectra of the Inrich Inx Al1−x N films exhibit infrared signals at 1.17–0.93 eV which almost coincide with the values calculated from the absorption spectra. It can be seen that both Eg values decrease by increasing the N2 flow rate. This reduction can be attributed to increase in indium composition of the films as shown in Fig. 1. However, it may be also related to the degraded crystallinity of the grown films at high nitrogen flow rates. High N2 flow rate-induced lattice damage creates defective energy level below the conduction band and as a result the band gap energy decreases [42]. The obtained values are around 0.20–0.47 eV higher than new band gap energy (0.70 eV) of InN due to the low concentration of Al incorporated within the film structure forming In-rich Inx Al1 x N. Such narrow band gap values are highly desirable for the application of full spectra solar cells. 4. Conclusion
Fig. 6. Transmittance (a) and Reflectance (b) spectra of the In-rich Inx Al1−x N thin films. (c): plots of versus h for the In-rich Inx Al1−x N thin films.
employed to measure band gap energy of amorphous films due to the lack of a linear portion in the (˛h)1/2 versus (h) plots [41]. The calculated band gap energy from the optical absorption spectra of the Inx Al1−x N films deposited at various nitrogen flow ratios are summarized in Table 1. The band gap value of the films grown at nitrogen flow rate of 40 sccm is estimated to be 1.08 eV. For the sample grown at N2 = 60, 80 and 100 the Eg value shifts to 0.95, 0.93 and 0.90 eV, respectively. The Eg values were also obtained from the
The effect of nitrogen flow rate on structural, morphological and optical properties of In-rich Inx Al1−x N thin films grown by plasma-assisted dual source reactive evaporation was investigated. It was shown that In incorporation (x) of the Inx Al1−x N thin films increases from 0.90 to 0.97 as the N2 flow rate is increased from 40 to 100 sccm, respectively. FESEM images of the surface and crosssectional microstructure of the Inx Al1−x N films showed that by increasing the N2 flow rate, the grown particles are highly agglomerated. Raman results indicated that by increasing nitrogen flow rate the In-rich Inx Al1−x N films tend to turn into amorphous state and E2 (high) and A1 (LO) peaks of the films are broadened at nitrogen flow rate of 80 and 100 sccm. Optical characterizations showed that the band gap energy of the films are in the range of 0.90-1.17 eV which is desirable for the application of full spectra solar cells. Acknowledgment This work was supported by University of Malaya, High Impact Research (HIR) fund of UM.C/HIR/MOHE/SC/06, the Ministry of Higher Education Fundamental Research Grant Scheme (FRGS) of FP009-2013B and University of Malaya postgraduate research Fund (PPP) of PG081-2012B.
156
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
References [1] I. Saidi, H. Mejri, M. Baira, H. Maaref, Superlattices Microstruct. 84 (2015) 113. [2] B. Monemar, P.P. Paskov, A. Kasic, Superlattices Microstruct. 38 (2005) 38. [3] I.M. Watson, C. Liu, E. Gu, M.D. Dawson, P.R. Edwards, R.W. Martin, Appl. Phys. Lett. 87 (2005) 151901. [4] C. Berger, A. Dadgar, J. Blasing, A. Krost, J. Cryst. Growth 370 (2013) 87. [5] V.Y. Davydov, A.A. Klochikhin, R.P. Seisyan, V.V. Emtsev, S.V. Ivanow, F. Bechstedt, J. Furthmuller, H. Harima, A.V. Mudryi, J. Aderhold, O. Semchinova, J. Graul, Phys. Status Solidi B 229 (2002) r1. [6] A. Yamamoto, Md.R. Islam, T.-T. Kang, A. Hashimoto, Phys. Status Solidi C 7 (2010) 1309. [7] K.S. Kim, A. Saxler, P. Kung, M. Razeghi, Appl. Phys. Lett. 71 (1997) 800. [8] K. Wang, R.W. Martin, D. Amabile, P.R. Edwards, S. Hernandez, E. Nogales, K. O’Donnell, K. Lorenz, E. Alves, V. Matias, A. Vantomme, D. Wolverson, I.M. Watson, J. Appl. Phys. 103 (2008) 073510. [9] M.J. Lukitsch, Y.V. Danylyuk, V.M. Naik, C. Huang, G.W. Auner, L. Rimai, R. Naik, Appl. Phys. Lett. 79 (2001) 632. [10] W. Terashima, S.B. Che, Y. Ishitani, A. Yoshikawa, Jpn. J. Appl. Phys. 45 (2006) L539. [11] W. Walukiewicz, S.X. Li, J. Wu, K.M. Yu, J.W. Ager, E.E. Haller, H. Lu, W.J. Schaff, J. Cryst. Growth 269 (2004) 119. [12] Q.X. Guo, T. Tanaka, M. Nishio, H. Ogawa, Jpn. J. Appl. Phys. 42 (2003) L141. [13] K. Kubota, Y. Kobayashi, K. Fujimoto, J. Appl. Phys. 66 (1989) 2984. [14] T.S. Yeh, J.M. Wu, W.H. Lan, Thin Solid Films 517 (2009) 3204. ˜ [15] L. Monteagudo-Lerma, S. Valdueza-Felip, A. Núnez-Cascajero, A. Ruiz, M. González-Herráez, E. Monroy, F.B. Naranjo, J. Cryst. Growth 434 (2016) 13. ˜ [16] S. Valdueza-Felip, J. Ibánez, E. Monroy, M. González-Herráez, L. Artús, F.B. Naranjo, Thin Solid Films 520 (2012) 2805. [17] T. Sasaoka, M. Mori, T. Miyazaki, S. Adachi, J. Appl. Phys. 108 (2010) 063538. [18] H. He, Y. Cao, R. Fu, W. Guo, Z. Huang, M. Wanga, C. Huang, J. Huang, H. Wang, Appl. Surf. Sci. 256 (2010) 1812. [19] X.-M. Cai, F. Ye, Y.-Q. Hao, D.-P. Zhang, Z.-H. Zhang, P. Fan, J. Alloys Compd. 484 (2009) 677. [20] M. Alizadeh, V. Ganesh, H. Mehdipour, N.F.F. Nazarudin, B.T. Goh, A. Shuhaimi, S.A. Rahman, J. Alloys Compd. 632 (2015) 741. [21] N.P. Tandian, E. Pfender, Plasma Chem. Plasma Proc. 17 (1997) 353.
[22] H. Kerstena, H. Deutscha, H. Steffena, G.M.W. Kroesenb, R. Hippler, Vacuum 63 (2001) 385. [23] K. Ostrikov, I. Levchenko, S. Xu, Pure Appl. Chem. 80 (2008) 1909. [24] M. Alizadeh, H. Mehdipour, B.T. Goh, S.A. Rahman, J. Appl. Phys. 114 (2013) 02430. [25] Z.X. Bi, R. Zhang, Z.L. Xie, X.Q. Xiu, Y.D. Ye, B. Liu, S.L. Gu, B. Shen, Y. Shi, Y.D. Zheng, Thin Solid Films 488 (2005) 111. [26] V. Lebedev, F.M. Morales, V. Cimalla, J.G. Lozano, D. Gonz´ıalez, M. Himmerlich, S. Krischok, J.A. Schaefer, O. Ambacher, Superlattices Microstruct. 40 (2006) 289. [27] Xu-Ping Kuang, Hua-Yu Zhang, Gui-Gen Wang, Lin Cui, Can Zhu, Lei Jin, Rui Sun, Jie-Cai Han, Appl. Surf. Sci. 263 (2012) 62. [28] M. Alizadeh, H. Mehdipour, V. Ganesh, A.N. Ameera, B.T. Goh, A. Shuhaimi, S.A. Rahman, Appl. Phys. A 117 (2014) 2217. [29] T. Suzuki, S. Hibino, R. Katayama, Y. Kato, F. Ohashi, T. Itoh, S. Nonomura, Jpn. J. Appl. Phys. 52 (2013) 11NG05. [30] A. Costales, M.A. Blanco, A. Martın Pendas, A.K. Kandalam, R. Pandey, J. Am. Chem. Soc. 124 (2002) 4116. [31] Qiang Jing, Hang Yang, Wancheng Li, Guoguang Wu, Yuantao Zhang, Fubin Gao, Yang Zhao, Guotong Du, Appl. Surf. Sci. 331 (2015) 248. [32] L. Rosenberger, R. Baird, E. McCullen, G. Auner, G. Shreve, Surf. Interface Anal. 40 (2008) 1254. [33] T.-T. Kang, A. Hashimoto, A. Yamamoto, Phys. Rev. B 79 (2009) 033301. [34] J.B. Wang, Z.F. Li, P.P. Chen, Wei Lu, T. Yao, Acta Mater. 55 (2007) 183. [35] M. Watanabe, C. Brauns, M. Komatsu, S. Kuroda, F. Gärtner, T. Klassen, H. Katanoda, Surf. Coat. Technol. 232 (2013) 587. [36] Y. Chen, X. Zhou, X. Zhao, X. He, X. Gu, Mater. Sci. Eng. B 151 (2008) 179. [37] V. Darakchieva, M.Y. Xie, F. Tasnádi, I.A. Abrikosov, L. Hultman, B. Monemar, J. Kamimura, K. Kishino, Appl. Phys. Lett. 93 (2008) 261908. [38] S.S. Lin, J.L. Huang, P. Sajgalik, Surf. Coat. Technol. 190 (2005) 39. [39] M. Di Giulio, G. Micocci, R.P. Rella Siliciano, A. Tepore, Thin Solid Films 148 (1987) 273. [40] T. Peng, J. Piprek, G. Qiu, J.O. Olowolafe, K.M. Unruh, C.P. Swann, E.F. Schubert, Appl. Phys. Lett. 71 (1997) 2439. [41] R. Weingartner, J.A. Guerra Torres, O. Erlenbach, G. Galvez de la Puente, F. De Zela, A. Winnacker, Mater. Sci. Eng. B—Solid 174 (2010) 114. [42] A. Shah, Arshad Mahmood, Physica B 407 (2012) 3987.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Effect of nitrogen flow rate on structural, morphological and optical properties of In-rich Inx Al1−x N thin films grown by plasma-assisted dual source reactive evaporation M. Alizadeh a,∗ , V. Ganesh a , B.T. Goh a , C.F. Dee b , A.R. Mohmad b , S.A. Rahman a,∗ a b
Low Dimensional Materials Research Centre (LDMRC), Department of Physics, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia Institute of Microengineering and Nanoelectronics (IMEN), Universiti Kebangsaan Malaysia, Bangi, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 24 December 2015 Received in revised form 26 February 2016 Accepted 24 March 2016 Keywords: Inx Al1−x N Plasma-assisted deposition Raman spectra Band gap
a b s t r a c t In-rich Inx Al1−x N thin films were deposited on quartz substrate at various nitrogen flow rates by plasmaassisted dual source reactive evaporation technique. The elemental composition, surface morphology, structural and optical properties of the films were investigated by X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FESEM), Raman spectroscopy, X-ray diffraction (XRD), UV–vis spectrophotometer and photoluminescence (PL) measurements. XPS results revealed that the indium composition (x) of the Inx Al1−x N films increases from 0.90 to 0.97 as the nitrogen flow rate is increased from 40 to 100 sccm, respectively. FESEM images of the surface and cross-sectional microstructure of the Inx Al1−x N films showed that by increasing the N2 flow rate, the grown particles are highly agglomerated. Raman and XRD results indicated that by increasing nitrogen flow rate the In-rich Inx Al1−x N films tend to turn into amorphous state. It was found that band gap energy of the films are in the range of 0.90–1.17 eV which is desirable for the application of full spectra solar cells. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Inx Al1−x N semiconductor has attracted so much attention for various applications such as distributed Bragg reflector, field effect transistors, facilitation layers in pseudomorphic epitaxy and solar cells [1–4]. The increasing interest is due to its tunable band gap varying remarkably from 0.7–6.2 eV depending on the amount of indium (or aluminum) constituent [5]. Therefore, varying the composition of Inx Al1−x N film makes it possible to customize the electronic structures of the alloy films for specific applications. For example, In-rich Inx Al1−x N thin films with tunable band gap of 0.7–2.4 eV can be used for multi-junction solar cell devices since this band gap (energy) range almost covers solar spectrum [6]. To date, there have been many reports on the growth of Inx Al1−x N alloys with different band gaps (Eg ) by metal organic chemical vapor deposition (MOCVD) [7,8], molecular beam epitaxy (MBE) [9–11] and sputtering-based deposition [12–14]. However, less works have been devoted to the effect of growth parameters on structural, morphological and optical properties of both In-rich
∗ Corresponding authors. E-mail addresses: alizadeh [email protected] (M. Alizadeh), [email protected] (S.A. Rahman). http://dx.doi.org/10.1016/j.apsusc.2016.03.174 0169-4332/© 2016 Elsevier B.V. All rights reserved.
and Al-rich Inx Al1−x N films. Investigation of Inx Al1−x N properties under different growth conditions is essential for understanding the defect mechanism and application of this trinary compound in the above-mentioned devices. This enables one to control the properties of the Inx Al1−x N films for desire purposes. On the other hand, there have been remarkable disagreements among the bang gap energies measured for Inx Al1−x N films grown using above-mentioned methods. The recent revision of InN bandgap energy from 1.9 eV to a much smaller value of 0.7 eV [5] is a possible reason for discrepancies among the reported results. Larger Eg values for InN (thus for In-rich Inx Al1−x N films) are usually obtained by sputtering [15–19] although the sputtered films are highly crystalline. Monteagudo-Lerma et al. [15] and ValduezaFelip et al. [16] obtained the Eg value of 1.7 eV from rf-sputtered InN films on AlN- and GaN-buffered sapphire substrate, respectively. Sasaoka et al. [17] deposited InN thin films by rf magnetron sputtering and the measured Eg value was approximately 1.4 eV. He et al. [18] reported the Eg value of 1.75 and 1.50 eV from highly crystalline rf-magnetron-sputtered Al0.18 In0.82 N and Al0.06 In0.94 N thin films, respectively. The optical measurement of the sputtered InN thin films reported by Cai et al. [19] showed an absorption edge at around 850 nm which may result in a larger band gap energy than 0.7 eV. These obtained large band gap energies limit the sputtered InN and In-rich Inx Al1−x N films for solar cell applications as
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
151
Fig. 1. In3d5/2 core-level photoelectron spectra of the In-rich Inx Al1−x N films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d). The inset of (a–d) are the corresponding Al2p core-level photoelectron spectra.
the films with narrow band gaps are highly desirable as the base materials in the solar cell devices. Plasma-assisted reactive evaporation method is employed here to deposit In-rich Inx Al1−x N films. Hot filaments are used for evaporation of aluminum and indium wires. In our previous work [20] it was shown that the rates of Al and In evaporation can be effectively controlled by varying an applied AC voltages to the filaments resulting in different Al(In) mole fractions in the obtained film structure. Moreover, the energized neutral species [21,22] along with electric-field-directed ion fluxes from the plasma ambient [23,24] contribute in local heating of top-most layers of the substrate, thus mobilizing the adsorbed adatoms and their incorporations in the structure of a growing film. Therefore, a controlled combination of reactive evaporation and plasma-based processing holds promise for deposition of Inx Al1−x N films. In this work, In-rich Inx Al1−x N thin films were successfully deposited by plasmaassisted dual source reactive evaporation technique and the effect of N2 flow rate on structural, optical and morphological properties of the grown films were investigated and discussed. 2. Experimental details In-rich Inx Al1−x N films were deposited on quartz substrate using a home-built plasma-assisted dual source reactive evaporation system. A schematic of the system can be seen in our previous work [20]. Prior to the deposition, the quartz substrate was ultrasonically degreased in decon 90 soap, rinsed in de-ionized water, ethanol and acetone and subsequently dried by nitrogen gas. Two separate tungsten wires were coiled and used as the hot filaments to evaporate aluminum and indium wires (with 99.999% purity). The filaments were clamped between the grounded substrate holder and a radio-frequency (RF) powered stainless steel electrode. In order to activate of the surface bonds (prior to the deposition), the
substrates were treated in H2 -plasma for 10 min at a fixed hydrogen flow rate of 100 sccm. The deposition process was conducted in three steps, as follows: (1) N2 plasma was generated in the chamber and stabilized for 5 min, (2) filament holding the Al wire (FAl ) was heated to the temperature required for evaporation of Al atoms (1650 ◦ C) and, then cooled down immediately to 1450 ◦ C to avoid high evaporation rates, (3) Immediately, filament holding the In wire (FIn ) was heated to around 1100 ◦ C and the Inx Al1−x N film growth was conducted at the evaporation of Al and In wires under N2 plasma (RF power: 250 W) for 5 min. The quartz substrates were initially heated to 200 ◦ C and during the deposition process the final (substrate holder) temperature was around 360 ◦ C. After the film deposition, the FIn was switched off, FAl was cooled down to 1300 ◦ C and the obtained samples were in situ annealed under N2 plasma (RF power: 50 W and N2 flow rate: 40 sccm) for 30 min to improve the quality of the films [18]. The N2 flow rate during the growth process was controlled as the growth parameter in this study. The deposition was carried out with different N2 flow rates of 40, 60, 80 and 100 sccm. The structural properties of the grown Inx Al1−x N films were investigated by Raman spectroscopy (Renishaw inVia Raman Microscope, with a laser excitation wavelength of 514 nm) and Xray diffraction (SIEMENS D5000 X-ray diffractometer, Cu K␣ X-ray ˚ The chemical compositions and bonding radiation = 1.54060 A). configuration of the samples were analyzed by X-ray photoelectron spectroscopy (XPS, PHI Quantera II). Field emission scanning electron microscopy (FESEM, JEOL JSM-7600F microscope) was employed to observe surface morphologies and cross-sectional microstructures of the as-gown films. The thickness of the films was measured using surface profiler (KLA-Tencor). The optical transmittance and reflectance for the grown Inx Al1−x N thin films was measured using an UV–vis spectrophotometer (Lambda 750, PerkinElmer). A 532-nm-line CW laser was used as an excitation
152
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
source, and photoluminescence (PL) signals were detected using an InGaAs photodiode array.
Table 1 The calculated compositional and optical values of the In-rich Inx Al1−x N films deposited at various nitrogen flow rates. N2 flow rate (sccm) XPS
3. Results and discussion XPS analysis was performed to investigate the Al (In) mole fraction in the as-prepared Inx Al1−x N films. Fig. 1(a)–(d) displays the In3d5/2 core-level photoelectron spectra of the films deposited at various N2 flow rates. The core-level photoelectron spectra of the Inx Al1−x N films were deconvoluted into three main sub-peaks centered at 445.4 ± 0.1, 444.3 and 443.6 eV which are attributed to In O, In N and In In bonds, respectively [25,26].The corresponding Al2p core-level photoelectron spectra of the films are presented in the inset of Fig. 1. The spectra were deconvoluted into two main components at 75.2 ± 0.2 and 74.1 eV which are assigned to Al O and Al N bonds, respectively [27,28]. In addition, Al2p core-level photoelectron spectra of the Inx Al1−x N samples deposited at N2 flow rate of 60 and 100 sccm exhibit a minor bump at the lower bonding energy side (73 eV) corresponding to Al Al bonding within the films [28]. The peak area of In N and Al N bonds as well as the estimated In composition (In mole fractions) in the Inx Al1−x N films (x = In N/(Al N + In N) [29]) are presented in Table 1. The analysis on the chemical composition of the Inx Al1−x N films showed that the x value increases from 0.90 to 0.97 as the nitrogen flow rate is increased from 40 to 100 sccm, respectively. The high values of x indicate that the Inx Al1−x N films are compositionally In-rich. The slight increase in indium composition of the Inx Al1−x N films may be attributed to gas phase reactions in the plasma and vicinity of the hot filament as well as to surface reactions at the growth sites on the substrates. By increasing the N2 flow rate, more excited atoms are generated in the plasma and in the vicinity of the hot filament. The increase in the N2 flow rate also results in the presence of more unreacted hot N2 molecules in the vicinity of the hot filament. More
40 60 80 100
Bandgap value (eV)
In-N peak area
Al-N peak area
In composition (x)
Optical absorption
PL
20607 27034 43817 50924
2347 2016 2168 1705
0.90 0.93 0.95 0.97
1.08 0.95 0.93 0.90
1.17 0.99 0.95 0.93
gas phase reactions occur due to collisions between the excited atoms and molecules. These reactions decrease the kinetic energy of ad-atoms/ions impinging on the growth surface of the film. As a result, lower mobility is imparted to the ad-atoms/ions at the growing surface of the thin films. The bonding energy of Al N bonds is much larger than that of In N [30]. Therefore, under a higher nitrogen flow rate ad-atoms/ions with lower energy are energetically favorable for the formation of In N bonds, rather than Al N resulting in higher In mole fraction of Inx Al1−x N alloys. Fig. 2(a)–(d) shows the N1s core-level photoelectron spectra of the Inx Al1−x N films grown at various N2 flow rates. The N1s core-level photoelectron spectra of all films were deconvoluted into two main sub-peaks centered at 396 ± 0.1 and 396.7 ± 0.1 eV which are assigned to N In [31] and N Al [27] bonds, respectively. The presence of N In and N Al components further confirms growth of Inx Al1−x N films. The spectra also exhibit a broad bump at the higher bonding energy side (398.5 ± 0.2 eV) corresponding to N Al O bond within the film [32]. This indicates that the unsatisfied bonds on AlN have higher affinity for O atoms compared to those on InN.
Fig. 2. N1s core-level photoelectron spectra of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
153
Fig. 3. FESEM images of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
Fig. 4. Cross-sectional images of the In-rich Inx Al1−x N thin films deposited at N2 flow rate of 40 sccm (a), 60 sccm (b), 80 sccm (c) and 100 sccm (d).
Fig. 3 displays the FESEM images of the surface of the In-rich Inx Al1−x N films grown at different N2 flow rates. It is evident that surface morphology of the films strongly depends on the gas flow rate. The surface morphology of the Inx Al1−x N films grown at N2 = 40 sccm (Fig. 3(a)) exhibits clusters of near-spherical-shaped nanoparticles over the surface. The FESEM image of the Inx Al1−x N films deposited at N2 = 60 sccm (fig. 3(b)) shows that the nanoparticles tend to agglomerate with increase in N2 flow rate. Further increase in nitrogen flow rate enhances agglomeration of the nanoparticles and a few excessively large grains with the size of ∼0.5–1 m are formed on the surface of the Inx Al1−x N films grown at N2 = 80 sccm (fig. 3(c)). The FESEM image of the Inx Al1−x N sample prepared at nitrogen flow rate of 100 sccm (fig. 3(d)) shows cauliflower-shaped features with appearance of dark pits in between. To better understanding the growth mechanism of the Inx Al1−x N films at various N2 flow rates, the cross-sectional microstructure of the films are displayed in Fig. 4. It can be seen
that the thickness of the films is around 1.5–1.7 m. Cross-sectional image of the Inx Al1−x N films grown at the N2 flow rate of 40 sccm (fig. 4(a)) shows columnar growth of nanoparticles at the early stages of the deposition process. These nanoparticles are clustered at the topmost layers resulting in formation of grains of dimension of 200–300 nm. A dense and compact microstructure is clearly observed from the cross-sectional image of the films deposited at N2 = 60 sccm. The boundaries of coalesced grains (shown by arrow) appear on the microstructure indicating agglomeration tendency of the particles at the higher N2 flow rate. Columnar growth of islands of micrometers scale, perpendicular to the substrate, is obviously seen from the cross-sectional image of the Inx Al1−x N films grown at N2 = 80 sccm (fig. 4(c)). Fig. 4(d) revealed that combining of these micro-islands resulted in formation of cauliflower-shaped features and appearance of wide pits (shown by arrow) in the micro structure of the films deposited at the N2 flow rate of 100 sccm.
154
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
spectrum of Inx Al1−x N films deposited at N2 = 60 sccm, the A1 (TO) phonon mode is highly broadened whereas the A1 (LO) mode experiences a blue shift which is possibly due to the induced residual compressive stress as a result of increased nitrogen flow rate [35]. However, the position of this peak is also close to that of the phonon associated with the bcc-structured In2 O3 [36]. The E2 (high) and A1 (LO) phonon modes of the thin films deposited at N2 = 80 and 100 sccm are very broad showing that the structure of the In-rich Inx Al1−x N thin films tends to turn into an amorphous state at high nitrogen flow rate. This is probably due to congelation of disorder state of ad-atoms/ions under high N2 flow rates, as explained earlier. The XRD patterns of the In-rich Inx Al1−x N thin films are shown in Fig. 5(b). The XRD spectra of the Inx Al1−x N thin films grown at nitrogen flow rates of 40 and 60 sccm show peaks at 2 = 31.30◦ and 31.25◦ , respectively, which are assigned to (002) plane of wurtzite In0.89 Al0.11 N and In0.91 Al0.09 N based on Vegard’s law: c Inx Al1−x N = xcInN + (1 − x)c AlN
Fig. 5. Micro Raman spectra (a) and XRD patterns (b) of the In-rich Inx Al1−x N thin films.
The dependence of surface morphology of the Inx Al1−x N thin films on the nitrogen flow rate can be also explained in terms of plasma dynamics. As mentioned earlier, at a high N2 flow rate lower kinetic energy (and hence lower mobility) is imparted to the ad-atoms/ions on the growing surface of the thin film. It is well known that during growth process the crystalline particle must move within a certain distance on the substrate surface to find an optimal site which matches its crystal structure. At high N2 flow rate conditions, before these atoms or ions have the opportunity to get the optimal energy site, the congelation of their disorder state has a higher probability of occurring due to the low mobility of the adatoms on the surface. As a result, enhanced agglomeration of particles and cauliflower–like morphology are observed on the surface of the Inx Al1−x N thin films grown at N2 = 80 and 100, respectively. The structural results of the Inx Al1−x N thin films are shown in Fig. 5(a) and (b). Fig. 5(a) displays the Raman spectra of the samples deposited under various nitrogen flow rates. For the films deposited at nitrogen flow rate of 40 sccm, two major peaks centered at 502 and 582 cm−1 are observed which are assigned to E2 (high) and A1 (LO) phonon modes of the In0.90 Al0.10 N films, respectively [33]. Also, one broad peak at around 457 cm−1 is seen in the Raman spectra of the films. Since the Al content in the sample is very low it can be easily attributed to A1 (TO) mode of the sample originating from that mode of InN (at 445 cm−1 [34]). In the Raman
(1)
where cInN = 5.793 A˚ and cAlN = 5.012 A˚ are the c lattice constant of crystalline InN and AlN, respectively [37]. The In incorporation (x) values calculated from XPS analysis are very close to those obtained from Vegard’s law. The spectra of the films grown at N2 = 60 and 80 sccm show no diffraction peaks assigned to crystalline Inx Al1−x N which is in good agreement with the Raman results (fig. 5(b)). Phases of Al and In crystalline grains are observed from XRD patterns of the films grown at the nitrogen flow rates of 40, 60 and 100 sccm. The presence of these metallic phases was also deduced from the XPS spectra of the films and is probably due to higher concentration of reactive In and Al atoms reaching the growth sites compared to the reactive N atoms. The particle size of the Inx Al1−x N thin films grown at N2 = 40 and 60 sccm was measured using the well-known Scherrer formula [38] which are 12 and 33 nm, respectively. The calculated values are close to the average particle size estimated from FESEM results (fig. 3(a) and (b)). Optical properties of the In-rich Inx Al1−x N films deposited at various nitrogen flow rates were studied by obtaining transmittance and reflectance spectra. Fig. 6(a) displays the transmittance curves of the In-rich Inx Al1−x N films grown at various nitrogen flow rates. It can be clearly observed that the absorption edge of the all samples is in near infrared (NIR) region which confirms that the Inx Al1−x N films are compositionally In-rich. The wavelength dependence of optical reflectance spectra of the Inx Al1−x N films is shown in Fig. 6(b). The reflectance spectra of the thin films deposited at N2 = 40 and 60 sccm show a minimum at NIR region. It is well-known that minimum value of the reflection occurs in the vicinity of the wavelength related to the optical bandgap energy. Hence, the band gap value of the samples is expected to be in NIR region. Insignificant reflectance (e. g., R ∼ 0%) is seen for the thin films grown at N2 = 80 and 100 sccm which is mainly due to very rough surface of the films as shown in Fig. 3(c) and (d). The absorption coefficient (␣) was also calculated for each film with the measured transmittance T, reflection R and film thickness d from equation [39]:
˛=
ln (1 − R)2 /T d
(2)
It is known that the squared absorption coefficient of a semiconductor with a direct band gap is a function of photon energy:
˛2 = A h − Eg (h)
2
(3)
where A and Eg are proportional constant and band gap energy, respectively [40]. Direct band gap of the Inx Al1−x N films at various growth condition was estimated by drawing the tangential line for the (˛h)2 versus (h) as shown in (Fig. 6(c)). This method has been
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
155
Fig. 7. PL spectra at room temperature for the In-rich Inx Al1−x N thin films grown at different N2 flow rates.
PL spectra of the Inx Al1−x N films (fig. 7). The exact peak positions of the PL spectra are also shown in Table 1. The PL spectra of the Inrich Inx Al1−x N films exhibit infrared signals at 1.17–0.93 eV which almost coincide with the values calculated from the absorption spectra. It can be seen that both Eg values decrease by increasing the N2 flow rate. This reduction can be attributed to increase in indium composition of the films as shown in Fig. 1. However, it may be also related to the degraded crystallinity of the grown films at high nitrogen flow rates. High N2 flow rate-induced lattice damage creates defective energy level below the conduction band and as a result the band gap energy decreases [42]. The obtained values are around 0.20–0.47 eV higher than new band gap energy (0.70 eV) of InN due to the low concentration of Al incorporated within the film structure forming In-rich Inx Al1 x N. Such narrow band gap values are highly desirable for the application of full spectra solar cells. 4. Conclusion
Fig. 6. Transmittance (a) and Reflectance (b) spectra of the In-rich Inx Al1−x N thin films. (c): plots of versus h for the In-rich Inx Al1−x N thin films.
employed to measure band gap energy of amorphous films due to the lack of a linear portion in the (˛h)1/2 versus (h) plots [41]. The calculated band gap energy from the optical absorption spectra of the Inx Al1−x N films deposited at various nitrogen flow ratios are summarized in Table 1. The band gap value of the films grown at nitrogen flow rate of 40 sccm is estimated to be 1.08 eV. For the sample grown at N2 = 60, 80 and 100 the Eg value shifts to 0.95, 0.93 and 0.90 eV, respectively. The Eg values were also obtained from the
The effect of nitrogen flow rate on structural, morphological and optical properties of In-rich Inx Al1−x N thin films grown by plasma-assisted dual source reactive evaporation was investigated. It was shown that In incorporation (x) of the Inx Al1−x N thin films increases from 0.90 to 0.97 as the N2 flow rate is increased from 40 to 100 sccm, respectively. FESEM images of the surface and crosssectional microstructure of the Inx Al1−x N films showed that by increasing the N2 flow rate, the grown particles are highly agglomerated. Raman results indicated that by increasing nitrogen flow rate the In-rich Inx Al1−x N films tend to turn into amorphous state and E2 (high) and A1 (LO) peaks of the films are broadened at nitrogen flow rate of 80 and 100 sccm. Optical characterizations showed that the band gap energy of the films are in the range of 0.90-1.17 eV which is desirable for the application of full spectra solar cells. Acknowledgment This work was supported by University of Malaya, High Impact Research (HIR) fund of UM.C/HIR/MOHE/SC/06, the Ministry of Higher Education Fundamental Research Grant Scheme (FRGS) of FP009-2013B and University of Malaya postgraduate research Fund (PPP) of PG081-2012B.
156
M. Alizadeh et al. / Applied Surface Science 378 (2016) 150–156
References [1] I. Saidi, H. Mejri, M. Baira, H. Maaref, Superlattices Microstruct. 84 (2015) 113. [2] B. Monemar, P.P. Paskov, A. Kasic, Superlattices Microstruct. 38 (2005) 38. [3] I.M. Watson, C. Liu, E. Gu, M.D. Dawson, P.R. Edwards, R.W. Martin, Appl. Phys. Lett. 87 (2005) 151901. [4] C. Berger, A. Dadgar, J. Blasing, A. Krost, J. Cryst. Growth 370 (2013) 87. [5] V.Y. Davydov, A.A. Klochikhin, R.P. Seisyan, V.V. Emtsev, S.V. Ivanow, F. Bechstedt, J. Furthmuller, H. Harima, A.V. Mudryi, J. Aderhold, O. Semchinova, J. Graul, Phys. Status Solidi B 229 (2002) r1. [6] A. Yamamoto, Md.R. Islam, T.-T. Kang, A. Hashimoto, Phys. Status Solidi C 7 (2010) 1309. [7] K.S. Kim, A. Saxler, P. Kung, M. Razeghi, Appl. Phys. Lett. 71 (1997) 800. [8] K. Wang, R.W. Martin, D. Amabile, P.R. Edwards, S. Hernandez, E. Nogales, K. O’Donnell, K. Lorenz, E. Alves, V. Matias, A. Vantomme, D. Wolverson, I.M. Watson, J. Appl. Phys. 103 (2008) 073510. [9] M.J. Lukitsch, Y.V. Danylyuk, V.M. Naik, C. Huang, G.W. Auner, L. Rimai, R. Naik, Appl. Phys. Lett. 79 (2001) 632. [10] W. Terashima, S.B. Che, Y. Ishitani, A. Yoshikawa, Jpn. J. Appl. Phys. 45 (2006) L539. [11] W. Walukiewicz, S.X. Li, J. Wu, K.M. Yu, J.W. Ager, E.E. Haller, H. Lu, W.J. Schaff, J. Cryst. Growth 269 (2004) 119. [12] Q.X. Guo, T. Tanaka, M. Nishio, H. Ogawa, Jpn. J. Appl. Phys. 42 (2003) L141. [13] K. Kubota, Y. Kobayashi, K. Fujimoto, J. Appl. Phys. 66 (1989) 2984. [14] T.S. Yeh, J.M. Wu, W.H. Lan, Thin Solid Films 517 (2009) 3204. ˜ [15] L. Monteagudo-Lerma, S. Valdueza-Felip, A. Núnez-Cascajero, A. Ruiz, M. González-Herráez, E. Monroy, F.B. Naranjo, J. Cryst. Growth 434 (2016) 13. ˜ [16] S. Valdueza-Felip, J. Ibánez, E. Monroy, M. González-Herráez, L. Artús, F.B. Naranjo, Thin Solid Films 520 (2012) 2805. [17] T. Sasaoka, M. Mori, T. Miyazaki, S. Adachi, J. Appl. Phys. 108 (2010) 063538. [18] H. He, Y. Cao, R. Fu, W. Guo, Z. Huang, M. Wanga, C. Huang, J. Huang, H. Wang, Appl. Surf. Sci. 256 (2010) 1812. [19] X.-M. Cai, F. Ye, Y.-Q. Hao, D.-P. Zhang, Z.-H. Zhang, P. Fan, J. Alloys Compd. 484 (2009) 677. [20] M. Alizadeh, V. Ganesh, H. Mehdipour, N.F.F. Nazarudin, B.T. Goh, A. Shuhaimi, S.A. Rahman, J. Alloys Compd. 632 (2015) 741. [21] N.P. Tandian, E. Pfender, Plasma Chem. Plasma Proc. 17 (1997) 353.
[22] H. Kerstena, H. Deutscha, H. Steffena, G.M.W. Kroesenb, R. Hippler, Vacuum 63 (2001) 385. [23] K. Ostrikov, I. Levchenko, S. Xu, Pure Appl. Chem. 80 (2008) 1909. [24] M. Alizadeh, H. Mehdipour, B.T. Goh, S.A. Rahman, J. Appl. Phys. 114 (2013) 02430. [25] Z.X. Bi, R. Zhang, Z.L. Xie, X.Q. Xiu, Y.D. Ye, B. Liu, S.L. Gu, B. Shen, Y. Shi, Y.D. Zheng, Thin Solid Films 488 (2005) 111. [26] V. Lebedev, F.M. Morales, V. Cimalla, J.G. Lozano, D. Gonz´ıalez, M. Himmerlich, S. Krischok, J.A. Schaefer, O. Ambacher, Superlattices Microstruct. 40 (2006) 289. [27] Xu-Ping Kuang, Hua-Yu Zhang, Gui-Gen Wang, Lin Cui, Can Zhu, Lei Jin, Rui Sun, Jie-Cai Han, Appl. Surf. Sci. 263 (2012) 62. [28] M. Alizadeh, H. Mehdipour, V. Ganesh, A.N. Ameera, B.T. Goh, A. Shuhaimi, S.A. Rahman, Appl. Phys. A 117 (2014) 2217. [29] T. Suzuki, S. Hibino, R. Katayama, Y. Kato, F. Ohashi, T. Itoh, S. Nonomura, Jpn. J. Appl. Phys. 52 (2013) 11NG05. [30] A. Costales, M.A. Blanco, A. Martın Pendas, A.K. Kandalam, R. Pandey, J. Am. Chem. Soc. 124 (2002) 4116. [31] Qiang Jing, Hang Yang, Wancheng Li, Guoguang Wu, Yuantao Zhang, Fubin Gao, Yang Zhao, Guotong Du, Appl. Surf. Sci. 331 (2015) 248. [32] L. Rosenberger, R. Baird, E. McCullen, G. Auner, G. Shreve, Surf. Interface Anal. 40 (2008) 1254. [33] T.-T. Kang, A. Hashimoto, A. Yamamoto, Phys. Rev. B 79 (2009) 033301. [34] J.B. Wang, Z.F. Li, P.P. Chen, Wei Lu, T. Yao, Acta Mater. 55 (2007) 183. [35] M. Watanabe, C. Brauns, M. Komatsu, S. Kuroda, F. Gärtner, T. Klassen, H. Katanoda, Surf. Coat. Technol. 232 (2013) 587. [36] Y. Chen, X. Zhou, X. Zhao, X. He, X. Gu, Mater. Sci. Eng. B 151 (2008) 179. [37] V. Darakchieva, M.Y. Xie, F. Tasnádi, I.A. Abrikosov, L. Hultman, B. Monemar, J. Kamimura, K. Kishino, Appl. Phys. Lett. 93 (2008) 261908. [38] S.S. Lin, J.L. Huang, P. Sajgalik, Surf. Coat. Technol. 190 (2005) 39. [39] M. Di Giulio, G. Micocci, R.P. Rella Siliciano, A. Tepore, Thin Solid Films 148 (1987) 273. [40] T. Peng, J. Piprek, G. Qiu, J.O. Olowolafe, K.M. Unruh, C.P. Swann, E.F. Schubert, Appl. Phys. Lett. 71 (1997) 2439. [41] R. Weingartner, J.A. Guerra Torres, O. Erlenbach, G. Galvez de la Puente, F. De Zela, A. Winnacker, Mater. Sci. Eng. B—Solid 174 (2010) 114. [42] A. Shah, Arshad Mahmood, Physica B 407 (2012) 3987.