In0.5Ga0.5N Layers by Atomic Layer Deposition

We present an ALD approach to metastable In 1‑x Ga x N with 0.1 < x < 0.5 based on co-sublimed solid In- and Ga-precursors that were co-sublimed into the deposition chamber in one pulse. A near In 0.5 Ga 0.5 N film with a bandgap of 1.94 eV was achieved on Si (100) substrate. Epitaxial In 1‑x Ga x N (0002) was successfully grown directly on 4H-SiC (0001). We present an ALD approach to metastable In 1-x Ga x N with 0.1 < x < 0.5 based on co-sublimed solid In- and Ga-precursors that were co-sublimed into the deposition chamber in one pulse. A near In 0.5 Ga 0.5 N film with a bandgap of 1.94 eV was achieved on Si (100) substrate. Epitaxial In 1-x Ga x N (0002) was successfully grown directly on 4H-SiC (0001).

Alloying the group 13-nitrides to ternary phases allows tuning of the bandgap from 6.2 eV for pure AlN 1 down to 0.7 eV for pure InN 2 . The bandgap of In1-xGaxN can theoretically span from UV to IR (3.4 -0.7 eV), including the whole visible light range by varying x, making a promising material for optoelectronic applications. 3,4 However, the practical ability to vary the composition of In1-xGaxN is limited by the theoretically predicted metastability of In1-xGaxN for 0.05 < x < 0.95 leading to phase separation into their binary materials. 5 The deposition of In1-xGaxN is also hindered by the low thermal stability of InN, which decomposes into In metal and N2 around 500 °C 6 .
Experimental results have lined up with the predicted metastability, albeit x = 0.8 (20 at. % In) has been shown in thin films deposited by chemical vapor deposition (CVD). 4,7,8,9,10 CVD is not ideal for In1-xGaxN due to the high temperatures (>500 °C) required to reach sufficient decomposition of NH3 11 and typically results in low In content, phase separation, and the appearance of In droplets. 8,10 A low temperature deposition technique is strongly preferred for In1-xGaxN with x close to 0.5. Atomic layer deposition (ALD) is a low temperature alternative to CVD, in which the precursors are pulsed sequentially into the reactor. We have recently shown that ALD is a promising technique to deposit InN thin films with excellent structural quality. 12 The sequential pulsing of the precursors in ALD presents a challenge to depositing a homogeneous ternary material as only one precursor can be pulsed into the reactor at a time. Ternary materials are therefore deposited by ALD as stacks of two binary materials.
In1-xGaxN could therefore be deposited as layers of InN and GaN in an ABAB…CBCB… super-cycle approach where A and C are In-and Gaprecursors, respectively, and B is the N-reactant. By varying the number of cycles for each binary material, the overall composition of the ternary material can be tuned. This approach relies on the diffusion of the two binary materials to form a homogeneous ternary phase. Otherwise, a multilayer of InN/GaN is obtained. This ALD approach has been used to obtain In1-xGaxN with x ranging from 0.15-0.85 using trimethylindium and trimethylgallium. 13 Herein, we present an alternative method to depositing ternary materials by introducing both metal precursors with a single pulse. This renders mixing of the metals in both the growth direction and in the growth plane. This was achieved by mixing and co-subliming two solid metal precursors into the ALD chamber.
We have previously investigated ALD of InN and GaN using tris(1,3diisopropyltriazenide)indium(III) 12 (1) and gallium(III) 14 (2) (Fig. S1), rendering stable ALD behavior with self-limiting deposition, wide temperature ranges where the growth per cycle is not affected by the temperature and a high growth per cycle, combined with good structural and electronic properties of the materials. In both studies, we found a sublimation temperature of 120-130 °C to be optimal for each precursor and epitaxial films were obtained at 350 °C for both InN and GaN on 4H-SiC (0001) substrate. In this study, we mixed both precursors in the sublimator and co-sublimed them into the reaction chamber in a single pulse. ALD of In1-xGaxN was then undertaken using the same previously optimized parameters as for the binary nitrides with NH3 plasma as nitrogen source (see supplementary information for experimental details). Initial trials were conducted Department of Physics, Chemistry and Biology, Linköping University, SE-58183 Linköping, Sweden *polla.rouf@liu.se Please do not adjust margins Please do not adjust margins on Si (001) and found that x in In1-xGaxN can be controlled (0.1 < x < 0.5) by the sublimation-and deposition temperatures, and by the mixing ratio of 1 and 2 (see supplementary information for details).
Growth of epitaxial In1-xGaxN was attempted on 4H-SiC (0001) as it is a suited substrate for both epitaxial InN and GaN. 12,14 When depositing In1-xGaxN directly on 4H-SiC, without a nucleation layer, using a 1:1 (In:Ga) precursor mix, sublimation temperature 130 °C and deposition temperature 350 °C, In1-xGaxN was obtained. In the -2 XRD measurement (Fig. 1a), the peak ascribed to In1-xGaxN (0002) is a wide, unsymmetrical triplet, indicating a compositional difference in the film. The peak at 32.0° is closer to InN (31.3°) indicating In-rich In1-xGaxN while the peak at 33.9° is closer to GaN  (Fig. 1b). The outer six poles with higher  value correspond to the 4H-SiC substrate while the inner six poles correspond to the In1-xGaxN film, confirming the epitaxial relation. It should be noted that the substrate poles have higher intensity than the In1-xGaxN poles. This could be due to a the compositional difference in the film, which was also observed in the -2 measurement. The similar, but There is no misalignment between the substrate and film poles as they line up precisely, indicating no twisting of the InGaN crystals with respect to the substrate.

and c) In18Ga82N layers. d) SAED pattern from the film and substrate. EDX maps of Ga e), In f) and Si g). EELS maps of N h) and C i).
and selected area electron diffraction (SAED). The XPS measurements showed the film composition was In0.55Ga0.45N (Table S1)

Conflicts of interest
PR and HP have applied for a patent on this method.  Precursor synthesis

Tris(1,3-diisopropyltriazenide)indium(III) (1) and tris(1,3-diisopropyltriazenide)gallium(III)
(2) were synthesized according to the literature procedures. 1,2 The reactions and handling of the precursors were undertaken in a dry nitrogen atmosphere on a Schlenk line and in a glove box (GS Gloveboxsystemtechnik). The precursor powders were mixed by a filling a glass vial with 1 and 2 to a total weight of ~1.0 g and further mixed by a spoon to obtain a uniform mixture.
The glass vial was placed in a stainless-steel container and inserted in a heated sublimator in the ALD reactor.

Film characterization
Film thickness was measured by X-ray reflectivity (XRR) and film crystallinity was measured by X-ray diffraction (XRD) in -2 mode using an PANalytical X'Pert PRO with a Cu-anode tube and Bragg-Brentano HD optics. To analyze the thickness, the software PANalytical X'Pert reflectivity and a two-layer model was used to fit the data, InGaN/substrate. PANalytical EMPYREAN MRD XRD with a Cu-anode X-ray tube and 5-axis (x-y-z-v-u) sample stage operating at 45 kV and 40 mA was used for the pole figures measurement using an X-ray lens and parallel plate collimator. Elemental compositions were obtained using XPS , RBS and ToF-ERDA. Kratos AXIS Ultra DLD X-ray photoelectron spectroscopy (XPS) equipped with Ar sputtering was used. The film composition was collected after Ar sputtering for 600 s with a beam energy of 0.5 keV with a sputtering area of 3 mm 2 . The RBS and ToF-ERDA measurements were carried out in a 5-MV NEC-5SDH-2 pelletron tandem accelerator. RBS employed 2 MeV 4 He + ions and detected in a scattering angle of 170°. Two different geometries, azimuth angles of 5°+tilt angle 2° and 40°+tilt angle 2°, were chosen to minimize channeling effects. In addition, suppression of the probable channeling effects was undertaken by multiple small random-angular movements around the equilibrium angles within a range of 2°. RBS spectra was fitted by SIMNRA 7.02 code 3 with an ~1% statistic uncertainty to determine elemental compositions. In ToF-ERDA, recoils were detected at 45° angle between the primary beam and a ToF-E detector telescope in a gas ionization chamber (GIC) using a 36 MeV 127 I 8+ beam incident at 67.5° with respect to the sample surface normal. The ToF-E detector telescope consists of two circular carbon foils with 8 and 5 µg/cm 2 thicknesses, 6 mm radius, a 0.05-msr solid angle (ΔΩ), and a flight distance of 425 mm between the foils. Utilizing a ToF-GIC setup provides a system with a good energy resolution and enhanced ion species separation in terms of mass and energy. 4 Average elemental compositions was also obtained from ToF-ERDA time-energy coincidence spectra using two different software packages, CONTES 5 and Potku 6 . Systematic uncertainties of the experiment, discussed in more detail elsewhere 7 in particular for light elements, was estimated to be maximum 5-10%, whereas statistic uncertainties arisen from the number of experimental counts was ˂ 2.3%. However, the relative elemental concentrations was obtained with higher accuracy. 8,9 The stopping power data required for both RBS and ERDA simulations was retrieved from SRIM2013 code. 10 Cross-sectional transmission electron microscope (TEM) samples were prepared by the traditional sandwich approach, which includes sample cutting, gluing, polishing and ion milling. A Gatan Precision Ion Polishing System Model 691 operated at 5kV and 40 mA, with an Ar ion source, was used to make the samples electron transparent. Scanning transmission electron microscopy (STEM), selective area electron diffraction (SAED), energy dispersive Xray analysis (EDX) and electron energy loss spectroscopy (EELS) characterization were performed using the Linköping double Cs corrected FEI Titan 3 60-300, operated at 300 kV.
The absorbance measurements were conducted using a custom fiber optical setup consisting of a light source (Ocean Optics DH-2000-BAL), a detector (Avantes AvaSpec-Dual) and a bifurcated optical fiber (Ocean Optics BIFBORO-2-1000). Absorption spectra for the films were collected using a custom software based LabView (National Instruments) with a Si(100) substrate used as a reference. A LEO 1550 scanning electron microscope (SEM) with an acceleration energy of 3 kV was used to study the morphology of the film.

In1-xGaxN composition control
Deposition of In1-xGaxN was first undertaken on Si(100) substrates to investigate the cosublimation approach where the metal precursor pulse was set to 10s. ALD of In1-xGaxN was undertaken using the same optimized parameters as our previously reported studies of InN and GaN using 1 and 2, respectively. 1,2 Initially, the sublimation temperature was varied from 90-170°C while the deposition temperature was set to 350 °C. XPS analysis showed that the In/Ga ratio depended on the sublimation temperature (Table S1). At lower sublimation temperatures (90-120 °C), the In content was approximately twice as much as Ga. The In content increased with the deposition temperature, reaching its peak between 130-140 °C with approximately 4 times more In than Ga. The In content decreased between 150-170 °C and a near 1:1 In/Ga ratio was found for a sublimation temperature of 150 °C. The general trend is the In content of the film is always higher, however, some control of the composition can be made by changing the sublimation temperature.
XRD analysis (Fig. S1) showed that all films were crystalline except for those using a sublimation temperature of 90 °C, which gave X-ray amorphous films. The amorphous films are thought to be due to insufficient precursor delivery into the chamber. For all crystalline films, polycrystalline In1-xGaxN was observed by several diffraction peaks. For sublimation temperatures of 110 °C and 170 °C, only one peak was observed indicating a preferred growth orientation along the c-axis.
The sublimation temperature was kept constant at 130 °C while the deposition temperature was varied. XPS analysis showed the deposition temperature has a large effect on the In/Ga ratio.
At lower deposition temperatures, a near 1:1 ratio of In/Ga was obtained while the In content increased drastically when increasing the deposition temperature (Table S2). The films deposited at 200 °C and 250 °C showed no peaks in the XRD measurement (Fig. S2), indicative of X-ray amorphous In1-xGaxN. The films were crystalline for deposition temperatures above 350 °C, displaying several XRD peaks and indicating a polycrystalline nature or compositional difference along the growth axis.
To further evaluate the composition of the films, different ratios of 1 and 2 were mixed in the sublimator and sublimed at 130 °C for film deposition at 350 °C. Table S3 shows the mixing ratio of precursors could control the In/Ga ratio of the In1-xGaxN films when higher amounts of 1 were used. Increasing the amount of 2 showed the In/Ga ratio of the In1-xGaxN film did not follow the mixing ratio. Although higher amounts of 2 led an increase of Ga in the film. The crystallinity was not affected significantly by the mixing ratio of 1 and 2, all rendering crystalline In1-xGaxN (Fig. S3).
As both In and Ga have a high affiliation to oxygen, avoiding surface oxidation prior to elemental analysis of the film composition was challenging. We previously capped InN with a thin layer of AlN to accurately determine the film composition with XPS. 1  in comparison to the ERDA measurements, which could be due to preferential sputtering. 12 In the measured films, 1 at% of carbon was also be detected. Due to long air exposure and the inability to cap the In1-xGaxN for ERDA measurements, as the nitrogen from the capping and the In1-xGaxN film cannot be distinguished, high oxygen content in the film of upwards to 20 at% was observed. Capping the In1-xGaxN would prevent post deposition oxidization and in turn drastically decrease the oxygen content, as we previously observed for InN. 1