1. Introduction
Transparent conducting oxides (TCOs) have attracted significant attention in recent years since they are a unique class of compounds that combine the properties of being transparent and electrically conducting at the same time [
1]. TCOs are semiconductors that exhibit a large bandgap and a high optical transmission in the visible wavelength region [
2,
3]. Owing to these properties, TCOs find applications in optoelectronic devices, light-emitting diodes, solar cells, and transparent thin-film transistors [
2,
4,
5,
6,
7]. TCOs have been extensively researched recently since their properties can be precisely controlled by varying their synthesis parameters. By precisely controlling the doping, TCOs such as SnO
2, In
2O
3, and ZnO can depict the electrical properties of a metal, semiconductor, or insulator while maintaining optical transparency [
8,
9,
10,
11]. Some of the well-known n-type TCOs include Sn-doped In
2O
3 (ITO) [
12,
13,
14], Sc-doped ZnO [
15,
16], SnO
2 [
17,
18], Al-doped ZnO (AZO) [
19,
20,
21,
22,
23], and CdSnO
2 [
24,
25,
26].
A combination of p- and n-type transparent conducting channels are used as active layers in transparent optoelectronic devices. Thus, it is mandatory to have a p-type counterpart. However, p-type TCOs have been identified to have either a good optical transparency (>80%) or a low electrical resistivity (<10
−2 Ω-cm) [
1,
2]. It has been difficult to synthesize p-type TCOs with both of the aforementioned properties. The root cause of this can be attributed to the intrinsic electronic structure of metal oxides, where strongly localized O-2p orbitals occupy most of the top of the valence band [
27,
28]. It has been shown that the valence band level of metal oxides is far lower lying than that of metals, which can be resolved by doping [
29]. However, the highly electronegative O
2 ions attract the holes, inhibiting their movement across the crystal lattice. Due to this, the localized holes need significant energy to cross the high barrier height to move across the crystal lattice [
4]. This is the prime reason for the reduced electrical conductivity in p-type TCOs [
30,
31].
To alleviate this issue, a new concept of ‘Chemical Modulation of Valence Band (CMVB)’ was proposed by Hosono et al. [
27]. They identified that the hybridization of O-2p orbitals is possible by the addition of appropriate cations such as Cu+. The reason stated that, since the energy level difference between Cu 3d
10 and O 2p
6 is very small, the former forms a strong covalent bond with the latter. During this process, the low-lying O-2p energy level is raised and the coulombic force of attraction between the oxygen ions and holes is reduced, which results in a lower energy requirement for holes to cross the barrier height. This ultimately results in an improvement in the electrical conductivity of p-type TCOs [
29]. It was also reported by Zhang et al. that the thin films obtained using the CMVB technique preserves the optical transparency for visible light since it was noticed that the closed shell of 3d
10 prevents the coloration of the deposited films [
30].
Thin films that are obtained using the CMVB technique are called delafossites. The appropriate cation addition during the CMVB process could include either Ag+ or Cu+. However, the previous literature has shown that the delafossites obtained using Ag+ cation had a very high electrical resistivity (10
4 to 10
6 Ω-cm) [
32]. This is due to the very low mobility of carriers which is a result of an unfavorable energy match between O-p and Ag-4d levels [
33]. The delafossites obtained using the addition of Cu+ cation are called Cu-based delafossites, which have a chemical formula of CuMO
2, where Cu is the positive monovalent cation (Cu
+), M = Al
3+, Cr
3+, In
3+, Ga
3+, Sc
3+, and Y
3+ (trivalent cation), and oxygen is the divalent anion (O
2−) [
34]. This understanding provided a new dimension that led to a recent spike in the interest in p-type TCOs, particularly Cu-based delafossites [
35,
36,
37]. It has been identified that Cu-based delafossites typically possess a high optical transparency (60%–85%) and a low electrical resistivity (5 × 10
−2 to 8 × 10
−3 Ω-cm) [
31,
38].
Among the p-type Cu-based delafossite candidates, CuCrO
2 is a promising material to be considered for transparent electronics applications since it presents a valuable trade-off between its optical and electrical properties [
31,
38]. CuCrO
2 has been reported to have a low synthesis temperature and acceptable thermal stability in the air [
31,
39]. The room temperature resistivity of RF magnetron sputtered CuCrO
2 has been previously reported to be 0.66 Ω cm with a transmittance of ~60% [
40]. Hence, the CuCrO
2-based p-type delafossite structure was chosen as the material of research for this work.
In this research, the delafossite structure of CuCrO2 was obtained for the first time using the dual-target RF magnetron sputtering technique of Cu2O and Cr2O3 targets. The RF sputtering powers of Cu2O and Cr2O3 targets were kept constant at 50 W and 200 W, respectively. Post-deposition, all the films were subjected to varying annealing temperatures in N2 ambiance which enabled the formation of single-phase p-type CuCrO2 delafossite thin films. The effect of an annealing temperature variation in the electrical, optical, structural, and morphological properties of the films has been investigated in this work. The superior optical and electrical characteristics of the phase pure CuCrO2 thin films instigated the research on the heterojunction studies using p-CuCrO2 and n-ITO as the active layers in this work.
2. Experimental Section
Deposition of CuCrO2 thin film. In this work, the radio frequency (RF) magnetron sputtering technique was adopted for the deposition of CuCrO
2 thin films. An AJA international ultra-high vacuum three-gun sputtering equipment was utilized for the deposition of the films analyzed in this research. The film depositions were carried out on fused quartz substrates. Before the film deposition, all the quartz substrates were cleaned with acetone, methanol, DI water, and blow-dried with nitrogen gas. A dual target RF magnetron sputtering technique from 3-inch diameter targets of Cu
2O (99.99% purity, Maideli Advanced Materials Co., Ltd., Jiangying, China) and Cr
2O
3 (99.99% purity, Maideli Advanced Materials Co., Ltd., Jiangying, China) was employed to deposit 125 nm thick CuCrO
2 thin films. Two separate RF sources (13.56 MHz) were put to use for the sputtering of the targets. The RF power to the Cr
2O
3 was maintained at 200 W, whereas the RF power to the Cu
2O target was held constant at 50 W. All the depositions were performed only after the base chamber pressure reached 5 × 10
−7 Torr. The deposition parameters maintained during this work for CuCrO
2 thin films have been listed in
Table 1. To ensure a uniform film thickness, the substrate holder was rotated at a constant speed of 20 rpm. The film depositions were carried out at 400 °C (sputtering tool limit) and were subsequently annealed in a tube furnace at varying annealing temperatures from 600 °C to 900 °C in N
2 ambiance for 4 h. The N
2 flow rate was held constant at 300 sccm for all the post-deposition annealing. The annealing temperatures were decided based on earlier research which suggested that the delafossite phase of CuCrO
2 can be formed at a much lower temperature of ~ 600 °C [
41,
42,
43,
44,
45,
46]. Further, the final annealing temperature of 900 °C was chosen as the film started to appear hazy and patchy above 900 °C, leading to a decrease in the optical transmittance.
The aforementioned parameters were maintained for the deposition of a 200 nm p-layer in the p-CuCrO
2/n-ITO heterojunction. The deposition of ITO used as the n-layer in the heterojunction involved the use of a 2-inch diameter ITO sputtering target (Composition: 90% In
2O
3, 10% SnO
2, 99.99% purity, ACI alloys, Inc.) in an ultra-pure grade Ar ambiance. The distance between the target and the substrate was maintained at 5 cm. A 400 nm thick ITO layer was used as the n-layer in the heterojunction study. The deposition parameters maintained for the sputtering of the ITO target have been listed in
Table 2. The aluminum contact pads in this research work have been deposited using the thermal evaporation technique.
Film Characterization. The thickness of the deposited films was checked using the Veeco Dektak 150 Surface Profiler. XRD analysis was carried out using a PANalytical Empyrean XRD system (Malvern Panalytical, Westborough, MA, USA), using radiation from a Cu source at 45 kV and 40 mA. The diffraction patterns recorded between 2θ angles of 25 and 70° assisted in the identification of the materials. Further, the phase information was analyzed using HighScore Plus Software (Malvern Panalytical, Westborough, MA, USA). ESCALAB 250 Xi + X-ray photoelectron spectroscopy (XPS) (ThermoFisher Scientific, Waltham, MA, USA) with a monochromatic Al Kα source (1486.7 eV) was used to determine the elemental composition of the films and the oxidation states of the elements. Prior to the XPS analysis, a surface oxide removal was carried out by the ion beam milling of the film surface using an in-built EX06 ion source. Thermo Fischer Scientific Avantage software (v5.9902) was used to perform elemental composition analyses, functional group analyses, and XPS peak fitting. Zeiss ULTRA-55 FEG SEM (Zeiss Microscopy, White Plains, NY, USA) was used to evaluate the surface morphology of the post-deposition annealed films. The optical transmission was measured using a Cary 100 UV-IR Spectrophotometer (Varian Analytical Instruments, Walnut Creek, CA, USA) at light wavelengths from 200 nm to 800 nm. The transmission data were used to conduct the bandgap studies of the films with the help of the Tauc plot. The hot probe method and four-point probe (Magne-tron Instruments Model M-700) were used to check the conductivity type of the thin films post-annealing. To measure the resistivity of the post-deposition annealed thin films, the parallel aluminum contact pads were patterned as shown in
Figure 1. A photolithography process was adopted to pattern the Al contact pad windows, which was followed by Al metal deposition using a thermal evaporation process and the removal of photoresist using the lift-off technique. The resistance (R) of the thin films was calculated using the IV characteristics obtained from the Keithley 2450 source meter unit. The resistivity (
ρ) of the thin films was identified using Equation (1):
where R is the resistance, L is the length, and A is the area of the cross-section.
3. Results and Discussion
XRD analysis. The XRD diffractograms of the films annealed at temperatures ranging from 600 to 900 °C have been shown in
Figure 2. Due to the use of quartz slides as substrates in this work, a broad amorphous quartz peak appeared in all the films at 2θ angles 18–25°. Hence, 2θ limits in
Figure 2 have been adjusted to omit the amorphous peak of quartz. A higher annealing temperature of ~ 600 °C is required for a nanocrystalline growth in the films. This is evident from the amorphous nature of films obtained post-annealing at 500 °C. It also confirms that higher thermal energy is required for the crystallization of the deposited films. Prominent peaks start to appear at annealing temperatures of 600 °C and higher. At an annealing temperature of 600 °C, peaks pertaining to CuO (indicated as ‘o’ in
Figure 2), Cr (indicated as ‘#’ in
Figure 2), and CuCr
2O
4 (indicated as ‘+’ in
Figure 2) were identified. However, the CuO and CuCr
2O
4 phases in the films are not acceptable for their use in TCO applications. Additionally, previous research has reported that a higher annealing temperature is required for the removal of CuO and CuCr
2O
4 phases and the promotion of CuCrO
2 phases in the films [
35,
43]. On a further increase in the annealing temperature to 650 °C, all the aforementioned peaks disappeared, leading to the promotion of the pure phase delafossite nature of CuCrO
2 (indicated as ‘*’ in
Figure 2) thin films. Further increases in the annealing temperatures led to the disintegration of the single-phase CuCrO
2. Additional peaks of Cr started to appear in the thin films annealed at 750 °C along with CuCrO
2 peaks. Since Cr is characteristically a high-temperature species, the single-phase CuCrO
2 disintegrates and the Cr peak is observed at 750 °C, which appears as Cr
2O
3 with a further increase in the annealing temperature [
47]. At 800 °C and 900 °C of annealing temperatures, Cr
2O
3 (indicated as ‘•’ in
Figure 2) peaks were identified in addition to CuCrO
2 peaks. It is worth mentioning that, since CuO and CuCr
2O
4 phases were not present in the films annealed above 600 °C, it is safe to conclude that both CuO and CuCr
2O
4 are comparatively lower temperature phases. The growth mechanism of CuCrO
2 can be attributed to the additive reaction between Cr
2O
3 and Cu
2O as shown in Equation (2) [
48,
49].
The appearance of single-phase CuCrO
2 at annealing temperatures ~650 °C has been reported in previous works [
41,
43,
46]. The diffraction peaks of the film annealed at 650 °C indexed well to the delafossite CuCrO
2 (ICDD: 98-016-3253) structure. The major diffraction peaks identified at 2θ angles of 31.343°, 35.186°, 36.377°, 40.845°, 55.812°, 62.353°, and 65.419⁰ were indexed to (006), (101), (012), (104), (018), (110), and (1010) of CuCrO
2, respectively. As evident from
Figure 2, an increase in the peak sharpness and peak intensity is confirmed as the annealing temperature is increased from 600 °C to 900 °C. This verifies a steady improvement in the crystallinity of the films with an increase in the annealing temperature. The XRD peaks identified using the diffractograms shown in
Figure 2 have been listed in
Table 3.
XPS Analysis. ESCALAB 250 Xi + X-ray photoelectron spectroscopy (XPS) (ThermoFisher Scientific, Waltham, MA, USA) with a monochromatic Al Kα source (1486.7 eV) was used to carry out the XPS study. The binding energy calibration was performed automatically by the spectrometer using the copper sample which was built into the instrument.
Figure 3a shows the recorded XPS survey spectra of the CuCrO
2 thin film obtained at an annealing temperature of 650 °C. The phase purity of the films can be confirmed as only peaks corresponding to Cu, Cr, and O were identified from the survey spectrum. The survey spectrum also shows the identified Auger peaks of Cu and O. Cu-2p, Cr-2p, and O-1s core level spectrums have been resolved and shown in
Figure 3b–d, respectively. The core level spectrums, oxidation states, bonding configurations, full-width half maximum (FWHM), peak positions, satellite peak positions, and their respective binding energy have been calculated and reported in
Table 4. Copper chromium oxide essentially exists in two phases: CuCr
2O
4 (spinal structure) and CuCrO
2 (delafossite structure) [
48,
50]. Previous research has shown that the oxidation state of Cu in CuCrO
2 is 1+ while Cu exists in a 2+ oxidation state in CuCr
2O
4 [
51]. The distinguishing feature between the core level spectra of the Cu
1+ and Cu
2+ species are the satellite peaks. In the case of the Cu
2+ species, strong satellite peaks (shake-up satellite features) are observed, while they are completely absent for the Cu
1+ species [
51,
52,
53,
54]. The presence of satellites has been attributed to the metal–ligand charge transfer that occurs during the photoemission process [
55].
The Cu-2p core level XPS spectrum has been shown in
Figure 3b. Two separate spin-orbit components of Cu classified as the main peaks of Cu (marked as peak 1 and peak 2 in
Figure 3b), Cu 2p
1/2 and Cu 2p
3/2 have been identified at 952.40 eV and 932.27 eV, respectively. The spin–orbit split energy for the Cu-2p core level reported in this work is Δ
Cu 2p = 20.13 eV. This result is close to the values previously reported in the literature for Cu existing in a 1+ oxidation state in CuCrO
2 phases [
54,
56]. It is important to note that there is an absence of satellite peaks corresponding to the Cu-doublet that appears towards the higher energy side of the main peaks. A similar absence of satellite peaks has been reported previously in the literature that report single-phase Cu-based delafossites [
35,
36,
57].
The Cr-2p core level XPS spectrum has been shown in
Figure 3c. The Cr-2p doublet, Cr 2p
1/2, and Cr 2p
3/2 (marked as peaks 1 and 2 in
Figure 3c) have been identified at 586.00 eV and 576.16 eV, respectively. The spin–orbit split energy for the Cr-2p core level obtained in this work is Δ
Cr 2p = 9.84 eV. This result is in good agreement with the previously reported values for Cr existing in a 3+ oxidation state in CuCrO
2 phases [
52,
54]. In the same range of the energies of the Cr core level spectrum, the Cu Auger peak (Cu LMM) is also identified. The Cu LMM peak is centered at 569.9 eV, which suggests the presence of Cu
1+ rather than metallic Cu
45, which is centered at 568.3eV [
49]. Similar findings were reported in the literature on the delafossite phase of CuCrO
2 [
49,
54,
58].
The resolved O-1s core level XPS spectrum has been shown in
Figure 3d. The main O-1s peak is centered at 530.12 eV (marked as peak 1 in
Figure 3d), which corresponds to lattice oxygen [
49,
59]. An energy-resolved O-1s peak identified at 532.08 eV (marked as peak 2 in
Figure 3d) pertains to the chemisorbed oxygen. These results indicate that Cu and Cr cations mainly exist in the oxidation state of 1+ and 3+, respectively. With the help of XPS analysis and the XRD results, it can be asserted that the pure phase delafossite structure of CuCrO
2 was successfully synthesized at an annealing temperature of 650 °C.
Morphology Studies.
Figure 4 shows the FESEM images at 50 K X magnification of the post-deposition annealed films obtained at varying annealing temperatures from 600 °C to 900 °C. A nanocrystalline growth has been confirmed through the FESEM images obtained for the aforementioned annealing temperatures, as clearly seen in
Figure 4a–e. An increase in the average grain size with an increase in the annealing temperature can be noticed. As the thermal energy increased, the grains obtain enough energy to undergo coalescence, which ultimately increases the average grain size. The average grain size escalated from 40.22 nm for a 600 °C annealed film to 105.31 nm for a 900 °C annealed film. It can be seen that at lower annealing temperatures of 600 °C and 650 °C, the FESEM images show a smooth matrix with no visible cracks. However, as the annealing temperature is increased above 650 °C, gaps start to appear between the grains due to higher thermal energy, which is also observed in the earlier literature [
60,
61]. At an annealing temperature of 900 °C, rod-like structures were observed along with increased grain sizes. Rod-like structures are a characteristic appearance of Cr
2O
3 nanostructures, as reported previously [
62,
63]. This explanation is in complete agreement with our XRD findings which report the presence of Cr
2O
3 with CuCrO
2 at an annealing temperature of 900 °C. It can be safely concluded that increments in the annealing temperature promote the average grain sizes in the films under study.
Figure 5 shows the increasing trend of the average grain size with an increase in the annealing temperature.
Optical Studies—
Optical Transmission. The optical transmission studies shown in
Figure 6 were conducted on all the thin films deposited on quartz substrates. The transmission data had been recorded using a UV–Visible spectrophotometer for the wavelength range of 200–800 nm. It is worth mentioning that the transmission data are affected by the structural changes in the film and has a correlation with the results of the XRD analysis. It can be seen that the film annealed at 600 °C shows the lowest optical transmission. In contrast, the film annealed at 650 °C shows the highest optical transmission for the aforementioned wavelength range. As observed from the XRD analysis, CuO and CuCr
2O
4 phases are identified in the film annealed at 600 °C. This might be a reason for a decreased optical transmission in the 200–800 nm wavelength range. Similar findings have been reported previously [
64,
65]. An optical transmission value of ~45% is observed at a wavelength of 700 nm in the film annealed at 600 °C. The optical transmission increased extensively to ~81% at 700nm as the annealing temperature was increased to 650 °C. This is due to the appearance of single-phase CuCrO
2 in the film as identified through XRD analysis. A similar increase in the optical transmission due to the appearance of phase pure CuCrO
2 in the films has been reported by Yu et al. [
42]. A further increase in the annealing temperature leads to a reduction in the optical transmission values. The observed optical transmission values at a 700 nm wavelength are ~78%, ~71%, and ~62% for the film annealed at 750 °C, 800 °C, and 900 °C, respectively. This is due to the disintegration of the single-phase CuCrO
2 and the appearance of Cr and Cr
2O
3 peaks seen in the XRD results. The decreased optical transmission with an increase in the annealing temperature can also be attributed to the increase in the grain sizes, thereby resulting in increased photon scattering [
66]. It is evident from the optical transmission data that at a 414 nm wavelength, the transmission spectra of the films annealed at temperatures above 600 °C steeply drops since the films are identified to have a single-phase structure (CuCrO
2). However, for the film annealed at 600 °C, the transmission spectra do not drop steeply, which might be possibly due to the two-phase structure (CuO and CuCr
2O
4). Similar results have been obtained by Tripathi et al. and Chiba et al. [
65,
67]. The average optical transmission for the visible wavelength region obtained at the annealing temperatures of 600 °C, 650 °C, 750 °C, 800 °C, and 900 °C was 40.46%, 61.32%, 57.17%, 54.51%, and 48.57%, respectively.
Optical Bandgap. The optical bandgap (E
g) of the post-deposition annealed films was calculated using the Tauc plot method [
68]. Using the transmission data, the absorption coefficient was calculated using the relation,
where ‘t’ is the thickness of the thin film post-annealing and ‘%T’ is the fraction of incident light transmitted through the film. The Tauc equation used to calculate the bandgap from the absorption coefficient is,
where ‘hν’ is the photon energy and ‘n’ denotes the nature of the sample’s transition. The value of n equals ½, 2, and 3/2 for the direct allowed, indirect allowed, and direct-forbidden transitions, respectively [
68,
69,
70]. However, the best linear fit of the (αhν)
(1/n) versus the photon energy curve was obtained for n = ½, which allows us to safely conclude that these materials exhibit a direct band gap transition. The resultant Tauc plot thus obtained for the films annealed at various temperatures has been shown in
Figure 7.
The linear portion of the Tauc plot for all the annealed films has been extrapolated to the x-axis to obtain the bandgap values of the films. The obtained bandgap values have been summarized in
Table 5.
As seen from
Table 5, the bandgap values appear to decrease with an increase in the annealing temperature. The bandgap values decreased from 3.52 eV for a 600 °C annealed film to 3.16 for 900 °C. A higher bandgap in a 600 °C annealed film can be attributed to the presence of the CuO phase in the film. Similar bandgap values in CuO-rich films have been reported previously by Vinothkumar et al. [
71]. As the annealing temperature is increased, the CuO phase in the film disappears and at the same time, the CuCrO
2 phase begins to appear. At 650 °C, a single-phase CuCrO
2 thin film is obtained with a bandgap of 3.21 eV, which is close to the previously reported values in the literature [
49,
65,
66,
72,
73,
74,
75]. With an increase in the annealing temperature, the single-phase CuCrO
2 dissociates and exists as two-phase structures. This might be one of the reasons for the increase in the bandgap. This study correlated well with the effect of the annealing temperature study on CuCrO
2 thin films conducted by Obulapathi et al. [
41].
Electrical Studies. Copper-based delafossites such as CuCrO
2, CuInO
2, CuGaO
2, and CuFeO
2 exhibit p-type conductivity due to intrinsic defects such as interstitial oxygen ions and/or Cu vacancies [
36]. The p-type conductivity of the CuCrO
2 thin films obtained in this research was verified with the help of the hot probe method and a four-point probe. The relation between the annealing temperature and the electrical resistivity of the films is shown in
Figure 8. The electrical resistivity was noted to decline from 17.19 Ω cm to 0.652 Ω cm as the annealing temperature was increased from 600 °C to 650 °C. This decrease can be attributed to the appearance of single-phase CuCrO
2 in the films. Among all the Cu-based delafossites, CuCrO
2 exhibits the lowest resistivity [
31]. The previously reported resistivity values for phase pure CuCrO
2 are close to the value of 0.652 Ω cm obtained in this research [
40,
53,
76]. With a further increase in the annealing temperature to 750 °C, the resistivity value is seen to increase to 24.67 Ω-cm. This increase in the resistivity is potentially due to the disintegration of single-phase CuCrO
2. A sudden increase in the resistivity to 162.7 Ω cm and 351.4 Ω cm is observed at annealing temperatures of 800 °C and 900 °C, respectively. This is likely due to the appearance of Cr
2O
3 phases in the films. Previous literatures have reported the resistivity of Cr
2O
3 to be greater than 1 kΩ cm [
77,
78]. However, since the films annealed at 800 °C and 900 °C are still CuCrO
2 phase-rich, comparatively lower values of resistivity have been observed in this research. It can be safely concluded that the lowest electrical resistivity of 0.652 Ω cm was obtained for films annealed at 650 °C exhibiting single-phase CuCrO
2.
Heterojunction Studies. Two configurations of diodes as shown in
Figure 9a,b were fabricated to perform the heterojunction studies of CuCrO
2 thin films. These diodes will henceforth be called diodes D1 and D2, respectively. For both the diodes, single-phase delafossite CuCrO
2 acts as the p-layer and ITO acts as the n-layer. The substrates used for the diodes are 1-20 Ω cm n-type Si. Prior to the deposition of the n and p layers, the Si wafer was dipped in BOE (5:1) for 2 min to remove the native oxide formation. For the heterojunction study, 200 nm thick p-CuCrO
2 and 400 nm thick n-ITO layers were used. The deposition parameters used for the n and p layers have been detailed in
Table 1 and
Table 2. The aluminum contact for both diodes have been deposited using the thermal evaporation technique. There are two differences between the two diode configurations. The first difference is the amount of overlap between the two layers. In D1, there is a complete overlap between the two layers. However, the overlap accounts for only 4 mm in the case of D2. The second difference is the way the two diodes are probed during the electrical characterization. D1 has aluminum contacts to be probed on the front and the back of the Si wafer. On the other hand, D2 has aluminum contacts on the top of the Si wafer. It is necessary to make sure that the current travel is through the p-CuCrO
2 and n-ITO layers and not through the Si. For this purpose, dry oxidation using a tube furnace was carried out on the Si wafer used for D2.
To eliminate the possibility of the formation of a Schottky junction between n-Si and p-CuCrO
2, the metal–semiconductor I–V characteristics were analyzed. The contact resistance of the metal–semiconductor junction was determined using the transmission line method (TLM) [
79,
80]. The contact resistance obtained in this research work is 6.14 × 10
−4 Ω-cm
2, which is close to the previously reported value [
81]. The I–V characteristics of Si/CuCrO
2 shown in
Figure 10a indicate a highly linear curve in both a forward and reverse bias. This indicates that the rectification mechanism in D1 and D2 is not due to a Schottky-type rectification. Diodes D1 and D2 show non-linear I–V characteristics as shown in
Figure 10b,c. The rectifying behavior of the diodes can be confirmed by the very low values of the reverse currents. The IV characteristics of the diodes show that there is a higher current flow in D1 as compared to D2. A possible explanation for this can be that looking at the diode’s structure, the current travel is purely in the vertical direction for D1. On contrary, in D2, there is a lateral as well as vertical propagation of the current which might lead to the lesser resultant current seen in the I–V characteristics of D2. The calculated diode parameters and their comparison with similar works have been listed in
Table 6. The rectification ratio of the diodes has been calculated by taking the ratio of the current for the forward and reverse bias voltages according to Equation (5) [
82,
83],
On comparing the rectification performance of the two diodes, it was noticed that D1 has a lower turn-on voltage of 0.85 V as compared to 1.29 V for D2. Further, the rectification ratio of D1 is 2375, which is substantially greater than the rectification ratio of 1664 for D2. The rectification ratio of D1 is two orders of magnitude greater compared to the reported values on the diodes prepared with CuCrO
2 as the p-layer [
83]. The rectification ratios indicate that D1 must have a lower leakage current as compared to D2, which is evident in the results. D1 has a leakage current of the order of 1.24 × 10
−8 A and that of D2 is of the order of 2.15 × 10
−8 A. The Ln (current density (J)) versus the voltage for both diodes have been shown in
Figure 10d,e. It can be seen from the current density plot that the rectification for the forward bias voltages is much larger in the case of D1. From the electrical characterization of the two diodes, it is safe to say that a superior device performance is obtained for D1 compared to D2.
The ideality factor of the diode is used to measure the performance of the diode in comparison to the ideal diode. The ideality factor can be calculated using the diode equation [
84],
The diode ideality factor was thus determined by calculating the slope of Ln (current density) versus the voltage graphs shown in
Figure 10d,e. The diode D1 exhibits an ideality factor of 4.13 while the ideality factor of the diode D2 was obtained as 5.8. In the case of an ideal pn-junction diode, the ideality factor is 1, where the conduction of the electrons due to thermionic diffusion is acting. The rise in the ideality factor observed in this research work can be attributed to the use of compounds as layers forming the diode, the recombination of carriers, the conduction of electrons by tunneling from the active layer to the p-region acting as the barrier layer, etc. [
84,
85,
86]. It is also observed that lattice matching plays an important role in the rectification ratio and the ideality factor of the diode. For homojunction diodes, the ideality factor is noticed to be below two, whereas the ideality factor rises above two for a heterojunction diode [
84,
87,
88,
89,
90].
To study the optical transparency of D1, the p-CuCrO
2/n-ITO stack without the Al layers was deposited on a cleaned quartz substrate. The transparency of D1 can be verified through the optical study results shown in
Figure 11a–c. It is evident that the p-CuCrO
2/n-ITO stack has greater than 60% transmission for the wavelengths of light greater than 550 nm. The optical bandgap of the stack was calculated from the Tauc plot and was obtained as 3.25 eV.
Figure 11c clearly shows the photomicrograph of the p-CuCrO
2/n-ITO stack deposited on the quartz substrate through which the details written on the paper can be clearly read. The results obtained through the heterojunction study confirmed that the diode D1 with a complete overlap between the semiconducting layers showed a superior performance over the diode D2, which had an overlap of only 4 mm.
Table 6.
Electrical characterization results of D1 and D2 and comparison with similar works.
Table 6.
Electrical characterization results of D1 and D2 and comparison with similar works.
Diode | Turn ON Voltage (V) | Leakage Current (A) | Rectification Ratio | Ideality Factor | Reference |
---|
D1 | 0.85 | 1.24 × 10−8 (at 0.9 V) | 2375 (at 2 V) | 4.13 | This work |
D2 | 1.29 | 2.15 × 10−8 (at 0.9 V) | 1664 (at 1.8 V) | 5.8 | This work |
n-ITO/p-CuCrO2 | ~0.87 | NR | >104 | 3.90 | [49] |
n-ZnO/p-Cu0.66Cr1.33O2 | 1.40 | NR | 13 | 6.30 | [90] |
n-ZnO/p-Mg:CuCrO2 | ~2.00 | NR | ~7–10 | NR | [91] |
n-ZnO/p-CuCrO2 | NR | NR | 107 | 1.56 | [92] |
n-ZnO/p-CuCrO2 | NR | NR | 20 | 2.20 | [93] |