Schottky barrier diodes on α − Ga 2 O 3⁠:Sn heteroepitaxial thin films grown by pulsed laser deposition on m-plane sapphire substrates are reported. Sets of co-planar diodes were fabricated with different metals and different deposition methods. The current rectification and effective Schottky barrier height of oxidized contacts realized by reactive sputtering significantly exceed the values of non-oxidized contacts realized by thermal evaporation or sputtering in an inert argon atmosphere. The best values obtained are rectification of about eight orders of magnitude (⁠ ± 2 V) and 1.3 eV effective barrier height. The current-voltage characteristics of selected non-oxidized and oxidized platinum diodes have been studied as a function of measurement temperature. The temperature dependence of the effective barrier height and the ideality factor of the diodes were fitted taking into account the lateral potential fluctuations of the barrier potential. The determined mean barrier heights and standard deviations are in the range of 1.76–2.53 and 0.2–0.33 eV, respectively, and are classified with respect to the literature and fulfill a well-established empirical correlation (Lajn’s rule) for a variety of Schottky barrier diodes on different semiconducting materials.

Topics
Schottky diodes, Current-voltage characteristic, Semiconductor device fabrication, Semiconductor structures, Epitaxy, Pulsed laser deposition, Sputter deposition, Thin films, Oxides, Potential energy barrier

Ga 2 O 3 is emerging as a viable alternative to GaN or SiC for the fabrication of high-power electronic devices.1–4 It may crystallize in various polymorphs,5 among which the monoclinic β-gallia structure is thermodynamically most stable. Moreover, the substrate and wafering technology is available for β − Ga 2 O 3⁠, so most of the scientific and technological studies are related to this polymorph. In addition, the orthorhombic modification, which is the only polymorph with spontaneous polarization,6 and rhombohedral α - Ga 2 O 3⁠, being isostructural to commonly used sapphire substrates, are also promising for high-power devices. For both, first Schottky barrier diodes (SBDs)7–13 and p n-heterodiodes have been demonstrated.13,14 So far, SBDs on α - Ga 2 O 3 have been realized on layers grown by MIST-CVD7–9,11,12,15 (with the exception of a UVC photo detector on α − Ga 2 O 3 grown by halide vapor phase epitaxy10). A rectification ratio of a lateral MESFET device structure of 6 × 10 6 was achieved for reactively sputtered AgO x SBD on Sn-doped α - Ga 2 O 3 on c-plane sapphire with Ti/Au ohmic contacts.7 A first vertical SBD was fabricated by Oda et al. by electron beam evaporation (EB-Ev) of Pt Schottky contacts and removal of the c-plane sapphire substrate used for epitaxy. Subsequent EB-Ev of Ti/Au ohmic contacts resulted in low on-resistance of 0.1 m Ω cm 2⁠.8,9 The properties of various Schottky contact metals have been investigated by Shiojima et al. as a function of annealing temperature.11 The contacts were deposited by EB-Ev on a MIST-CVD-grown α - Ga 2 O 3 layer stack on c-plane sapphire. The stack consisted of a n + bottom and n −-top layer allowing a quasi-vertical diode layout (ohmic contact metals were not specified). As-deposited Ti, Pt, Fe SBDs have an effective barrier height (ideality factor) of 0.78⁠, 1.18⁠, and 1.39 eV (1.04, 1.38, and 1.32), respectively, as determined from current-voltage (⁠ I V⁠) measurements considering thermionic emission as the transport mechanism. For annealing temperatures of 400  °C and higher, interfacial reactions were observed for all contact metals resulting in higher leakage current, lower effective barrier height, and higher ideality factor. Temperature-dependent current-voltage (⁠ I V T⁠) measurements were conducted for Ti SBDs on a similar α- Ga 2 O 3 layer stack (with Ti ohmic contacts to the n + bottom layer) on m-plane sapphire15 for 298  ≤ T ≤ 423 K. The forward current is well described by the thermionic emission (TE) theory the reverse current follows the thermionic field emission (TFE) theory. The effective barrier height and the ideality factor were independent of temperature with values of 0.86 eV and 1.03, respectively. Despite these partially encouraging results, further research on (i) different Schottky contact metals for SBD performance optimization, (ii) the influence of the Schottky contact metal deposition process, and (iii) the temperature-dependent electrical properties of SBDs on α - Ga 2 O 3 are needed to accelerate the exploitation of this polymorph in applications.

The electrical properties of SBDs on α - Ga 2 O 3⁠:Sn grown by pulsed laser deposition (PLD) on m-plane sapphire substrates are discussed in this study. The use of m-plane sapphire substrates was shown to be advantageous for the growth and stabilization of the metastable corundum phase of Ga 2 O 3⁠.16 On c-plane sapphire or on c-facets formed during growth on r-plane sapphire, β - Ga 2 O 3(⁠ 2 ¯ 01⁠) is obtained by physical vapor deposition. For growth on m-plane sapphire, the c-direction is in-plane (perpendicular to the growth direction), allowing the growth of phase-pure α - Ga 2 O 3⁠.17,18 The room temperature (RT) current-density-voltage (⁠ j V⁠) characteristics of SBDs with different contact metals on α - Ga 2 O 3⁠:Sn(10.0) are compared and discussed first. Then, for Pt-SBDs, the properties of thermally evaporated contacts are compared to inert and reactively sputtered layers. Reactive sputtering produces oxidized Schottky contacts. The effects of oxidation on the properties of Schottky contacts have been investigated and described in detail for various semiconducting oxides and also for different contact metals (e.g., Ag, Au, Ir, Rh, Pd, and Pt) and show (i) significant lower leakage current, (ii) increased rectification, and (iii) improved thermal and temporal stability as demonstrated for ZnO,19,20 In 2 O 3⁠,21,22 zinc-tin-oxide,23,24 and β - Ga 2 O 3⁠.25,26 In general, near-surface oxygen vacancies, which are donor-like defects in such materials, are saturated due to the transfer of oxygen from the contacts to the semiconductor (reduction of oxidized metal electrode) and/or the surface downward band bending and electron accumulation layers are removed during reactive sputtering or oxygen plasma treatment.19,22,23 Furthermore, the deposition conditions influence the rectification and the homogeneity of the Schottky barrier.20,25 For SBDs on heteroepitaxial β - Ga 2 O 3 thin films the highest rectification and highest barrier height are obtained for oxidized contacts realized by reactive sputtering while most homogeneous Schottky barriers are obtained for deposition methods with low kinetic energy of the deposited species such as evaporation and to some extent long-throw reactive sputtering.25 Finally, the temperature dependence of the effective barrier height and the ideality factor are evaluated by the model of Splith et al.27,28 and Werner and Güttler,29 respectively, and the derived parameters are classified in relation to the literature, and it is shown that Lajn’s rule30 also applies to SBDs on the group-III sesquioxides. Lajn’s rule is an empirically derived correlation between the homogeneous barrier height and the corresponding standard deviation of laterally inhomogeneous Schottky barrier potentials.

α - Ga 2 O 3⁠:Sn was grown by PLD on an α - Ga 2 O 3 buffer layer on a 2-in.-diameter m-plane sapphire substrate. The thickness of the active and the buffer layer are 700 and 100 nm, respectively. The deposition parameters were optimized by Vogt et al.18 The wafer was cut into 5  × 5 mm 2 sized pieces for maximum comparability of results. Co-planar ohmic and Schottky contacts were patterned by photolithography and lift-off. Prior to the deposition of the Schottky contact metals the ohmic Ti/Al/Au (30/30/30 nm) layer stack, realized by thermal evaporation, was rapidly thermally annealed for 2 min at 400  °C in air using a CO 2 laser. Schottky contact metals were thermally evaporated (Au, Cu, Ni, Pt), sputtered in inert (Nb, Pt), or reactive (Pt) ambient, respectively. Furthermore, sets of Pt contacts were prepared as a function of oxygen and argon volume flow ratio O 2/Ar ranging from 0%/100% to 100%/0%. Current-voltage characteristics were acquired using an Agilent Semiconductor Parameter Analyzer 4155C and 4156C, respectively, inside a SUESS WaferProber PA200 for RT measurements or inside a Lake Shore Cryotronics Cryogenic Probestation for IVT measurements.

On β - Ga 2 O 3 various Schottky contact materials including Ag, Au, Co, Cu, Ir, Mo, Ni, Pd, Pt, Ru, Ti, and W as well as oxidized Ag, Au, Ir, Pd, Pt, Ru, Sn, and graphite have been investigated.31 For α - Ga 2 O 3⁠, as noted above, little information is available and a comparison of different Schottky contact materials is required. We have evaluated the RT j V-characteristics of sets of Au, Cu, Ni, Nb, and Pt Schottky barrier diodes assuming thermionic emission across a laterally inhomogeneous Schottky barrier as the dominant charge transport process,
(1)
with
(2)
and where the ideality factor is η⁠, the effective barrier height is Φ B , eff⁠, the Richardson constant is A ∗⁠, the Boltzmann constant is k B⁠, the absolute temperature is T⁠, the applied bias is V⁠, and the effective and free electron mass are m eff and m 0⁠, respectively. The derived effective barrier height Φ B , eff is plotted vs the corresponding ideality factor η in Fig. 1 for each set of diodes. The highest ideality factors (⁠ η > 4.5⁠) are obtained for the sputtered Nb contacts. All thermally evaporated contacts have much lower η⁠, but their values are also significantly larger than the ideality factors reported for β - Ga 2 O 3 thin films.25,27 The lateral device layout, resulting in bias-dependent lateral and vertical electric field components, is one source of the comparatively large η values. Nevertheless, for all Schottky contact metals the inverse correlation between the effective barrier height and the ideality factor is evident, which is typical for thermionic emission across a barrier with lateral barrier height fluctuations.32,33 The effective barrier heights are clustered for all contacts in a range between 0.65 and 0.75 eV, so no influence of the metal work function on Φ B , eff is found, similar to the results of Shiojima et al.;11 however, their values for effective barrier height and ideality factor were 1.18 eV and 1.38 for a vertical Pt/ α - Ga 2 O 3 diode. For monoclinic heteroepitaxial thin films reported in the literature, typical effective barrier height values of Cu (i-MS)34 and Pt(Ev)25 are about 0.9 and about 1.0 eV, respectively, and are also higher than those reported here for similar diodes.
FIG. 1.
Effective barrier height vs ideality factor of Schottky contact metals on heteroepitaxial α - Ga 2 O 3(010). The contacts are made by either thermal evaporation (Ev) or inert magnetron sputtering (i-MS) in argon ambient.
View largeDownload slide

Effective barrier height vs ideality factor of Schottky contact metals on heteroepitaxial α - Ga 2 O 3(010). The contacts are made by either thermal evaporation (Ev) or inert magnetron sputtering (i-MS) in argon ambient.

FIG. 1.
Effective barrier height vs ideality factor of Schottky contact metals on heteroepitaxial α - Ga 2 O 3(010). The contacts are made by either thermal evaporation (Ev) or inert magnetron sputtering (i-MS) in argon ambient.
View largeDownload slide

Effective barrier height vs ideality factor of Schottky contact metals on heteroepitaxial α - Ga 2 O 3(010). The contacts are made by either thermal evaporation (Ev) or inert magnetron sputtering (i-MS) in argon ambient.

Close modal

The influence of deposition conditions on RT Schottky diode j V-characteristics is shown in Fig. 2 for typical Pt-SBDs realized by evaporation, inert sputtering, and reactive sputtering. The reverse current is the highest for the evaporated and inert sputtered contact in agreement with other semiconducting oxides.19–26 Ohmic behavior is observed for Pt contacts sputtered in an inert Ar ambient. The reverse current decreases by orders of magnitude for oxygen-containing contacts deposited by reactive sputtering. We note that the forward current density of all types of sputtered Pt-contacts is below the evaporated layer, indicating an increase in the series resistance induced by the sputtering process. The logarithm of the current rectification ratio

(3)
of sets of differently prepared Pt SBDs is compared by box plots depicted in Fig. 3(a). For the evaporated contacts the median of the rectification is S ~ V = 4.4⁠. The highest rectification is found for sputter ambients with 25%/75% and 50%/50% O 2/Ar gas mixtures with S ~ V = 6.3 and S ~ V = 6.5⁠, respectively. A further increase of the oxygen content in the sputter gas leads to a decrease of the current rectification with S ~ V = 5.5 and S ~ V = 4.0 for 75%/25% and 100%/0% O 2/Ar gas mixtures, respectively. The SBD with highest rectification of almost eight orders of magnitude was achieved for 25%/75% O 2/Ar. The effective barrier height Φ B , eff and the ideality factor η of all functional diodes were determined using the thermionic emission model and are shown in Fig. 3(b). As discussed above for the different contact metals, we observe an inverse correlation between Φ B , eff and η for all deposition methods, confirming that thermionic emission across an inhomogeneous barrier potential is the dominant current transport process. The lowest Φ B , eff values are determined for the non-oxidized thermally evaporated Pt contacts in agreement with investigations on monoclinic Ga 2 O 3⁠.25 The creation of oxidized contacts by the reactive sputtering process leads to a significant increase of Φ B , eff for all O 2/Ar gas mixtures. However, for the 100%/0% O 2/Ar gas mixture, the highest observed ideality factors among the entire sample set indicate the highest barrier potential variations resulting in the lowest rectification among the oxidized contacts. For this sample set the effective barrier height varies from 0.8 to about 1 eV. For the other sets of oxidized contacts, the ideality factor is lower and ranges from 2 to about 3.5. The effective barrier heights are highest for the 50%/50% O 2/Ar gas mixture with values between 0.9 and about 1.3 eV. The set of diodes fabricated with 25%/75% O 2/Ar gas mixtures performs better than the set fabricated with 75%/25% O 2/Ar gas mixture. In summary, oxidized Schottky contact metals typically have significantly higher current rectification than non-oxidized contacts. For the diodes investigated in this study, a 50%/50% O 2/Ar gas mixture gives the best diode performance at room temperature.
FIG. 2.
RT j V characteristics of evaporated and oxidized Pt SBDs prepared by reactive sputtering with oxygen admixture as labeled. Solid (dashed) lines represent the forward (reverse) voltage sweep direction.
View largeDownload slide

RT j V characteristics of evaporated and oxidized Pt SBDs prepared by reactive sputtering with oxygen admixture as labeled. Solid (dashed) lines represent the forward (reverse) voltage sweep direction.

FIG. 2.
RT j V characteristics of evaporated and oxidized Pt SBDs prepared by reactive sputtering with oxygen admixture as labeled. Solid (dashed) lines represent the forward (reverse) voltage sweep direction.
View largeDownload slide

RT j V characteristics of evaporated and oxidized Pt SBDs prepared by reactive sputtering with oxygen admixture as labeled. Solid (dashed) lines represent the forward (reverse) voltage sweep direction.

Close modal
FIG. 3.
(a) Room-temperature current rectification and (b) effective barrier height vs ideality factor of sets of Pt-SBDs on heteroepitaxial α - Ga 2 O 3(010) for deposition conditions as labeled.
View largeDownload slide

(a) Room-temperature current rectification and (b) effective barrier height vs ideality factor of sets of Pt-SBDs on heteroepitaxial α - Ga 2 O 3(010) for deposition conditions as labeled.

FIG. 3.
(a) Room-temperature current rectification and (b) effective barrier height vs ideality factor of sets of Pt-SBDs on heteroepitaxial α - Ga 2 O 3(010) for deposition conditions as labeled.
View largeDownload slide

(a) Room-temperature current rectification and (b) effective barrier height vs ideality factor of sets of Pt-SBDs on heteroepitaxial α - Ga 2 O 3(010) for deposition conditions as labeled.

Close modal
FIG. 4.
Dependence of the effective barrier height on the reciprocal temperature and corresponding fits (dashed lines) according to Eq. (5).
View largeDownload slide

Dependence of the effective barrier height on the reciprocal temperature and corresponding fits (dashed lines) according to Eq. (5).

FIG. 4.
Dependence of the effective barrier height on the reciprocal temperature and corresponding fits (dashed lines) according to Eq. (5).
View largeDownload slide

Dependence of the effective barrier height on the reciprocal temperature and corresponding fits (dashed lines) according to Eq. (5).

Close modal
Selected diodes of the differently prepared ensembles of Pt/ α - Ga 2 O 3 Schottky contacts were investigated by current-density-voltage measurements as a function of the measurement temperature T between 50 and 350 K in steps of 10 K. The characteristics of selected temperatures are shown in the supplementary material58 for evaporated and reactively sputtered Pt contacts. For the evaporated contact, the increase in series resistance with decreasing measurement temperature is significantly lower than for the oxidized contacts. In addition, the reverse current of both the evaporated and the pure oxygen sputtered contact is much higher than that of the other reactively sputtered diodes. All the characteristics were fitted by the TE theory to derive the effective barrier height Φ B , eff ( T ) and the ideality factor η ( T )⁠. According to the simple model of Werner and Güttler, which describes the barrier height variation by a Gaussian distribution,29 a decrease of Φ B , eff is expected for decreasing temperature and a linear relationship should be observed in a plot of the effective barrier height vs the reciprocal temperature for intermediate and higher temperatures. An extended model by Splith et al.28 allows to fit Φ B , eff ( T ) without restrictions on the considered temperature range by
(4)
(5)
For high temperatures,
(6)
the models of Werner and Güttler and Splith et al. coincide. Irrespective of the preparation method of our diode sets, an almost linear dependence is observed for the highest measuring temperatures (lowest reciprocal temperature), which is explained by the reduced thermal energy of the charge carriers confining the current transport more and more to regions of the metal-semiconductor interface with low barrier height. For T < 200 K (⁠ 1 / T > 0.005 K − 1⁠), the dependence of the effective barrier height on the reciprocal temperatures decreases as expected by the extended model of Splith et al. A fit of the data according to formula (5) yields the mean barrier height Φ ¯ B , 0 and the standard deviation σ 0 of the Gaussian-broadened barrier potential that are summarized in Table I. As reported for various metal-semiconductor interfaces,30 the standard deviation is about 1/10 of the mean barrier height (empirical Lajn’s rule). For Cu/ β - Ga 2 O 3⁠, a homogeneous barrier height and a standard deviation of 1.32 and 0.126 eV, respectively, have been reported34 also in accordance with Lajn’s rule as discussed in detail below.
TABLE I.

Mean barrier height Φ ¯ B , 0⁠, standard deviation σ0, and voltage coefficients ρ2 and ρ3 of Gaussian-broadened inhomogeneous Schottky barriers deposited by thermal evaporation (Ev) or reactive sputtering in an O2/Ar mixture as indicated. The parameters were determined from the temperature dependence of the effective barrier height and the ideality factor depicted in Figs. 4 and 5.

Method Φ ¯ B , 0 (eV)σ0 (eV)ρ2ρ3 (eV)
Ev 1.985 ± 0.1 0.273 ± 0.015 0.22 ± 0.01 −0.02 ± 0.005 
25/75 O2/Ar 2.57 ± 0.034 0.284 ± 0.0035 0.042 ± 0.015 −0.037 ± 0.000 84 
50/50 O2/Ar 2.46 ± 0.017 0.25 ± 0.002 0.19 ± 0.013 −0.019 ± 0.000 5 
75/25 O2/Ar 1.76 ± 0.039 0.20 ± 0.005 0.27 ± 0.0075 −0.021 ± 0.000 4 
100/0 O2/Ar 2.53 ± 0.55 0.33 ± 0.075 0.48 ± 0.013 −0.012 ± 0.000 6 
Method Φ ¯ B , 0 (eV)σ0 (eV)ρ2ρ3 (eV)
Ev 1.985 ± 0.1 0.273 ± 0.015 0.22 ± 0.01 −0.02 ± 0.005 
25/75 O2/Ar 2.57 ± 0.034 0.284 ± 0.0035 0.042 ± 0.015 −0.037 ± 0.000 84 
50/50 O2/Ar 2.46 ± 0.017 0.25 ± 0.002 0.19 ± 0.013 −0.019 ± 0.000 5 
75/25 O2/Ar 1.76 ± 0.039 0.20 ± 0.005 0.27 ± 0.0075 −0.021 ± 0.000 4 
100/0 O2/Ar 2.53 ± 0.55 0.33 ± 0.075 0.48 ± 0.013 −0.012 ± 0.000 6 
View Large
The shape of the potential distribution at the metal/semiconductor interface depends on the bias applied to the diode. This is immediately obvious when considering the distribution under flatband bias conditions V FB or higher forward bias conditions, where all lateral barrier inhomogeneities disappear. As soon as the bias is reduced below V FB⁠, barrier potential fluctuations begin to affect charge transport and these fluctuations increase in amplitude with decreasing applied bias. This causes the effective barrier height to be lower than the mean barrier height, since current transport preferentially occurs at lower barrier regions, as already discussed above. However, the dependence of the barrier distribution on the applied bias also leads to changes in the mean barrier height and the standard deviation. For a bias-independent ideality factor, this variation is described in the model of Werner and Güttler by the voltage coefficients ρ 2 and ρ 3⁠, which can be determined from the temperature dependence of the ideality factor,
(7)
These voltage coefficients express the variation of the mean barrier height and the standard deviation with bias,
(8)
(9)

In Fig. 5, η − 1 − 1 is plotted against the reciprocal temperature for the determination of the voltage coefficients by linear fitting. The fits and the derived parameters are depicted in the figure (dashed lines) and are summarized in Table I, respectively. Non-oxidized Cu contacts to heteroepitaxial β − Ga 2 O 3 have voltage coefficients ρ 2 = 0.15 ± 0.04 and ρ 3 = ( − 0.013 ± 0.001 ) eV that are very similar to those of non-oxidized Pt/ α - Ga 2 O 3⁠, although the RT ideality factor and the standard deviation of the latter are significantly higher. Similar ρ 3 values, together with the different values of σ 0 and η⁠, are attributed to the significantly larger mean barrier height Φ ¯ B , 0 = 1.98 eV of the non-oxidized Pt/ α - Ga 2 O 3 compared to the non-oxidized Cu/ β - Ga 2 O 3 diode with Φ ¯ B , 0 = 1.32 eV, which means that the forward bias of the Pt/ α - Ga 2 O 3 diode must be 660 mV higher than that of the Cu/ β - Ga 2 O 3 diode to achieve flatband conditions (for this bias, the barrier potential is by definition homogeneous). Among the oxidized platinum contacts, the lowest bias-dependent variation of the mean barrier height is found for an ambient during sputtering of 25%/75% O 2/Ar. Combined with the highest mean barrier value and high current rectification at room temperature, these preparation conditions are the most promising for achieving the highest diode performance.

FIG. 5.
Plot of η ( T ) − 1 − 1 vs the reciprocal temperature and fits (dashed lines) for the determination of voltage coefficients according to Eq. (7). The range, indicated by the dashed lines, corresponds to the temperature range considered for the corresponding fit.
View largeDownload slide

Plot of η ( T ) − 1 − 1 vs the reciprocal temperature and fits (dashed lines) for the determination of voltage coefficients according to Eq. (7). The range, indicated by the dashed lines, corresponds to the temperature range considered for the corresponding fit.

FIG. 5.
Plot of η ( T ) − 1 − 1 vs the reciprocal temperature and fits (dashed lines) for the determination of voltage coefficients according to Eq. (7). The range, indicated by the dashed lines, corresponds to the temperature range considered for the corresponding fit.
View largeDownload slide

Plot of η ( T ) − 1 − 1 vs the reciprocal temperature and fits (dashed lines) for the determination of voltage coefficients according to Eq. (7). The range, indicated by the dashed lines, corresponds to the temperature range considered for the corresponding fit.

Close modal

The derived barrier parameters are plotted in Fig. 6 along with data compiled from the literature.17,29,30,35–57 The published data cover various semiconductor materials, including both covalently and more ionically bonded single crystals and thin films, as well as amorphous semiconductors such as zinc-tin oxide. Also included are various diode arrangements and a variety of contact deposition methods. The figure clearly demonstrates that the correlation between the standard deviation σ 0 and the mean barrier height Φ ¯ B , 0⁠, originally established by Lajn et al.,30 also holds for the group-III sesquioxides. The figure includes data of the α - Ga 2 O 3 discussed here and of diodes to β - Ga 2 O 334,57 and bixbyite In 2 O 3⁠.54 The line in the figure corresponds to Lajn’s empirical rule σ 0 = L F Φ ¯ B , 0 with Lajn’s factor L F = 0.114⁠.30 

FIG. 6.
Compilation of mean barrier height and standard deviation of numerous Schottky barrier diodes realized on various materials. Data are compiled from this work (labeled as α - Ga 2 O 3) and from Refs. 17,37–52, and 53–57. The line represents σ 0 = L F Φ ¯ B , 0 with L F = 0.114 (Lajn’s rule).
View largeDownload slide

Compilation of mean barrier height and standard deviation of numerous Schottky barrier diodes realized on various materials. Data are compiled from this work (labeled as α - Ga 2 O 3⁠) and from Refs. 17,37–52, and 53–57. The line represents σ 0 = L F Φ ¯ B , 0 with L F = 0.114 (Lajn’s rule).

FIG. 6.
Compilation of mean barrier height and standard deviation of numerous Schottky barrier diodes realized on various materials. Data are compiled from this work (labeled as α - Ga 2 O 3) and from Refs. 17,37–52, and 53–57. The line represents σ 0 = L F Φ ¯ B , 0 with L F = 0.114 (Lajn’s rule).
View largeDownload slide

Compilation of mean barrier height and standard deviation of numerous Schottky barrier diodes realized on various materials. Data are compiled from this work (labeled as α - Ga 2 O 3⁠) and from Refs. 17,37–52, and 53–57. The line represents σ 0 = L F Φ ¯ B , 0 with L F = 0.114 (Lajn’s rule).

Close modal

In conclusion, we have compared the properties of sets of Au, Cu, Nb, Ni, and Pt Schottky contacts with heteroepitaxial PLD α - Ga 2 O 3⁠:Sn on m-plane sapphire. The lateral Schottky diodes were realized by thermal evaporation, inert, and reactive sputtering, respectively. In agreement with literature results, highest RT current rectification (about eight orders of magnitude) and highest RT effective barrier heights (about 1.3 eV) are obtained for oxidized platinum Schottky contacts. Selected non-oxidized and oxidized platinum Schottky barrier diodes were investigated by current-voltage measurements as a function of temperature. The variation of the effective barrier height and the ideality factor is consistent with thermionic emission across laterally varying potential barriers as introduced by Werner et al.29 and refined by Splith et al.27,28 The mean barrier height and standard deviation of oxidized (non-oxidized) platinum diodes ranges between 1.76 and 2.53 eV (is 1.98 eV) and 0.2 and 0.33 eV (0.27 eV), respectively. The values of the standard deviation σ 0 and the mean barrier height Φ ¯ B , 0 follow the empirical rule of Lajn et al.:30  σ 0 = L F Φ ¯ B , 0 with L F = 0.114⁠.

This work was partially performed in the framework of GraFOx a Leibniz-Science Campus partially funded by the Leibniz Association.

The authors have no conflicts to disclose.

S. Köpp: Data curation (lead); Formal analysis (equal); Investigation (lead). C. Petersen: Conceptualization (equal); Methodology (equal); Supervision (lead); Validation (equal); Writing – review & editing (equal). D. Splith: Data curation (equal); Formal analysis (equal); Software (lead); Validation (equal); Writing – review & editing (equal). M. Grundmann: Funding acquisition (equal); Project administration (equal); Writing – review & editing (equal). H. von Wenckstern: Conceptualization (equal); Data curation (equal); Funding acquisition (equal); Project administration (equal); Supervision (equal); Visualization (lead); Writing – original draft (lead); Writing – review & editing (lead).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
S. I.
Stepanov
,
V. I.
Nikolaev
,
V. E.
Bougrov
, and
A. E.
Romanov
,
Rev. Adv. Mater. Sci.
44
,
63
(
2016
).
Google Scholar
 
2.
H. von
Wenckstern
,
Adv. Electron. Mater.
3
,
1600350
(
2017
).
https://doi.org/10.1002/aelm.201600350
Google Scholar
Crossref
Search ADS
 
3.
S. J.
Pearton
,
F.
Ren
,
M.
Tadjer
, and
J.
Kim
,
J. Appl. Phys.
124
,
220901
(
2018
).
https://doi.org/10.1063/1.5062841
Google Scholar
Crossref
Search ADS
 
4.
A. J.
Green
 et al.,
APL Mater.
10
,
029201
(
2022
).
https://doi.org/10.1063/5.0060327
Google Scholar
Crossref
Search ADS
 
5.
H. Y.
Playford
,
A. C.
Hannon
,
E. R.
Barney
, and
R. I.
Walton
,
Chem. Eur. J.
19
,
2803
(
2013
).
https://doi.org/10.1002/chem.201203359
Google Scholar
Crossref
Search ADS
 
6.
M. B.
Maccioni
 and
V.
Fiorentini
,
Appl. Phys. Express
9
,
041102
(
2016
).
https://doi.org/10.7567/APEX.9.041102
Google Scholar
Crossref
Search ADS
 
7.
G. T.
Dang
,
T.
Kawaharamura
,
M.
Furuta
, and
M. W.
Allen
,
Electron Devices IEEE Trans.
62
,
3640
(
2015
).
https://doi.org/10.1109/TED.2015.2477438
Google Scholar
Crossref
Search ADS
 
8.
M.
Oda
,
J.
Kikawa
,
A.
Takatsuka
,
R.
Tokuda
,
T.
Sasaki
,
K.
Kaneko
,
S.
Fujita
, and
T.
Hitora
, “Vertical Schottky barrier diodes of α - Ga 2 O 3 3 fabricated by MIST epitaxy,” in
2015 73rd Annual Device Research Conference (DRC)
(IEEE, New York,
2015
), pp.
137
–
138
.
https://doi.org/10.1109/drc.2015.7175593
Crossref
Search ADS
 
9.
M.
Oda
,
R.
Tokuda
,
H.
Kambara
,
T.
Tanikawa
,
T.
Sasaki
, and
T.
Hitora
,
Appl. Phys. Express
9
,
021101
(
2016
).
https://doi.org/10.7567/APEX.9.021101
Google Scholar
Crossref
Search ADS
 
10.
J.
Bae
,
J.-H.
Park
,
D.-W.
Jeon
, and
J.
Kim
,
APL Mater.
9
,
101108
(
2021
).
https://doi.org/10.1063/5.0067133
Google Scholar
Crossref
Search ADS
 
11.
K.
Shiojima
,
Y.
Kawasumi
,
F.
Horikiri
,
Y.
Narita
,
N.
Fukuhara
,
T.
Mishima
, and
T.
Shinohe
,
Jpn. J. Appl. Phys.
60
,
108003
(
2021
).
https://doi.org/10.35848/1347-4065/ac2917
Google Scholar
Crossref
Search ADS
 
12.
D.
Yang
 et al.,
ACS Appl. Mater. Interfaces
14
,
5598
(
2022
).
https://doi.org/10.1021/acsami.1c21845
Google Scholar
Crossref
Search ADS
PubMed
 
13.
M.
Kneiß
,
D.
Splith
,
P.
Schlupp
,
A.
Hassa
,
H. von
Wenckstern
,
M.
Lorenz
, and
M.
Grundmann
,
J. Appl. Phys.
130
,
084502
(
2021
).
https://doi.org/10.1063/5.0056630
Google Scholar
Crossref
Search ADS
 
14.
S.-I.
Kan
,
S.
Takemoto
,
K.
Kaneko
,
I.
Takahashi
,
M.
Sugimoto
,
T.
Shinohe
, and
S.
Fujita
,
Appl. Phys. Lett.
113
,
212104
(
2018
).
https://doi.org/10.1063/1.5054054
Google Scholar
Crossref
Search ADS
 
15.
T.
Maeda
,
M.
Okigawa
,
Y.
Kato
,
I.
Takahashi
, and
T.
Shinohe
,
AIP Adv.
10
,
125119
(
2020
).
https://doi.org/10.1063/5.0028985
Google Scholar
Crossref
Search ADS
 
16.
R.
Jinno
 et al.,
Sci. Adv.
7
,
eabd5891
(
2020
).
https://doi.org/10.1126/sciadv.abd5891
Google Scholar
Crossref
Search ADS
 
17.
J.
Yu-Long
,
R.
Guo-Ping
,
L.
Fang
,
Q.
Xin-Ping
,
L.
Bing-Zong
,
L.
Wei
, and
L.
Ai-Zhen
,
Chin. Phys. Lett.
19
,
553
(
2002
).
https://doi.org/10.1088/0256-307X/19/4/332
Google Scholar
Crossref
Search ADS
 
18.
S.
Vogt
,
C.
Petersen
,
M.
Kneiß
,
D.
Splith
,
T.
Schultz
,
H. von
Wenckstern
,
N.
Koch
, and
M.
Grundmann
,
Phys. Status Solidi A
220
,
2200721
(
2022
).
https://doi.org/10.1002/pssa.202200721
Google Scholar
Crossref
Search ADS
 
19.
M. W.
Allen
,
R. J.
Mendelsberg
,
R. J.
Reeves
, and
S. M.
Durbin
,
Appl. Phys. Lett.
94
,
103508
(
2009
).
https://doi.org/10.1063/1.3089871
Google Scholar
Crossref
Search ADS
 
20.
S.
Müller
,
H. von
Wenckstern
,
F.
Schmidt
,
D.
Splith
,
R.
Heinhold
,
M.
Allen
, and
M.
Grundmann
,
J. Appl. Phys.
116
,
194506
(
2014
).
https://doi.org/10.1063/1.4901637
Google Scholar
Crossref
Search ADS
 
21.
H. von
Wenckstern
,
D.
Splith
,
F.
Schmidt
,
M.
Grundmann
,
O.
Bierwagen
, and
J. S.
Speck
,
APL Mater.
2
,
046104
(
2014
).
https://doi.org/10.1063/1.4870536
Google Scholar
Crossref
Search ADS
 
22.
J.
Michel
 et al.,
ACS Appl. Mater. Interfaces
11
,
27073
(
2019
).
https://doi.org/10.1021/acsami.9b06455
Google Scholar
Crossref
Search ADS
PubMed
 
23.
T.
Schultz
,
S.
Vogt
,
P.
Schlupp
,
H. von
Wenckstern
,
N.
Koch
, and
M.
Grundmann
,
Phys. Rev. Appl.
9
,
064001
(
2018
).
https://doi.org/10.1103/PhysRevApplied.9.064001
Google Scholar
Crossref
Search ADS
 
24.
S.
Vogt
,
H. von
Wenckstern
, and
M.
Grundmann
,
Appl. Phys. Lett.
113
,
133501
(
2018
).
https://doi.org/10.1063/1.5038941
Google Scholar
Crossref
Search ADS
 
25.
S.
Müller
,
H. von
Wenckstern
,
F.
Schmidt
,
D.
Splith
,
H.
Frenzel
, and
M.
Grundmann
,
Semicond. Sci. Technol.
32
,
065013
(
2017
).
https://doi.org/10.1088/1361-6641/aa6a8d
Google Scholar
Crossref
Search ADS
 
26.
C.
Hou
,
R. M.
Gazoni
,
R. J.
Reeves
, and
M. W.
Allen
,
Appl. Phys. Lett.
114
,
033502
(
2019
).
https://doi.org/10.1063/1.5079423
Google Scholar
Crossref
Search ADS
 
27.
D.
Splith
,
S.
Müller
,
H. von
Wenckstern
, and
M.
Grundmann
,
Proc. SPIE
10533
,
105330C
(
2018
).
https://doi.org/10.1117/12.2301370
Google Scholar
Crossref
Search ADS
 
28.
D.
Splith
,
S.
Müller
,
H.
von Wenckstern
, and
M.
Grundmann
,
Phys. Status Solidi A
218
,
2100121
(
2021
).
https://doi.org/10.1002/pssa.202100121
Google Scholar
Crossref
Search ADS
 
29.
J. H.
Werner
 and
H. H.
Güttler
,
J. Appl. Phys.
69
,
1522
(
1991
).
https://doi.org/10.1063/1.347243
Google Scholar
Crossref
Search ADS
 
30.
A.
Lajn
 et al.,
J. Vac. Sci. Technol. B
27
,
1769
(
2009
).
https://doi.org/10.1116/1.3086718
Google Scholar
Crossref
Search ADS
 
31.
H.
Sheoran
,
V.
Kumar
, and
R.
Singh
,
ACS Appl. Electron. Mater.
4
,
2589
(
2022
).
https://doi.org/10.1021/acsaelm.2c00101
Google Scholar
Crossref
Search ADS
 
32.
R. F.
Schmitsdorf
,
T. U.
Kampen
, and
W.
Mönch
,
J. Vac. Sci. Technol. B
15
,
1221
(
1997
).
https://doi.org/10.1116/1.589442
Google Scholar
Crossref
Search ADS
 
33.
R. F.
Schmitsdorf
 and
W.
Monch
,
Eur. Phys. J. B
7
,
457
(
1999
).
https://doi.org/10.1007/s100510050634
Google Scholar
Crossref
Search ADS
 
34.
D.
Splith
,
S.
Müller
,
F.
Schmidt
,
H. von
Wenckstern
,
J. J. V.
Rensburg
,
W. E.
Meyer
, and
M.
Grundmann
,
Phys. Status Solidi A
211
,
40
(
2014
).
https://doi.org/10.1002/pssa.201330088
Google Scholar
Crossref
Search ADS
 
35.
H.
von Wenckstern
,
G.
Biehne
,
R. A.
Rahman
,
H.
Hochmuth
,
M.
Lorenz
, and
M.
Grundmann
,
Appl. Phys. Lett.
88
,
092102
(
2006
).
https://doi.org/10.1063/1.2180445
Google Scholar
Crossref
Search ADS
 
36.
H.
Kim
,
H.
Kim
, and
D.-W.
Kim
,
J. Appl. Phys.
108
,
074514
(
2010
).
https://doi.org/10.1063/1.3493261
Google Scholar
Crossref
Search ADS
 
37.
F.
Cimilli
,
M.
Sağlam
,
H.
Efeoğlu
, and
A.
Türüt
,
Phys. B
404
,
1558
(
2009
).
https://doi.org/10.1016/j.physb.2009.01.018
Google Scholar
Crossref
Search ADS
 
38.
M.
Siva Pratap Reddy
,
A.
Ashok Kumar
, and
V.
Rajagopal Reddy
,
Thin Solid Films
519
,
3844
(
2011
).
https://doi.org/10.1016/j.tsf.2011.01.258
Google Scholar
Crossref
Search ADS
 
39.
Ö.
Güllü
,
M.
Biber
,
S.
Duman
, and
A.
Türüt
,
Appl. Surf. Sci.
253
,
7246
(
2007
).
https://doi.org/10.1016/j.apsusc.2007.03.002
Google Scholar
Crossref
Search ADS
 
40.
M.
Hudait
,
P.
Venkateswarlu
, and
S.
Krupanidhi
,
Solid-State Electron.
45
,
133
(
2001
).
https://doi.org/10.1016/S0038-1101(00)00230-6
Google Scholar
Crossref
Search ADS
 
42.
İ.
Dökme
,
Microelectron. Reliab.
51
,
360
(
2011
).
https://doi.org/10.1016/j.microrel.2010.08.017
Google Scholar
Crossref
Search ADS
 
43.
M.
Soylu
 and
B.
Abay
,
Microelectron. Eng.
86
,
88
(
2009
).
https://doi.org/10.1016/j.mee.2008.09.045
Google Scholar
Crossref
Search ADS
 
44.
A.
Ashok Kumar
,
V.
Janardhanam
,
V.
Rajagopal Reddy
, and
P.
Narasimha Reddy
,
Superlattices Microstruct.
45
,
22
(
2009
).
https://doi.org/10.1016/j.spmi.2008.10.016
Google Scholar
Crossref
Search ADS
 
45.
B.
Abay
,
G.
Cankaya
,
H. S.
Güder
,
H.
Efeoglu
, and
Y. K.
Yogurtcu
,
Semicond. Sci. Technol.
18
,
75
(
2002
).
https://doi.org/10.1088/0268-1242/18/2/302
Google Scholar
Crossref
Search ADS
 
46.
P.
McCafferty
,
A.
Sellai
,
P.
Dawson
, and
H.
Elabd
,
Solid-State Electron.
39
,
583
(
1996
).
https://doi.org/10.1016/0038-1101(95)00162-X
Google Scholar
Crossref
Search ADS
 
47.
S.
Acar
,
S.
Karadeniz
,
N.
Tugˇluogˇlu
,
A.
Selcuk
, and
M.
Kasap
,
Appl. Surf. Sci.
233
,
373
(
2004
).
https://doi.org/10.1016/j.apsusc.2004.04.011
Google Scholar
Crossref
Search ADS
 
48.
S.
Karadeniz
,
M.
Sahin
,
N.
Tuǧluoǧlu
, and
H.
Safak
,
Semicond. Sci. Technol.
19
,
1098
(
2004
).
https://doi.org/10.1088/0268-1242/19/9/005
Google Scholar
Crossref
Search ADS
 
49.
A.
Gümüş
,
A.
Türüt
, and
N.
Yalçin
,
J. Appl. Phys.
91
,
245
(
2002
).
https://doi.org/10.1063/1.1424054
Google Scholar
Crossref
Search ADS
 
50.
M.
Mamor
,
J. Phys.: Condens. Matter
21
,
335802
(
2009
).
https://doi.org/10.1088/0953-8984/21/33/335802
Google Scholar
Crossref
Search ADS
PubMed
 
51.
H.
Kim
,
S.
Kim
,
K.-K.
Kim
,
S.-N.
Lee
, and
K.-S.
Ahn
,
Jpn. J. Appl. Phys.
50
,
105702
(
2011
).
https://doi.org/10.1143/JJAP.50.105702
Google Scholar
Crossref
Search ADS
 
52.
S.
Zhu
,
R. V.
Meirhaeghe
,
C.
Detavernier
,
F.
Cardon
,
G.-P.
Ru
,
X.-P.
Qu
, and
B.-Z.
Li
,
Solid-State Electron.
44
,
663
(
2000
).
https://doi.org/10.1016/S0038-1101(99)00268-3
Google Scholar
Crossref
Search ADS
 
53.
D. H.
Lee
,
K.
Nomura
,
T.
Kamiya
, and
H.
Hosono
,
IEEE Electron Device Lett.
32
,
1695
(
2011
).
https://doi.org/10.1109/LED.2011.2167123
Google Scholar
Crossref
Search ADS
 
54.
H.
von Wenckstern
,
D.
Splith
,
F.
Schmidt
,
M.
Grundmann
,
O.
Bierwagen
, and
J. S.
Speck
,
APL Mater.
2
,
046104
(
2014
).
https://doi.org/10.1063/1.4870536
Google Scholar
Crossref
Search ADS
 
55.
P.
Schlupp
,
H.
von Wenckstern
, and
M.
Grundmann
,
Phys. Status Solidi A
214
,
1700210
(
2017
).
https://doi.org/10.1002/pssa.201700210
Google Scholar
Crossref
Search ADS
 
56.
Y.
Son
 and
R. L.
Peterson
,
Semicond. Sci. Technol.
32
,
12LT02
(
2017
).
https://doi.org/10.1088/1361-6641/aa95d2
Google Scholar
Crossref
Search ADS
 
57.
A.
Hassa
,
H.
von Wenckstern
,
D.
Splith
,
C.
Sturm
,
M.
Kneiß
,
V.
Prozheeva
, and
M.
Grundmann
,
APL Mater.
7
,
022525
(
2019
).
https://doi.org/10.1063/1.5054394
Google Scholar
Crossref
Search ADS
 
58.
See supplementary material online for current-density-voltage measurements of oxidized and non-oxidized platinum Schottky barrier diodes for various measuring temperatures.
© 2023 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Published open access through an agreement withTechnische Informationsbibliothek

Supplementary Material

Supplementary materials- zip file