Research Article

1.534 µm Er electroluminescence in the RF magnetron deposited In₂O₃ : Er films on a Si substrate

Konstantin V. Feklistov^‡§, Aleksey G. Lemzyakov^|¶, Demid S. Abramkin^‡, Kirill A. Svit^‡, Alexey M. Pugachev^#, Vladimir A. Volodin^‡¤, Denis V. Marin^¤, Evgeny V. Spesivsev^‡, Ivan A. Azarov^¤‡, Leonid N. Safronov^‡, Sergey A. Kochubey^‡, Kirill S. Ershov^«, Aleksandr V. Kapishnikov^»¤, Alexander N. Shmakov^|, Alexander A. Shklyaev^¤‡, Yuriy A. Zhivodkov^‡

‡ Rzhanov Institute of Semiconductor Physics Siberian Branch of Russian Academy of Sciences, Novosibirsk, Russia

§ Academ Infrared LLC, Novosibirsk, Russia

| Synchrotron Radiation Facility “SKIF”, Koltsovo, Russia

¶ Budker Institute of Nuclear Physics, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia

# Institute of Automation and Electrometry, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia

¤ Novosibirsk State University, Novosibirsk, Russia

« Voevodsky Institute of Chemical Kinetics and Combustion, Novosibirsk, Russia

» Boreskov Institute of Catalysis, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia

Corresponding author: Konstantin V. Feklistov ( kofeklistov@yandex.ru )

© 2026 Konstantin V. Feklistov, Aleksey G. Lemzyakov, Demid S. Abramkin, Kirill A. Svit, Alexey M. Pugachev, Vladimir A. Volodin, Denis V. Marin, Evgeny V. Spesivsev, Ivan A. Azarov, Leonid N. Safronov, Sergey A. Kochubey, Kirill S. Ershov, Aleksandr V. Kapishnikov, Alexander N. Shmakov, Alexander A. Shklyaev, Yuriy A. Zhivodkov.

This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Citation: Feklistov KV, Lemzyakov AG, Abramkin DS, Svit KA, Pugachev AM, Volodin VA, Marin DV, Spesivsev EV, Azarov IA, Safronov LN, Kochubey SA, Ershov KS, Kapishnikov AV, Shmakov AN, Shklyaev AA, Zhivodkov YA (2026) 1.534 µm Er electroluminescence in the RF magnetron deposited In₂O₃ : Er films on a Si substrate. Modern Electronic Materials 12(2): 95-102. https://doi.org/10.3897/j.moem.12.2.205390

Abstract

In₂O₃ : Er films were synthesized on silicon substrates by the RF magnetron sputter deposition. The solid solution ((In_1-xEr_x)₂O₃) is formed here. The 1.534 µm erbium electroluminescence is observed by the current through the investigated heterostructure substrate-n-Si\In₂O₃ : Er-film\ITO-contact. The Er excitation model by the electron-hole recombination is proposed as follow. The electron current flows at the indium oxide conduction band. And the hole current flows at the channel in the middle of the In₂O₃ : Er bandgap. The hole channel is formed by the defect state density spreading from the valence band edge into the bandgap. Therefore the electron-hole recombination energy is lower than the indium oxide bandgap and equals to the 1.56 eV. Then the electron-hole recombination excites in resonance the third excited state of Er^{3+ 4}I_9/2 (1.53 eV). Then the nonradiative relaxation to the first excited state ⁴I_13/2 (0.81 eV) occurs. And finally the 1.534 µm radiative emission into ground state ⁴I_15/2 occurs.

Keywords

silicon, indium oxide, erbium, thin films, heterostructure, band structure, electrons, holes, electroluminescence

1. Introduction

The inter- and intro-chip data transfer by the optical interconnections is the ongoing perspective step in the development of computing systems [1, 2]. The key challenge here is the integration within chips, i.e. on the silicon (Si) substrate, of the Light Emitting Diodes (LEDs) with the 1.5 μm emission wave length appropriate for minimal optical fiber losses.

A small review of the main approaches to and features of LED on silicon integration is presented in paper [3]. The short list is presented here as follows. The mainstream approach is to transfer of the well-known A^IIIB^V (In_xGa_1-xAs) films on a silicon substrate [4–10] either by the bonding technique [4, 5] or by the direct growth on Si by the molecular-beam epitaxy (MBE) technique [4, 6, 7].

The alternative approach is to use the Er³⁺ inneratomic optical transition ⁴I_13/2 → ⁴I_15/2 with emission wavelength 1.54 μm [11, 12] well suited for minimal optical fiber losses. The straightforward doping of Si with erbium (Er) [13–17] gives inappropriate low quantum efficiency values [14, 16–21]. Nonetheless the Er optical fiber lasers are quite spread for now [22]. But the optical Er excitation by the same A^IIIB^V LEDs is used in them. There arises a question: is it possible to reach an efficient Er excitation straightforward by the electrical current? The Er and other rare-earth metal electroluminescence (EL) was observed in silicon oxide [23, 24] and other dielectrics, such as Si₃N₄ [25] and TiO₂ [26]. But the Er excitation mechanism was realized by hot electrons in high electric fields. It unlikely gives the practically high quantum efficiency. The Er EL excited by the electron-hole recombination mechanism was demonstrated in some optical transparent conducting oxides, e.g., in ZnO [27] and TiO₂ [28] for example. But the ZnO has the defect related energy levels within the bandgap, which gives the luminescence in the visual range [27]. TiO₂ has the defect related energy levels within the bandgap as well. And Er excites the high levels through it. It gives the Er luminescence in the visual range [28]. The defect related energy levels above are the reason for the low efficiency of the main infrared (IR) emission at the 1.54 μm wavelength. In the previous [3, 29], In₂O₃ : Er is proposed to obtain the Er EL. The Er photoluminescence (PL) in the In₂O₃ was observed earlier [29–31]. And the relative ITO (In₂O₃ : SnO₂) is well known as an optically transparent conducting oxide for electrical contacts [32, 33]. That were the reasons for considering In₂O₃ : Er.

In the previous papers, the barriers for the current injection from the Si substrate into the In₂O₃ film were found [3, 29]. They are equal the 0.14 eV for electrons and 0.3 eV for holes. Together with the Si bandgap E_gSi = 1.12 eV [34] the electron-hole recombination energy should be the 1.56 eV. It is quite lower than the In₂O₃ bandgap E_gIn2O3 = 2.69–2.93 eV [35, 36]. In [3] it is described by the suggestion of the hole transport channel related to defect states within the In₂O₃ : Er bandgap. Whether the found current barriers capable to provide injection and transport of electrons and holes toward each other and get Er EL? The aim of this work is to check the hypothesis above.

2. Materials and methods

The In₂O₃ : Er films were deposited on the n-type 6∙10¹⁴ cm^-3 phosphorus-doped Cz-Si (100) wafers with resistivity 7.5 Ohm·cm. The wafers were back-implanted with As⁺ 100 keV, 1·10¹⁵ cm^-2 and annealed at 1000 °С during 1 h in the Ar ambient in order to get the back-contact doping.

The Si wafer surface was cleaned by the RCA process [37] before the film deposition. The film deposition was performed by means of the RF magnetron sputtering on the BOC Edwards AUTO 500 system using the In₂O₃ and Er₂O₃ powder compound magnetron target. The Er concentration in the deposited films was fond by the X-ray photoelectron spectra (XPS) and equals 1 at.% [29]. The main deposition parameters were slightly different from [3, 29]:

– Ar flow rate 20 sccm, O₂ flow rate 2 sccm;

– working pressure P = 6∙10^-3 mbar;

– RF power W_RF = 100 Wt;

– supply power frequency 13.56 MHz;

– substrate temperature 100 °C;

– deposition time t = 5 h.

The deposited film thickness was 1 µm.

The EL emission was observed through top electric ITO (Indium Tin Oxide [32, 33]) contacts deposited through a 0.7×0.7mm² mask. The back-side contact was produced by the metallic In deposition over the whole back surface area.

The I–V curves for the structures Si/In₂O₃ : Er/Contact on n- and p-type conductivity silicon substrates have rectifying patterns as shown in former works [3, 29]. The EL emission was measured in the forward positive bias at the top ITO-contact. The stationary forward current +6 mA, i.e. the +1.2 A/cm² current density, and the voltage drop at the sample +6 V were mainly used for EL measurements.

The EL emission was measured at the Fourier-transform (FT) spectrometer Brucker Vertex 70. The spectrometer is equipped with the near-infrared range Raman module and liquid nitrogen-cooled germanium detector sensitive up to 1.7 µm including 1.54 µm required. The spectral resolution of the FTIR spectrometer is 4 cm^-1 or 4 A. The measurements were performed at room temperature. First, the Raman FTIR spectrum from the sample was measured excited with the apparatus neodymium laser at the 1064 nm wavelength and power 10 mWt. Focusing into 300×300 µm² area it gives power density of 14 Wt/cm². The Stocks part of Raman FTIR spectrum was measured. Second, the current was put through the sample and the EL spectrum was measured.

In addition to the microscopy data of [29] the X-ray diffraction (XRD) measurements were performed here using the X-ray powder diffractometer ARL X’tra (ThermoFisher, Switzerland, Basel) with Cu-Kα source (λ = 0.15418 nm).

3. Experimental results

3.1. X-ray diffraction

As it was shown in [29], the microstructure of the RF magnetron-deposited Si\In₂O₃ : Er films represents an array of ~7–10 nm diam. nanowires densely grouped into bunches of discrete 50–100 nm diam. nanocolumns. They are spread from the substrate to the film surface at the same small angle to the surface normal. All the nanowires have a single-crystalline body-centered cubic (bcc) lattice of typical bixbyite In₂O₃. But each nanowire has an individual lattice orientation within [29]. In addition to the microscopy analysis above, the XRD analysis is performed here. The films for XRD were deposited with the same parameters as in [3, 29], i.e., with the next difference from the main parameters above: Ar flow 8 sccm, O₂ flow 2 sccm, RF power W_RF = 120 Wt. Despite the slight difference of deposition parameters, the film microstructure is the same as above, i.e., nanowires bunched into nanocolumns.

Figure 1 presents the XRD of the sample. The XRD pattern is typical of a nanocrystalline film. There are reflections of only one phase on the pattern. The XRD pattern peaks are close to the bcc In₂O₃ Ia-3 space group (PDF-2 No. 6-416). However the In₂O₃ : Er phase peaks are shifted slightly to the left from the In₂O₃ ones. Therefore, the lattice constant of In₂O₃:Er is slightly bigger than that of In₂O₃. There are the In₂O₃ (222) peak position presented by the blue line and that of Er₂O₃ (PDF-2 No. 8-50) shown by the read line in Fig. 1 for visual perception. Both indium oxide and erbium oxide have the same crystalline structure except the lattice constants are different: the 1.0118 nm for In₂O₃ against the 1.0548 nm for Er₂O₃. The peaks at the XRD pattern are quite wide but still single ones, and there are no the double peaks in the XRD pattern. Therefore, the obtained In₂O₃ : Er films consist of the solid solution (In_1-xEr_x)₂O₃. There is no content of impurity phases like In, Er or pure oxides Er₂O₃, In₂O₃ observed in the XRD pattern.

Figure 1.

XRD of the In₂O₃ : Er film

The lattice constant was found from the experimental XRD pattern in Fig. 1 by the Polycrystal software package [38]. The obtained In₂O₃ : Er film has a lattice constant 1.0225±0.0004 nm. It is a bit lager than the one for In₂O₃ (1.0118 nm), but smaller than the one for Er₂O₃ (1.0548 nm). Since the solid solution (In_1-xEr_x)₂O₃ is obtained, then the lattice constant should linearly depend on the contend. For example, the solid solution (In_0.5Er_0.5)₂O₃ (PDF-2 No. 25-1145) has lattice constant 1.0338 nm exactly between In₂O₃ and Er₂O₃. Other solid solutions (In_1-xEr_x)₂O₃ with the lattice parameter 1.01213 nm in accordance with Er contend 1 at.% are observed in [39]. The linear extrapolation of the lattice constant obtained here presumes the Er₂O₃ content of 25 at.%. It obviously exceeds the Er contend in the magnetron target. Moreover, the Er content of 1 at.% was checked for the same films in [29] by the XPS method. Therefore, the oversized lattice constant here is not related to the Er contend. One possible reason could be the lattice vacancy defects. They could weaken the interatomic bonds in the lattice, increase the lattice constant and therefore shift the XRD pattern peaks. The other possible reason could be the residual extension strains in the crystalline lattice. They could shift the crystallographic plane distances and, as a result, the XRD pattern peaks. The lattice residual strain is often observed in the different magnetron deposited films [40]. An additional research is required to clarify the question of the oversized lattice constant here.

The blue lines at Fig. 1 indicate the strongest peaks of the theoretical In₂O₃ powder XRD pattern. They correspond, in their position, to the obtained experimental XRD pattern. Except for the maximal peak 222, there is some insignificant disagreement in peak amplitudes between the theoretical and experimental XRD pattern. Therefore, in addition to the complex film morphology (nanowires bunched into nanocolumns [29]), the crystallographic plane orientation in nanowires (texture) is similar mainly to disordered nanocrystals (powders). However, some experimental reflections, such as 400, 440, 622 and 642, have significantly different amplitude than the theoretical XRD pattern for powders, probably because of the some texture in the obtained films.

Another feature of the In₂O₃ : Er film experimental XRD pattern is large width of peaks. It evidences a small coherent scattering area (CSA). CSA could depend of the film nanocrystal size. The CSA size was estimated with the Sheerer formula [41]:

$D_{h k l} = \frac{n λ}{β \cos θ},$ (1)

where D_hkl is the CSA size (in angstroms) perpendicular to the hkl plain; n ≈ 1 is the nanocrystal shape coefficient; λ = 0.15418 nm is the apparatus wave length used; θ (in radians) is the scattering angle; β (in radians on 2θ scale) is the peak full width at a half maximum (FWHM). The estimation was performed for each key XRD peak and is presented in the Table 1. The obtained CSA size 6–10 nm fits the nanowire diameter obtained by the SEM [29]. It supports the nanocrystalline structure of the films.

Table 1.

Download as

CSV

XLSX

The reflexes parameters and CSA size

Reflex	Position (deg)	FWHM (deg)	CSA (nm)
211	21.375	1.629	5.8
222	30.32	1.008	9.8
400	35.08	1.692	5.7
440	50.49	1.788	5.7
642	69.12	3.369	3.3

3.2. Optical measurements

In Fig. 2 is the total combined spectrum of Raman, PL and EL obtained in different excitation conditions: 1) only the apparatus laser with the 1064 nm wavelength and 10 mWt power (black dashed line), 2) the laser together with the direct 6 mA current flow (green solid line), 3) only current without laser (red dashed line). The X-axis is the wavelength in nm units. The Y-axis is the emission intensity in arbitrary units. The upper X-axis gives some key points in energy units eV.

Figure 2.

Combined spectrum of Raman, silicon PL and Er EL

The black dashed line shows the typical silicon spectra obtained under laser excitation only. Since both the deposited In₂O₃ : Er film and the top ITO-contact are optically transparent under the exciting laser, then there is strong and narrow Raman peak out of silicon substrate observed at the 1126.3 nm. This Raman peak is due to the inelastic scattering by silicon optical phonons with wave vector k = 520 cm^-1 and energy 64.5 meV [34, 42, 43]. Since the Stocks part of the Raman FTIR spectrum was measured at the energy less than that of the excitation laser, then the observed k = 520 cm^-1 Raman peak is related to optical phonon emission.

Since the excitation laser energy (λ = 1064 nm, E = 1.17 eV) exceeds the silicon bandgap (E_g = 1.12 eV at T = 300 K [34]), then there is a wide peak of the silicon PL observed at the combined spectrum. This peak was also observed both for the silicon PL and EL in the literature [44, 45]. The observed silicon PL peak consists of the main peak and a secondary peak. The secondary peak is about one order amplitude less than the main one and is slightly shifted to the infrared direction. Therefore, it looks like a tail out of the main peak toward the IR region (Fig. 2). Since the silicon has the indirect bandgap, then the electron-hole recombination and photon emission requires the participation of one or two phonons for the main and secondary PL peak, accordingly [34]. The first one is the optical phonon with the wave vector within the X-valley of Brillouin zone. The electron conducting valley minimum in silicon is there [34]. So that the first phonon provides the impulse conservation during the electron-hole recombination. The first short-wavelength optical phonon has the energy of 58 meV [34, 42, 43]. Then the main PL peak edge is shifted toward IR at the E_g – 58 meV (see Fig. 2). The secondary PL peak edge is shifted toward IR at the E_g – 58–64 meV (see Fig. 2). The second phonon here is the above-mentioned Raman long-wavelength optical phonon within the Г-valley with k = 520 cm^-1 and energy 64.5 meV. The two-phonon process probability and, consequently, secondary PL peak amplitude are about an order of magnitude less than those of one-phonon ones. The silicon bandgap E_g = 1.12 eV edge at 300 K and that shifted by one- and two-phonon positions, in comparison with the silicon PL edges, are shown with blue vertical lines in Fig. 2. The observed discrepancy could be accounted for with the sample heating and consequent bandgap reduction [34]. The assumption of the sample temperature T = 340 K gives the bang gap E_g = 1.11 eV, and one- and two-phonon shifted positions (Fig. 2, red vertical lines) coincide well with the experimental main and second PL edges. The such heating under the laser excitation used looks quite possible.

Under the stationary direct current flow through the sample the multiplet peak at the 1534 nm wavelength appears at the combined spectrum (see Fig. 2, green solid line). It corresponds to the Er³⁺ ion radiative transition from the first excited state to the ground one: ⁴I_13/2 → ⁴I_15/2 [11, 12]. The multiplet shape of peak is explained by the Stark splitting of Er³⁺ energy levels in the electric fields of nearby atoms of the In₂O₃ crystal (Fig. 4). The control measurements under the current flow but without excitation laser, were performed as well. Here the multiplet Er EL peak is observed, but Raman and silicon PL peaks are absent (see Fig. 2, red dashed line).

Figure 3.

Schematic band diagram, injection barriers, current transport and Er³⁺ excitation mechanism in heterostructure n-Si\In₂O₃ : Er at forward bias +V at the top contact

Figure 4.

Excited states of Er³⁺ ion in accordance with the literature data [11, 12]. The Er excitation and emission mechanism

As it is shown at Fig. 2 by a black dashed line and a green solid line, when the direct current is turned on, then the silicon PL peak at the E_g region is increased. As well the backward current reduces the silicon PL. This is not shown at Fig. 2 since the most interesting Er EL was not found at the backward currents in the samples. The silicon PL dependence on the current dependence is explained by the space charge region (SCR) at the silicon surface. In the SCR, the inner electric field separates the electrons and holes, prevents them from recombination and reduces the silicon PL. In the case of direct current flow, the SCR is reduced and the silicon PL is increased. In the case of backward current flow, vice versa, the SCR is increased and the silicon PL is reduced.

As it is shown at Fig. 2, the Er EL peak amplitude is compared with that of the silicon PL one. The silicon PL is excited by the 10 mWt laser. The Er EL is excited by the 6 mA current and about a 6 V voltage drop at the sample, i.e. the 36 mWt electric excitation power is used. The silicon has the indirect bandgap and a small probability of the phonon-assisted photon emission. Therefore, despite the Er EL is observed at room temperature, the Er EL efficiency is still low in the obtained samples.

4. Discussion: the Er EL excitation mechanism

The barriers for the currents injection from the silicon substrate into the RF-magnetron deposited In₂O₃ : Er films were obtained in [3, 29] by the temperature dependence measurements of forward and backward currents. In Fig. 3 is the schematic band diagram of the investigated heterostructure in the forward bias (+V at the top contact) together with the electron and hole barriers found in [3, 29] put on it. Here the E_c and E_v is, accordingly, the conductance and valence bands, individual for silicon and erbium-doped indium oxide (In₂O₃ : Er). E_f is the Fermi level and E_g is the bandgap: E_gSi = 1.12 eV for silicon [34] and E_gIn2O3 = 2.69-2.93 eV for In₂O₃ [35, 36]. The forward barrier for the electron injection from the conduction band of n-silicon into the conduction band of indium oxide was found Φ_ef = 0.14 eV (see Fig. 3) [3, 29]. The forward barrier for the hole injection from the valence band of p-Si into indium oxide was also found Φ_hf = 0.3 eV [3]. This is a lot less than the heterostructure valence band discontinuity (E_v in Fig. 3). This is explained in [3] by the assumption of the defect state density (D_ds) spreading from the indium oxide valence band to the bandgap. This defect state density is formed by the defects of nano-crystalline indium oxide structure and creates the hole current channel inside the bandgap (E_ds in Fig. 3) [3].

In this case the electron-hole pairs (e^– – h⁺) are formed in indium oxide, when the electron is located in the conduction band (E_c) and the hole is in the defect channel (E_ds) inside the bandgap. The energy of such an electron-hole pair equals the sum of forward barrier for electrons, one for holes and the silicon bandgap (Φ_ef + Φ_hf + E_g(Si)) which gives E_c – E_ds = 0.14 + 0.3 + 1.12 = 1.56 eV (see Fig. 3, [3]). This electron-hole pair recombination energy excites the nearby Er ion. The hν designation is the IR photon emitted by excited erbium. The IR photon energy equals 0.81 eV (see Fig. 2), which is less than the energy transferred to excite the Er (1.56 eV, Fig. 3). This discrepancy will be discussed further. Thus, the small forward barrier for the electron injection from silicon into the In₂O₃ : Er film (Φ_ef = 0.14 eV, Fig. 3) and a relatively small barrier for the hole injection from the surface contact into the film (Φ_hb = 0.5 eV, Fig. 3, [3]) support the Er excitation mechanism as follows. The electrons and holes are injected over the barriers into the In₂O₃:Er film at the forward bias, flow in the electric field toward each other, recombine with each other and excite the Er³⁺ ions (see Fig. 3).

As it was mentioned above, the Er³⁺ ions are excited by electron-hole pairs recombination energy E_c – E_ds = 1.56 eV (see Fig. 3). But the erbium photonic emission has the energy 0.81 eV (see Fig. 2). The erbium excitation mechanism is proposed as follows. In Fig. 4 are the Er³⁺ excited states obtained from the literature data [11, 12]. As it is seen, the electron-hole recombination energy E_c – E_ds = 1.56 eV (see Fig. 3) coincide in resonance with the third excited Er³⁺ state ⁴I_9/2 1.53 eV (Fig. 4). Therefore, the electron-hole recombination excites this Er³⁺ state. Then the nonradiative relaxation without photon emission to the second ⁴I_11/2 and first ⁴I_13/2 excited Er³⁺ states occurs. And, finally, the radiative transition from the first excited state ⁴I_13/2 (0.81 eV) to the ground state ⁴I_15/2 occurs with the photon emission with the 1.54 µm wavelength.

Thus, the observed 1.534 µm Er EL supports the proposed Er excitation mechanism by means of the electron-hole recombination. The strong confirmation of the proposed mechanism requires further investigations.

Conclusion

The magnetron sputtered In₂O₃ : Er films were found to be the solid solution ((In_1-xEr_x)₂O₃). The obtained ((In_1-xEr_x)₂O₃) solid solution has the bcc In₂O₃ Ia-3 space group. The lattice constant of the ((In_1-xEr_x)₂O₃) solid solution was found equal to a = 1.0225 ± 0.0004 nm. The observed lattice constant exceeds the one for the given Er concentration 1 at.% in the film. This discrepancy could be related to the vacancy type defects or to the tensile lattice strain.

The EL at room temperature was observed for the In₂O₃ : Er films on the n-Si substrate with the top ITO-contact at the forward bias applied. The observed EL spectrum demonstrates the radiative transition from the first excited state Er^{3+ 4}I_13/2 to the ground state ⁴I_15/2 with the photon emission at the 1.534 µm wavelength.

The Er excitation mechanism proposed by the electron-hole recombination in the indium oxide and the electron-hole pair energy (1.56 eV) transfer to excite in resonance the third excited Er³⁺ state ⁴I_13/2 (1.53 eV), and by a subsequent nonradiative relaxation to the first excited state ⁴I_13/2 (0.81 eV) followed by the radiative transfer to the ground state ⁴I_15/2 with the photon emission.

Acknowledgments

The research was supported by FSI (grant No. 4235GS1/70543 from 27.10.2021) and partly by the Ministry of Education and Science of the Russian Federation. V.A.V., M.D.V. and K.A.V. (NSU) acknowledge the financial support from the Ministry of Education and Science of the Russian Federation by the grant: FSUS-2024-0020., S.A.A. (ISP SB RAS): FWGW-2025-0011, E.K.S. (ICKC SB RAS): FWGF-2026-0011, P.A.M. (IA&E SB RAS): FWNG-2024-0023, S.A.N (BIC SB RAS): FWUR-2024-0042. The XRD measurements were performed on the equipment of the Shared research center ATICKP NSU and CKP “SKIF”. Оptical measurements were realized on the equipment of CKP “High resolution spectroscopy of gases and condensed media” IA&E SB RAS. The film sputtering was performed at the shared research center SSTRC on the basis of the VEPP-4 – VEPP-2000 complex at BINP SB RAS. The sputtering target was manufactured by Phildal Holding Co., Ltd., China. The authors acknowledge E.D. Zanaev and N.V. Dudchenko for their chemical and thermal treatments.

References

1. Sun C., Wade M., Lee Y., Orcutt J.S., Alloatti L., Georgas M.S., Waterman A.S., Shainline J.M., Avizienis R.R., Lin S., Moss B.R., Kumar R., Pavanello F., Atabaki A.H., Cook H.M., Ou A.J., Leu J.C., Chen Y.-H., Asanović K., Ram R.J., Popović M.A., Stojanović V.M. Single-chip microprocessor that communicates directly using light. Nature. 2015; 528(7583): 534–538. https://doi.org/10.1038/nature16454

2. Atabaki A.H., Moazeni S., Pavanello F., Gevorgyan H., Notaros J., Alloatti L., Wade M.T., Sun Ch., Kruger S.A., Al Qubaisi H.M.K., Wang I., Zhang B., Khilo A., Baiocco Ch.V., Popović M.A., Stojanović V.M., Rajeev J.R. Integrating photonics with silicon nanoelectronics for the next generation of systems on a chip. Nature. 2018; 556(7701): 349–354. https://doi.org/10.1038/s41586-018-0028-z

3. Feklistov K.V., Lemzyakov A.G., Shklyaev A.A., Protasov D.Yu., Deryabin A.S., Spesivsev E.V., Gulyaev D.V., Pugachev A.M., Esaev D.G. The barriers for electron and hole injection from Si substrate into the RF magnetron-deposited In₂O₃: Er films. Izvestiya Vysshikh Uchebnykh Zavedenii. Materialy Elektronnoi Tekhniki = Materials of Electronics Engineering. 2023; 26(3): 234–247. (In Russ.). https://doi.org/10.17073/1609-3577j.met202305.529

4. Cornet Ch., Léger Y., Robert C. Integrated lasers on silicon. Elsevier Ltd.; 2016. 178 p. https://doi.org/10.1016/C2015-0-01237-0

5. Di L., Kurczveil G., Huang X., Zhang C., Srinivasan S., Huang Z., Seyedi M.A., Norris K., Fiorentino M., Bowers J.E., Beausoleil R.G. Heterogeneous silicon light sources for datacom applications. Optical Fiber Technology. 2018; 44: 43–52. https://doi.org/10.1016/j.yofte.2017.12.005

6. Norman J.C., Jung D., Wan Y., Bowers J.E. Perspective: The future of quantum dot photonic integrated circuits. APL Photonics. 2018; 3(3): 030901. https://doi.org/10.1063/1.5021345

7. Jung D., Norman J., Wan Y., Liu S., Herrick R., Selvidge J., Mukherjee K., Gossard A.C., Bowers J.E. Recent advances in InAs quantum dot lasers grown on on-axis (001) silicon by molecular beam epitaxy. Physica Status Solidi (A). Applications and Materials Science. 2019; 216(1): 1800602. https://doi.org/10.1002/pssa.201800602

8. Jung D., Herrick R., Norman J., Turnlund K., Jan C., Feng K., Gossard A.C, Bowers J.E. Impact of threading dislocation density on the lifetime of InAs quantum dot lasers on Si. Applied Physics Letters. 2018; 112(15): 153507. https://doi.org/10.1063/1.5026147

9. Mukherjee K., Selvidge J., Jung D., Norman J., Taylor A.A., Salmon M., Liu A.Y., Bowers J.E., Herrick R.W. Recombination-enhanced dislocation climb in InAs quantum dot lasers on silicon. Journal of Applied Physics. 2020; 128(2): 025703. https://doi.org/10.1063/1.5143606

10. Shang C., Hughes E., Wan Y., Dumont M., Koscica R., Selvidge J., Herrick R., Gossard A.C., Mukherjee K., Bowers J.E. High-temperature reliable quantum-dot lasers on Si with misfit and threading dislocation filters. Optica. 2021; 8(5): 749–754. https://doi.org/10.1364/OPTICA.423360

11. Carnall W.T., Fields P.R., Rajnak K. Electronic energy levels in the trivalent lanthanide aquo ions. I. Pr³⁺, Nd³⁺, Pm³⁺, Sm³⁺, Dy³⁺, Ho³⁺, Er³⁺, and Tm³⁺. The Journal of Chemical Physics. 1968; 49(10): 4424–4442. https://doi.org/10.1063/1.1669893

12. Gruber J.B., Henderson J.R., Muramoto M., Rajnak K., Conway J.G. Energy levels of single-crystal erbium oxide. The Journal of Chemical Physics. 1966; 45(2): 477–482. https://doi.org/10.1063/1.1727592

13. Ennen H., Schneider J., Pomrenke G., Axmann A. 1.54 mkm luminescence of erbium implanted III-V semiconductors and silicon. Applied Physics Letters. 1983; 43(10): 943–945. https://doi.org/10.1063/1.94190

14. Polman A. Erbium implanted thin film photonic materials. Journal of Applied Physics. 1997; 82(1): 1–39. https://doi.org/10.1063/1.366265

15. Kenyon A.J. Topical review: Erbium in silicon. Semiconductor Science and Technology. 2005; 20(12): R65–R84. https://doi.org/10.1088/0268-1242/20/12/R02

16. Coffa S., Franz'o G., Priolo F. Mechanism and performance of forward and reverse bias electroluminescence at 1.54 μm from Er-doped Si diodes. Journal of Applied Physics. 1997; 81(6): 2784–2793. https://doi.org/10.1063/1.363935

17. Coffa S., Franzò G., Priolo F. High efficiency and fast modulation of Er‐doped light emitting Si diodes. Applied Physics Letters. 1996; 69(14): 2077–2079. https://doi.org/10.1063/1.116885

18. Polman A., van den Hoven G.N., Custer J.S., Shin J.H., Serna R., Alkemade P.F.A. Erbium in crystal silicon: Optical activation, excitation, and concentration limits. Journal of Applied Physics. 1995; 77(3): 1256–1262. https://doi.org/10.1063/1.358927

19. Gusev O.B., Bresler M.S., Pak P.E., Yassievich I.N., Forcales M., Vinh N.Q., Gregorkiewicz T. Excitation cross section of erbium in semiconductor matrices under optical pumping. Physical Review B. 2001; 64(7): 075302. https://doi.org/10.1103/PhysRevB.64.075302

20. Priolo F., Franzo G., Coffa S., Carnera A. Excitation and nonradiative deexcitation processes of Er³⁺ in crystalline Si. Physical Review B. 1998; 57(8): 4443. https://doi.org/10.1103/PhysRevB.57.4443

21. Coffa S., Franz G., Priolo F., Polman A., Serna R. Temperature dependence and quenching processes of the intra-4f luminescence of Er in crystalline Si. Physical Review B. 1994; 49(23): 16313. https://doi.org/10.1103/PhysRevB.49.16313

22. Bradley J.D.B., Pollnau M. Erbium‐doped integrated waveguide amplifiers and lasers. Laser & Photonics Reviews. 2011; 5(3): 368–403. https://doi.org/10.1002/lpor.201000015

23. Wang S., Eckau A., Neufeld E., Carius R., Buchal Ch. Hot electron impact excitation cross-section of Er³⁺ and electroluminescence from erbium-implanted silicon metal-oxide-semiconductor tunnel diodes. Applied Physics Letters. 1997; 71(19): 2824–2826. https://doi.org/10.1063/1.120147

24. Krzyzanowska H., Ni K.S., Fu Y., Fauchet P.M. Electroluminescence from Er-doped SiO₂/nc-Si multilayers under lateral carrier injection. Materials Science and Engineering: B. 2012; 177(17): 1547–1550. https://doi.org/10.1016/j.mseb.2011.12.032

25. Berencen Y., Illera S., Rebohle L., Ramirez J.M., Wutzler R., Cirera A., Hiller D., Rodríguez J.A., Skorupa W., Garrido B. Luminescence mechanism for Er³⁺ ions in a silicon-rich nitride host under electrical pumping. Journal of Physics D: Applied Physics. 2016; 49(8): 085106. https://doi.org/10.1088/0022-3727/49/8/085106

26. Zhu C., Lv C., Gao Z., Wang C., Li D., Ma X., Yang D. Multicolor and near-infrared electroluminescence from the light-emitting devices with rare-earth doped TiO₂ films. Applied Physics Letters. 2015; 107(13): 131103. https://doi.org/10.1063/1.4932064

27. Yang Y., Li Y., Xiang L., Ma X., Yang D. Low-voltage driven ~1.54 μm electroluminescence from erbium-doped ZnO/p⁺-Si heterostructured devices: Energy transfer from ZnO host to erbium ions. Applied Physics Letters. 2013; 102(18): 181111. https://doi.org/10.1063/1.4804626

28. Yang Y., Jin L., Ma X., Yang D. Low-voltage driven visible and infrared electroluminescence from light-emitting device based on Er-doped TiO₂/p⁺-Si heterostructure. Applied Physics Letters. 2012; 100(3): 031103. https://doi.org/10.1063/1.3678026

29. Feklistov K.V., Lemzyakov A.G., Prosvirin I.P., Gismatulin A.A., Shklyaev A.A., Zhivodkov Y.A., Krivyakin G.K., Komonov A.I., Kozhukhov А.S., Spesivsev E.V., Gulyaev D.V., Abramkin D.S., Pugachev A.M., Esaev D.G., Sidorov G.Yu. Nanowired structure, optical properties and conduction band offset of RF magnetron-deposited n-Si/In₂O₃: Er films. Materials Research Express. 2020; 7(12): 25903. https://doi.org/10.1088/2053-1591/abd06b

30. Kim H.K., Li C.C., Nykolak G., Becker P.C. Photoluminescence and electrical properties of erbium-doped indium oxide films prepared by RF sputtering. Journal of Applied Physics. 1994; 76(12): 8209–8211. https://doi.org/10.1063/1.357882

31. Xiao Q., Zhu H., Tu D., Ma E., Chen X. Near-infrared-to-near-infrared downshifting and near-infrared-to-visible upconverting luminescence of Er³⁺-doped In₂O₃ nanocrystals. The Journal of Physical Chemistry C. 2013; 117(20): 10834–10841. https://doi.org/10.1021/jp4030552

32. Tahar R.B.H., Ban T., Ohya Y., Takahashi Y. Tin doped indium oxide thin films: Electrical properties. Journal of Applied Physics. 1998; 83(5): 2631–2645. https://doi.org/10.1063/1.367025

33. Hamberg I., Granqvist C.G. Evaporated Sn-doped In₂O₃ films: Basic optical properties and applications to energy-efficient windows. Journal of Applied Physics. 1986; 60(11): R123–R159. https://doi.org/10.1063/1.337534

34. Sze M. Physics of semiconductor devices. In 2 vol. New York: Wiley; 1981. Vol. 1. 300 p.

35. Weiher R.L., Ley R.P. Optical properties of indium oxide. Journal of Applied Physics. 1966; 37(1): 299–302. https://doi.org/10.1063/1.1707830

36. King P.D.C., Veal T.D., Fuchs F., Wang Ch.Y., Payne D.J., Bourlange A., Zhang H., Bell G.R., Cimalla V., Ambacher O., Egdell R.G., Bechstedt F., McConville C.F. Band gap, electronic structure, and surface electron accumulation of cubic and rhombohedral In₂O₃. Physical Review B. Condensed Matter. 2009; 79(20): 205211. https://doi.org/10.1103/PhysRevB.79.205211

37. Kern W., Puotinen D.A. Cleaning solutions based on hydrogen peroxide for use in silicon semiconductor technology. RCA Review. 1970; 31: 187–206. Available at: https://www.americanradiohistory.com/ARCHIVE-RCA/RCA-Review/RCA-Review-1970-Jun.pdf

38. Tsybulya S.V., Cherepanova S.V., Soloviyova L P. Polycrystal software package for IBM/PC. Journal of Structural Chemistry. 1996; 37(2): 332–334. https://doi.org/10.1007/BF02591064

39. de Brito A.S., Oliveira A.A., Valerio-Cuadros M.I., Tupan L.F.S., Gualdi A.J., Barco R., Ivashita F.F., de Araújo J.H., Torres M.A.M., Jr A.P. Structural, optical, and magnetic characterization of Er-doped In₂O₃ nanoparticles. Journal of Alloys and Compounds. 2024; 990: 174353. https://doi.org/10.1016/j.jallcom.2024.174353

40. Shayapov V.R., Bogoslovtseva A.L., Chepkasov S., Kapishnikov A., Mironova M.I., Geydt P. Highly textured AlN films deposited by pulsed DC magnetron sputtering with optimized process parameters. Solid State Communications. 2025; 397(4): 115821. https://doi.org/10.1016/j.ssc.2024.115821

41. Patterson A. L. The Scherrer formula for X-ray particle size determination. Physical Review. 1939; 56(10): 978. https://doi.org/10.1103/PhysRev.56.978

42. Nilsson G., Nelin G. Study of the homology between silicon and germanium by thermal-neutron spectroscopy. Physical Review B. 1972; 6: 3777. https://doi.org/10.1103/PhysRevB.6.3777

43. Brockhouse B.N. Lattice vibrations in silicon and germanium. Physical Review Letters. 1959; 2: 256. https://doi.org/10.1103/PhysRevLett.2.256

44. Green M., Zhao J., Wang A., Reece P.J., Gal M. Efficient silicon light-emitting diodes. Nature. 2001; 412: 805–808. https://doi.org/10.1038/35090539

45. Sveinbjornsson E.O., Weber J. Room temperature electroluminescence from dislocation-rich silicon. Applied Physics Letters. 1996; 69(18): 2686. https://doi.org/10.1063/1.117678