SEARCH

SEARCH BY CITATION

Keywords:

  • crystal growth;
  • physical vapor transport;
  • single crystals;
  • SnO2;
  • transparent semiconducting oxides

Abstract

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

SnO2 is a semiconductor with a wide optical bandgap (3.5 eV), which makes it an attractive transparent semiconducting oxide (TSO) for electronic and opto-electronic applications. At elevated temperatures it is, however, much more unstable than other TSOs (such as ZnO, Ga2O3, or In2O3). This leads to a rapid decomposition even under very high oxygen pressures. Our experiments showed that stoichiometric SnO2 does not melt up to 2100 °C, in contradiction to earlier published data. Bulk SnO2 single crystals, that could provide substrates for epitaxial growth, have not been reported so far. Hereby we report on truly bulk SnO2 single crystals of 1 inch diameter grown by physical vapor transport (PVT). The most volatile species during SnO2 decomposition is, in addition to oxygen, SnO, which is stable in the gas phase at high temperature and reacts again with oxygen at lower temperatures to form SnO2. We identified a relatively narrow temperature window, temperature gradients and a ratio of SnO/O2 for providing the best conditions for SnO2 single crystal growth. X-ray powder diffraction (XRD) proved the single SnO2 phase. Moreover, by selecting a suitable SnO/O2 ratio it was possible to obtain either n-type conductivity with electron concentrations up to 2 × 1018 cm−3 and electron mobilities up to 200 cm2 V−1 s−1, or insulating behavior. The crystals exhibited an optical absorption edge located at 330–355 nm, depending on the crystal orientation, and a good transparency over visible and near infrared (NIR) spectra.

1 Introduction

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

SnO2 has been known as a transparent semiconducting or conducting oxide (TSO or TCO), which in combination with its wide optical bandgap of about 3.5 eV attracts interest for numerous applications, such as transparent electrodes (e.g., for solar cells [1] and touch screens [2]), light emitting devices (e.g., LEDs [3]), transparent electronics (e.g., transparent FET transistors [4]), power electronics (Schottky diodes, MISFET or MESFET transistors), spintronics [5], architectural windows [6], gas sensors [7], and catalysis [6].

SnO2 crystallizes in the tetragonal rutile structure with P42/mnm space group and lattice constants of a = b = 4.7374 Å, c = 3.1864 Å. In this configuration, each Sn atom is surrounded by a distorted octahedron of O atoms [8].

Attempts to grow SnO2 by various techniques date back several decades. Most of undoped and doped SnO2 were grown in form of amorphous, crystalline, or nano-crystalline thin films or layers by deposition or epitaxial techniques, such as coating [9], sol–gel [10], spray pyrolysis [11], electron beam evaporation [12], magnetron sputtering [13], chemical vapor deposition (CVD) [14], pulsed laser deposition (PLD) [15], molecular beam epitaxy (MBE) [16], and like.

Bulk SnO2 single crystals in the form of platelets, rods, or needles were obtained from the gas phase by the physical vapor transport (PVT) method [17-19] and the chemical vapor transport (CVT) method [20, 21], as well as from flux [22-24]. Single crystals obtained from the PVT method [17-19] were grown at temperatures of 1350–1650 °C in the presence of oxygen, while those by the CVT method [22, 23] were grown at temperatures of 900–1250 °C by using I2, S, and/or Cl2 as transport gas. The crystals grown from the vapor phase were typically colorless upon slow cooling [17-19], but they were brown when quenched to room temperature (RT) [25, 26]. The as-grown crystals had usually high resistivity, but they could be doped with Sb [20, 25-28] in order to induce semiconducting behavior. Sb-doped SnO2 single crystals were blue [27, 28] with free electron concentration ranging from about 1016 cm−3 [20] to about 2 × 1019 cm−3 [28]. When using the flux method, the crystals were grown at temperatures in the range 1250–1300 °C from Cu2O [22, 24] and Bi2O3/V2O5 [23] fluxes in a Pt crucible. The crystal shapes were similar to those characteristic for the vapor phase (platelets, rods, needles). Electrical measurements of flux-grown SnO2 revealed very high resistivity of 109–1010 Ω cm [22] suggesting electrically insulating behavior of that material. Further electrical and/or optical properties of bulk SnO2 single crystals can be found in Refs. [19, 20, 22, 25, 27-30].

Relatively small crystals of size typically below 5 × 5 mm2 in cross-section and less than 20 mm in length which were grown by vapor and flux methods, are not suitable for preparation of substrates for homo- and heteroepitaxy. In fact, SnO2 substrates are currently not available except those prepared from natural cassiterite and moreover, both as-grown and undoped bulk SnO2 crystals are typically electrical insulators, which limit their applicability. As a result, single crystalline thin films of SnO2 are grown heteroepitaxially, e.g., on sapphire or rutile substrates.

Recently we have undertaken a research activity aimed at investigating growth conditions for growing truly bulk, pure SnO2 single crystals with semiconducting behavior. In particular, we carried out a number of experiments by using different growth equipment and wide spectrum of conditions, which in combination with thermodynamic investigations led us to understand the behavior of SnO2 at very high temperatures (up to 2100 °C) and find optimal PVT conditions. The optimized conditions yielded 1 inch diameter SnO2 single crystals. The thermodynamic investigations along with experimental data allowed a better estimate of thermodynamic entities, e.g., the melting point, which was often improperly quoted in the literature. Additionally, the study provided data on crystal quality, thermal stability, as well as fundamental electrical and optical properties of the grown SnO2 single crystals, including their sensitivity to heat treatments. The results presented in this paper open the way to the application of bulk SnO2 single crystals as substrates for homo- and heteroepitaxy and thus for the realization of the above-mentioned transparent electronic devices.

2 Thermodynamics

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

Before considering the thermodynamic behavior of SnO2 we shall first discuss the melting point of that compound. We performed a number of experiments in order to melt SnO2 under various conditions, including oxygen partial pressures ranging from 0.05 to 140 bar and temperatures up to 2100 °C. High oxygen-pressure experiments (5–140 bar) at temperatures up to about 1900–1950 °C were performed on a crucible-free, high-pressure optical floating zone of the company ScIDre at the Leibniz Institute for Solid State and Materials Research (IFW), Dresden. Despite the extremely high oxygen partial pressure and high temperature, there were no signs of melting of SnO2 while the evaporation rate remained quite high (about 10 wt% h−1). Higher temperatures of up to 2100 °C and 0.3 bar of oxygen partial pressure were achieved with use of an iridium crucible and CO2 growth atmosphere. Also in this case no melting was observed. Our experiments clearly evidences that the melting point or nearly-stoichiometric SnO2 (if exists) well exceeds 2100 °C. This is in contrast to values occasionally quoted in the literature of e.g., 1630 °C [31]. In fact, thermodynamic calculations (with use of the FactSage software) as shown in Table 1 revealed that the minimum oxygen partial pressure (inline image) required for stabilization of SnO2 is about 500 times higher than that required for stabilization of ZnO, which is considered to be almost the limit for melt growth techniques. Thus, melting of near-stoichiometric SnO2 is practically impossible, because it would require an oxygen partial pressure higher than that achievable in current growth furnaces. This also explains why SnO2 single crystals have never been grown from the melt. If the melting point of SnO2 was actually below 2000 °C, as proposed in some references, growing single crystals from the melt would have been possible.

Table 1. Minimum oxygen partial pressure required for stabilization of selected TSOs compounds at their melting points: β-Ga2O3, ZnO, In2O3, SnO2, normalized with respect to β-Ga2O3 (FactSage calculations)
TSOβ-Ga2O3ZnOIn2O3SnO2
relative oxygen partial pressure1∼100∼1000∼50 000

From the thermodynamic study it becomes clear, that growing truly bulk SnO2 single crystals is a great challenge from the technical and industrial point of view, although the first reports on growth of very small crystals dates back about half a century.

Figure 1 shows the stability diagram of SnO2 separating areas of SnO2 solid, Sn liquid, and gas phase. Here SnO2 liquid does not appear. At low and moderate temperatures (<1800 °C) oxygen released from CO2 (when used as the growth atmosphere) is sufficient to stabilize, at least partly, the solid SnO2, however, at higher temperatures only the gas phase is stable even at high oxygen partial pressures, which confirms our experiments. This corresponds to SnO2 decomposition and massive evaporation.

image

Figure 1. Stability diagram of SnO2 versus oxygen partial pressure and oxygen partial pressure provided by the self-adjusting dynamic growth atmosphere of CO2 (total pressure of 1 bar).

Download figure to PowerPoint

The impossibility of melting and easy decomposition and evaporation of SnO2 led us to the conclusion that the PVT method is possibly the most appropriate technique for obtaining truly bulk SnO2 single crystals. This has been experimentally recognized long time ago (see Section 'Introduction'). The decomposition of SnO2, occurs via liberation of oxygen and tin monoxide, the latter being stable in the gas phase at high temperatures:

  • display math(1)

The reaction (1) actually indicates the route to the growth of SnO2 by PVT via recombination of SnO and oxygen as it will be discussed below. Figure 2 shows the partial pressure of SnO as the result of the SnO2 decomposition for two oxygen partial pressures (obtained by the FactSage software), i.e., for oxygen released from CO2 and pure oxygen at 100 bar overpressure. An increase of the oxygen partial pressure does not lead to a decrease of the SnO partial pressure by a similar factor. For example, the increase of the oxygen partial pressures by a factor of 104–105 with respect to O2 provided by CO2 leads to a decrease of SnO partial pressure by about 102 only. To achieve an effective growth process from the vapor phase [according to reaction (1)] two crucial parameters must be adjusted, namely oxygen partial pressure and temperature. For high oxygen partial pressures high temperatures are required (>1700 °C), but on the other hand at too low temperatures (<1200 °C) the SnO partial pressure might be too low for effective transport. We have found that CO2 is a very good oxygen source to grow SnO2 by the PVT method, since its partial pressure changes non-linearly with temperature and crosses the SnO partial pressure line at moderate temperature (just below 1500 °C). Therefore, when using CO2 as growth atmosphere a temperature around or just above 1500 °C seem to be optimal for the PVT growth.

image

Figure 2. Relation between partial pressure of gaseous SnO and SnO2 as a function of temperature for two different oxygen partial pressures: 100 bar and that obtained from decomposition of CO2 at 1 bar (nonlinear).

Download figure to PowerPoint

To obtain SnO2 single crystals from the vapor phase, SnO must be oxidized back to SnO2. This can be achieved by creating a sufficient temperature difference, ΔT, between the SnO2 source at THIGH and a substrate on which the deposition takes place, according to the reaction:

  • display math(2)

If additionally carbon dioxide as oxygen supplying agent of the growth atmosphere is taken into account, Eqs. (1) and (2) can be expanded to:

  • display math(3)

and the Gibbs free energy balance ΔG is positive for small T and negative for large T. Equilibrium (ΔG = 0) is reached for 1486 °C which corresponds to the intersection of the SnO(g) and O2(g) functions in Fig. 2 at inline image = 1 bar.

Figure 3 shows the amount of solid SnO2 created by oxidation of gaseous SnO as a function of source and substrate temperatures. For the calculation it was assumed that 1 mole SnO2 reacts at the source temperature with CO2 and the resulting gas phase is transported to the substrate temperature where crystallization of SnO2 takes place. The higher the temperature difference ΔT, the higher the oxidizing rate, however, a too high oxidizing rate may lead to an excessive growth rate which in turn affects negatively the crystal quality.

image

Figure 3. Amount of solid SnO2 obtained from oxidation of gaseous SnO as a function of source and substrate temperatures for 1 mole of source SnO2.

Download figure to PowerPoint

According to the present study it appears that the temperature difference ΔT between the SnO2 source and the substrate should be maintained in the range 100–200 K in order to assure an optimal growth rate and a full conversion of gaseous SnO into solid SnO2.

3 Crystal growth

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

The growth of SnO2 single crystals was performed by PVT taking into account the results of the above-mentioned thermodynamic calculations. The growth furnace included an iridium crucible (also Pt crucible is appropriate, depending on the operating temperature) of 40 mm diameter and CO2 growth atmosphere at atmospheric pressure, although other pressures between 0.5 and 2 bar were tried as well. A growth chamber was filled up with the gaseous growth atmosphere at desire pressure with no gas flow therethrough. The source material was SnO2 raw material from Fox Chemicals of 99.99% purity (4 N). We employed 30 mm diameter epi-ready R-plane sapphire and 25 mm diameter (110) rutile (TiO2) as substrates. The substrate was located on the top of the crucible by means of a ring-shaped iridium holder with a seat for the substrate. The initial distance between the substrate and the source material was 10–15 mm and it was increasing with time due to evaporation. The crucible with the substrate was surrounded by a thermal insulation designed and configured such that the temperature difference between the SnO2 starting material within the crucible and the center of the substrate, ΔT, was about 100–130 K, which enabled full conversion of gaseous SnO into solid SnO2. Moreover, the top insulation was so configured, that the radial temperature gradient across the substrate was about 2 K mm−1 (with the lowest temperature at the center and the highest at the rim of the substrate) assuring inverse dome-shaped temperature profile and thus similar crystal shape. The outer diameter of the thermal insulation was 90 mm, while its thickness below and above the crucible was 50–60 mm. In this arrangement an RF coil was used for inductive heating, although resistive heating is a good alternative as well.

The growth experiments were performed at substrate temperatures on both sides of the crossing point between oxygen and SnO partial pressures (which is just below 1500 °C), as demonstrated in Fig. 4. Therefore, on the left-hand side of that crossing point the growth conditions could be defined as oxygen-rich, while on the right-hand side as SnO rich conditions.

image

Figure 4. Relation between free electron concentration and growth conditions: substrate temperature and p(SnO)/p(O2) ratio. The temperature difference, ΔT, between the source and the substrate was 100–130 K. The black square at 1400 °C indicates insulating state.

Download figure to PowerPoint

Figure 4 reports a most remarkable finding of the present study, that under oxygen-rich conditions the grown SnO2 crystals are electrical insulators, while the crystals grown under SnO-rich conditions showed n-type semiconducting behavior with the free electron concentration ranging between 3 × 1017 and 2 × 1018 cm−3. Under oxygen-rich conditions the growth rate was very low (<0.2 g h−1) and the resulting crystal quality was not very good. On the other hand, when SnO partial pressure was too high, the growth rate was also too high (>1 g h−1) and the crystal quality deteriorated. The best results were obtained for SnO partial pressures approximately equal or a bit higher than the oxygen partial pressure, correspondingly to substrate temperatures ranging between 1480 and 1580 °C and growth rates of about 0.5–0.6 g h−1.

An example of a bulk, 1-inch diameter SnO2 single crystal grown under such conditions, along with one-side epi-polished wafers, is shown in Fig. 5.

image

Figure 5. Bulk SnO2 single crystal of 1 inch diameter grown on an R-plane sapphire substrate and one-side epi-ready SnO2 wafers.

Download figure to PowerPoint

It is to be noted that the crystals obtained on R-plane sapphire were of better quality than those obtained on (110)-plane rutile substrates, that is surprising when thinking that SnO2 and TiO2 have the same crystallographic structure. The orientation of SnO2 obtained on R-plane sapphire substrate was close to 〈100〉 or 〈001〉.

It is important to note, that the observed crystal coloration is directly correlated with the ratio of O2 and SnO partial pressures: p(SnO)/p(O2). For p(SnO)/p(O2) < 1 (oxygen-rich conditions), the crystal color was orange or pink-orange, for p(O2)/p(SnO) ≈ 1 it was brownish, while for p(SnO)/p(O2) > 1 it was dark violet or almost black. In other words, the higher the SnO partial pressure the darker the crystal appearance. The differences in the coloration can probably be explained in terms of stoichiometry deviations and defect density.

Since SnO is a main precursor in the growth of SnO2 single crystals by PVT method, it was important to check whether SnO had been incorporated as such into SnO2. X-ray powder diffraction (XRD) performed on SnO2 crystals obtained under different conditions proved that within the accuracy of that method, SnO2 was of single phase, as clearly visible from Fig. 6. This is not surprising, because SnO is not stable at low temperatures in the solid phase, as already concluded from the thermodynamic study.

image

Figure 6. XRD pattern of powdered SnO2 single crystal.

Download figure to PowerPoint

The quality of the obtained SnO2 single crystals was found to be relatively good. This can be inferred from the rocking curve shown in Fig. 7, the full width at half maximum (FWHM) of which is typically below 100 arc sec and sometimes as small as about 35 arc sec. Further confirmation of relatively good crystal quality is well demonstrated by the high resolution transmission electron microscopy image in Fig. 8 (taken on FEI Titan 80–300 microscope with aberration correction) showing a defect-free area of the (001) plane of the SnO2 single crystal. Moreover, HRTEM investigations of different samples obtained from crystals grown under different conditions did not reveal any presence of SnO phase.

image

Figure 7. Rocking curve of bulk SnO2 220 peak (slit width = 2 mm).

Download figure to PowerPoint

The following residual impurities identified in the nominally undoped SnO2 single crystals by the electrothermal vaporization inductively-coupled plasma optical emission spectroscopy ETV-ICP OES are as follows (concentrations in ppm): Al = 10 ± 3, Si < 5, Fe < 2, Cu < 1, Ca < 1, Mg < 0.4, Zr < 0.3, Mn < 0.1. The most important contaminant, Al, comes likely from alumina insulation.

image

Figure 8. HRTEM image of (001) face of bulk SnO2 single crystal. Green and red spots are simulated atom positions.

Download figure to PowerPoint

We performed an analysis of the thermo-chemical stability of obtained bulk SnO2 single crystals, as this is a key question in view of applications as epitaxial substrates. We carried out thermal annealing in three different atmospheres: oxidizing (oxygen), neutral (argon), and reducing (5% H2 + Ar) at temperatures up to 1400 °C for 10 or 20 h. Figure 9 shows the mass change in percentage with respect to the initial sample mass as a function of annealing temperature (negative mass change indicates decomposition). The crystals are stable up to 1200 °C under oxidizing and neutral atmospheres and 20 h annealing time (no signs of decomposition), and up to about 700 °C and 10 h annealing time under highly reducing atmosphere containing hydrogen, beyond which a remarkable decomposition takes place. This suggests that SnO2 single crystals grown from the vapor phase possess a sufficient thermo-chemical stability to be used as substrates for various epitaxial techniques. These results are quite similar to those obtained for melt-grown bulk In2O3 single crystals [32].

image

Figure 9. Stability of bulk SnO2 single crystals vs. temperature at three different annealing atmospheres: 5%H2 + Ar, Ar and O2. Negative mass change indicates decomposition.

Download figure to PowerPoint

4 Electrical and optical properties

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

We performed Hall effect measurements on as-grown SnO2 single crystals in van-der-Pauw configuration using In[BOND]Ga Ohmic contacts. A summary of the free electron concentration, electron mobility and resistivity at RT in as-grown crystals is reported in Table 2. SnO2 crystals grown under oxygen-rich conditions are electrical insulators, while those grown under SnO-rich conditions are n-type semiconductors with free electron concentrations of 3 × 1017–2 × 1018 cm−3 and electron mobilities between 125 and 200 cm2 V−1 s−1. These values are comparable with those reported for other bulk TSO single crystals, such as In2O3 [32, 33], β-Ga2O3 [34], and ZnO [35].

Table 2. Electrical properties at RT of as-grown bulk SnO2 single crystals
p(SnO)/p(O2)carrier conc. (cm−3)carrier mobility (cm2 V−1 s−1)resistivity (Ω cm)conduction type
≥13 × 1017–2 × 1018125–2020.03–0.11n-type semiconductor
<1very highinsulator

The bulk SnO2 single crystals present a remarkable sensitivity to heat treatments in terms of electrical properties. When annealed in oxygen (or any other oxidizing atmosphere) at temperatures not exceeding 800 °C (Fig. 10), the free electron concentration decreases by about one order of magnitude and saturates for long annealing times. On the other hand, annealing at higher temperatures, such as 1000 °C (Fig. 10), turns SnO2 single crystals from semiconducting to insulating state already after relatively short annealing time (about 7.5 h). Figure 11 shows the relation between the free electron concentration and annealing temperature for 20 h annealing time in two atmospheres: neutral (argon) and oxidizing (oxygen). From these results it is clear that SnO2 single crystals switch from semiconducting to insulating state at about 1000 °C in case of neutral and at about 800 °C in case of oxidizing atmosphere. It is worth to mention, that insulating SnO2 single crystal could be brought from the insulating back to the semiconducting state after subsequent annealing in a reducing atmosphere containing hydrogen (will be reported separately).

image

Figure 10. Free electron concentration of bulk SnO2 single crystals versus annealing time in oxygen at 800 and 1000 °C.

Download figure to PowerPoint

image

Figure 11. Free electron concentration of bulk SnO2 single crystals versus annealing temperature in argon and oxygen for 20 h.

Download figure to PowerPoint

The optical transmission of SnO2 single crystals was measured at RT using a commercial spectrophotometer (Lambda 19, Perkin-Elmer) in the wavelength range 200–2500 nm with the incident light possessing a preferred, instrumentally induced polarization of the electric field vector E horizontal with respect to the spectrometer setup. Figure 12 illustrates the transmittance spectra of as-grown SnO2 single crystals cut and polished on different surface orientations: (100), (001), and (110). In case of the (001) sample E was perpendicular to the c-axis, while for the other two surface orientations E was preferentially parallel to c. The strong dichroism of the fundamental absorption edge (see inset of Fig. 12) is in accordance with literature reports of experimental data [36, 37] and their recent, theoretical interpretation [38]. The difference in absorption edge between (001) and other orientations arises from the tetragonal structure having the optical axis along 〈001〉. From more detailed absorption measurements (not shown here) we could determine an energy gap of 3.49 eV (355 nm) at RT for E ⊥ c, while for E ∥ c the absorption is suppressed so that the onset of the absorption seems to be shifted to 3.76 eV (330 nm). Although some residual absorption is present the transmission curves indicate a good transparency of the crystals in the visible range. The absorption in the near infrared (NIR) spectrum (>800 nm) is approximately proportional to the square of the wavelength and hence due to free carriers, in the present case due to free electrons of a density of about 1018 cm−3. SnO2 shows full transparency in the visible and NIR spectra (dashed line in Fig. 12) when in insulating state, obtained after annealing in oxygen as demonstrated in Figs. 10 and 11.

image

Figure 12. Transmittance spectra (not corrected for reflection) of as-grown bulk SnO2 single crystals of different orientations and having free carrier concentration around 1018 cm−3. Annealed SnO2 exhibiting insulating state is shown by the dashed line. Sample thickness was 0.5 mm.

Download figure to PowerPoint

Figure 13 shows the transmittance spectra of SnO2 wafers, the free electron concentrations of which have been presented in Fig. 10 (red curve), after annealing in oxygen at 1000 °C for different annealing times. As expected, the annealed SnO2 single crystals with high resistivity show full transmittance in the NIR spectrum as there is substantially no NIR-absorption due to free electrons. This happens for annealing durations longer than 5 h. Moreover, the annealed insulating crystals became colorless, although some minor, defect-dependent extinction in the visible spectrum may still be present.

image

Figure 13. Transmittance spectra of bulk SnO2 single crystals annealed at 1000 °C in oxygen corresponding to different annealing times.

Download figure to PowerPoint

5 Conclusions

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

We have successfully grown truly bulk SnO2 single crystals of 1 inch diameter by the PVT method. By choosing an appropriate oxygen partial pressure with respect to SnO partial pressure and operating temperature it was possible to obtain either n-type semiconducting or insulating SnO2 single crystals. Moreover, switching between insulating and semiconducting state could also be achieved by suitable post-growth annealing at temperatures exceeding 800 °C in oxidizing or hydrogen-containing atmosphere. The free electron concentration of as-grown semiconducting SnO2 single crystals was between 3 × 1017–2 × 1018 cm−3 with an electron mobility in the range of 125–200 cm2 V−1 s−1. Transmittance spectra show a steep absorption edge at 330–355 nm depending on the crystal orientation and a high transparency in the visible and NIR wavelength regions. Additionally, we also proved experimentally and theoretically that growing of SnO2 single crystals from the melt is practically impossible and that the melting point of stoichiometric SnO2 (if at all existing) well exceeds 2100 °C.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References

The authors express their gratitude to Michael Schulze from IFW Dresden for Optical Floating Zone experiments. This work was supported by the Deutsche Forschungsgemeinschaft grant No. FO 558/3-1.

References

  1. Top of page
  2. Abstract
  3. 1 Introduction
  4. 2 Thermodynamics
  5. 3 Crystal growth
  6. 4 Electrical and optical properties
  7. 5 Conclusions
  8. Acknowledgements
  9. References