【作者】 王超；

【导师】 季振国；

【作者基本信息】 浙江大学 ， 材料物理与化学， 2006， 博士

【摘要】 ZnO为Ⅱ-Ⅵ族化合物半导体材料，具有压电、热电、气敏、光电导等多种性能，在许多领域都有广泛的应用。近年来ZnO在光电领域的应用引起了人们的很大关注，这是由于ZnO在室温下禁带宽度为3.37eV，可以用来制备蓝光或紫外发光二极管（LEDs）和激光器（LDs）等光电器件。尤其是ZnO具有较高的激子束缚能（60meV），大于GaN的24meV和室温下的热能26meV，完全有可能在室温下实现有效的激子发射，因此在光电领域具有极大的发展潜力。ZnO在光电领域的应用依赖于高质量的n型和p型薄膜的制备。目前人们通过掺杂Ⅲ族元素已经获得了具有较好电学性能的n型ZnO。然而本征ZnO在内部容易产生各种施主型缺陷，发生自补偿作用使得p型ZnO薄膜难以制备，这种情况很大程度上限制了ZnO薄膜在光电器件方面的发展。因此如果通过理论和实验找到合适的受主杂质实现高质量的p型掺杂将对ZnO的实际应用起到极大的推动作用。尽管许多研究小组对ZnO各种可能的受主元素进行了理论研究，但获得的结论却不尽相同，并且有的与实验还存在着矛盾。因此本论文基于密度泛函理论进一步对ZnO的Ⅰ族、Ⅴ族掺杂元素进行了分析，试图找出最优的掺杂元素。在Ⅴ族元素当中，计算出N的受主电离能为0.31eV，远小于P（0.77eV）和As（0.89eV），并且N的受主形成能在这三种元素当中也是最小的，因此可以说N是Ⅴ族元素中比较理想的p型掺杂元素。Ⅰ族元素的情况相对Ⅴ族元素来说比较复杂。理论计算出Li、Na的受主电离能分别0.11和0.16eV，两者相差很小。而K的受主电离能则相对较大，为0.29eV，所以从电离能角度可以首先排除K元素。在对形成能研究时发现，与Na元素相比，Li替代位受主难以形成，而其间隙位施主形成则比较容易。综合考虑，Na作为ZnO的p型掺杂元素较好。而Na和N相比较，Na掺杂时会有间隙位施主补偿的影响，因此从理论上来说在所有的掺杂元素当中N是最理想的掺杂元素，Na掺杂效果相对于N来说稍差。在理论分析的指导下，我们首先制备了N掺杂p型ZnO薄膜。本论文中主要采用了两种方法实现N掺杂。一种是热氧化Zn₃N₂法，此方法可以克服N不易掺入ZnO的困难，并且工艺比较简单，具有较好的重复性。我们在实验中研究了热氧化温度和热氧化时间对薄膜性能的影响。XRD和紫外吸收光谱测试结果表明Zn₃N₂完全转变为ZnO所需温度为350℃。Hall测试表明热处理温度为400℃经过3个小时热处理后获得的ZnO薄膜具有最好的p型性能，载流子浓度为1.22×10¹⁷cm^-3，电阻率为83.2Ωcm。另外一种方法是以N₂为掺杂源的直流反应磁控溅射法。研究发现在一定的衬底温度下，空穴浓度会随着N₂流量的增加而增大，N₂流量为20sccm时获得的p型ZnO薄膜具有最高的空穴浓度。但N₂流量超过此值时，薄膜的导电类型会由p型转变为n型。而当N₂流量固定为20sccm时，400℃的衬底温度下制备的ZnO薄膜具有最优的p型性能。利用本方法获得的p型ZnO薄膜的空穴浓度为5.58×10¹⁷cm^-3，电阻率为8.44Ωcm。实验中发现N掺杂p型ZnO电学性能不稳定，为此本文对Na掺杂ZnO进行了研究。通过XRD、SEM和Hall等测试手段研究了衬底温度和靶材中Na含量对ZnO薄膜性能的影响。实验中发现靶材中Na含量为0.2％、衬底温度550℃时制备的p型ZnO薄膜具有较好的晶体质量，并且电学性能也很好，载流子浓度可达1.07×10¹⁸cm^-3左右，电阻率为27.6Ωcm。在对Na掺杂薄膜的电学稳定性测试时发现，存放一个月后，薄膜的电学性能并没有发生明显的退化，优于采用N掺杂制备的p型ZnO薄膜。更多还原

【Abstract】 Zinc oxide （ZnO） is II-VI compound semiconductor. It can be used in many areas due to its piezoelectric, thermoelectric, gas sensing and photoconducting multiple properties. Recently, the application of ZnO in optoelectronic devices has attracted great attention because ZnO can be used to fabricate blue or ultraviolet light emitting diodes （LED） and laser diodes （LD） due to its direct wide-bandgap （3.37 eV） at room temperature. In particular, ZnO has higher exciton binding energy （60 meV）, compared with 24 meV for GaN and 26 meV for the thermal activation energy at room temperature, which permits efficient excitonic emission processes at room temperature, therefore ZnO has great development potential in the field of optoelectronic devices. Its practical applications in this field depend on the fabrication of high quality n-type and p-type ZnO thin films. To date, high quality n-type ZnO thin films have already been achieved by doping with group-III elements. Howerver, undoped ZnO can easily generate some defects acting as donors, so it is difficult to obtain p-type ZnO due to self-compensating effect, which heavily limits the development of ZnO in the field of optoelectronic devices. Hence, it will add a considerable impetus to the development of ZnO if we find some appropriate acceptor impurities to realize high quality p-type ZnO by theories and experiments.Many research groups theoretically calculated various possible acceptor elements, but the results are not uniform and even some are conflicting with the experiments. Based on the density functional theory （DFT）, group-I and group-V impurities are further studied in this dissertation in order to find the best dopant for p-type ZnO among them. In group-V elements, the calculated acceptor ionization energy of N is 0.31 eV, which is much smaller than that of P （0.77 eV） and As （0.89 eV）, and the acceptor formation energy of N is lowest among them. So N element is ideal dopant for p-type ZnO in group-V elements. Compared with group-V elements, the group-I elements are complicated. The acceptor ionization energies of Li and Na calculated by DFT are 0.11 eV and 0.16 eV, respectively. The difference between them is small. Butfor K element, the acceptor ionization energy is relatively high and its value is about 0.29 eV. In the view of ionization energy, K element can be excluded. The calculated formation energies show that it is easy to form interstitial Li instead of substitutional Li, compared with Na element. Considering ionization energies and formation energies of group-I elements, Na is the best candidate for p-type ZnO. Compared with N, however, Na is worse elemental dopant source for p-type ZnO because of the formation of compensating interstitials.Thus N is the best dopant for p-type ZnO in theory.Guided by theoretical analysis, N-doped p-type ZnO thin films were first prepared in our experiments. In this dissertation, there are two methods to realize N-doped ZnO. One is thermal oxidization of Zn₃N₂ thin films. It can overcome the difficulty that N is not easy to dope into ZnO, and also has simple process and good reproducibility. Oxidation temperature and time were studied in experiments. The results of XRD and UV-Vis spectra showed that Zn₃N₂ completely changed to ZnO at 350℃. Hall results indicated that ZnO thin film, prepared at 400℃ for 3 hour, had the best p-type properties. Carrier concentration and resistivity of the film was 1.22×10¹⁷ cm^-3 and 83.2 Ωcm, respectively. The other is DC reactive magnetron sputtering with N₂ as dopant source. Hall results showed that when substrate temperature was fixed, hole concentration of as-grown thin films incrcased with the N₂ flow increasing and the thin film prepared at 20 sccm had the highest hole concentration. But if N₂ flow was higher than 20 sccm, the conduction type would change to n-type. When N₂ flow was fixed at 20 sccm, ZnO thin films obtained at 400℃ showed the best electrical properties. The hole concentration of ZnO thin films prepared by this way was 5.58×10¹⁷ cm^-3, with the lowest resistivity of 8.44 Ωcm.The electrical properties of N-doped p-type ZnO thin film were not stable in our experiments. So Na-doped ZnO thin films were also studied in this dissertation. The effects of substrate temperature and Na contents in target on properties of ZnO thins films were studied by XRD, SEM and Hall. It was found that the p-type ZnO thin film prepared at 550℃ and 0.2% Na contents in target had better crystallinity and electrical properties. Hall results showed that the highest carrier concentrication was1.07×10¹⁸ cm^-3 and the lowest resistive was 27.6 Ωcm. The stability of p-type conductive for Na-doped ZnO was better than that of N-doped ZnO and the electrical properties didn’t show obvious degradation after one month.更多还原