We report on a comprehensive analysis of titanium boride thin films by x-ray photoelectron spectroscopy (XPS). Films were grown by both direct current magnetron sputtering and high-power impulse magnetron sputtering from a compound TiB₂ target in Ar discharge. By varying the deposition parameters, the film composition could be tuned over the wide range $1.3\lesssim \mathrm{B}/\mathrm{Ti}\lesssim 3.0$, as determined by elastic recoil detection analysis and Rutherford backscattering spectrometry. By comparing spectra over this wide range of compositions, we can draw original conclusions about how to interpret XPS spectra of TiB_x. By careful spectra deconvolution, the signals from Ti–Ti and B–B bonds can be resolved from those corresponding to stoichiometric TiB₂. The intensities of the off-stoichiometric signals can be directly related to the B/Ti ratio of the films. Furthermore, we demonstrate a way to obtain consistent and quantum-mechanically accurate peak deconvolution of the whole Ti 2p envelope, including the plasmons, for both oxidized and sputter-cleaned samples. Due to preferential sputtering of Ti over B, the film B/Ti ratio is best determined without sputter etching of the sample surface. This allows accurate compositional determination, assuming that extensive levels of oxygen are not present in the sample. Fully dense films can be accurately quantified for at least a year after deposition, while underdense samples do not give reliable data if the O/Ti ratio on the unsputtered surface is $\gtrsim 3.5$. Titanium suboxides detected after sputter etching is further indicative of oxygen penetrating the sample, and quantification by XPS should not be trusted.

中文翻译：

---

TiBx（1.3≤x≤3.0）薄膜的X射线光电子能谱分析

我们报告通过X射线光电子能谱（XPS）对硼化钛薄膜进行全面分析。通过氩气放电中的化合物TiB ₂靶，通过直流磁控溅射和高功率脉冲磁控溅射两种方法生长薄膜。通过改变沉积参数，可以在很宽的范围内调节膜组成$1.3\lesssim \mathrm{乙}/\mathrm{钛}\lesssim 3.0$由弹性后坐力检测分析和卢瑟福背散射光谱法确定。通过比较在如此广泛的组成范围内的光谱，我们可以得出有关如何解释TiB _x的XPS光谱的原始结论。通过仔细的光谱去卷积，可以从与化学计量的TiB ₂对应的信号中分辨出Ti-Ti和B-B键的信号。偏离化学计量的信号的强度可以与膜的B / Ti比直接相关。此外，我们展示了一种方法，可为氧化和溅射清洗的样品获得包括等离子体激元在内的整个Ti 2p包膜的一致且量子力学上准确的峰去卷积。由于Ti比B优先溅射，因此可以在不对样品表面进行溅射蚀刻的情况下确定B / Ti膜的最佳比例。假设样品中不存在大量的氧气，则可以进行准确的成分确定。在沉积后至少一年内，可以完全定量完全致密的薄膜，而如果未溅射表面上的O / Ti比为0，则低密度样品无法提供可靠的数据。$\gtrsim 3.5$。溅射蚀刻后检测到的低价钛氧化物进一步表明了氧气渗透到样品中，因此不应相信通过XPS进行的定量分析。