Surface analysis insight note: Analysis of X‐ray photoelectron spectroscopy images with summary statistics

IF 1.6 4区化学 Q4 CHEMISTRY, PHYSICAL

Surface and Interface Analysis Pub Date : 2023-07-17 DOI:10.1002/sia.7248

Behnam Moeini, M. Linford

{"title":"Surface analysis insight note: Analysis of X‐ray photoelectron spectroscopy images with summary statistics","authors":"Behnam Moeini, M. Linford","doi":"10.1002/sia.7248","DOIUrl":null,"url":null,"abstract":"Developments in X‐ray photoelectron spectroscopy (XPS) instrumentation and the need for spatial information in surface characterization have led to advances in XPS imaging and related image processing techniques. In this Insight Note, we demonstrate the use of summary statistics as simple, but effective, tools for understanding XPS hyperspectral images (data cubes) prior to more advanced image processing. An XPS image obtained from a silicon surface patterned with different thicknesses of oxide was analyzed with three summary statistics: a tool in the MATLAB programming environment, the mean, and pattern recognition entropy (PRE). The MATLAB tool largely separates the spectra into two groups. The mean does a somewhat better job differentiating between the spectra, and PRE is even more effective. The results of the MATLAB summary statistic are confirmed by plotting the average and standard deviation spectra of different regions of the image it produces. The results of the mean and PRE summary statistics are confirmed by evenly segmenting the results and examining the average and standard deviation spectra of these segments. Fitting these average spectra demonstrates the greater effectiveness of the PRE summary statistic in segmenting the spectra into chemically distinct groups.","PeriodicalId":22062,"journal":{"name":"Surface and Interface Analysis","volume":"55 1","pages":"789 - 797"},"PeriodicalIF":1.6000,"publicationDate":"2023-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Surface and Interface Analysis","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1002/sia.7248","RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}

引用次数: 3

Abstract

Developments in X‐ray photoelectron spectroscopy (XPS) instrumentation and the need for spatial information in surface characterization have led to advances in XPS imaging and related image processing techniques. In this Insight Note, we demonstrate the use of summary statistics as simple, but effective, tools for understanding XPS hyperspectral images (data cubes) prior to more advanced image processing. An XPS image obtained from a silicon surface patterned with different thicknesses of oxide was analyzed with three summary statistics: a tool in the MATLAB programming environment, the mean, and pattern recognition entropy (PRE). The MATLAB tool largely separates the spectra into two groups. The mean does a somewhat better job differentiating between the spectra, and PRE is even more effective. The results of the MATLAB summary statistic are confirmed by plotting the average and standard deviation spectra of different regions of the image it produces. The results of the mean and PRE summary statistics are confirmed by evenly segmenting the results and examining the average and standard deviation spectra of these segments. Fitting these average spectra demonstrates the greater effectiveness of the PRE summary statistic in segmenting the spectra into chemically distinct groups.

查看原文本刊更多论文

表面分析洞察注:分析X射线光电子能谱图像汇总统计

X射线光电子能谱（XPS）仪器的发展和表面表征中对空间信息的需求导致了XPS成像和相关图像处理技术的进步。在本Insight Note中，我们展示了在更高级的图像处理之前，使用摘要统计作为简单但有效的工具来理解XPS高光谱图像（数据立方体）。利用MATLAB编程环境中的一个工具、平均值和模式识别熵（PRE）对从具有不同厚度氧化物图案的硅表面获得的XPS图像进行了分析。MATLAB工具在很大程度上将光谱分为两组。平均值在区分光谱方面做得更好，PRE甚至更有效。MATLAB汇总统计的结果通过绘制其产生的图像的不同区域的平均光谱和标准差光谱得到了证实。平均值和PRE汇总统计的结果通过均匀分割结果并检查这些片段的平均值和标准差谱来确认。拟合这些平均光谱表明PRE汇总统计在将光谱分割成化学上不同的组方面具有更大的有效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Surface and Interface Analysis 化学-物理化学

CiteScore

3.30

自引率

5.90%

发文量

130

审稿时长

4.4 months

期刊介绍： Surface and Interface Analysis is devoted to the publication of papers dealing with the development and application of techniques for the characterization of surfaces, interfaces and thin films. Papers dealing with standardization and quantification are particularly welcome, and also those which deal with the application of these techniques to industrial problems. Papers dealing with the purely theoretical aspects of the technique will also be considered. Review articles will be published; prior consultation with one of the Editors is advised in these cases. Papers must clearly be of scientific value in the field and will be submitted to two independent referees. Contributions must be in English and must not have been published elsewhere, and authors must agree not to communicate the same material for publication to any other journal. Authors are invited to submit their papers for publication to John Watts (UK only), Jose Sanz (Rest of Europe), John T. Grant (all non-European countries, except Japan) or R. Shimizu (Japan only).