Development of micro-dielectric barrier discharge mass spectrometry for fast surface analysis of solid samples and insights into its underlying mechanisms via optical emission spectroscopy
MetadataShow full item record
Surface analysis in analytical chemistry is the technique that aims to characterize the composition and structure of the solid surface. It has been developed for many years as a powerful tool for analysis and quality management of materials with different features. Depth profiling is one of the surface analysis techniques that allows exploring the chemical information along the depth dimension. Nowadays, polymers have become one of the most widely used materials in the world for a variety of different fields, because of the polymers’ various composition and structure result in their excellent properties. Since the performance of many applications relies on the surface properties of the polymer, surface characterization and analysis for composition and structure is becoming a more and more important for fundamental research as well as practical applications. Among the many techniques used for surface analysis of polymers, ambient mass spectrometry (AMS) has risen great attention in recent years. It allows the direct desorption and ionization under atmospheric pressure with minimum sample preparation, high sample throughput, and the ability to reveal elemental, molecular, structural, and isotopic information, making it extremely suited for polymer surface analysis. However, the conventional AMS ionization source can only perform surface desorption and lacks the ability for depth profiling. Herein, a micro-dielectric barrier discharge (µDBD) ionization source coupled with mass spectrometry is developed for fast surface analysis of solid samples, including polyethylene (PE), polyethylene glycol (PEG), and Cu. The fingerprint mass spectra of PE show a complex oxidation reaction occurs on the surface with the interaction of the ambient plasma effluent. The dominating ionization mechanism is hydride abstraction reaction, thus leading to a series of fragment ions consist of PE backbones with oxygen whith a loss of hydrogen. The fingerprint mass spectra of PEG suggest that the decomposition of PEG ions is the major fragmentation mechanisms and a series of ions with C-O or C-C bond-cleavage observed with a mass spacing of 44, corresponding to the mass of the PEG monomer (C2H4O). Furthermore, the produced erosion crater shows the potential for depth profiling analysis. However, the energy transfer pathways inside the µDBD and the underlying mechanisms for plasma-sample interaction are not fully understood. Therefore, it is necessary to perform plasma diagnostics on µDBD to obtain critical information such as distribution of excited species, electron temperature, and gas temperature, etc. Optical emission spectroscopy (OES) imaging is often used for plasma diagnostics as it allows obtaining the distribution of excited species in the plasma, which can yield valuable insights into the energy distribution pathways as well as the access to the calculation of vibrational and rotational temperatures. Here in this study, a pushbroom hyperspectral imaging (Pb-HSI) OES system is used for its high sample throughput and multi-dimensional information accessibility. Spatial resolution improvement for Pb-HSI is beneficial in many instances, however, typical solutions suffer from the limitation of geometric extent, lowering light throughput, or reducing the field-of-view (FOV). Sub-pixel shifting (SPS) acquires higher-resolution images, compared to typical imaging approaches, from the deconvolution of low-resolution images acquired with a higher sampling rate. Furthermore, SPS is particularly suited for Pb-HSI due to its scanning nature. In this study, an SPS approach is developed and implemented on a Pb-HSI system for plasma optical emission spectroscopy. Preliminary results showed that a periodic deconvolution error was generated in the final SPS Pb-HSI images. The periodic error was traced back to random noise present in the raw/convoluted SPS data and its frequency displays an inverse relationship to the number of sub-pixel samples acquired. Computer modeled data allows studying the effect of varying the relative standard deviation (RSD) in the raw/convoluted SPS data on the final reconstructed SPS images and optimization of noise filtering. The resolution improvement of the optimized SPS Pb-HSI technique is evaluated by the USAF 1951 resolution target and an improvement of approximately two times in spatial resolution is observed. Typical spectral images, however, contain intensity maps that are integrated along the line-of-sight (LOS). A widespread method to extract the important radially resolved information is Abel’s inversion but most algorithms result in accumulation of error toward the plasma axial position, which is often the region of most interest. Here, a Fourier-transform based Abel’s inversion algorithm, which spreads the error evenly across the radial profile, is optimized for OES images collected on a Pb-HSI with SPS sampling algorithm. This method allows the reconstruction of the radial profile from the LOS emission images, and the SPS algorithm allows improved radial reconstruction fidelity from the increased number of data points. The accuracy and fidelity of the protocol are characterized and optimized with a software-based 3-dimensional hyperspectral model datacube. A systematic study of the effects of varying levels of representative added noise, different noise filters, the number of data points and cosine expansions used in the inversion, as well as the spatial intensity distribution shapes of the radial profile, are presented. Optimum conditions include: 3D median noise filter with 3-pixel radius, as well as a minimum of 50 points and 8 cosine expansions needed to keep the relative root mean squared error (rRMSE) <10%. The optimized protocol is implemented for the first time on optical emission spectral datacubes of the µDBD source obtained with the SPS Pb-HSI system and the extracted radially resolved emission maps of different plasma species. The radially resolved emission distribution of plasma species of (He I, N2, N2+, He2, O I) and the calculated vibrational and rotational temperatures yield some insight into the energy transform pathways. The higher vibrational temperatures inside the capillary indicate that electron impact reactions are more significant there, where the excited helium species, including helium metastables (Hem*) and helium dimer ions (He2+), are generated. These excited helium particles are then carried out with the plasma effluent and react with the neutral nitrogen coming from the open air. The nitrogen is ionized and excited by Hem* or He2+. Recombination reactions are proposed to be the dominating mechanisms in the µDBD effluent exposed to the atmosphere where the excited neutral nitrogen species N2* are formed from the N2+ and free electrons. The exited helium dimer is also detected in the afterglow of the plasma, which has high enough energy to ionize or excite the neutral nitrogen gas, indicating another possible energy transfer pathway from the helium dimer to the ambient gas. The surface is believed to have a role as indicated by emission peaks at Cu sample surface, while absent with polymer samples or when the plasma effluent is not exposed to a sample surface. The erosion crater characterization is also studied and compared with their corresponding emission profiles. There is a clear correlation between the crater width and the diameter of the N2+ radially resolved emission profile, which indicates the nitrogen ion could play an important role in the surface erosion process via the µDBD jet. It is worth noting that the optimized Abel's Inversion algorithm as well as the proposed mechanisms here may be applicable to other atmospheric pressure plasma jets, for example for ones utilized for plasma-based ambient mass spectrometry.