Nitroaromatic explosives' detection and quantification using an attention-based transformer on surface-enhanced Raman spectroscopy maps.

Rapidly and accurately detecting and quantifying the concentrations of nitroaromatic explosives is critical for public health and security. Among existing approaches, explosives' detection with Surface-Enhanced Raman Spectroscopy (SERS) has received considerable attention due to its high sensitivity. Typically, a preprocessed single spectrum that is the average of the entire or a selected subset of a SERS map is used to train various machine learning models for detection and quantification. Designing an appropriate averaging and preprocessing procedure for SERS maps across different concentrations is time-consuming and computationally costly, and the averaging of spectra may lead to the loss of crucial spectral information. We propose an attention-based vision transformer neural network for nitroaromatic explosives' detection and quantification that takes raw SERS maps as the input without any preprocessing. We produce two novel SERS datasets, 2,4-dinitrophenols (DNP) and picric acid (PA), and one benchmark SERS dataset, 4-nitrobenzenethiol (4-NBT), which have repeated measurements down to concentrations of 1 nM to illustrate the detection limit. We experimentally show that our approach outperforms or is on par with the existing methods in terms of detection and concentration prediction accuracy. With the produced attention maps, we can further identify the regions with a higher signal-to-noise ratio in the SERS maps. Based on our findings, the molecule of interest detection and concentration prediction using raw SERS maps is a promising alternative to existing approaches.

Raman spectrum matching with contrastive representation learning.

Raman spectroscopy is an important, low-cost, non-intrusive technique often used for chemical identification. Typical approaches identify a spectrum by comparing it with a reference database using supervised machine learning, which usually requires careful preprocessing and multiple spectra available per analyte. We propose a new machine learning technique for spectrum identification using contrastive representation learning. Our approach requires no preprocessing and works with as little as a single reference spectrum per analyte. We have significantly improved or are on par with the existing state-of-the-art analyte identification accuracy on two Raman spectral datasets and one SERS dataset that include a single component. We demonstrate that the identification accuracy can be further increased by slightly increasing the candidate set size using conformal prediction on the SERS dataset. Based on our findings, we believe contrastive representation learning is a promising alternative to the existing methods for Raman spectrum matching.

Quantitative SERS Assay on a Single Chip Enabled by Electrochemically Assisted Regeneration: A Method for Detection of Melamine in Milk.

Reusability of sensors is relevant when aiming to decrease variation between measurements, as well as cost and time of analysis. We present an electrochemically assisted surface-enhanced Raman spectroscopy (SERS) platform with the capability to reverse the analyte-surface interaction, without damaging the SERS substrate, allowing for efficient sensor reuse. The platform was used in combination with a sample pretreatment step, when detecting melamine from milk. We found that the electrochemically enhanced analyte-surface interaction results in significant improvement in detection sensitivity, with detection limits (0.01 ppm in PBS and 0.3 ppm in milk) below the maximum allowed levels in food samples. The reversibility of interaction enabled continuous measurement in aqueous solution and a complete quantitative assay on a single SERS substrate.

Process optimization of ultrasonic spray coating of polymer films.

In this work we have performed a detailed study of the influence of various parameters on spray coating of polymer films. Our aim is to produce polymer films of uniform thickness (500 nm to 1 µm) and low roughness compared to the film thickness. The coatings are characterized with respect to thickness, roughness (profilometer), and morphology (optical microscopy). Polyvinylpyrrolidone (PVP) is used to do a full factorial design of experiments with selected process parameters such as temperature, distance between spray nozzle and substrate, and speed of the spray nozzle. A mathematical model is developed for statistical analysis which identifies the distance between nozzle and substrate as the most significant parameter. Depending on the drying of the sprayed droplets on the substrate, we define two broad regimes, "dry" and "wet". The optimum condition of spraying lies in a narrow window between these two regimes, where we obtain a film of desired quality. Both with increasing nozzle-substrate distance and temperature, the deposition moves from a wet state to a dry regime. Similar results are also achieved for solvents with low boiling points. Finally, we study film formation during spray coating with poly (D,L-lactide) (PDLLA). The results confirm the processing knowledge obtained with PVP and indicate that the observed trends are identical for spraying of other polymer films.

Machine learning methods for the detection of explosives, drugs and precursor chemicals gathered using a colorimetric sniffer sensor.

Colorimetric sensing technology for the detection of explosives, drugs, and their precursor chemicals is an important and effective approach. In this work, we use various machine learning models to detect these substances from colorimetric sensing experiments conducted in controlled environments. The detection experiments based on the response of a colorimetric chip containing 26 chemo-responsive dyes indicate that homemade explosives (HMEs) such as hexamethylene triperoxide diamine (HMTD), triacetone triperoxide (TATP), and methyl ethyl ketone peroxide (MEKP) used in improvised explosives devices are detected with true positive rate (TPR) of 70-75%, 73-90% and 60-82% respectively. Time series classifiers such as Convolutional Neural Networks (CNN) are explored, and the results indicate that improvements can be achieved with the use of kinetics of the chemical responses. The use of CNNs is limited, however, to scenarios where a large number of measurements, typically in the range of a few hundred, of each analyte are available. Feature selection of important dyes using the Group Lasso (GPLASSO) algorithm indicated that certain dyes are more important in discrimination of an analyte from ambient air. This information could be used for optimizing the colorimetric sensor and extend the detection to more analytes.

Mathematical Model for Biomolecular Quantification Using Large-Area Surface-Enhanced Raman Spectroscopy Mapping.

Surface-enhanced Raman spectroscopy (SERS) based on nanostructured platforms is a promising technique for quantitative and highly sensitive detection of biomolecules in the field of analytical biochemistry. Here, we report a mathematical model to predict experimental SERS signal (or hotspot) intensity distributions of target molecules on receptor-functionalized nanopillar substrates for biomolecular quantification. We demonstrate that by utilizing only a small set of empirically determined parameters, our general theoretical framework agrees with the experimental data particularly well in the picomolar concentration regimes. This developed model may be generally used for biomolecular quantification using Raman mapping on SERS substrates with planar geometries, in which the hotspots are approximated as electromagnetic enhancement fields generated by closely spaced dimers. Lastly, we also show that the detection limit of a specific target molecule, TAMRA-labeled vasopressin, approaches the single molecule level, thus opening up an exciting new chapter in the field of SERS quantification.