Real and Simulated Datasets for Automated Background Noise Removal in Raman Spectra Using Dynamic Time Warping

Description of 1st Dataset

Name: 1st Dataset_MicroPlastics_PDMS substrate_PET collected by Monica Quarato

1st data set was generated to identify microplastics spiked in ultrapure water and pumped into a microfluidic device capable of pre-concentrating and subsequently detecting them by Raman spectroscopy using the same device. This microfluidic device was fabricated in glass-PDMS. The plastic detected was PET. This dataset included a PDMS background spectrum along with several Raman spectra of PET microplastics. Files beginning with Background_ correspond to the PDMS background spectrum. All files are provided in .dat format.

Each filename encodes the following information:

·         Pillar array where the Raman spectrum was acquired. The microfluidic chamber contains different pillar arrays that trap microplastics of varying sizes. These appear in the filenames as 1umpillars, 3umpillars, 5umpillars, or 10umpillars.

·         Plastic type and size, indicated for example as _PET_su300um_.

·         Concentration, reported in either µg/mL (ugmL) or ng/mL (ngmL).

Most spectra in this dataset were acquired with a confocal Raman microscope using a 20x objective, indicated in the filename by _20x_.

Description of 2nd Dataset

Name: 2nd Dataset_NanoPlastics_Si substrate_PET_PS collected by Monica Quarato

2nd data set was acquired for the identification and detection of different nanoplastics in artificial and natural water using a silicon wafer as support. We focused on PS and PET spiked water samples. This data set included a Si background spectrum, accompanied by several Raman spectra of PS and/or PET nanoplastics. The files’ format is .txt. This dataset contains two folders one related to PS and the other related to PET. A file starting with Si_ exist in each folder, representing Si background spectrum.

Each filename encodes the following information:

·         Plastic type and size, indicated for example as PET with 36nm (_36PET_) or vPET with 450nm (_vPET_) and 33nm spherical PS nanoparticles (_33PS_) or 161nm PS nanoparticles (_161PS_).

·         Concentration, reported in either µg/mL (ugmL) or ng/mL (ngmL).

The spectra in this dataset were acquired with a confocal Raman microscope using a 50x objective, indicated in the filename by _50x. Here, we have 5 different types of water: two artificial (artificial freshwater and artificial seawater) and three natural water (seawater from North sea, freshwater from Domitz (Elbe river) and freshwater from Thames river).

Description of Simulated Dataset:

Name: Simulated_Dataset generated by Ensieh Iranmehr

This folder contains the simulated data used to evaluate the proposed algorithm. It includes 10 .txt files, each named according to the corresponding figure and row in the supplementary information. Every file has two columns: primary signal and background signal. Each file contains results from 50 runs, with runs separated and labeled by their respective run name at the end.

---

## Creators

- Ensieh Iranmehr  - Monica Quarato- Begoña Espiña,- Laura Rodriguez-Lorenzo

---

## Citation

Please cite this dataset as:

Iranmehr, E., Quarato, M., Espiña, B., & Rodriguez-Lorenzo, L. (2025).  *Real and Simulated Datasets for Automated Background Noise Removal in Raman Spectra Using Dynamic Time Warping* [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.17091616

---

## License

© 2025 Espiña Group (Water Quality), INL – International Iberian Nanotechnology Laboratory.  This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License.  You are free to share and adapt the dataset for any purpose, even commercially, as long as appropriate credit is given.

---

Thank you for using our dataset!