Colorimetric Signal Readout for the Detection of Volatile Organic Compounds Using a Printable Glass-Based Dielectric Barrier Discharge-Type Helium Plasma Detector

Jingqin Mao, L. D. Liu, Yahya Atwa, Junming Hou, Zhenxun Wu, Hamza ShakeelACS 2023 · DOI

TL;DR

This paper introduces a novel, low-power helium plasma photoionization detector (µHDBD-PID) that uses a smartphone camera to capture color changes in the plasma as a readout for volatile organic compounds (VOCs). This colorimetric approach, processed into light intensity plots, offers a new avenue for gas analysis, successfully detecting various VOC classes with a promising limit of detection.

What Problem Does This Paper Address?

Traditional gas chromatography (GC) detectors, like photoionization detectors (PIDs), often rely on electrical signals for detection. While effective, these systems can be complex and "blind" to the observer, meaning they don't inherently provide information about the specific chemical identity beyond retention time. The authors aim to address this by developing a more visually intuitive and potentially simpler readout method for micro-scale PIDs, specifically for detecting volatile organic compounds (VOCs). They also tackle the challenge of miniaturizing these detectors while maintaining good performance and explore a novel fabrication method for the detector's core component.

Methods

The core innovation lies in the colorimetric readout of the µHDBD-PID. Instead of measuring electrical current, the device uses a smartphone camera to record the visual changes in the helium plasma when VOCs are introduced.

Here's a breakdown of the key methods:

  • Printable Glass-Based DBD Plasma Chamber: The detector's plasma chamber is fabricated using a printable glass suspension and 3D printing. This allows for the creation of complex, millimeter-sized geometries. The fabrication process involves multiple thermal debinding and sintering steps.
  • Helium Plasma Generation: A custom-built AC power supply (900 V, 83.6 kHz) generates a non-thermal helium plasma within the chamber. The device operates at a low power of ~5 W.
  • Colorimetric Signal Acquisition: A smartphone camera captures video of the plasma. As VOCs are introduced, they interact with the plasma, causing a visible change in its color and intensity.
  • Image Processing and Data Analysis: The recorded video is processed frame-by-frame using Python. The authors explored three color space models (RGB, HSL, HSV) to convert the visual color changes into quantitative data. They found that the RGB color space model provided the best performance, with the lowest baseline fluctuations, making it easier to detect subtle changes. The processed data is converted into an "image light intensity vs retention time" plot, essentially a gas chromatogram.
  • GC Integration: The µHDBD-PID is coupled with a standard 30-m gas chromatography column for separation of VOC mixtures.

Datasets

The paper does not explicitly mention training datasets. The evaluation was performed using known VOCs and mixtures.

  • VOCs Tested:
    • Straight-chain alkanes: n-pentane, n-hexane, n-heptane, n-octane, n-nonane
    • Aromatics: benzene, toluene, ethylbenzene
    • Polar compounds: acetone, ethanol, dichloromethane
  • Mixtures: The device was tested with mixtures of three and five compounds.

Key Insights and Ideas

  • Visualizing Gas Detection: The most significant insight is the successful demonstration of using visible plasma color changes as a quantitative readout for VOC detection. This opens up possibilities for simpler, more intuitive gas sensing systems.
  • Smartphone as a Detector Component: Leveraging a smartphone camera as a key component for data acquisition highlights the potential for low-cost, portable analytical devices.
  • RGB Color Space Superiority: The finding that the RGB color space model yielded the most robust and interpretable results for baseline fluctuation and signal-to-noise ratio is crucial for practical implementation.
  • Printable Glass Fabrication: The ability to 3D print complex glass structures for the plasma chamber is a key enabler for miniaturization and potentially mass production of these detectors.
  • Broad Applicability: The device demonstrated its capability to detect diverse classes of VOCs, including those with higher ionization potentials, which can be challenging for traditional PIDs.

Key Results

  • Detection of VOC Classes: The µHDBD-PID successfully detected straight-chain alkanes, aromatics, and polar organic compounds.
  • Limit of Detection (LoD): The best LoD achieved was 10 ng for benzene at room temperature.
  • Mixture Analysis: The device performed excellently with mixtures of three and five compounds.
  • Response Time: The detector showed a quick response time, with plasma light intensity increasing to a peak within approximately 1 second after n-heptane injection and returning to baseline in about 6 seconds.
  • Device Reliability: While the initial design showed some etching of the viewing window after prolonged use, the authors note that replacement is simple and inexpensive.

Limitations and Future Work

  • Limit of Detection (LoD): While promising, the LoD of 10 ng is still an order of magnitude higher than state-of-the-art µHDBD-PIDs that use electrical signal outputs (pg-level).
  • Peak Splitting: The authors observed peak splitting in some chromatograms, which they hypothesize is due to nonuniform VOC and helium flow within the chamber.
  • Viewing Window Durability: The viewing glass window showed signs of etching and particle accumulation after 20 hours of plasma operation, necessitating regular replacement.
  • Camera Limitations: The current setup uses a visible light camera, limiting the detectable wavelength range. The micro-focusing capability of the smartphone camera also impacts light collection efficiency.
  • Electromagnetic Interference: The proximity of the camera to the high-voltage electrodes can lead to electromagnetic interference.

Future Work suggested by the authors includes:

  • Improving LoD: Exploring more sensitive cameras (e.g., high-resolution multispectral cameras) and optimizing the distance between the camera and the device.
  • Enhancing Detector Lifetime: Developing more resilient solutions for bonding the viewing window and plasma chamber, and selecting more chemically resistant glass substrates.
  • Optimizing Chamber Design: Addressing peak splitting by optimizing the chamber design for better gas flow dynamics.
  • Developing a Comprehensive Database: Creating a database of RGB values for various VOCs to enable compound identification beyond retention time.
  • Improving Power Supply: Working on an improved power supply design to reduce power consumption and increase device lifetime.

Summary updated

Abstract

In this paper, we report on a printable glass-based manufacturing method and a new proof-of-concept colorimetric signal readout scheme for a dielectric barrier discharge (DBD)-type helium plasma photoionization detector. The sensor consists of a millimeter-sized glass chamber manufactured using a printable glass suspension. Plasma inside the chip is generated using a custom-built power supply (900 V and 83.6 kHz), and the detector uses ∼5 W of power. Our new detection scheme is based on detecting the change in the color of plasma after the introduction of target gases. The change in color is first captured by a smartphone camera as a video output. The recorded video is then processed and converted to an image light intensity vs retention time plot (gas chromatogram) using three standard color space models (red, green, blue (RGB), hue, saturation, lightness (HSL), and hue, saturation, value (HSV)) with RGB performing the best among the three models. We successfully detected three different categories of volatile organic compounds using our new detection scheme and a 30-m-long gas chromatography column: (1) straight-chain alkanes (<i>n</i>-pentane, <i>n</i>-hexane, <i>n</i>-heptane, <i>n</i>-octane, and <i>n</i>-nonane), (2) aromatics (benzene, toluene, and ethylbenzene), and (3) polar compounds (acetone, ethanol, and dichloromethane). The best limit of detection of 10 ng was achieved for benzene at room temperature. Additionally, the device showed excellent performance for different types of sample mixtures consisting of three and five compounds. Our new detector readout method combined with our ability to print complex glass structures provides a new research avenue to analyze complex gas mixtures and their components.
BibTeX
@inproceedings{mao2023colorimetric,
  author = {Jingqin Mao and L. D. Liu and Yahya Atwa and Junming Hou and Zhenxun Wu and Hamza Shakeel},
  title = {{Colorimetric Signal Readout for the Detection of Volatile Organic Compounds Using a Printable Glass-Based Dielectric Barrier Discharge-Type Helium Plasma Detector}},
  booktitle = {ACS},
  year = {2023},
  doi = {10.1021/acsmeasuresciau.3c00012},
}
Figure