Mobile spectrometer considerations
Context
Maybe you’ve read about mobile spectrometers in the latest issue of your favorite journal, or maybe the lab down the hall can’t stop talking about them at the department coffee hour. Regardless of how you first heard about them, mobile near-infrared spectrometers can seem too good to be true. Can you really predict the quality of your grain, fruit, or root non-destructively and without investing in a benchtop spectrometer that costs more than a brand new car?
The answer is, of course, it depends. Which traits are you interested in predicting? What is your required accuracy and precision? Is there existing literature that successfully predicts your trait of interest with near-infrared spectroscopy? How much time are you willing to invest in protocol development? Do you have access to model training software or the coding skills to use a free R package? Do you have a plan for managing your data?
And on top of these, there are spectrometer-specific questions to ask as well: What is your budget? Do you already have access to a laboratory-grade benchtop spectrometer through a collaborator or shared equipment at your organization? Given that most mobile spectrometers have more limited spectral ranges, can your trait of interest be predicted within the range measured by your desired mobile spectrometer (a brief literature search may give you a ballpark to aim for)? Is shipping available to your area? Can you get it in time for your planned experiment?
Consider yourself warned – model development can take a lot of time and effort. If you’re short on time but flush with cash, you may want to consider using a more expensive spectrometer or another method entirely. If you’re committed to the mobile spectrometer train, decide on your plan of attack early. Conduct a few pilot experiments to get your methods nailed down. Trust me, it’s worth the upfront effort to avoid wasting time and resources with your full-scale data collection.
Hardware comparison
| Spectrometer | Range (nm) | Firsthand experience | Connection | Software | Approximate price | Other considerations |
|---|---|---|---|---|---|---|
| SCiO | 740-1070 | Works well for cassava root water and starch content, squash brix. Currently testing for phenomic selection | Bluetooth | Comes with app for Apple and Android | $ 2,000.00 | Requires license for access to raw spectra. Comes with one year license, but $3k/yr after that |
| LinkSquare1 | ~400-1000 | Do not recommend | WiFi | Prospector | $ 550.00 | |
| LinkSquareNIR | ~700-1050 | Do not recommend | WiFi | Prospector | $ 650.00 | |
| InnoSpectra | 900-1700 | Purchased but not yet tested | USB and Bluetooth | Prospector | $ 2,000.00 | Also marketed as TellSpec |
| Nix Spectro 2 | 400-700 | Purchased but not yet tested | Bluetooth | Comes with app for Apple and Android | $ 1,300.00 | Not sure if you can export raw data, but could reach out to company to find out. We may explore adding it to Prospector in the future |
| GoyaLab IndiGo | 380-720 | USB and Bluetooth | Android, Windows | May add to Prospector | ||
| GoyaLab NIR | 720-1040 | USB and Bluetooth | May add to Prospector. Company is working on one with a larger range: 950-1650nm | |||
| QualitySpec Trek | 350-2500 | Works well for cassava root water, starch, and carotenoid content | Records spectra internally. Connect with USB for data transfer direct to computer | Windows? | $ 50,000.00 | Can’t input sample names so matching can be a pain |
| MicroNIR | 950-1650 | USB and Bluetooth | Windows | $ 20,000.00 | ||
| Ocean Optics | Various | USB | Windows? | $ 2,000.00 | Requires fiber optic cable. Not marketed to consumers so may be less user friendly? |
Software options
Free & open source
- waves (R package)
- Prospector (Android app for spectral data collection)
- Orange
Proprietary/purchase required
- WinISI (Windows only)
- Unscrambler
Further reading
Broad overview
- Pioneer Agronomy Resources: Stretch Your Sampling Budget with NIRS
- Pioneer Agronomy Resources: NIRS Analysis Has Long and Credible History
In-depth overview
- Ciurczak, E. W., Igne, B., Workman Jr, J., & Burns, D. A. (Eds.). (2021). Handbook of near-infrared analysis. CRC press.
- Cortés, V., Blasco, J., Aleixos, N., Cubero, S., & Talens, P. (2019). Monitoring strategies for quality control of agricultural products using visible and near-infrared spectroscopy: A review. Trends in Food Science & Technology, 85, 138-148. (Sections 1-3 are general NIRS overview)
- Pasquini, C. (2003). Near infrared spectroscopy: fundamentals, practical aspects and analytical applications. Journal of the Brazilian chemical society, 14, 198-219.