Effects of Filter Composition, Spectral Bandwidth, and Pathlength on Stray Light Levels in the Near-Infrared Region

Importance of the topic

Stray light is a critical source of error in UV-Vis-NIR spectrophotometry, especially when measuring highly absorbing samples in the near-infrared (NIR) region. It degrades the linearity of absorbance versus concentration, reduces photometric accuracy at high optical densities, and can falsify quantitative results. Reliable quantification of stray light is therefore essential for applications in materials science, pharmaceuticals, environmental analysis and quality control.

Study objectives and overview

This technical overview examines how the composition of optical filters, the choice of spectral bandwidth, and the optical pathlength influence stray light measurements on Agilent Cary 5000 and 7000 UV-Vis-NIR spectrophotometers. The work evaluates traditional chloroform and water filters alongside a new dibromomethane filter for stray light assessment in the 1,380–2,385 nm range, aiming to identify optimal measurement parameters and demonstrate instrument performance under different conditions.

Methodology and instrumentation

The stray light studies employed an Agilent Cary 5000 UV-Vis-NIR spectrophotometer (applicable to the Cary 7000 UMS). Three liquid filters with well-defined cut-off wavelengths were used: dibromomethane (1,690 nm), chloroform (2,365 nm), and water (1,420 nm). Measurements were carried out using quartz cuvettes of 10 mm and 50 mm pathlength, rear beam attenuation filters (2 Abs for dibromomethane, 3 Abs for chloroform and water), a fixed data interval of 1 nm, and a signal averaging time of 1 s. Spectral bandwidths from 6 to 20 nm were tested for dibromomethane, while chloroform and water used automatic bandwidth selection.

Main results and discussion

• Dibromomethane filter:

• 10 mm pathlength, 2 Abs RBA: stray light rose from 4.08 × 10⁻⁴ %T at 6 nm SBW to 3.28 %T at 20 nm SBW.
• 50 mm pathlength, 2 Abs RBA: stray light values fell below the detection limit (negative %T) at SBW ≥ 12 nm, indicating instrument noise floor.

• Chloroform and water filters (10 mm, 3 Abs RBA, auto SBW): stray light of 2.13 × 10⁻⁴ %T at 2,365 nm and 3.50 × 10⁻⁵ %T at 1,420 nm, demonstrating very low background in the NIR.

These results confirm that narrower spectral bandwidths and shorter pathlengths improve stray light performance until the detector noise limit is reached. Longer pathlengths enhance sensitivity but require careful adjustment of bandwidth and averaging time to avoid negative readings.

Benefits and practical applications

• High accuracy in quantitative NIR analyses of highly absorbing samples (e.g., polymers, oils).
• Reliable photometric performance up to high optical densities, preserving Beer-Lambert linearity.
• Methodological guidelines for selecting filters, pathlengths, and bandwidths to optimize stray light rejection.
• Enhanced quality control in R&D and regulated environments.

Future trends and potential applications

Advances may include development of novel liquid and solid-state filters with steeper cutoffs, integration of real-time stray light correction algorithms, and expanded stray light testing across emerging detectors (e.g., InGaAs arrays). Coupling these improvements with automated instrument diagnostics will further extend the reliable dynamic range for complex samples in biotechnology, energy materials, and process monitoring.

Conclusion

This overview demonstrates that the Agilent Cary 5000/7000 instruments exhibit exceptionally low stray light in the NIR when tested with dibromomethane, chloroform, and water filters. Optimal measurement requires balancing spectral bandwidth, pathlength and averaging time to minimize stray light while avoiding detector saturation or noise limits. Implementing these best practices ensures precise quantitative spectroscopy across challenging sample matrices.

Reference

1. W. Alwan, T. Burt; Agilent Technologies Inc. “Effects of Filter Composition, Spectral Bandwidth, and Pathlength on Stray Light Levels in the Near-Infrared Region”, Technical Overview 5994-4982EN, June 2022.