Near-infrared spectroscopy technology is used for L-valine

Near-infrared (NIR) spectroscopy technology can be effectively used for online monitoring of the L-valine fermentation process. It has advantages such as rapidity and non-destructiveness, and can provide key parameter information for optimizing the fermentation process. The details are as follows:

1. Monitoring Principle

The wavelength range of near-infrared light is 780-2526 nm. Overtones and combination bands of vibrations of hydrogen-containing groups (such as C-H, N-H, O-H) in organic molecules absorb light in this 波段 (waveband). Substances in the L-valine fermentation broth, such as L-valine itself, substrate glucose, and metabolite lactic acid, exhibit different spectral characteristics due to differences in their absorption of near-infrared spectra by hydrogen-containing groups. By scanning the near-infrared spectrum of the fermentation broth, characteristic information of these substances can be obtained, thereby realizing the monitoring of the fermentation process.

2. Monitoring Content

It can monitor changes in the concentration of major metabolites during the L-valine fermentation process. For example, by constructing detection models for substances such as L-valine, glucose, and lactic acid, the real-time content of these substances in the fermentation broth can be known. At the same time, combined with metabolic flux analysis, it can also analyze changes in metabolic fluxes at key metabolic nodes in the fermentation process, such as the glycolytic pathway, pentose phosphate pathway, and tricarboxylic acid cycle, helping to understand the cellular metabolic state.

3. Advantages and Characteristics

This technology is a "green detection technology" that overcomes the shortcomings of traditional analytical methods, such as slow detection speed and complex sample processing steps. Near-infrared spectrometers can be equipped with detection terminals such as fiber optic probes, which can directly perform in-situ online detection of fermentation broth. Generally, no pretreatment or only simple treatments such as degassing and filtration are required. Moreover, the determination speed is fast, multiple physical and chemical parameters can be analyzed simultaneously, and the results are reliable, which can meet the requirements of Process Analytical Technology (PAT) in production.

4. Modeling and Analysis Process

First, representative fermentation broth samples are selected, their near-infrared spectral data are collected, and reference values such as the concentration of relevant substances in the samples are measured. Then, the spectral data are preprocessed, such as removing abnormal samples, correcting spectral backgrounds, and denoising. Next, feature engineering is performed to compress the information scale. Finally, based on chemometric methods such as partial least squares (PLS), a prediction model is established. By substituting the spectral data of unknown samples into the model, parameters such as the concentration of relevant substances in the fermentation broth can be predicted.

5. Application Example

Wei Hongbo and others monitored the L-valine fermentation process based on near-infrared spectroscopy combined with metabolic flux analysis, and constructed detection models for major metabolites such as L-valine, glucose, and lactic acid. This provided strong support for the optimal control of the L-valine fermentation process and helped improve fermentation efficiency and product quality.