Method for identifying chemical and structural variations through terahertz time-domain spectroscopy

1. A method for identifying chemical and structural irregularities in a substrate, comprising:

providing a sample comprising a plurality of points and having chemical or structural variations;

providing a terahertz scanner comprising:

at least one terahertz source generating a terahertz beam along a terahertz beam path, and

at least one terahertz detector, wherein

the terahertz scanner produces a characteristic signal when the terahertz beam is detected by the at least one terahertz detector after interacting with a point of the plurality of points of the sample; and

the terahertz scanner produces a reference signal when the terahertz beam is not intercepted by the sample;

obtaining a characteristic signal and a reference signals for at least one point of the plurality of points of the sample;

extracting one or more characteristic features from a characteristic data set afforded by the characteristic signal, and extracting one or more corresponding reference features from a reference data set afforded by the reference signal; wherein at least one of the characteristic features and the corresponding reference features are selected from a list consisting of: peak intensity, average intensity, delay, pulse width, spectral power, phase of the characteristic signals, and any combination thereof;

analyzing the characteristic features and the corresponding reference features using a machine learning data analysis model to find chemical and or structural information about the sample; and

recognizing irregularities in or differences between the characteristic features and the corresponding reference features to identify the chemical or structural variations in the sample.

2. The method of claim 1 , wherein the machine learning data analysis model is selected from a list consisting of: supervised model, unsupervised model, and any combination thereof.

3. The method of claim 1 , wherein the machine learning data analysis model is trained to recognize the chemical or structural variations in the sample.

4. The method of claim 1 , wherein the terahertz scanner is a terahertz time-domain spectroscopy scanner.

5. The method of claim 4 wherein the terahertz scanner further comprises:

a femtosecond laser generating an optical beam, and

a plurality of optical lenses for focusing the optical beam along an optical beam path.

6. The method of claim 5 , wherein the femtosecond laser is selected from a list of: a single femtosecond laser combined with a delay stage, and a dual femtosecond laser.

7. The method of claim 5 , wherein the femtosecond laser generates the optical beam at a wavelength selected from: ˜800 nm, 1550 nm.

8. The method of claim 5 , wherein the plurality of optical lenses adjusts a spot size of the optical beam to cover a active area of the terahertz source and the terahertz detector.

9. The method of claim 4 , wherein the terahertz scanner further comprises a plurality of terahertz lenses and or off-axis parabolic mirrors for focusing and or collimating the terahertz beam along the terahertz beam path.

10. The method of claim 9 , wherein the plurality of terahertz lenses is made from a material selected from a list of: polyethylene, TPX, ceramic, Teflon, and any combination thereof.

11. The method of claim 4 , wherein the characteristic signal and the reference signal are in time domain.

12. The method of claim 11 , wherein the characteristic signal and the reference signal in time domain are used to find their corresponding signals in frequency-domain; and extracting the characteristic features and the corresponding reference features in both a time domain and a frequency domain.

13. The method of claim 1 , wherein the chemical or structural variations are selected from a list consisting of: chemical contamination, material defects, other irregularities in a material, and any combination thereof.

14. The method of claim 13 , wherein the chemical contamination comprises a toxin.

15. The method of claim 14 , wherein the toxin is selected from a group consisting of: mycotoxins, aflatoxins, or a combination thereof.

16. The method of claim 13 , wherein the material defects are further selected from a list consisting of: thickness variations, voids, cracks, non-uniformities, and any combination thereof.

17. The method of claim 1 , wherein the sample comprises a battery electrode and the chemical or structural variations are material defects.

18. The method of claim 1 , wherein the sample comprises a composite material and the chemical or structural variations are material defects.

19. The method of claim 1 , wherein the at least one terahertz detector is selected from a list consisting of: a single-pixel terahertz detector, a focal plane 1D or 2D array comprising a plurality of terahertz detectors.

20. The method of claim 1 , wherein the at least one terahertz detector is capable of taking three-dimensional terahertz snapshots of an area as large as ˜10 cm 2 within less than a 100 ms.

21. The method of claim 1 , wherein the at least one terahertz source and the at least one terahertz detector comprise plasmonic electrodes.

22. The method of claim 21 , wherein the at least one terahertz source and the at least one terahertz detector are fabricated on a semiconductor substrate and optimized for operation at 800 nm or 1550 nm wavelengths.

23. The method of claim 21 , wherein the at least one terahertz source and the at least one terahertz detector are fabricated on a GaAs, InGaAs and InAs substrate.

24. The method of claim 1 , wherein the sample is placed on a sample holder selected from a list of: a moving conveyor belt, a single or multi-well plate or multiple plates mounted on a motorized translational stage, and any combination thereof.