8/3/2019 WS_Roos

1/11

Near Infrared SpectroscopyNear Infrared Spectroscopy

Introduction into the method

8/3/2019 WS_Roos

2/11

NIR Spectroscopy: Somecharacteristics

Normal modevibration

(MIR)

Combination

and overtones

(NIR)

High degree of

excitation

Low degree of

excitation

NIR: = 780 2500 nm (12500 4000 cm-1)

Combination and overtones of CH-, OH-, and NH- vibration

8/3/2019 WS_Roos

3/11

NIR Spectroscopy: Some

characteristics

In contrast to sharp absorption peaks in the

MIR region, NIR spectra show less intensityand broad bands

A assignment of peaks to individual vibrations

is not possible

Reflection intensity in the NIR region is higher

than in mid infrared.

8/3/2019 WS_Roos

4/11

User Spectra

1/cm

5000 6000 7000 8000 9000

-0.25

0.00

0.25

0.50

0.75

100048 Ethanol 96%139492 Ethanol absolut021687 TriethanolaminIsoprop.119866 Propylenglykol140273 Triethylcitrat (Citroflex 2)

B

Example spectra

Kombin.

C-H 1.O

N-H, O-H 1.O

C-H 2.O

8/3/2019 WS_Roos

5/11

Advantages

Liquids, powders and even high-viscositysubstances can be measured without sample

preparation

Silica can be used as transmittance material

for cuvettes, fibers and probes.

Measurement period!

One NIR spectra is a characteristicfingerprint of an individual extract andcovers all organic compounds!

8/3/2019 WS_Roos

6/11

Evaluation of NIR spectra

Qualitative classification

Sufficiently big sample setwith a certain varietyconcerning to the expecteddetermination range

Development of aqualitative model, e.g.principal component

analysis

Quantitative Analysis

Reference data of allsamples have to bedetermined

Spectral data andrefernce data arecombined in a calibration,e.g. PLS

International Chemometrics Society:Chemometrics is the science of relating measurements made on a chemisystem or process to the state of the system via application of mathemator statistical methods.

8/3/2019 WS_Roos

7/11

Identification procedure

Original SpectraAll Spectra

1/cm

Tr

ansflectance

5000 6000 7000 8000 9000

0,50

0,75

1,00

1,25

1,50

NIRCal:Fettele1.nirFettl0,93SNVdg12/519.08.200310:22:49gr

WeizenkeimlMandellJojobal

Pretreated SpectraAll Spectra

1/cm

Tr

ansflectance

5000 6000 7000 8000 9000

-1,5

-1,0

-0,5

0,0

0,5

1,0

NIRCal:Fettele1.nirFettl0,3SNVdg12/519.08.200310:25:17gr

WeizenkeimlMandellJojobal

Pretreated SpectraAll Spectra

1/cm

Tra

nsflectance

5000 6000 7000 8000 9000

-0,1

0,0

0,1

0,2

NIRCal:Fettele1.nirFettl0,93SNVdg

12/519.08.200310:26:24gr

WeizenkeimlMandellJojobal

Loadings / LoadingsAll Spectra

Loading Value 1

Lo

ading

Value

2

-0,3 -0,2 -0,1 -0,0 0,1 0,2 0,3

-0,4

-0,2

-0,0

0,2

0,4

NIRCal:Fettele1.nirFettl0,93SNV

dg12/519.08.200310:27:42gr

WeizenkeimlMandellJojobal

8/3/2019 WS_Roos

8/11

Principles for NIR measurements

Diffuse reflection

Transmission

TransflectionNIR detectorNIR detector

NIR detectorNIR detector

NIR detectorNIR detector

8/3/2019 WS_Roos

9/11

Spectrometer types

Dispersive spectrometer (e.g. Foss NIRSystems5000):

referencesample via l

sample

4 PbS detectors

(set at 45)

holographic

grid

mirror

halogen

lamp

mea surement

window

8/3/2019 WS_Roos

10/11

Spectrometer types

Fourier transform spectrometer (e.g. Bchi NIRflex)

Diode array spectrometer

Interferometer

Fourier transformation

Interferogramm

8/3/2019 WS_Roos

11/11

Interim statement

In NIRS single compounds influence not onlyindividual regions, but often the whole

spectrum Visual analysis of the spectra is impossible

NIR spectra have no significant peaks, but

they are unique reproductions of asubstance or a mixture

Data have to be combined with multivariate

mathematical methods

We need a different thinking, when workingwith NIRS !