1

__________________________________

_________________________

H. W. Siesler Department of Physical Chemistry University of Duisburg-Essen D 45117 Essen, Germany

hw.siesler@uni-due.de

Near-Infrared Spectroscopy: Analytical Applications, Research and

Process Control

22

__________________________________

_________________________

Outline of the Presentation

Introduction

Some qualitative aspects of NIR spectroscopy

Miniaturization

Conclusions

Selected research and quality/process control applications by NIR spectroscopy

Introduction

3

___________________________________________________________

Today NIR Spectroscpy is Ubiquitous

Agriculture Food Chemistry Polymers Petrochemistry Pharmaceuticals Medicine Environment Geology

This is also reflected in the number of globally sold units (2011):near-IR: ~ 8500 (~ $ Mil 225)mid-IR: ~ 10000 (~ $ Mil 450)

Notwithstanding these facts there is still a lot of persuading to do.

4

__________________________________

_________________________

The Principles of MIR NIR

m1 m2

NEAR-INFRARED

OVERTONES/COMBINATIONS12500 – 4000 cm-1

0.8 – 2.5 m

V

q

ANHARMONICITYm2 << m1

0

q /

FUNCTIONALITIESCH / OH / NH

MID-INFRARED

FUNDAMENTALS4000 – 400 cm-1

2.5 – 25 m

V

qn = 0n = 1n = 2

n = 3

0q

ALL POLARFUNCTIONALITIES

5

___________________________________________________________

100

50

Tran

smitt

ance

(%)

Wavenumber (cm-1)40000

3000 2000 1500 1000 500

(C-F)

2 x (C-F)(CH2)

The Most Striking Case: MIR NIR-(CF2-CF2)n - (Teflon ™)

log

1/R

Wavenumber (cm-1)12000 8000 400010000 6000

00.

20.

4

(CH2) + (CH2)

Nonabsorbing referencefor NIR spectroscopy !!!

66

__________________________________

_________________________

Outline of the Presentation

Some qualitative aspects of NIR spectroscopy

Miniaturization

Conclusions

Selected research and quality/process control applications by NIR spectroscopy

Introduction

__________________________________

_________________________

Interpretation of MIR/NIR Spectra(CH3CH2COCH3 + 5% H2O)

J. Workman Jr., L. WeyerPractical Guide and Spectral Atlas

for Interpretive Near-Infrared Spectroscopy (2nd ed.)CRC Press, Boca Raton (2012)

MIR

1.0

0.8

0.6

0.4

0.2

0.0

NIR

88

___________________________________________________________

“NIR spectra have only broad absorption bands“

0.0

1.0

2.0A

bsor

banc

e

Wavenumber (cm-1)

Wavenumber (cm-1)

0.0

0.2

0.4

0.6

Abs

orba

nce

O Si

CH3

CH3n

Poly(dimethylsiloxane)

(Si-O)

99

___________________________________________________________

a

b

a-b

PDMS/PU70/30 %(w/w)

PDMS/PU57/43 %(w/w)

Interpretive Aspects of NIR Spectra

_________________________

Interpretive Aspects of MIR/NIR Spectra

Wavenumber (cm-1)10000 8000 6000 4000

Abs

orba

nce

2.0

1.0

0.0

NIRCH2-specific

CH3-specific

MIRButylglycolacetate

Butyldiglycolacetate

__________________________________

(CH3) (CH2) (CH2)

(CH3)

1111

__________________________________

_________________________

Outline of the Presentation

Some qualitative aspects of NIR spectroscopy

Miniaturization

Conclusions

Selected research and quality/process control applications by NIR spectroscopy

Introduction

12

__________________________________

_________________________

Driving Forces for the Implementation of NIR

The acceptance of NIR-spectroscopy for research and quality/process control has steadily increased because of:

••• time savings in comparison to other analyticaltechniques

••• simplified or no sample handling of toxic samples

••• measurement flexibility in hostile environments

••• speed of data acquisition

The following exemplary applications will demonstrate theversatility of NIR spectroscopy for these applications.

••• occasionally higher selectivity than MIR

1313

___________________________________________________________

Organic OH-Functionality H2OContrary to MIR spectroscopy in the NIR spectra the bands of organic OH-groups are separated from the water bands.

This is of importance for polymer quality control applications:Simultaneous determination of OH-number and water content.

How can you discriminate: Upon peracetylation the band of the organic OH-group disappears.

NIR spectra of a polyetherwith different OH-number

___________________________________________________________

OH/NH-Functionalities

Contrary to MIR spectroscopy in the NIR spectra the bands of organic OH- and NH-groups are separated.

15

___________________________________________________________

Polymorphism Pharmaceutical Industry

• Polymorphism is the ability of a substance to exist in two or more crystalline forms.

• More than half of the drug substances described in monographs crystallize in more than one solid form.

• Solid state properties of active ingredients have a direct impact on: bioavailability/quality/efficacy safety/patent situation

• What analytical techniques are available for the rapid discrimination of different polymorphs ?

16

_________________________

__________________________________A Case Study of Polymorphism in a Pharmaceutical Active

(cooperation with A. P. Ayala, Fortaleza, Brazil)

Mebendazole

• Broad-spectrum anthelmintic against worms

• Three polymorphs (A, B, C): solubility B > C > A

• Therapeutic trials: A is inactive Publications: B is toxic

• USP test is not able to distinguish between forms A and C.

• polymorph C is pharmaceuticallypreferred

WHO estimates that one quarter of the world’s population is infected with one or more of the parasites.

__________________________________

_________________________

Detection of Polymorphism in Mebendazole: X-Ray Diffraction, Raman, NIR and MIR Spectroscopy

Form C

Form B

Form A

2

X-Ray

Inte

nsity

2

Form A

Form B

Form CFT-NIR

log

1/R

Form A

Form B

Form C

FT-IR

500 1500

FT-RamanForm C

Form B

Form A

18

___________________________________________________________

Analysis of Phosgene in o-Dichlorobenzene Solutions

Because it is a highly toxic gas, it had to be analyzed off-line under extreme precaution measures (gas mask etc.)

NIR in-line control has alleviated the situation significantly.

Phosgene (COCl2) is used extensivelyin the production of isocyanates:

Synthesis: CO + Cl2 COCl2

COCl2

3 x ν(C=O)

- 2 HClR-NH2 + COCl2 R-N=C=O

Toxic Reaction: COCl2 + H2O 2 HCl + CO2

19

__________________________________

_________________________

NIR-Spectroscopic On/In-Line Determination of HF and HNO3in Si Wafer Etching Solutions of the Semiconductor Industry

Teflon is transparent in the NIR.

ABB BomemFT-NIR spectrometer

Teflon Tubing

Fiber Optic Cables

ClippirTransmission Cell

The etching solution is circulated in an all-Teflon™ closed loop.

tube diameter can be squeezedto the suitable thickness

__________________________________

_________________________

The Spectral Calibration Range Aquaphotomics !!!

Wavenumbers (cm-1)

Abs

orba

nce

3 x ν(OH) 2 x ν(OH) + δ(OH)

HNO3 69%(w/w)

HF 48 % (w/w)

H2O

industrial etching solution

2121

__________________________________

_________________________

Polymer Recycling by NIR Spectroscopy

NIR-Spectrometer

NIR-Source

Conveyor Belt

( 3 m/s)

Monitoring Lamps

Separation Control

Air Jets

Unidentified Objects

Separation Modules

PE

PE

PP PVC PS PET

PETPSPVCPP

Approximately 95% of the polymer waste is correctly identified.

No other technique could cope with this task.

__________________________________

_________________________

256 Diode Array Wavelength Range: 1100 – 2100 nm Integration time: typical: 3 -10 ms Background: Spectralon

VisioNIR

In a Continuous Mixer (Blend Uniformity) On the Tablet Press (Content Uniformity)

On-Line Control of Pharmaceutical Processes by NIR Spectroscopy

___________________________________________________________

Technical Information: Continuous Mixer Continuous Mixer: KM5

Lödige GmbH, Paderborn, Germany Flow Rate: 10 - 300 kg/h Dwell Time: 10-30 s

Side View Front View

light fiber optics behindsapphire window

24

__________________________________

_________________________

Technical Information: Tablet Press

Tablet Press: Fette 102i Fette Compacting, Schwarzenbek, Germany

Tableting Velocity: 18.000 – 216.000 Tab/h (-120 Rot./min)Standard: 150.000 Tab/h (80 Rot./min)

Time between Tablets: app. 18 - 50 ms (20 - 55 Hz)

Total View

25

__________________________________

_________________________

The VisioNIR Fiber Probe for the Tablet Press

Collection Fiber

Illumination Fiber

Air Tubes

2626

__________________________________

_________________________

Implementation of the VisioNIR Probe into the Tablet Press

Illumination Fiber

Air Tubes

Collection Fiber

27

___________________________________________________________

The Overall Test Set-Up

Dosing Head (Schenk)

Continuous Mixer (Lödige)

VisioNIR for Mixer (VisioTec)

VisioNIR for Tablet Press (VisioTec)

Tablet Press (Fette)

2828

___________________________________________________________

Method Development for the Mixer

API content in the specified range (e.g. 20 - 30%, target content was 25%)

On-line monitoring of the mixed powders with the typical process variations

NIR spectra acquisition of the mixed product in 1% intervals (20 - 30%)

Sampling for Reference Analysis (UV)

74% Lactose

25% Paracetamol

1% MgSt

Investigated Formulation

2929

___________________________________________________________

NIR Spectra of the Powder from Mixer

1100 2100 Wavelength (nm)

Data Pretreatment: SNVIn

tens

ity

10000

15000

20000

25000

30000

35000

40000

45000

1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101

106

111

116

121

126

131

136

141

146

151

156

161

166

171

176

181

186

191

196

201

206

211

216

221

226

231

236

241

246

251

256

Intensity

Diodes

Paracetamol

30

___________________________________________________________

PLS Model for Active Ingredient (Mixer)Regression Coefficient

1100 nm Wavelength 2100 nmPredicted vs. Reference Plot

3131

___________________________________________________________

Method Development for the Tablet Press

On-line monitoring of the compressed tablets with the typical processvariations and variation of the compression force and the tableting speed

API content in the specified range between 20 - 30% (target content 25%)

NIR spectra acquisition of tablets with API contents 20 - 30% (1% intervals)

Sampling for Reference Analysis (UV)

74% Lactose

25% Paracetamol

1% MgSt

Investigated Formulation

___________________________________________________________

PLS Model for Active Ingredient (Tablet Press)

Predicted vs. Reference Plot

33

___________________________________________________________

Prediction of the API During the Test Run

0

5

10

15

20

25

30

35

0:00 0:10 0:20 0:30 0:40 0:50 1:00 1:10 1:20 1:30

API con

tent [%

]

Time [h:min]

Phase 1             Phase 2              Phase 3            Phase 4           Phase 5        Phase 6       Phase 7    Phase 8      Phase 9         Phase10   Ph.11    Ph.12

0:00 0:10 0:20 0:30 0:40 0:50 1:00 1:10 1:20 1:30

Time (h:min)1:40

API

Con

tent

(% w

/w)

35

30

25

20

15

10

5

0

Continuous Mixer / Tablet Press RMSEP > 1%

3434

__________________________________

_________________________

Outline of the Presentation

Introduction

Some qualitative aspects of NIR spectroscopy

Miniaturization

Conclusions

Selected research and quality/process control applications by NIR spectroscopy

Introduction

_________________________

__________________________________Miniaturization of IR/NIR-Spectrometers 1975 2015

~ 250 kg

Nicolet FT-IR 1975

VIAVI Solutions (formerly JDSU)

2015

On-Site Analysis

___________________________________________________________

i-Spec Nano by B&W Tek

Preliminary Specifications

calibrationvalidation

PLS calibration for % fat in soybeans spectral range: 1350 - 1700 nm

Further New Development

37

__________________________________

_________________________

https://www.consumerphysics.com/myscio/

A Word of Caution

38

__________________________________

_________________________

http://tellspec.com/

Beam your health up !

A Word of Caution

39

__________________________________

_________________________

A Word of Caution

http://biozoom.net/de/A Health Data Revolution

4040

__________________________________

_________________________

Conclusions

NIR spectroscopy has become a mature and extremely versatile technique that can be applied to solve a multiplicity of researchand quality/process control problems.

Despite this positive development a further improvement of theacceptance and recognition of this technique can be achieved by:

■ intensified communication between spectroscopists, agro- and food-oriented users and chemometricians

■ better understanding of the basic issues (band assignments, spectra pretreatments, use of alternative techniques, etc.).

4141

__________________________________

_________________________

Acknowledgements

Frank Pfeifer, Laima Thimm, Sylvia Iwascek, Sedakat Altinpinar Damir Sorak, Uwe Hoffmann, Lars Herberholz, Olga KolomietsUniversity of Duisburg-Essen, Essen, Germany

ABB-Bomem, Quebec, CanadaBruker Optik GmbH, Ettlingen, GermanyBüchi GmbH, Essen, GermanyUhlmann VisioTec, Laupheim, GermanyVIAVI Solutions (formerly JDSU), Santa Rosa, USA

W. Hoehe, Uhlmann VisioTec GmbH, Laupheim, Germany

Kristiina Järvinen, Maiju Järvinen, School of Pharmacy, Faculty of HealthSciences University of Eastern Finland, Kuopio, Finland

M. Tondar, Glatt GmbH, Binzen, Germany

Fette Compacting, Schwarzenbek, Germany

Gebrueder Loedige Maschinenbau GmbH, Paderborn, Germany