Vibrational Spectroscopy of Polymers - ACS Publications - American

2 Vibrational Spectroscopy of Polymers Analysis, Physics, and Process Control H . W. Siesler Department of Physical Chemistry, University of Essen, D 4300 Essen, Germany

Improvements in rapid-scanning Fourier-transform infrared (FTIR) spectroscopy, the recent introduction of Fourier-transform Raman spectroscopy, and the more efficient exploitation of the near-infrared region launched vibrational spectroscopy into a new era of polymer chemical and physical applications. On the one hand, increased sensitivity led to breakthrough sampling techniques such as photoacoustic spectroscopy and diffuse reflectance measurements; on the other hand, improved time resolution largely enhanced the potential of FTIR spectroscopy for on-line combination with other techniques such as gas and liquid chromatography or thermal analysis. Significant progress also has been made in the characterization of time-dependent phenomena by the simultaneous acquisition of spectral and other relevant physical data (e.g., during mechanical measurements). Furthermore, multivariate data evaluation, which was restricted to near-IR multicomponent analysis of agricultural products for more than a decade, is increasingly applied in the field of polymer analysis. Last, but not least, the development of fiber optics for the near-IR wavelength range has opened up completely new areas for process control and remote sensing.

\ ^ I B R A T I O N A L SPECTROSCOPY IS AN IMPORTANT TOOL FOR CHARACTERIZATION of the chemical a n d physical nature o f polymers. I n p r i n c i p l e , the c o m p l e mentary techniques o f infrared a n d R a m a n spectroscopy provide qualitative 0065-2393/93/0236-0041$ 12.75/0 © 1993 American Chemical Society

42

STRUCTURE-PROPERTY RELATIONS IN POLYMERS

a n d quantitative i n f o r m a t i o n about the following structural details o f a polymer: • C h e m i c a l n a t u r e a n d c o m p o s i t i o n : structural units, type a n d degree o f branching, e n d groups, additives, impurities • S t e r i c o r d e r : cis/trans ( Z / E ) isomerism, stereoregularity • C o n f o r m a t i o n a l o r d e r : physical arrangement of the p o l y m e r chains (planar or nonplanar), regular conformations • T h r e e - d i m e n s i o n a l s t a t e o f o r d e r : crystalline a n d amorphous phases, n u m b e r o f chains per unit cell, intermolecular forces, lamellar thickness • O r i e n t a t i o n : type a n d degree o f preferential p o l y m e r chain a n d side group alignment i n anisotropic materials D e s p i t e the uncontested importance o f vibrational spectroscopy for the characterization o f macromolecular structure, it should be emphasized that only a l i m i t e d n u m b e r o f problems may be solved b y its exclusive application. I n the majority of cases m a x i m u m i n f o r m a t i o n o n the structural details i n question can be obtained only b y an appropriate choice a n d combination o f c h e m i c a l and physical methods. I n this respect the i n t r o d u c t i o n o f rapid-scan n i n g F o u r i e r - t r a n s f o r m I R ( F T I R ) a n d F T - R a m a n spectroscopy and the advantages o f these methods over conventional dispersive instrumentation have revitalized the utilization o f vibrational spectroscopy i n p o l y m e r re search. I n combination w i t h other techniques a n d the application o f destruc tion-free sampling procedures, F T I R a n d F T - R a m a n spectroscopy allow a better correlation o f the spectroscopic data w i t h the results obtained b y other methods f r o m the original sample. T h e intention o f the present chapter is to review some aspects o f m i d a n d near-infrared a n d R a m a n spectroscopy i n p o l y m e r research w i t h special emphasis o n selected examples o f diffuse reflectance measurements, compar ison o f photoacoustic F T I R a n d F T - R a m a n spectroscopy, dynamic F T I R characterization o f time- a n d temperature-dependent p h e n o m e n a , a n d the application o f light-fiber optics to process control and remote sensing.

Experimental Details T h e principles, theory, a n d instrumentation o f F T I R spectroscopy have b e e n covered i n detail i n several books a n d reviews (1-4). T h e F T I R spectra presented here were measured o n spectrometers ( N i c o l e t 7199 a n d R r u k e r IFS88). Since the i n t r o d u c t i o n o f laser sources d u r i n g the 1960s, a vast amount o f w o r k has b e e n undertaken using R a m a n spectroscopy to p r o b e a w i d e range o f p o l y m e r i c structures (4, 5-7). C o n v e n t i o n a l R a m a n spectrometers that

2.

SlESLER

Vibrational Spectroscopy of Polymers

43

were used to carry out this w o r k generally consist o f a visible laser source, an efficient double o r triple monochromator, a n d a p h o t o m u l t i p l i e r detection system. H o w e v e r , a n u m b e r o f problems are inherent i n the application o f conventional laser R a m a n spectroscopy to p o l y m e r systems. F o r a large p r o p o r t i o n o f samples, irradiation w i t h visible light caused strong fluores cence o f additives a n d impurities that was superimposed o n the weak R a m a n signal. A n u m b e r o f (sometimes very time-consuming) approaches were attempted to circumvent this p r o b l e m , i n c l u d i n g p r i o r purification, b u r n i n g out the fluorescence, shifting the excitation fine, o r using p u l s e d lasers. These approaches, at best, have b e e n only partially successful. A further p r o b l e m w i t h conventional R a m a n spectroscopy is that highly c o l o r e d samples o r polymers that contain fillers m a y absorb the R a m a n photons w h i c h prevents the photons f r o m reaching the detector a n d leads to t h e r m a l degradation o f the investigated polymer. T h e foregoing difficulties are some o f the reasons w h y the R a m a n technique is not as familiar as infrared techniques, despite the fact that, i n certain circumstances, R a m a n spectroscopy has a n u m b e r o f advantages over infrared spectroscopy. T h e recent development o f F T - R a m a n spectroscopy (8-12), however, increases the l i k e l i h o o d that this tech nique w i l l b e c o m e standard instrumentation i n spectroscopic laboratories. T h e m a i n features o f a n F T - R a m a n instrument are as follows: • a n e o d y m i u m - y t t r i u m a l u m i n u m garnet ( N d - Y A G ) laser, oper ated at 1064 n m w i t h a p o w e r output range o f 0 - 2 W • an efficient

dielectric

filter

system to remove the Rayleigh

component o f the scattered radiation • a F T I R spectrometer e q u i p p e d w i t h a quartz o r c a l c i u m fluo ride b e a m splitter f o r operation i n the near-infrared region • a photoelectric detector; usually a cooled g e r m a n i u m unit (at 77 K ) o r a n i n d i u m - g a l h u m - a r s e n i d e ( I n G a A s ) detector operating at ambient temperature • a R a m a n sampling compartment w i t h 1 8 0 ° o r 9 0 ° lens-based optics T h e use o f near-infrared excitation confers a n u m b e r o f advantages o n a F T - R a m a n system. B o t h fluorescence a n d self-absorption o f the R a m a n signal are very m u c h r e d u c e d and, d u e to the lower energy o f the exciting fight, t h e r m a l degradation is also less o f a p r o b l e m . A further breakthrough i n this field is the development o f a F T - R a m a n microscope system for small samples i n the m i c r o m e t e r range that offers the application o f m a p p i n g procedures (13). H o w e v e r , c o m p a r e d to conventional R a m a n , F T - R a m a n does have some disadvantages and limitations. It is less sensitive because, d u e to the proportionality o f the R a m a n scattering cross section to the f o u r t h

44


p o w e r o f the exciting frequency, the shift f r o m the visible to the near-infrared region reduces the intensity o f R a m a n scattering. F u r t h e r m o r e , as a conse quence o f detector a n d filter limitations, the R a m a n shift c a n presently only be measured between ~ 3500 a n d 100 c m . This long-wavelength cutoff excludes the investigation o f longitudinal acoustic modes i n lamellar p o l y m e r structures (4, 7 ) . - 1

T h e R a m a n spectra presented here were carried out using a F T - R a m a n accessory ( B r u k e r F R A 1 0 6 ) that was interfaced to the F T I R unit ( B r u k e r IFS88).

Diffuse Reflectance FTIR Studies of the Glass Powder-Coupling Agent Interface of a Composite T h e interface between the reinforcement a n d the p o l y m e r material is o f critical importance to the mechanical properties a n d performance o f compos ites, a n d c o u p l i n g agents are u s e d to i m p r o v e the interfacial b o n d i n g (14, 15). Because o f its nondestructive sample preparation procedure, a n d h i g h sensitivity to surfaces, diffuse reflectance F T I R spectroscopy has p r o v e d to b e one o f the most successful techniques to study the reaction of a c o u p l i n g agent w i t h either the resin matrix o r the reinforcement material (16, 17). F r e q u e n t l y , organofunctional silanes are used to promote the adhesion b e tween the inorganic reinforcement a n d the organic p o l y m e r . I n feasibility studies concerning the substitution o f amalgam-based dental filling materials b y glass p o w d e r - p o l y a c r y l a t e composites, vibrational spectroscopy p r o v e d to be an extremely important m e t h o d for investigation o f the glass p o w d e r - c o u p l i n g agent interface a n d the progress o f the photoinitiated c u r i n g o f the m o n o m e l i c diacrylate (18). F o r the composite u n d e r investigation, an a p p l i e d 7-methacryloxypropyltrimethyloxysilane c o u p l i n g agent reacted w i t h the S i O H functionalities o f the glass surface b y condensation a n d f o r m e d bonds w i t h the diacrylate b y polymerization o f the C = C d o u b l e bonds (Structure 1). W i t h respect to a more detailed understanding o f the microstructural changes at the glass p o w d e r - c o u p l i n g agent interface, the diffuse reflectance F T I R spectra o f A e r o s i l O X 50 (average particle diameter 50 n m ; 50 m / g ) were studied as a function o f increasing c o u p l i n g agent coating. A t t e n t i o n was focused o n spectroscopic changes o f n o n - h y d r o g e n - b o n d e d a n d associated S i O H groups o f the glass surface, the increase o f O C H a n d C H functional ities, a n d the variation o f the p r o p o r t i o n o f h y d r o g e n - b o n d e d a n d n o n - h y d r o gen-bonded C = 0 groups. T h e amount o f c o u p l i n g agent was characterized b y the organic carbon content d e t e r m i n e d independently b y elemental analy sis ( 1 % C corresponded to approximately 100 m m o l o f c o u p l i n g agent o n 100-g glass powder). 2

3

2

T h e spectra were measured i n a diffuse reflectance c e l l ( H a r r i c k ) w i t h K C l as the inert reference matrix [sample:matrix about 1:2 (w/w)] a n d 500

3

3

\ 2

3

II I

DCH 3

agent p o l y mer i nteriace

2

Structure 1. The glass powder-coupling agent-polydiacrylate composite interfaces.

coupling

CH 0-Si-(CH ) -D-C-C=CH / CH Q

i nteriace

g 1 ass

/ ι

-Si-OH

3

CH 0

46


scans w e r e accumulated at a resolution o f 2 c m ( 0 . 9 6 4 - c m " data point interval). Spectra o f three selected samples w i t h organic carbon contents o f 0, 0.95, and 1.8% are shown i n F i g u r e 1 and reflect the f o l l o w i n g intensity a n d frequency changes w i t h increasing coating ( F i g u r e 2; ν denotes stretching vibration) (18): - 1

1

• T h e intensity of the sharp v ( O H ) absorption (3740 c m " ) o f n o n - h y d r o g e n - b o n d e d S i O H groups decreases drastically ( F i g ure 2a). 1

• T h e intensity o f the v ( O C H ) absorption o f the S i O C H tionalities at 2848 c m increases ( F i g u r e 2b). 3

3

func

- 1

• T h e intensity o f the v ( C H ) a n d v ( C = 0 ) ( F i g u r e 2c) ab sorptions i n the 2900- a n d 1 7 0 0 - 1 7 2 5 - c m regions, due to the c o u p l i n g agent, naturally increase. A n absorption that can be assigned to free C = C resonance stabilized carbonyl groups gradually increases at 1718 c m " at the expense o f the initial v(C=0) b a n d at 1705 c m ( F i g u r e 2d). 2

t o t a l

- 1

1

- 1

a s s o c i a t e d

• T h e v ( O H ) absorption o f h y d r o g e n - b o n d e d S i O H functionali ties at 3666 c m "

1

slightly decreases.

T h u s , the mechanism for the chemisorption o f the c o u p l i n g agent at the glass p o w d e r interface can be outlined. I n a first step, the free O H groups o f the glass react completely w i t h the available S i O C H functionalities o f the siliconemethacrylate. A s c o n s u m p t i o n o f free S i O H groups increases, the h y d r o g e n - b o n d e d S i O H groups take over this reaction role (although to a smaller degree) and the percentage o f unreacted S i O C H groups increases. D u r i n g this process the percentage of n o n - h y d r o g e n - b o n d e d C = 0 groups o f the c o u p l i n g agent increases at the cost o f the h y d r o g e n - b o n d e d moieties. 3

3

T h e application o f near-IR spectroscopy to determine the residual C = C d o u b l e - b o n d content a n d enable estimation o f the progress o f the diacrylate c u r i n g process w i l l be discussed i n the section o n near-IR spectroscopy.

FTIR Photoacoustic Spectroscopy Compared

to FT-Raman Spectroscopy

T h e request for r a p i d identification o f u n k n o w n p o l y m e r i c materials w i t h varying morphologies is a very c o m m o n p r o b l e m i n analytical laboratories. D u e to the broader availability o f F T I R instrumentation, the task is often accomplished using infrared spectroscopy even though there are frequently difficulties associated w i t h finding a suitable sampling technique. Photoacous tic F T I R as w e l l as F T - R a m a n spectroscopy offer a simple approach to this p r o b l e m . Independently of the morphology, almost any p o l y m e r sample can

2.

SIESLER


47

α in

a

* * κ ·

α *

0%C

{lien Sen 3

£° _J Q UJ »~i CD · ID *-«

α α 4000

36QQ

3200

2800 2400 WAVENUMBERS

2GGQ

16QQ 12QQ

2000

1600

α in in α Q

jii

0.95%C

(η S Sen ID 1 ° Σ: CM

cc

-j α CD

·

α α 4000

3600

3200


1200

Figure 1. Diffuse reflectance FTIR spectra of Aerosil OX 50 coated with increasing amounts (0, 0.95, and 1.8% C) of coupling agent. Continued on next page-

American Chemical Society Library 1155 16th St.. H.W. Washington, DC 20036

48


α in ι Q

α *

3

c

—ι α CQ ·

a

α 10(30

3600

3200

2800 2400 WRVENUMBEBS

2000

1600

1200

Figure 1. Continued. be p l a c e d i n t h e photoacoustic ( P A ) c e l l o r i n the laser b e a m to collect a spectrum i n a matter o f minutes. T o demonstrate the preferential applicabil ity o f F T I R - P A a n d F T - R a m a n spectroscopy f o r the destruction-free i d e n t i fication o f p o l y m e r samples, some selected examples w i l l b e discussed after a short i n t r o d u c t i o n to the basic principles o f photoacoustic F T I R spec troscopy. T h e photoacoustic effect was first reported b y B e l l i n 1881 ( 1 9 ) , b u t the m e t h o d r e m a i n e d i n the b a c k g r o u n d u n t i l a transducer m o r e sensitive a n d selective than the ear c o u l d b e f o u n d . It was not u n t i l 1973 that a renaissance o f photoacoustic spectroscopy o c c u r r e d ( 2 0 ) , a n d since then numerous interesting applications o f this technique have b e e n r e p o r t e d i n the ultravio let, visible, a n d near-infrared regions (21-25). T h e introduction o f F T I R spectroscopy expanded the availability o f P A spectroscopy to the m i d - i n f r a r e d region (25-28) w h e r e it is a p p l i e d quite frequently to the investigation o f polymers ( 2 9 - 3 0 ) . I n F T I R - P A spectroscopy the solid sample is p l a c e d i n a n enclosed c e l l w i t h a n IR-transparent entrance w i n d o w a n d a b u i l t - i n sensitive m i c r o p h o n e ( F i g u r e 3). T h e c e l l contains air or, for increased sensitivity, another c o u p l i n g gas, such as h e l i u m , u n d e r atmospheric pressure ( 2 5 , 31, 32). T h i s c e l l is m o u n t e d i n the sample compartment o f the F T I R spectrometer a n d t h e m i c r o p h o n e takes over the f u n c t i o n o f a conventional F T I R detector. T h e

2.

SIESLER

49

Vibrational Spectroscopy of Polymers ω

t/i h£

CsJ

v(OH) f r e e

ω

Q

Q

Ο

α

.00

1.00

2.00

CARBON

3. 0 0

4 .00

m

(a)

ο

en

v(0CH ) 3

Q

α

ο ο

.00

1.00

2.00 3, 00 CARBON t'A)

4.00

(b) Figure 2. Intensity changes of selected absorption bands as a function of the amount of coating represented by the content of organic carbon. Continued on next page.

50


Figure 2. Continued

2.

SIESLER


modulated IR-radiation

51

IR-transparent window microphone

pressure fluctuations

air-tight cell

gas _ (e.g. air, He)

w

intermittent heating

sample

Figure 3. Scheme of a PA cell and principle of the photoacoustic effect. sample is irradiated w i t h polychromatic light that is m o d u l a t e d w i t h fre quency / : f=2vv

(1)

where ν is the w a v e n u m b e r (per centimeter) and ν is the m i r r o r velocity (centimeters per second) o f the M i c h e l s o n interferometer. W i t h a typical m i r r o r velocity o f 0.1 c m / s , the modulation frequencies vary between 800 and 80 s i n the 4 0 0 0 - 4 0 0 - c m " region, respectively. - 1

1

W h e n radiation o f a certain wavelength is absorbed, the sample is excited to a higher energy level whence it can return to the g r o u n d state b y emission of light or (more c o m m o n l y ) b y nonradiative decay processes or heat release. T h e photoacoustic effect is based o n the heat release p h e n o m e n o n . T h e released heat is transferred to the sample surface at a rate that depends o n the t h e r m a l characteristics o f the sample. D u e to the m o d u l a t i o n of the absorbed light, the nonradiative deactivation processes lead to a p e r i o d i c sample heating that is c o u p l e d into the s u r r o u n d i n g fill gas at the sample-gas interface a n d produces a pressure fluctuation w i t h i n the P A cell. T h e acoustic wave is then propagated through a small connecting passage to the m i c r o phone chamber where the resulting P A s i g n a l — r e c o r d e d as a f u n c t i o n o f w a v e n u m b e r — i s correlated w i t h the optical spectrum. P A spectroscopy has some limitations insofar as the frequency positions of absorption bands are r e p r o d u c e d but the b a n d intensities cannot be interpreted according to the conventional rules o f transmission spectroscopy. T h e signal saturation is a p h e n o m e n o n that is important for understanding the P A effect a n d the interpretation of P A spectra (21, 23, 26, 32-34). T h e amplitude o f the previously m e n t i o n e d periodic heating is decreased b y thermal d a m p i n g processes d u r i n g propagation i n the investigated m e d i u m . It

52

S T R U C T U R E - P R O P E R T Y R E L A T I O N S IN P O L Y M E R S

can b e shown (21, 23) that only the part o f the sample w i t h i n a d e f i n e d distance f r o m the sample surface contributes to the heat transfer to the surrounding gas (23) a n d the intensity o f the P A signal depends o n the coefficients o f optical absorption a n d t h e r m a l diffusion. I n the practical application o f the P A spectroscopy m e t h o d the spectrum o f carbon-black is r e c o r d e d as b a c k g r o u n d a n d then ratioed against the P A spectrum o f the sample u n d e r investigation. Because o f the lower signal-tonoise ratio f o r P A spectroscopy, longer scanning times ( i n the range o f minutes) are r e q u i r e d c o m p a r e d to transmission spectroscopy. T h e P A spec tra presented i n Figures 4, 5a, a n d 6 were taken i n a G i l f o r d - N i c o l e t photoacoustic c e l l w i t h a m i r r o r velocity o f 0.11 c m / s a n d about 500 scans were accumulated w i t h a resolution o f 4 c m (1.928-cm data point interval). Recent r e p r o d u c t i o n o f the data i n a P A c e l l ( M T E C M o d e l 200) o n a F T I R spectrometer ( P e r k i n - E l m e r 1760X) w i t h a m i r r o r velocity o f 0.10 c m / s a n d a resolution o f 8 c m demonstrated that, despite a comparable signal-to-noise ratio, t h e n u m b e r o f scans c o u l d b e p u s h e d d o w n to 32, thereby r e d u c i n g the analysis t i m e considerably (35). -

1

_ 1

- 1

T o demonstrate the potential o f photoacoustic F T I R a n d F T - R a m a n spectroscopy, t h e spectra o f selected polymers i n different morphologies

4000

3000

2000 1600 1200 WAVENUMBERS

800

4 00

4000

3000

2000 1600 1200 WfWENUMBERS

800

400

Figure 4. PA spectroscopy spectra of poly (vinylidene fluoride) granulate in the II (a) modification and fibers in the ΐ(β) modification. Asterisks denote characteristic absorption bands of the 11(a) form.

2.

SIESLER

53


tooo

3Q00

2θδϋ 1500 WAVENUMBERS (a)

1000

500

> ϋ ζ in h

ζ

< < 3500

3000

2500

2000

1500

WAVENUMBER

10

(cm ) -1

(b) Figure

5. FTIR-PA

1000

(a) and FT-Raman (b) spectra of polyi^^phenylene terephthalamide) fibers.

3000

20'00

1500

1000

500

WAVENUMBERS Figure 6. FTIR-PA spectrum of a carbon-black-filled polycarbonate sheet.

54


were examined "as r e c e i v e d . " N o t w i t h s t a n d i n g t h e simplicity o f sample preparation f o r b o t h techniques, t h e F T I R - P A a n d F T - R a m a n spectra o f poly(vinylidene

fluoride)

granulate a n d fibers shown i n Figures 4 a n d 7,

respectively, clearly reflect absorption bands characteristic o f specific confor mational states o f order d u e to the differences i n the mechanical pretreatment o f the p o l y m e r d u r i n g the p r o d u c t i o n process. Whereas the granulate occurs p r i m a r i l y i n the T G T G conformation o f the 11(a) f o r m , the fibers have b e e n transformed largely to the all-trans conformation o f the Ι ( β ) modifica tion.

i500

3000

2500

2000

1500

1000 500

WAVENUMBER

(cm )

WAVENUMBER

(cm" )

-1

1

Figure 7. FT-Raman spectra of poly (vinylidene fluoride) granulate in the 11(a) form and fibers in the l( β) modification.

2.

SIESLER

55


T h e p h e n o m e n o n o f saturation is demonstrated b y the strongly r e d u c e d intensities ( c o m p a r e d to a transmission spectrum) o f the v ( C = 0 ) a n d v ( C - N ) + δ ( Ν - Η ) absorptions o f the P A spectrum o f poly(p-phenylene terephthalamide) fibers ( F i g u r e 5a). T h e excellent quality o f the correspond i n g F T - R a m a n spectrum shown i n F i g u r e 5 b recommends F T - R a m a n spec troscopy as the m e t h o d o f choice f o r such samples where saturation effects dominate the P A spectrum. D e p e n d i n g o n the nature o f the filler, either F T I R - P A o r F T - R a m a n spectroscopy has advantages. I n the case o f carbonblack-filled polymers, F T - R a m a n spectroscopy usually fails to y i e l d useful spectra, whereas the excellent F T I R - P A spectrum obtained f r o m a sheet o f carbon-black-filled polycarbonate ( F i g u r e 6) emphasizes its value as a valu able alternative. O n the other h a n d , highly i n f r a r e d active filler materials such as glass fibers w i l l strongly overlap the actual p o l y m e r spectrum f r o m 1400 cm" to longer wavelengths, whereas they d o not significantly contribute to the F T - R a m a n spectrum ( 3 5 ) . 1

Depth-Profiling in Polymers by Attenuated-Total-Reflection FTIR Spectroscopy A l t h o u g h attenuated-total-reflection ( A T R ) spectroscopy is a well-established m e t h o d for the characterization o f p o l y m e r surfaces ( 3 , 4, 36), comparatively few available publications elucidate the detailed theoretical a n d practical b a c k g r o u n d f o r d e p t h profiling w i t h this technique ( 3 7 - 3 9 ) . T h e p r i n c i p l e o f A T R spectroscopy is based o n the p h e n o m e n o n o f total internal reflection. F o r the experiment the sample (refractive index n ) is brought into direct contact w i t h the surface o f a reflection element o f h i g h refractive index n (n > n ) . W h e n the angle o f incidence o f the light b e a m at the reflection e l e m e n t - s a m p l e interface exceeds the critical angle, total internal reflection takes place a n d the radiation slightly penetrates the optically rarer m e d i u m (4, 36). T h e penetration o f the radiation into the sample is usually o n the order o f a f e w micrometers, w h i c h is sufficient to constitute a short absorbing path. Therefore, total internal reflection w i l l b e attenuated i n the wavelength regions o f sample absorption a n d the r e c o r d e d spectrum w i l l b e very similar to the transmission spectrum. 2

l

2

l

T h e penetration d e p t h (d ) for w h i c h the electric field c o m p o n e n t decreased to 1 /e o f its value i n the reflection e l e m e n t - s a m p l e interface was derived theoretically b y H a r r i c k ( 3 6 ) : P

λ/η,

=

d v

7

1

2ττγ sin Θ 2

(%/%)

(2)

H e r e , Θ is the angle o f incidence, λ is the wavelength o f radiation i n v a c u u m , a n d n a n d n are the refractive indexes. T h i s equation d e m o n 1

2

56

S T R U C T U R E - P R O P E R T Y R E L A T I O N S IN POLYMERS

/

Figure 8. Setup of the two-layer ATR measurements: surface layer of variable thickness, A; and detection layer, B.

strates the dependence o f penetration d e p t h o n the wavelength o f the radiation, the angle o f incidence a n d the refractive indexes o f the sample, a n d the reflection element. F o r the practical application o f the A T R m e t h o d i n p o l y m e r - p o l y m e r interdiffusion experiments, this theoretically d e r i v e d pene tration d e p t h is o f l i m i t e d value. W h a t is actually r e q u i r e d is a measure o f the d e p t h to w h i c h the composition o f the investigated p o l y m e r c a n b e reliably d e t e r m i n e d b y A T R spectroscopy u n d e r specified experimental conditions. T o this e n d , p o l y m e r films w i t h varying, definite thicknesses (surface layer) and a coating detection layer o f " i n f i n i t e " thickness were measured b y A T R - F T I R spectroscopy to determine the thickness o f the surface layer (later referred to as i n f o r m a t i o n d e p t h d^) at w h i c h characteristic absorption bands o f the detection layer are n o longer observable (40). T h e evaluation o f these systematic measurements should t h e n reveal the dependence o f the i n f o r m a tion d e p t h o n the wavelength o f the i n c o m i n g radiation. T h e arrangement o f the surface a n d detection layers o n the A T R crystal is shown i n F i g u r e 8, a n d detailed results w i l l b e presented here w i t h reference to the system w i t h a polystyrene ( P S ) surface layer a n d a polyamide 12 detection layer. T h e reflection element was a K R S 5 parallelepiped w i t h a refractive index o f 2.39, a n angle o f incidence o f 4 5 °, a n d a total n u m b e r o f six reflections. T h e intensities o f the v ( N H ) , amide I, amide II, a n d δ ( Ν Η ) absorption bands o f the polyamide 12 detection layer w e r e d e t e r m i n e d as a function o f increasing thickness o f the polystyrene surface layer ( F i g u r e 9) and their n o r m a l i z e d intensities (Table I) plotted versus P S film thickness. Examples o f such plots are shown for the v ( N H ) a n d amide I bands i n F i g u r e 10. A linear wavelength dependence o f the i n f o r m a t i o n d e p t h d c a n b e d e r i v e d f r o m these data ( F i g u r e 11), a n d comparison o f the theoretically d e r i v e d penetration d e p t h d f o r polystyrene (1.59 refractive index) w i t h the experimentally d e t e r m i n e d i n f o r m a t i o n d e p t h d yields the relation (40) x

T

l

d

l

= 1.29 d

v

(3a)

2.

SlESLER

57


T h i s relation is very close to the relation d e t e r m i n e d f o r p o l y ( 2 , 6 - d i m e t h y l 1,4-phenylene ether) ( P P E ) w i t h polyamide 12 as the detection layer (40, 41),

(3b)

= 1.33 ά

ά

λ

Έ

a n d shows that a proportionality constant o f about 1.31 can b e used f o r any p o l y m e r system u n d e r the specified experimental conditions. I n more d e tailed investigations this i n f o r m a t i o n d e p t h was used as the experimental basis for interdiffusion experiments

o f low-molecular weight polystyrene

into

p o l y ( 2 , 6 - d i m e t h y l - l , 4 - p h e n y l e n e ether) (41).

PS-FILM THICKNESS

3Ê00

3^00

3έϋϋ

3000

WAVENUMBERS

2§00

0

μπα

0.12

μm

0.31

μπι

0.64

μπι

6.78

μπη

2ΪθΟ

Figure 9. ATR spectra of polystyrene (PS) surface layers with different thickness and a polyamide 12 detection layer. Asterisks denote evaluated absorption hands. Continued on next page.

58

S T R U C T U R E - P R O P E R T Y RELATIONS IN POLYMERS

'l7Q0

lèSu

lèûO

ië5G

lèOÛ

TÎ5Ô

Roo

WflVENUMBERS Figure 9. Continued

Variable-Temperature FTIR Measurements and Hydrogen Bonding T h e role p l a y e d b y hydrogen bonds i n the structure a n d properties o f polymers has b e e n the subject o f numerous investigations ( 3 , 4,

42-44).

H y d r o g e n bonds are o f particular importance for the chemical, physical, a n d mechanical properties o f polymers that contain amide (polypeptides, proteins, polyamides), urethane (polyurethanes),

a n d hydroxyl [cellulose, poly(vinyl

alcohol), p o l y (acrylic acid)] functional groups. A l t h o u g h the energies o f hydrogen bonds are weak ( 2 0 - 5 0 k j / m o l ) i n comparison to covalent bonds ( o f the order o f 400 k j / m o l ) , this type o f molecular interaction is large enough to p r o d u c e appreciable frequency a n d

2.

59


SIESLER

Table I. Normalized Integral (») or Peak-Maximum (p) Absorbances of Selected Polyamide 12 Absorption Bands as a Function of Polystyrene Film Thickness (x and y are integration limits)

Sample Number

PS Film Thickness

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 a

c d

v(N-H)

0 0.05 0.07 0.12 0.31 0.63 0.64 1.09 1.33 1.35 1.49 1.53 1.79 2.13 2.31 2.43 2.45 6.78

χ= χ= x = v

m a x

1 0.49 0.51 0.47 0.10 0.04 0.09

— — — — — — — — — — —

v(C=0)

v(C-N) + δ(Ν-Η)

1 0.75 0.82 0.65 0.34 0.26 0.27 0.08 0.09 0.09 0.03 0.02 0.04

1 0.79 0.72 0.64 0.33 0.24 0.25 0.05 0.08 0.08 0.02 0.01 0.01

— — — — —

Ap/Ap δ(Ν-Η) max

1 0.77 0.84 0.84 0.53 0.49 0.53 0.37 0.42 0.42 0.37 0.32 0.28 0.30 0.23 0.28 0.23

— — — — —

—

3412 cm ~ y = 3160 cm , A4 ^'max = 24.28 1687 cm ~ y = 1610 cm A4 = 26.05 1590 cm - y = 1503 c m " A i = 27.29 = 720.6 c m " A p = 0.43 l

- 1

l

- 1

1

1

1

m a x

m a x

intensity changes i n the vibrational spectra. I n fact, the disturbances are so significant, that m i d - i n f r a r e d a n d R a m a n spectroscopy have b e c o m e one o f the most informative sources of criteria for the presence a n d strength hydrogen bonds ( 3 , 4, 44-46).

A n y variation of spectroscopic

of

parameters,

such as intensity, w a v e n u m b e r position, a n d b a n d shape, directly reflects the temperature

dependence

o f the vibrational behavior of the

p o l y m e r as a consequence of changes i n the inter- a n d

investigated

intramolecular

interactions. Unfortunately, only few vibrational spectroscopists

are

aware

that near-infrared spectroscopy offers a p o w e r f u l alternative to gain insight into these phenomena. T o demonstrate the potential of variable temperature measurements for the elucidation of hydrogen-bonding effects, the

results

obtained f r o m m i d - and near-infrared spectroscopic studies of thermoplastic polyamide 12 w i l l be outlined. I n polyamides a n d polyurethanes hydrogen b o n d i n g involves the p r o t o n donating group Ν H a n d the p r o t o n acceptor C = 0 .

M a n y o f the investiga

tions so far reported deal w i t h the observed frequency shift and intensity increase of the v ( N H ) stretching vibration b a n d w h e n hydrogen b o n d i n g

60


I.OQ

Ai/Ai

0.0 0.0

O.S

2.0

1.0

PS-FILM THICKNESS [ μ m3 (a)

Ai/Ai,

1.0

1.5

2.0

PS-FILM THICKNESS

2.5

3.0

3.5

Εμπι3

(b) Figure 10. Normalized integrated absorbanees (A/Ai ) of the v(NH) (a) and v(C = 0) (b) absorption bands of polyamide 12 versus polystyrene (PS) film thickness. The arrow marks the information depth d j . max

4.0

2.

SIESLER

61


10

WAVELENGTH λ c μ m ι Figure 11. Information depth dj of polystyrene as a function of wavelength (the detection layer is polyamide 12).

occurs (44-46). T h e frequency shift Δ ν ( Ν Η ) , i n particular, has b e e n corre lated w i t h various c h e m i c a l a n d physical properties o f the hydrogen b o n d (e.g., enthalpy o f formation, h y d r o g e n - b o n d distance) ( 3 , 4, 45, 46). Increas i n g hydrogen b o n d strength decreases the Ν · · · Ο distance, a n d this decrease is reflected b y an increase i n the shift between the v ( N H ) absorption frequency o f the associated a n d nonassociated N H functionalities. T h e v ( N H ) region o f the m i d - i n f r a r e d spectrum o f the investigated polyamide is shown i n F i g u r e 12. A c c o r d i n g to differential scanning calorimetric ( D S C ) measurements polyamide 12 melts at about 450 Κ a n d the following spectral changes o f the v ( N H ) b a n d are observed w i t h increasing temperature: • T h e intensity o f the v ( N H )

a s s o c i a t e d

b a n d decreases.

• T h e peak m a x i m u m o f the v ( N H )

a s s o c i a t e d

b a n d shifts toward

higher wavenumbers. • T h e half-bandwidth of the ν ( Ν Η )

ω 8 0 { ; ΐ α ί β < 1

b a n d increases consid

erably. • A v(NH)

f r e e

b a n d appears at higher wavenumbers.

T h e w a v e n u m b e r shift a n d increase i n b a n d w i d t h o f the v ( N H ) b a n d at higher temperatures are the result o f a general weakening o f the hydrogen bonds a n d a concomitant broader distribution of the hydrogen b o n d energies. Unfortunately, the intensity of the v ( N H ) b a n d is very l o w i n the m i d - i n f r a r e d range and, additionally, at elevated temperatures the v ( N H ) absorptions o f the free a n d associated functionalities overlap. A total v ( N H ) absorbance procedure for the quantitative assessment o f hydrogen a s s o c i a t e d

f r e e

2

Figwre 22. Mid-infrared spectra of polyamide 12 in the v(NH) and v(CH ) temperature.

absorption region as a function of

50

w

Ο

S

δ

•υ w

>-d so ο

SO M I

to C Ο Η

H

cz>

to

2.

SlESLER


63

b o n d i n g i n polyamides and polyurethanes has b e e n p r o p o s e d b y Srichatrapimuk and C o o p e r ( 4 7 ) . H o w e v e r , for aliphatic polyamides, C o l e m a n et al. (48) have shown that w h e n the strong dependence of the absorption coefficient u p o n b a n d frequency (and, i n t u r n , the strength o f the hydrogen bond) is neglected, the p r o p o r t i o n of n o n b o n d e d N H groups at elevated temperatures d e r i v e d w i t h the foregoing procedure f r o m m i d - i n f r a r e d spectra i n the v ( N H ) region is too high. Inspection of F i g u r e 13 shows that near-infrared spectroscopy certainly offers an alternative b y evaluating the high-frequency w i n g of the 2 X v ( N H ) overtone absorption. F o r a quanti tative evaluation of n o n b o n d e d Ν H groups, however, the p h e n o m e n o n p o i n t e d out b y C o l e m a n et al. (48) as w e l l as the reversal o f the intensity ratio between the fundamentals of the free a n d associated Ν H groups and their first overtones must be taken into account (44). Additionally, contrary to m i d - i n f r a r e d spectroscopy, where the v ( O H ) absorption o f water is s u p e r i m posed b y the dominating v ( N H ) b a n d o f the polyamide, the loss of water d u r i n g t h e r m a l treatment o f the p o l y m e r can readily be quantified b y the intensity decrease of the ν ( O H ) + δ ( Ο Η ) combination b a n d at about 5150 c m " . f r e e

1

Recent investigations (49; H . W . Siesler, u n p u b l i s h e d results) showed that variable-temperature F T I R spectroscopy w i t h p o l a r i z e d radiation can also be advantageously a p p l i e d to study the temperature-induced reorientational m o t i o n o f the mesogenic groups i n liquid-crystalline side-chain poly mers w i t h polyacrylate m a i n chains (50). T h e structure o f the investigated polymer, p o l y ( 1 -(6-(4 ' -cyanophenyl-4-benzyloxy)hexyloxycarbonyl)-ethylene), is shown i n F i g u r e 14 along w i t h the polarization spectra measured at r o o m temperature of a sample whose mesogenic groups were oriented i n the friction direction o f polyimide-coated a n d mechanically pretreated K R r windows. T h e molecular weight o f the p o l y m e r was 20,000 a n d the transition temperature f r o m the nematic to the isotropic phase was 133 °C (49). To study the relative orientational m o t i o n o f the mesogenic groups, the spacer a n d the m a i n chain as a f u n c t i o n of heating, the v ( C = N ) a n d ô ( C H ) absorption bands ( F i g u r e 14) were evaluated i n terms of their orientation functions (see e q 5). T h e sample was subjected to a p r o g r a m m e d heating o f l ° C / m i n and 64 scans were taken i n 1 - m i n intervals w i t h the polarization direction alternately parallel a n d perpendicular to the mechanical reference (friction) direction. D e p e n d i n g o n the t h e r m a l pretreatment, i n repeated measurements a strong reorientational behavior was observed between the glass transition (33 °C) and clearing point (133 °C) b e y o n d w h i c h the sample became isotropic. A l t h o u g h this behavior is not clearly understood, the orientation f u n c t i o n - t e m p e r a t u r e plot proves that, c o m p a r e d to the meso genic groups, the m a i n chains exhibit only relatively small orientational effects obviously as a consequence of small coupling. T h e spacers d e m o n strate an intermediate behavior c o m p a r e d to the mesogenic groups a n d the m a i n chain. 2

Figure 13. Near-infrared spectra of polyamide 12 as a function of temperature. ta

S

δ

I

g

Su Ο M

*d

3d M I

ci

Η

α η

H ta

2

2.

SIESLER


65

Eheooptical FTIR Spectroscopy T h e mechanical properties o f p o l y m e r i c materials are of considerable i m p o r tance to their engineering applications. A p a r t f r o m the chemical structure a n d the thermal history, molecular orientation has a major influence o n the mechanical properties of a polymer. T h e increased n e e d for more detailed data a n d a better understanding of the mechanisms involved i n p o l y m e r deformation has l e d to the search for n e w experimental techniques to characterize transient structural changes d u r i n g mechanical processes. W i t h the advent o f rapid-scanning F T I R spectroscopy, simultaneous vibrational spectroscopic a n d mechanical (so-called rheooptical) measurements d u r i n g the deformation o f polymers emerged. These rheooptical measurements were used as informative probes for the study o f deformation and relaxation p h e n o m e n a i n p o l y m e r films i n the late 1970s; since then they have b e e n applied to obtain data o n strain-induced crystallization and orientational a n d conformational changes d u r i n g mechanical treatment of a w i d e variety o f polymers (4, 46, 51). C o n t r a r y to the technique of time-resolved F T I R spectroscopy, w h i c h has i m p r o v e d t i m e resolution d o w n to the microsecond range due to o r d e r e d interferometric sampling techniques (52), the application of rheooptical F T I R spectroscopy is not restricted to the characterization of reversible structural changes caused b y small-amplitude oscillatory strains; therefore, irreversible p h e n o m e n a over large elongation and recovery scales may be studied (46, 51). T h e experimental p r i n c i p l e of rheooptical F T I R spectroscopy is illus trated i n F i g u r e 15. T h e technique is restricted to a film geometry of the sample and the specimen to be tested is uniaxially drawn a n d recovered i n the sample compartment o f the F T I R spectrometer. D u r i n g the mechanical treatment interferograms can be acquired i n small time intervals ( d o w n to 50 ms). U p o n c o m p l e t i o n of the experiment the interferograms are transformed to the corresponding spectra for further processing of the conventional data. T h e electromechanical apparatus constructed for the deformation a n d relax ation measurements is shown i n F i g u r e 16. F o r orientation measurements the polarization direction of the incident radiation is alternately adjusted parallel a n d perpendicular to the stretching direction b y a pneumatically rotatable w i r e - g r i d polarizer that is also controlled by the computer. T h e construction o f a heating c e l l as a closed, nitrogen-purged system allows the deformation and stress relaxation to be studied u n d e r controlled temperature conditions ( ± 0 . 5 K ) up to 523 K . Presently w e are testing the application of F T - R a m a n spectroscopy for rheooptical measurements. F o r this purpose, the stretching machine shown i n F i g u r e 16 is u t i l i z e d i n back-scattering geometry b y rotating the clamp mechanism for 90 ° w i t h the R a m a n b e a m i m p i n g i n g onto the sample through the front w i n d o w o f the stretching machine. T o improve the back-scattering

C/î

M

*d ο

δ

Η

tu

Μ

*d

Ο

5ΰ

I

ο

Η

ζΛ

Γ0

CD

φ

3obo

zobQ

1 ® iabo

if bo

120Q

WAVENUMBERS

i6bo

2

IGQO

υ yy •® 0) at higher elongations. T h e orientation function o f the v ( C - O - C ) absorption characterizes the soft segments that orient into the direction o f stretch u p o n elongation a n d disorient u p o n recovery. T h e contribution o f h a r d segments to the v ( C H ) absorption a n d the interfacial character o f the C = O functionality results i n deviations f r o m the orientation function behavior of the p u r e soft seg ments. T h e irreversible structural changes—elongations a n d orientations—are reflected b y the spectroscopic as w e l l as the mechanical data. - 1

2

e s t e r

a m i d e

a m i d e

2

e s t e r

T h e ability to m o n i t o r strain-induced crystallization i n the soft-segment phase o f the investigated p o l y m e r is demonstrated i n F i g u r e 21. F o r this purpose the integral structural absorbance ratio of the 9 9 7 - c m conformational-regularity b a n d of crystalline ofigotetrahydrofuran segments ( 5 6 ) a n d the 2 8 0 0 - c m thickness-reference b a n d has b e e n plotted along w i t h the intensity o f the reference b a n d versus strain. A p a r t f r o m thickness changes, the graph clearly demonstrates the appearance a n d disappearance o f straini n d u c e d crystallization as a function of l o a d i n g a n d unloading, respectively. - 1

- 1

Near-Infrared Spectroscopy A l t h o u g h the wavelength region of the near-infrared ( 1 0 , 0 0 0 - 4 0 0 0 c m or 1 0 0 0 - 2 5 0 0 n m ) has b e e n used over m a n y decades for the quantitative analysis o f compounds containing O H , N H , a n d C H functionalities (e.g., determination o f O H n u m b e r , water, protein, or residual d o u b l e - b o n d c o n tent) (4, 57-59), it has never b e e n established as a w i d e s p r e a d analytical a n d physical tool comparable to m i d - i n f r a r e d spectroscopy. - 1

72


I υ

I ζ

3600


2>i00

U ι

0

1800

i>ioo looo WAVENUMBERS

"èoo

Figure 17. Spectral assignment of the IR spectrum of the investigated poly (ether-block-amide).

I n the course o f the 1970s a n d 1980s however, t w o n e w developments initiated a renaissance of near IR-spectroscopy i n analytical chemistry. O n the one hand, chemometric data evaluation techniques i n combination w i t h d i f f u s e - r e f l e c t i o n measurements have o p e n e d u p the possibility o f nonde structive, rational multicomponent analysis o f solid samples. O n the other hand, the introduction of optical light fibers has contributed to an enormous instrumental expansion o f conventional near-IR spectroscopy i n terms o f remote control. F i b e r - o p t i c probes allow a separation o f the spectrometer a n d the sample measurement location over several h u n d r e d meters a n d greatly alleviate the analysis o f toxic or otherwise critical samples i n c l u d i n g process a n d reaction control. T h e basic difference between near- a n d m i d - i n f r a r e d ( 4 0 0 0 - 4 0 0 c m ) spectroscopy is attributed to the fact that the absorption bands observed i n the m i d - I R spectrum can (with f e w exceptions) be assigned to fundamental vibrations o f the investigated molecule, whereas the near-IR absorptions b e l o n g to overtone a n d c o m b i n a t i o n vibrations, p r i m a r i l y o f O H , N H , a n d C H functionalities. T h i s statement, however, may lead to an underestimation of the information content o f a near-IR spectrum because ( i n analogy to a m i d - I R spectrum) the near-IR spectrum also contains i n f o r m a t i o n o n the (1) temperature, (2) inter- a n d intramolecular interactions, (3) t h e r m a l a n d mechanical pretreatment, (4) ionic concentration (aqueous solutions), (5) viscosity/molar mass (polymers), (6) density, a n d (7) particle s i z e - f i b e r d i a m eter ( i n the case o f d i f f u s e - r e f l e c t i o n spectra) o f the sample u n d e r investiga tion. - 1

T h e intensity o f an absorption b a n d decreases b y a factor o f about 1 0 - 1 0 0 i n going f r o m the fundamental to the first overtone, w h i c h necessi tates larger optical pathlengths for the near-IR spectral region (about 1 - 2 m m for u n d i l u t e d samples; u p to 100 m m for solutions). H o w e v e r , this leads

74


to a considerable advantage i n sample h a n d l i n g c o m p a r e d to conventional m i d - I R spectroscopy

(4, 57). T h u s , Figures 2 2 a n d 2 3 demonstrate t h e

m o n i t o r i n g o f the U V - i n d u c e d polymerization o f a diacrylate b y m i d - a n d n e a r - I R spectroscopy, respectively. I n b o t h cases, information o n the amount o f residual double bonds i n the c u r e d resin c a n b e readily derived f r o m the decrease o f the 1 6 0 0 - c m

3500

- 1

p ( C = C ) absorption i n the m i d - I R region a n d the

3000

2600

WAVENUMBERS

WAVENUMBERS Figure 19. Polarization spectra recorded during elongation of the investigated poly (ether-block-amide) up to 300% strain.

2.

SIESLER


75

WAVENUMBERS Figure 19. Continued

2 v ( = C - H ) absorption at 1625 n m i n the n e a r - I R region w i t h the advantage o f a 200-fold increase i n thickness for the n e a r - I R sample a n d a concomitant ease o f sample h a n d l i n g . F u r t h e r m o r e , glass o r quartz, w h i c h are insensitive to water, may be used as w i n d o w materials i n the near-IR region. Instrumen tal advantages o f the light source a n d the detector allow near-IR spectra to be r e c o r d e d w i t h a signal-to-noise ratio signal-to-noise ratio

»

10,000 c o m p a r e d to the m i d - I R

< 10,000. T h i s observation supports the application o f

76


strain

G»

1

recover ,

A

A

α*

Α

δ

Α-ΕΡ

strain

A

A

, recover ,

a

strain

A

A

51

S -EH

V

ο ο "

°

v(NH) (CH ) 2

-A-V(C-O-C)

J

100

200

1 300

r-h 200 200

, 300

h 400

300

300

400

strain % () strain

recover .

strain

, recover .

strain

V c .2 I

§

n(C=0)

AMIDE

-A-V(C=0)

Ο ο

100

200

300

200 200

300

400

300

e s t e r

300

400

strain {%) Figure 20. Orientation function-strain plots of selected absorption bands corresponding to the mechanical treatment outlined in Figure 19.

2.

S IE s LE R

77


CO

strain

recover

strain

recover

• AAA

ι

100

200

300

~1——Γ 200 200

strain

*AAA

-δ le: ο L

A

CD

4300

400

300

300

400

strain (50 Figure 21. Monitoring of strain-induced crystallization in a poly (ether-blockamide) by rheooptical FTIR spectroscopy. chemometrie data evaluation techniques (60). Last, b u t n o t least, the materi als conventionally used as light fibers (glass, quartz) have a n attenuation m i n i m u m i n the near-infrared region (61, 62).

Near-Infrared Diffuse-Reflection Spectroscopy.

Near-infrared

spectroscopy has b e e n u s e d f o r almost t w o decades i n diffuse-reflection measurements f o r the analysis o f agricultural a n d f o o d products b y filter instruments (58, 5 9 ) . T e c h n o l o g i c a l progress has developed to a point w h e r e this technique c a n b e a p p l i e d w i t h scanning instruments i n the area o f c h e m i c a l a n d pharmaceutical m u l t i c o m p o n e n t analysis o f l i q u i d a n d solid formulations a n d i n p o l y m e r analysis. I n p r i n c i p l e , the m e t h o d belongs to the discipline o f chemometrics, w h i c h has b e e n recognized since the mid-1970s (63, 64). T h e purpose o f this discipline is to generate correlations between experimental data (e.g., absorption intensities i n the present case) a n d the c h e m i c a l composition o r physical properties o f the investigated samples b y mathematical a n d statistical procedures [e.g., multilinear wavelength regres sion ( M L W R ) , p r i n c i p a l component analysis ( P C A ) , partial least squares (PLS)]. T h e p r i n c i p l e o f the measurement p r o c e d u r e f o r quantitative analysis is based o n r e c o r d i n g the near-IR transmittance o r diffuse-refl ection spectra o f reference samples (the n u m b e r d e p e n d i n g o n the n u m b e r o f components o r parameters to b e determined) o f k n o w n composition. These spectra are

78


«t000

3000


1000

500

polymer 8/um

h «*000

3000


1000

500

Figure 22. Mid-infrared spectra of the UV-induced polymerization of a diacrylate.

stored i n the c o m p u t e r o f the spectrometer, a n d the levels o f the constituents or the physical parameters are d e t e r m i n e d b y independent, conventional analytical o r physical methods. T h e n t h e set o f reference spectra a n d the independently d e t e r m i n e d values o f the parameters u n d e r investigation are used b y the selected statistical m e t h o d to b u i l d a calibration (57-60). U s e o f this existing standardization a n d t h e near-IR spectra enables the u n k n o w n samples to b e evaluated w i t h regard to the individual parameters o f interest. O n c e t h e calibration has b e e n p e r f o r m e d , t h e analysis t i m e r e q u i r e d f o r a n u n k n o w n sample is drastically r e d u c e d to a f e w minutes i n comparison to t h e several hours o r even days previously r e q u i r e d . P r o m i s i n g applications have b e e n reported ( 5 7 ) for the destruction-free analysis o f polymers o f w i d e l y varying m o r p h o l o g y i n c l u d i n g t h e determina t i o n o f characteristic c h e m i c a l a n d physical parameters o f synthetic fibers. T h e experimental results o n polyacrylonitrile fibers demonstrated here were obtained o n a near-IR spectrometer ( B r a n a n d L u e b b e 500) w i t h the fiber bundles m o u n t e d i n a sample c u p that is positioned u n d e r the integrating sphere. T h e measurement o f opaque solids is p e r f o r m e d i n diffuse reflection, whereas highly transparent solids, pastes, a n d liquids c a n b e measured w i t h the transflection m e t h o d , w h i c h is based o n a diffusely reflecting c e l l b o t t o m

2.

SIESLER

79


ι H ; ιι;j' r

1

11 ΓΓ

monomer ' : ι · Γ. : ι : I

:uj: :Q:

-< :m: -GC:

o

2mm ±2

:

:CQ-

US

S.4 5^

polymer :

Ο - ζ -30.8-

3mm

:

•co-

6pQ4izt8fâQrz2OO0

2200.Γ

WAVELENGTH (nm) Figure 23. Near-infrared spectra of the U\ -induced polymerization of a diacrylate. 7

80


that passes the radiation t h r o u g h t h e sample twice. T h e radiation directly reflected f r o m the gold-coated i n n e r surface o f the integrating sphere serves as a reference intensity ( F i g u r e 24). A r o u n d 100 s are r e q u i r e d to take a spectrum i n the 1 1 0 0 - 2 5 0 0 - n m wavelength range. T h e concentration p r o p o r tional value o f l o g 1 /R is used as t h e ordinate scale:

logl/B = l / X > s

(6)

i C i

w h e r e R = I/I represents the intensity ratio o f light reflected b y the sample a n d t h e reference, respectively, s is the scattering coefficient, a are the absorptivities, a n d c are the concentrations o f the individual components. F r e q u e n t l y the first derivative o f l o g 1 /R is used to reduce the influence o f sample inhomogeneity ( 5 7 , 5 9 , 60). 0

{

{

T h e calibration was based o n t h e l o g 1 /R spectra o f about 70 reference samples. T h e spectrum o f such a reference sample is shown i n F i g u r e 25. I n T a b l e I I , the actual ( A ) a n d n e a r - I R - p r e d i c t e d ( P ) values a n d their residuals ( R ) f o r the parameters u n d e r investigation are shown f o r three test samples. These values demonstrate the potential o f this technique f o r rational c h e m i c a l a n d physical m u l t i c o m p o n e n t analysis.

SAMPLE REFERENCE MEASUREMENT

Figure 24. Simplified optical scheme of a scanning near-infrared diffusereflectance spectrometer.

2.

SIESLER


81

.700 +

.600

0.000 J

Figure

I 1200

1

1 1400

1

1 1600

1 ' 1 ' 1800 2000 WAVELENGTH (nm) 1

25. Near-infrared diffuse-reflectance spectra poly acrylonitnle fiber bundles.

1 ' 2200

of

1 2400

as-received

Table II. Actual (A) and Near-IR-Predicted (P) Values and Their Residuals R (P - A) of the Dimethylformamide (DMF) Solvent Residues and the Preparation, Shrinkage, and Strain of Polyacrylonitrile Fibers Measured as Fiber Bundles in Diffuse Reflection Sample

DMF (%)

Preparation (%)

Shrinkage (%)

Strain (%)

0.9342

0.2715

4.7389

22.9778

A

0.9200

0.2800

5.0000

22.8000

R

0.0142

-0.0085

-0.2611

0.1778

0.6661

0.2655

4.0705

20.4069

A

0.7000

0.2600

4.0000

21.0000

R

-0.0339

0.0055

0.0705

-0.5931

0.7107 0.7300

0.2797

A

0.2700

3.9485 3.9000

22.8066 23.0000

R

0.0193

0.0097

0.0485

-0.1934

IP

2P

3P

Near-Infrared Light-Fiber Spectroscopy. I n conventional transmission spectroscopy, the sample o f interest is measured i n the sample compartment o f the spectrometer, whereas the p r i n c i p l e o f fiber-optic spec troscopy is based o n the transfer o f fight f r o m the spectrometer v i a a suitable d e v i c e — t h e fight fiber—to the sample a n d back to the spectrometer after transmission of, o r reflection f r o m , t h e sample. S u c h light fibers usually consist o f quartz w i t h a length ranging u p to several h u n d r e d meters. Details

82


POLYMER TEST SAMPLE (b) Figure 26. Optical scheme of a near-infrared light-fiber spectrometer with two applications of at-line monitoring: (a) At-line monitoring of process streams and (b) at-line monitoring of molecular orientation.

2.

SIESLER

83


ft i.25

1580 1600 1620 1640 1660 1680 1700 1720 1740 1760 1780 1880 1823

WAUELENGTHCNM] Figure 27. Near-infrared spectra of polystyrene-CCl solutions with different polystyrene concentrations measured on the flow. 4

o f light-fiber geometries, substrates, a n d optical properties are available i n the literature ( 6 1 , 62). T h e obvious advantage o f such a geometry is the ability to separate a n d vary the location o f measurement f r o m the spectrome ter w i t h i n the limits given b y the length o f the fight fiber. Various probes that can b e integrated into different reaction vessels or bypasses offer a m u l t i p l i c ity o f remote sensing o n - a n d at-line process-control applications ( 6 5 ) . T h e optical scheme o f one o f the instruments used f o r o u r studies (the G u i d e d W a v e M 200 spectrometer) is shown i n F i g u r e 26. F i g u r e 26 also includes two applications o f transmission m o n i t o r i n g l i q u i d streams i n quartz tubes (a) a n d molecular orientation i n solid p o l y m e r test samples (b). I n b o t h cases t h e transmitted light is refocused into t h e return waveguide v i a a biconvex lens. T h e spectra obtained b y m o n i t o r i n g p o l y s t y r e n e - C C l solutions o f vari ous concentrations o n the flow i n a 1 9 - m m quartz tube are shown i n F i g u r e 27 ( 5 7 , 65). I n combination w i t h a near-IR polarizer ( G l a n - T h o m p s o n p r i s m ) the same optical configuration offers a very elegant approach for the characterization o f anisotropy i n p o l y m e r i c solids. T h i s anisotropy characteri zation is achieved b y evaluation o f the dichroic effects o f selected functionali ties measured w i t h light p o l a r i z e d parallel a n d perpendicular to t h e d r a w i n g direction o f the investigated p o l y m e r . F i g u r e 28 shows the near-IR polariza tion spectra o f a 2 6 0 % d r a w n polyamide 12 specimen (thickness 1.3 m m ) . T h i s spectra reflect significant dichroic effects o n t h e i n d i v i d u a l absorption bands ( 5 7 , 6 5 ) . C o m p a r i s o n o f the results d e r i v e d f r o m samples w i t h different draw ratios shows that the d i c h r o i c ratios obtained f r o m o n e overtone [e.g., 2 v ( C H ) ] c a n also b e transferred to t h e next overtone ( 3 v ( C H ) ] . N e a r - I R spectroscopy is u n i q u e for this purpose i n that i t offers a destruction-free investigation even o f thick specimens i n their original morphology. 4

2

2

84


ft 0,8026 Β 0.6548 0.5076 3 χ V(CH ) 0.3604

/

2

0.2131 0.0659 0.0813

ι—ι—ι—ι—ι—ι—ι—ι—ι—ι—ι—I

1200 1267 1333 1400 1467 1533 1600 1667 1733 1800 1867 1933 2000 HAUELEHGTHtNNl Figure 28. Near-infrared polarization spectra of a 260% drawn polyamide 12 test specimen measured with light polarized parallel (II ) and perpendicular ( ±) to the stretching direction.

Figure 29. Near-infrared light-fiber ATR sensor (see text). NA denotes the numerical aperture; n , n and n are the refractive indexes of air, the core, and the cladding, respectively. 0

l 5

2

2.

SIESLER


85

A n application that shows promise is the modification o f a c o m m e r c i a l light fiber based o n a quartz core, a poly(dimethylsiloxane) cladding, a n d a polyamide coating ( F i g u r e 29) to act as a n A T R sensor i n the near-infrared region ( 6 6 ) . R e m o v a l o f the polyamide coating b y dissolution (over distances o f several meters) allows the r e m a i n i n g light fiber to b e w o u n d a r o u n d a T e f l o n holder a n d i m m e r s e d into the solution f o r analysis. This m e t h o d is based p r i n c i p a l l y o n the effect that the c o m p o u n d to b e d e t e r m i n e d c a n b e e n r i c h e d i n the p o l y (dime thylsiloxane) c l a d d i n g (with n o interference f r o m water, f o r example) a n d detected via total internal reflection o f the near-IR b e a m at the c o r e - c l a d d i n g interface. T o compensate f o r the absorptions o f the poly(dimethylsiloxane) cladding, a b a c k g r o u n d has to b e measured against air. A similar application has b e e n described f o r the m i d - i n f r a r e d region w i t h a polymer-coated internal reflection element ( 6 7 ) a n d a T e f l o n - c l a d d i n g fluoride glass optical fiber (68).

Acknowledgments T h e financial and instrumental support o f D e u t s c h e Forschungsgemeinschaft, R o n n , G e r m a n y , M i n i s t e r i u m f u r Wissenschaft u n d F o r s c h u n g des Landes N o r d r h e i n - W e s t f a l e n , Dusseldorf, G e r m a n y , F o n d s d e r C h e m i s c h e n Indus trie, F r a n k f u r t , G e r m a n y , H i i l s A G , M a r l , G e r m a n y , Bayer A G , D o r m a g e n , G e r m a n y , a n d B r a n a n d L u e b b e G m b H , Norderstedt, G e r m a n y , are grate fully acknowledged. T h e author also thanks S. D e k i e r t , B . Feldhâuser, U . Becker, P . W u , a n d I. Grose for experimental assistance.

References 1. Geick, R. In Topics in Current Chemistry; Springer: Berlin, Germany, 1975; Vol. 58, p 73. 2. Griffiths, P. R.; deHaseth, J. A. Fourier-Transform Infrared Spectroscopy; Chem ical Analysis Series 83; Wiley: New York, 1986. 3. Koenig, J. L. Adv. Polym. Sci. 1983, 54, 87. 4. Siesler, H . W.; Holland-Moritz, K. Infrared and Raman Spectroscopy of Polymers; Dekker: New York, 1980. 5. Gilson, T. R.; Hendra, P. J. Laser-Raman Spectroscopy; Wiley: New York, 1970. 6. Koenig, J. L. Appl. Spectrosc. Rev. 1971, 4(2), 233. 7. Cutler, D. J.; Hendra, P. J.; Fraser, G. In Developments in Polymer Characterization; Dawkins, J. V., Ed.; Applied Science Publishers: London, England, 1980; V o l . 2, p 71. 8. Hirschfeld, T.; Chase, B. Appl. Spectrosc. 1986, 40, 133. 9. Zimba, C. G.; Hallmark, V. M.; Swalen, J. D.; Rabolt, J. F. Appl. Spectrosc. 1987, 41, 721. 10. Hendra, P. J., Ed. Spectrochim. Acta 1990, 46A(2), 121. 11. Schrader, B.; Hoffmann, Α.; Simon, Α.; Sawatzki, J. Vibrational Spectrosc. 1991, 1, 239. 12. Hendra, P. J.; Jones, C.; Warnes, G. Fourier Transform Raman Spectroscopy; Ellis Horwood: Chichester, England, 1991. 13. Sawatzki, J.; Simon, A. In XXIIth International Conference on Raman Spec-

86


troscopy, August 13-17, 1990, Columbia, South Carolina; Durig, J. R.; Sullivan, J. F., Eds.; Wiley: Chichester, England, 1990. 14. Plueddemann, E. P. Silane Coupling Agents; Plenum: New York, 1983. 15. Ishida, H.; Koenig, J. L. Polym. Eng. Sci. 1978, 18, 128. 16. Graf, T. R.; Koenig, J. L.; Ishida, H . Anal. Chem. 1984, 56, 773. 17. Davis, J. Α.; Sood, A. Makromol. Chem. 1985, 186, 1631. 18. Siesler, H . W. Mikrochim. Acta (Wien) 1988, I, 319. 19. Bell, A G. Philos. Mag. 1881, 11, 510. 20. Rosencwaig, A. Opt. Commun. 1973, 7, 305. 21. Rosencwaig, A. Photoacoustics and Photoacoustic Spectroscopy; Wiley: New York, 1980. 22. Pao, Y.-H., Ed. Optoacoustic Spectroscopy and Detection; Academic: Orlando, FL, 1977. 23. Somoano, R. B. Angew. Chem. 1978, 90, 250. 24. Hunter, T. F.; Turtle, P. C. In Advances in IR and Raman Spectroscopy; Clark, R. J. H.; Hester, R. E., Eds.; Heyden: London, England, 1980. 25. Coufal, H.; McClelland, J. F. J. Mol. Struct. 1988, 173, 129. 26. Graham, J. Α.; Grim, W. M., III; Fateley, W. G. In FTIR Spectroscopy—Applica tions to Chemical Systems; Ferraro, J. R.; Basile, L. J., Eds.; Academic: Orlando, FL, 1985; Vol. 4,p345. 27. Vidrine, D. W. Appl. Spectrosc. 1980, 34, 314. 28. Rockley, M . G. Appl. Spectrosc. 1980, 34, 405. 29. Yang, C. Q.; Fateley, W. G. Anal. Chim. Acta 1987, 194, 303. 30. Urban, M . W. J. Coat. Technol. 1987, 59, 29. 31. Adams, M . J.; King, Α. Α.; Kirkbright, G. F. Analyst 1976, 101, 73. 32. McClelland, J. F.; Kniseley, R. N. Appl. Opt. 1976, 13, 2658. 33. Rosencwaig, Α.; Gersho, A. Science 1975, 19, 556. 34. Rosencwaig, Α.; Gersho, A. J. Appl. Phys. 1976, 47, 64. 35. Grose, R. I.; Hvilsted, S.; Siesler, H . W. Makromol. Chem., Makromol. Symp. 1991, 52, 175. 36. Harrick, N . J. Internal Reflection Spectroscopy; Interscience: New York, 1980. 37. Gidaly, G.; Kellner, R. Fresenius Z. Anal. Chem. 1980, 302, 257. 38. Ohta, K.; Iwamoto, R. Appl. Spectrosc. 1985, 39, 418. 39. Blackwell, C. S.; Degen, P. J.; Osterholtz, F. D. Appl. Spectrosc. 1978, 32, 480. 40. Becker, U. M.Sc. Thesis, University of Essen, Essen, Germany, 1988. 41. Machate, Ch., Ph.D. Thesis, University of Münster, Münster, Germany, 1989. 42. Painter, P. C.; Coleman, M . M . ; Koenig, J. L. The Theory of Vibrational Spectroscopy and Its Applications to Polymeric Materials; Wiley-Interscience: New York, 1982. 43. The Hydrogen Bond; Schuster, P.; Zundel, G.; Sandorfy, C., Eds.; North-Holland: New York, 1976. 44. Vinogradov, S. N.; Linnell, R. H . Hydrogen Bonding; Van-Nostrand Reinhold: New York, 1972. 45. Murthy, A. S. N.; Rao, C. N . R. Appl. Spectrosc. Rev. 1968, 2, 69. 46. Siesler, H . W. Adv. Polym. Sci. 1984, 65, 1. 47. Srichatrapimuk, V. W.; Cooper, S. L. J. Macromol. Sci. Phys. 1978, B15, 267. 48. Skrovanek, D. J.; Painter, P. C.; Coleman, M . M . Macromolecules 1985, 18, 299 and 1676. 49. Bürkle, K.-R. Ph.D. Thesis, University of Ulm, Ulm, Germany, 1987. 50. Zentel, R.; Benalia, M . Makromol. Chem. 1987, 188, 665. 51. Siesler, H . W. Makromol. Chem., Makromol. Symp. 1992, 53, 89. 52. Noda, I. Appl. Spectrosc. 1990, 44(4), 550.

2.

SIESLER


87

53. Ishihara, H.; Kimura, I.; Saito, Κ.; Ono, H . J. Macromol. Sci. Phys. 1974, B10, 591. 54. Lohmar, J.; Meyer, K.; Goldbach, G. Makromol. Chem. 1988, 189, 2053. 55. Bonart, R.; Hoffmann, K. Colloid Polym. Sci. 1982, 260, 268. 56. Tadokoro, H.; Kobayashi, M . In Polymer Spectroscopy; Hummel, D. O., Ed.; Verlag Chemie: Weinheim, Germany, 1974; p. 1. 57. Siesler, H . W. Makromol. Chem., Makromol. Symp. 1991, 52, 113. 58. Weyer, L. G. Appl. Spectrosc. Rev. 1985, 21, 1. 59. Osborne, B. G.; Fearn, T. Near Infrared Spectroscopy in Food Analysis; Wiley: New York, 1986. 60. Hirschfeld, T.; Stark, E. In Analysis of Food and Beverages; Charamboulos, G., Ed.; Academic: Orlando, F L , 1984; p 505. 61. Herbrechtsmeier, P. Chem. Ing. Tech. 1987, 59, 637. 62. Frank, W. Fernmelde-Ing. 1990, 44(3), 1. 63. Sharaf, Μ. Α.; Illman, D. L.; Kowalski, B. R. Chemometrics; Wiley-Interscience: New York, 1986. 64. Martens, H.; Naes, T. Multivariate Calibration; Wiley: Chichester, England, 1989. 65. Feldhäuser, Β. M.Sc. Thesis, University of Essen, Essen, Germany, 1988. 66. Bürck, J.; Conzen, J.-P.; Ache, H.-J. Fresenius J. Anal. Chem. 1992, 342, 394. 67. Heinrich, P.; Wyzgol, R.; Schrader, B.; Hatzilazaru, Α.; Lübbers, D. W. Appl. Spectrosc. 1990, 44(10), 1641. 68. Ruddy, V.; McCabe, S. Appl. Spectrosc. 1990, 44(9), 1461. RECEIVED for review May 14, 1991. ACCEPTED revised manuscript September 22, 1992.

Vibrational Spectroscopy of Polymers - ACS Publications - American

Recommend Documents