Farrow

The Measurement of Crystallinity in Polypropylene Fibres by X-ray Diffraction G. FARROW A method has been devised [or the measurement o[ crystallinity in oriented polypropylene fibres by an X-ray technique. The general level of crystallinity is higher than in similar oriented poly(ethylene terephthalate) fibres. The results also show that the crystallinity so measured is lower than values in[erred from density measurements and that no correlation exists between the two methods. A possible method is suggested [or the determination of the atactic content of polypropylene polymer by X-ray diffraction.

BECAUSE of its considered importance in determining physical properties of textile fibres various physical methods have been proposed and developed for the measurement of crystallinity in polymers and fibres. Amongst these was an X-ray method for the measurement of the degree of crystallinity in oriented poly(ethylene terephthalate) fibre#. This has now been adapted to the measurement of the degree of crystallinity in oriented polypropylene fibres. EXPERIMENTAL

(i)

Sample preparation

The material used for this work was polypropylene filaments and chip of varying crystallinity. The filaments used in this investigation were prepared by a conventional two-stage process ~. Molten polymer is first extruded and collected as filaments with slight molecular orientation. Filaments of high orientation are then produced by the application of continuous extension to a pre-set limit. This second drawing stage is carried out by a machine which, in essence, consists of two pairs of roils with either one or two heating devices in between; the temperature of one heating device is maintained at 95°C and the other, when used, at 145°C. One pair of rolls feeds the undrawn filaments from the spinning machine forward at a constant rate and the second pair moves at a higher velocity, drawing the filaments over the heated surface(s), Filaments of different draw ratios (i.e. ratios of velocity of feed roils to velocity of draw rolls) were produced by this method. A further treatment of all the filaments, to give an additional set of samples, was carried out in a vacuum oven at 145°C for half an hour. (ii) X-ray determination of crystallinity The experimental technique is similar to that already described in detail for the measurement of the degree of crystallinity in oriented poly(ethylene terephthalate) fibres 1. Only a brief summary is, therefore, given here. Oriented fibres are chopped and made into a randomized sample by a pelleting technique. The preparation of the pellet is rather more critical than 409

G. FARROW with oriented poly(ethylene terephthalate) fibres for if the fibre is insufficiently randomized quite significant changes occur in the ratios between the principal X-ray reflections when compared with similar measured ratios from a sample of heat crystallized unoriented polypropy~lene yarn, used as a standard. The prepared sample is placed on the circumference of a focusing camera (evacuated to eliminate 'air scatter') and exposed to a strictly monochromatic beam of X-rays (quartz bent crystal monochromator). The X-ray diffraction lines are recorded on film, which, after processing and drying, is scanned by a microdensitometer. From the resultant trace the crystallinity is determined by measuring the integrated area of the crystalline reflections and the integrated area of the non-crystalline background and comparing the two. With the microdensitometer i n use the total area of the trace can be determined automaticallys. In order to determine the non-crystalline or amorphous background, however, when the X-ray reflections from the crystalline and amorphous regions overlap as in polypropylene, it is necessary to obtain a diffraction pattern of an amorphous specimen of the same type of material. The trace from this is then used as a template for drawing in the non-crystalline background. X-ray photographs of polypropylene samples at high temperatures in the melt were obtained and used for this purpose. The samples were held in the inner chamber of the X-ray furnace4 (in' vacuo) in a small cylindrical container made of aluminium foil 0"0005 in. in thickness with a thermocouple placed in contact with one end of the sample. With Cu Ka X-ray radiation (strictly monochromatic, in this case), the absorption of the primary X-ray beam by this thickness of aluminium is negligible. Furthermore, any X-ray diffraction pattern produced from the aluminium foil falls outside the principal X-ray reflections of polypropylene. Two microdensitometer traces, from high temperature X-ray diffraction photographs, are shown in Figure 1.

Samples 'Atactic'

.... ....

/

Integrated

///

~

~'-

{

I

2 O = 0°

Figure

'Commercia('

]--Traces

28-_32° o f t w o samples o f p o l y p r o p y ] o n e iB the m e l t fron~ h i g h

temperature X-ray photography One is a commercial sample of polypropylene (intrinsic viscosity 2"7) and the other a so-called atactic sample of polypropylene (hot heptane extract) of high molecular weight which showed some crystallinity at room temperature. It can be seen that there is a difference in shape between the two curves although not a big difference in areag The cur~e from the sample of polypropyleno is asymmetrical in shape and is similar to published curves used by Natta et al. ~ for measurements of crystallinity, by X-ray diffraction, on unoriented specimens of polypropylene. This curve has been used for our crystallinity measurements. 410

T H E M E A S U R E M E N T OF CRYSTALL1NITY IN P O L Y P R O P Y L E N E FIBRES

The drawing in of the non-crystalline background is illustrated in

Figure 2 for some drawn polypropylene filaments. 110

131 041

Integrated intensity

I

• 20:0" 20=34 o Figure 2--Microdensitometer trace of X-ray diffraction photograph of drawn polypropylene filaments. Shaded area: non-crystalline background; rise in background : incoherent scatter

Points A and B are used as reference points for determining the height of the non-crystalline background, and the crystallinity of a sample is then equal to area of crystalline fraction area of crystalline fraction + area of the amorphous fraction (y) where y is a factor, necessary to correct for the non-coincidence of the centres of gravity of the amorphous and crystalline reflectionsL A factor of 0-98 was obtained for polypropylene with the present focusing camera, the crystalline reflections being measured over the range 20=6 ° to 20= 38 ° . In this set of measurements on drawn fibres of polypropylene the following four assumptions have been made. (1) The shape of the non-crystalline background deduced from the high temperature X-ray photographs is the same as that which would exist at room temperature. (2) The shape of the non-crystalline background is invariant and only reduced in proportion in a partially crystall'ine sample even with drawn yarns. (3) The thermal motions of the atoms in the crystalline regions do not make an appreciable contribution to the non-crystalline background. (4) The scattering efficiencies of the crystalline and amorphous regions are equivalent. It appears, from Figure 3, that the effect of high temperature is not to change the shape of the amorphous curve over the angular range considered but only to increase the mean distance between molecular chains by approximately 0'5A. An amorphous sample of poly(ethylene terephthalate) polymer when examined at room temperature and in the melt also behaved in a manner analogous to the polypropylene sample. ' ' Trace at room temp. referred to A - - - T r a c e at 170~ referred to A Integrated

.

I~,~"

intensity 20 :l~a"

/

t

!

I

~ -

--

'

'

, ~

~'~ 0-5~ ]'~ 219=32° Displacement "" Figure 3--Microdensitometer trace of X-ray diffraction photograph of polypropylene in the melt and its correct angular displacement at room temperature

411

G. FARROW A number of highly crystalline specimens of drawn and undrawn fibres of polypropylene were examined over a wide range of temperature ( - 150°C up to the melt). The resolution on the X-ray photographs was sufficiently good to enable the non-crystalline backgrounds to be drawn in to a reasonable approximation without making any assumptions about shape at all. On scaling these curves to the one deduced from the melt they were found to be almost identical except for the angular displacement already referred to in the previous paragraph. This result lends support to assumptions (2) and (3). In particular, if assumption three was invalid the deduced noncrystalline backgrounds should show significant differences at different temperatures. The fourth assumption appears reasonable from an examination of the experimental data. The X-ray diffraction photographs taken under standard conditions and scanned by a microdensitometer gave total integrated areas which were approximately constant irrespective of the actual crystalline/amorphous ratio. (iii) Density In the density method of measuring crystallinity in a polymer it is assumed that only two kinds of material exist, viz. crystalline and amorphous. If the densities of a wholly crystalline and a wholly amorphous sample are known, then, by measuring the density of the sample of unknown crystallinity and applying simple proportion the crystallinity of the specimen can be inferred. With polypropylene it is impossible to achieve either of the above two states. The density of the crystallites can, however, be calculated from the crystal structure ~ and has the value of 0.936 g/cm. ~ On the other hand rapid quenching of polypropylene samples does not produce an amorphous specimen as occurs with poly(ethylene terephthalate) polymer. In order to determine an amorphous density it is necessary to use indirect methods. The method followed here was to assume that a linear relationship exists [as with unoriented poly(ethylene terephthalate) samples 1] between the crystallinity measured by X-rays and density for unoriented specimens of polypropylene. Samples of different crystallinity were produced by heat crystallization in a vacuum oven at different temperatures. The best straight line was obtained by a 'least squares' analysis and the results obtained are shown in Figure 4. The value deduced for the amorphous density, by extrapolation, is given in Table 1 together with three other values obtained from the literature. It can be seen that the value deduced from the X-ray measurements agrees remarkably well with the value obtained by extrapolation for an Table 1 Polypropylene sample

Amorphous 'Extrapolated' Amorphous 'Experimentalwholepolymer' Amorphous 'Ether extract' Amorphous 'Ether extract'

Density, g/cm3

0 ~870 0.870

Method

X-ray as described DilatometryS Extrapolated from melt 0"855 DilatometryS 0"85 to 0"855 Densitygradient column9 412

THE MEASUREMENT OF CRYSTALLINITY IN POLYPROPYLENE FIBRES

100

Density from unit c e t t - / ~ dimension

-

80-

60 "E

20'

// / / /

i

0.86~ 0-88 ExtrapoLated/

density

I

i

0.90 092 0.94 Density g/crr9

Figure 4--Crystallinity (X-ray) versus density of heat crystallizedsamples of polypropylenespun fibre 'experimental whole polymer' (such a polymer would be expected to be similar to the polymer used in these experiments) but disagrees with the other quoted values. It is considered, however, that the value of 0"870 g/cm ~ is a reasonable figure to use for the calculation of crystallinities from density measurements. Specimens obtained at the Fibres Division, by similar extraction techniques, tended to be of low molecular weight and showed X-ray diffraction lines which were probably due to low molecular weight oligomers of polypropylene1°. Furthermore, it is not necessarily true that the density of an atactic polypropylene should have the same density as the non-crystalline regions in a partially crystalline but essentially isotactic polypropylene. It was shown in the previous section that X-ray diffraction photographs of polypropylenes of high and low atactic content respectively showed differences when scanned by a microdensitometer. (iv) Optical measurements The optical birefringences of samples used in Figure 5 were measured using polarized light and a Berek compensator.

Results--The results are presented in Tables 2 and 3 which list crystallinities measured both by X-ray diffraction and density for a variety of polypropylene filaments. Taking the X-ray measure of crystallinity as the yardstick and knowing the density of the crystalline regions ~ the density of the non-crystalline material in oriented filaments can be calculated. These values are given in the columns labelled 'oriented non-crystalline'. Some additional results are also presented in graphical form in Figure 5 and will be referred to in the discussion. 413

G. FARROW

The yams were kept in a refrigerator after drawing and allowed to condition for 48 hours at room temperature before measurements were taken. The densities were measured in a graded density column 11 maintained at 30°C and the results are the means of two separate measurements. Table 2 Filament draw ratio

Density g/cm 3

(1) (2) (3) (4) (5)

5.0 5- 5 6.0 6.5 6- 9

0.903 0-898 0-902 0"903 0"905

49 44 48 49 52

37 39 37 37 39

(6) (7)

5-0 5'5 6"0 6-5 6-9

0-909 0'912 0.913 0"920 0-920

58 63 64 75 75

47 48 47 47 49

(8) (9) (It))

F i l a m e n t s 1 to Filaments 6 to

°//oCrystallinity (density)

°//oCrystallinity ] Density (X-ray) j orientednoncrystaHine , ~

0"878 0.873 0"877 0"878 0"879

j :

0"878 0"880 0"880 0.889 0"887

S p u n yarn erystallinity I(X-~ay)=33 per cent. 5 p a s s e d over heating device at a m a x i m u m tempera.ture o f device at a m a x i m u m temperature o f f o l l o w e d by o n e at a m a x i m u m temperature ol 145°C

10 passed over heating

95"C 95°C

Table 3 Filament draw ratio

Density g/cm 3

(1) (2) (3) (4) (5)

5.0 5- 5 6.0 6.5 6.9

0.906 0.910 0.911 0.914 0,916

(6) (7) (9) (10)

5.0 5.5 6- 5 6.9

0" 910 0"914 0"916 0-917

I

.!.

% Crystallinity (density)

% Crystallinity (X-ray)

54 60 61 66 69

52 53 54 58 58

0"871 0" 875 0 ' 875 0" 875 0" 878

60 66 69 70

54 58 57 55

0" 874 0"875 0- 878 0"882

Table,2 b u t heat crystallized in a vacuum oven for ½ hour Heat crystallized chip at 145"C, 66 per cent crystalline.

Y a r n c o n d i t i o n s as in

Density oriented noncrystalline

at

145"C.

The X - r a y crystallinities h a v e an estimated a c c u r a c y of + 3 per cent. DISCUSSION

(i)

X-ray crystallinity

It can be seen from the results that the method of measuring the amount of crystalline material by X-ray diffraction, developed for oriented polyethylene terephthalate fibres can easily be applied to oriented polypropylene filaments. The general level of crystallinity in oriented polypropylene fibres is higher than that found for similarly treated polyethylene terephthalate fibres; furthermore, polypropylene spun fibre has a crystallinity of 33 per cent which is almost the same as that found in the polypropylene filaments, 414

THE MEASUREMENT OF CRYSTALLINITY IN POLYPROPYLENE FIBRES

Table 2, 1 to 5. There is, however, a marked difference in the widths of the X-ray reflections between the spun fibre and these filaments. The X-ray diffraction lines in the former case are broader indicating that the crystallites in the spun fibre are much smaller than those found in these filaments. Furthermore the general level of crystallinity in the heat-treated drawn filaments (~-~ 56 per cent) is less, by X-ray diffraction, than is found for heat-treated unoriented fibre or chip (,~ 65 per cent) at the same temperature and under the same conditions. Some recent work has suggested that this is simply a 'rate effect', for on repeated annealing of drawn filaments the crystallinity gradually rises to the same level as that found in undrawn fibres ~°. Molecular chains in the oriented state are presumably less mobile than in the unoriented state. (ii) X-ray crystallinity and density Examination of Tables 2 and 3 shows that the relationship between X-ray crystallinity and density is different for unoriented and oriented filaments. In the latter case the degree of crystallinity measured by density is higher than the corresponding crystallinity measured by X-ray diffraction. If a lower value (Table 1) was used for the density of the amorphous regions of polypropylene then the difference between the results would be even greater. It is considered that the discrepancy arises between the two methods because [as with oriented poly(ethylene terephthalate) fibres] it is no longer permissible, in drawn filaments, to regard the non-crystalline material as having a constant density ~, i.e. regions of semi-order exist which are noncrystalline but have a higher density. The higher the molecular orientation the higher is the discrepancy between the two methods up to birefringence values of ~ = 20 × 10 -3. This is illustrated in Figure 5 for filaments which cover a range of A -- 5 > 2 2 x 1 0 -3 . 6o

_..~"'~'0" .

.

t..)

o 2G

l

5

Figure .5--Differences

in crystallinity measured by X-ray diffraction and density with increasing molecularorientation

....... . . I

{

X-ray Density {

10 15 20 ,4 x 10 -3 Birefringence

25

The difference in crystallinity measured by the two methods is quite variable, particularly for the filaments 6 to 10 in Table 2. This is reflected in the calculation of an 'oriented non-crystalline density'. In view of these results, the use of density as a measure of the degree o f crystallinity of oriented polypropylene should, therefore, be applied with caution. (iii) Atactic content of polypropylene There is clearly some connection between the degree of crystallinity and the atactic content of a polymer, for, if considerable lengths of a molecular chain are regularly substituted, a n d if a number of such chains are brought 415

G. FARROW

together under the correct conditions, crystallization would be expected to occur to a certain degree. It should, therefore, be possible under certain circumstances to relate the degree of crystaUinity to the degree of tacticity of the polymer. It is obvious, however, that if the stereo-blocks, as they may be called, are low in number and short in length, then the chances of a number of them coming together to form a three-dimensional network are remote and consequently an estimate of the tacticity, from crystallinity measurements, in such circumstances could be highly inaccurate. Such an extreme situation does not appear to exist with the present polymer in use and it is of interest to calculate a value for the atactic content derived from X-ray measurements. In order to make this estimate of atactic content it is necessary to crystallize polymer to a maximum degree (a similar idea has been put forward by Nielsen13). With the polypropylene used in these experiments values between 65 and 70 per cent have been arrived at. It is also necessary to know the crystallinity of a polymer which is completely regular (i.e. 100 per cent isotactic material). Some estimate for this can be made from measurements of crystallinity of heat-crystallized low pressure polyethylenes1" (i.e. where very little branching occurs) where values of approximately 85 to 90 per cent have been arrived at. Natta 5 has also quoted a crystallinity value of 82 per cent for 'isotactic' polypropylene. If a value of 85 to 90 per cent is assumed for the highest expected crystallinity of a completely isotactic polypropylene (this seems a reasonable assumption because dislocations, entanglements, etc. could account for the other 10 to 15 per cent of the material) a value for the atactic content of the commercial polypropylene, used in these experiments, can then be calculated. Values are assumed intermediate between the limits set out in the previous paragraph. We have, therefore Maximum attainable degree of Maximum attainable degree of crystallinity of a complete iso- - crystallinity of commeretia.1 tactic polypropylene (87"5 per sample of polypropylene (67"5 cent) per cent) = atactic content of commercial polypropylene, ---20 per cent. It is hoped to check this method of measurement against values obtained by a proposed infra-red method of measurement recently put forward by McDonald and Ward is. Others, notably Brader 1~, have also suggested that certain absorption bands in the infra-red spectra of polypropylene are related to tacticity. McDonald and Ward have, furthermore, defined tacticity in their paper. They state that a monomer unit can be considered in an isotactic environment when both its adjacent neighbours on the chain are substituted in a stereo identical manner. It should be pointed out that this definition, which is perfectly adequate for infra-red purposes, would not be so for X-ray measurements. A randomly substituted polypropylene which might be considered atactic would almost certainly have some isotactie content (on statistical grounds alone) when such short lengths of the molecular chain are considered. Such a polymer examined by X-ray diffraction would almost certainly be amorphous (or atactic). 416

THE MEASUREMENT OF CRYSTALLINITY IN POLYPROPYLENE FIBRES CONCLUSIONS A n X - r a y m e t h o d has been devised for the m e a s u r e m e n t of crystallinity in o r i e n t e d p o l y p r o p y l e n e filaments. T h e levels of crystallinity in these filaments are high c o m p a r e d with c o m p a r a b l e oriented poly(ethylene t e r e p h t h a l a t e ) filaments. A m e t h o d of d e t e r m i n i n g the atactic content of p o l y p r o p y l e n e b y X - r a y diffraction has also been suggested.

The author would like to thank Dr K. W. Hillier for helpful criticisms o~ the manuscript. Fibres Division, I.C.I, Ltd, Hookstone Road, Harrogate, Yorkshire (Received August 1961) REFERENCES 1 FARROW,G. and PRESTON, D. Brit. J. appl. Phys. 1960, 11, 353 a HILL, R. (Ed.). Fibres from Synthetic Polymers, Chap. 14. Elsevier: London, 1953 3 FARROW,G. and PRESTON, D. J. sci. lnstrum. 1960, 37, 347 4 FARROW,G. and PRESTON, D. J. sci. lnstrum. 1960, 37, 305 5 NATrA, G., CORRADINI,P. and CESARI,M. R.C..4ccad. Lincei, 1957, 22, 11 e HOWELLS,E . R . I . C . I . (Plastics Division). Unpublished r NATrA, G., CORRADINLP. and CESARLM. R.C. Accad. Lincei, 1956, 21, 365 s NEWMAN,S. J. Polym. Sci. 1960, 47, 111 MILLER, R. L. Polymer, Land. 1960, 1, 135 10 FARROW,G. To be published ~1 KOLB, H. J. and IZARD,E. F. J. appL Phys. 1949, 20, 564 12 FARROW,G. and WARD,I. M. Polymer, Lond. 1960, 1, 330 12 NIELSON,L. E. J. appl. Polym. Sci. 1959, 2, 351 14 FARROW, G. Unpublished is McDONALD~M. P. and WARD, I. M. Polymer, Lond. 1961, 2, 341 is BRADER,J. J. J. appl. Polym. Sci. 1960, 3, 370

417