close

Вход

Забыли?

вход по аккаунту

?

The study of propylene-styrene graft and block copolymer polydispersity by fractionation and light scattering.

код для вставкиСкачать
Die Angewandte Makromokkulare Chenaie 18 (1970) 43-54 (Nr. 174)
From the Polymer Institute of the Slovak Academy of Sciences,
Bratishva, Czechoslovakia
The Study of Propylene-Styrene Graft and
Block Copolymer Polydispersity by Fractionation
and Light Scattering
By ~ T G P A NFLORL~N,
DIETERLATHand ZDENGK M A % ~ E K
(Eingegmgen am 2. MIirz 1970)
SUMMARY:
The properties of graft and block propylene-styrene copolymers with varying
chemical composition have been studied by fractionation, light scattering, and
osmometric methods.
The low values of composition polydispersity parameters (P and Q) calculated
from the results of light scattering measurements of copolymers in various solvents
as well as the conformity of the polydispersity parameters y, a and /3 of copolymers
and their components confirm a high composition homogeneity of these copolymers.
On the basis of experimental data a correlation between the lengths of polypropylene and polystyrene chains may be supposed.
The agreement of studied properties of graft and block copolymers with equal
chemical compositiongives evidence of their structural similarity. This result is conform with the idea, obtained on the basis of grafting efficiency data, that mostly one
or two branches of polystyrene are supposed to be on the main polypropylene chain.
ZUSAMMENFASSTJNG:
Die Eigenschaften von Propylen-Styrol-Pfropf-und Blockcopolymeren mit verschiedener chemischer Zusammensetzung m d e n mit Hilfe der Fraktionierung,
Lichtstreuung und Osmometrie studiert.
Die niedrigen Werte der Parameter der Zusammensetzungspolydispersitiit,die
aus den Ergebnissen der Lichtstreuungsmetx3ungender Copolymeren in verschiedenen Losungsmitteln berechnet wurden, sowie die Vbereinstimmung der Polydispersitiitsparameter y , a, und /3 der Copolymeren und deren Komponenten beweisen
eine hohe Zusammensetzungshomogenitiit dieser Copolymeren. Auf Grund der
experimentellen Angaben ist es m6glich anzunehmen, daI3 eine Komlation zwischen den Liingen der Polypropylen- und Polystyrolketten existiert.
Die ffbereinstimmung der studierten Eigenschaften der Pfropf- und Blockcopolymeren der gleichen chemischen Zusammensetzung bestiitigt eine h l i c h k e i t
ihrer Strukturen. Diese Tatsache stimmt mit der am den kinetischen Angaben hervorgehendenVorstellung iiberein, daB die Grundkette des Polypropylens vorwiegend
nur einen oder zwei Polystyrolzweige besitzt.
43
g. FLORIAN,
D. LATHand Z. MAGASEK
General considerations
There are three kinds of polydispersities which can be defmed for copolymers,
namely the polydispersity in molecular weight, chemical composition, and
molecular structure. It is very important to know these polydispersities because they affect a great number of physico-chemical and technological properties of polymer substance.
The polydispersity in molecular weight is commonly characterized by the
- polydispersity parameter M,/M~ wherea, and Enare the weight-average and
the number-average molecular weights resp.
I n case of binary copolymers the following polydispersity paramster may be
considered :
- ,
- -
y - for the entire copolymer = M,/M,
a - for the component A
= M$/Mt,
,8 - for the component B
=
@/E.
For these polydispersity parameters the subsequent relationship (1) is valid
y = a?:
+ b(1 -
?A)2
+ 2 XA(1 - CA)[1 + r(a - 1)'/2(b -
(1)
where %A is the chemical composition expressed as average weight fraction of
component A and r is the correlation coefficient,which indicates quantitatively
the degree of dependence between the stoichiometrically bound phenomena.
The first two terms on the right side of the equation (1) express the polydispersity ofthe copolymer synthesized from the two components A and B (mutually
not dependent) while the third term expresses a contribution t o the copolymer
polydispersity due to some correlation between the phenomena during the
synthesis, e. g. when the number or length of branches of the graft copolymer
depends on the length of main chain. I n case of maximum dependence (r = 1) the
copolymer obtained is monodisperse with respect to composition and if the
polydispersities a and ,8 are equal also ,,strict" composition monodispersity of
copolymer is obtained.
The relationship (1) involves the experimentally accessible quantities the
determination of which enables us to fmd the value of correlation coefficient and
thus t o enhance the elucidation of the character of a given synthesis.
I n contrast to the number-average molecular weight as a n additive contribution of the components A and B, the determination of weight-average molecular weight of copolymer is more difficult. I n this case the additivity of molecular weights does not exist, but E
, is increased by a contribution due to the
composition homogeneity of copolymer. The problem of weight-average molecular determination by the method of light scattering has been pursued by
STOCKMAYERa t 01.2 and later by BTJSHUK
and BE NO IT^^^. For the molecular
44
Propykm-Styrene Graft and Block Copolymers
weight of a binary copolymer BUSHTJK
and BENOIT
have deduced the following
expression :
map,= K w+ 2 P (
+Q
~
VA;
VB)
(
VA
V B ) ~
~
where Zapp is the apparent molecular weight determined by usual methods
v, V A and V B refractive index increments of the copolymer and its components
resp. , and P and Q are parameters of composition heterogeneity of copolymer :
2P
=
(1 - EA) *
(W,- -B
M,)
Q = %A (1 - ?A)
*
- %A (Kw
- N$)
(mt+ B t - g w ) .
(3)
(4)
The parameter P characterizes the trend of macromolecule composition
change with their increasing molecular weight and the parameter Q expresses the
average composition heterogeneity of copolymer.
Thus the equation (2)enables us to calculate
P and Q (or
@) of a
copolymer by measuring Bapp in three solvents.
Sometimes such characterization of a copolymer may, however, be insufficient
for certain purposes. A more detailed information on the copolymer heterogeneity can be obtained on the basis of fractionation results. But in this case
such experimental conditions have to be found that only a single polydispersity
should have a decisive influence on the fractionation results.
The problems of propylene-styrene graft copolymer fractionation were
followed in 59 6. The observed fact of low composition heterogeneity of the
copolymer mentioned will be completed in this contribution by the study of the
copolymer samples, with different chemical compositions by the methods of
light scattering, fractionation and osmometry.
zw,
zw,
at,
Experimental
Materials
The propylene-styrene graft copolymer (PP-S) waa prepared by styrene (S) polymerization in the presence of atactic polypropylene (PP) by using the transfer
reactions and initiation with benzoyl peroxide 7. The grafting of A-D samples was
performed in sealed ampules under nitrogen. The conditions of grafting are summarized in Table 1.
The polymerization of each sample took 6 hours. Thus polymerizationwas finished
the ampules were broken and the polymer mixture wag dissolved in toluene and
subsequently precipitated by methanol. The copolymer was isolated from reaction
mixtfire by selective extraction and its purity checked turbidimetrically5.
The propylene-styreneblock copolymer (PP-PS) waa prepared mechanochemicallys by plastication of the polypropylene fraction with initial molecular weight
= 4.3 105 in a laboratory plasticator in the presence of styrene vapour for one
aw
45
s. FLORIAN,D. LATHand Z. MAGASEK
hour at 300 rev./min and - 8 "C. The amount of styrene added was approximately
1 ml/hour/g polymer. Homopolymerswere removed from the reaction mixture by an
extraction carried out similarly as it has been performed in cam of the graft copolyme^-5.
All samples were rid of the peroxide rests by dissolvingthem a t 40 "C in a saturated
solution of sodium nitrite in tetrahydrofurane. The solvents used were purified by
usual procedures and their purity waa checked by gaa chromatography.
Sample
APP-S
BPP-S
CPP-s
DPP-S
Viscosity measurements
The viscosities of solutions were measured in an UBBELOHDE
dilution viscosimeter
modified according to SEIDE-DECKERTg.The capillary bore waa so chosen that the
corrections for the kinetic energy and capilktry curvature should be negligible. The
influence of shear rate on the measured values was also negligible since the deviation
did not exceed 1% for highest molecular weights.
The viscosity meaaurements were evaluated by use of HUGQINS
relation. The
temperature was kept at 25 "C with an accuracy of 0.01 "C.
Osmometry
The osmotic pressure n of the polymer solution in toluene was measured in an
automatic membrane osmometer Mechrolab, Model 502, by using a membrane of
Schleicher-Schulltype. Measurements were carried out a t 37 "C, number average
molecular weight was estimated by the usual graphical extrapolation.
Refractive index increment and l i g h t scattering
The refractive index increment was measured by means of an interferometer,
Zeiss, Jena.
A survey of refractive index increment valqes is given in Table 2 for polypropylene and polystyrene in some common solvents.
Refractive indices of solventswere measured by means of a dipping refractometer,
Zeiss. Our measurements showed that v waa an additive property of v p p and vps for
the sample BPP-S. This additivity was then assumed for other samples of copolymer too.
Light scattering was measured on the Brice-Phoenix scattering photometer,
Model 2000 by using unpolarized monochromatic light with wave length of 646 nm
46
Propylene-Styrene Uraft and Block Copolymers
and a cylindrical cell of C-105 type. The influence of depolarization waa negligible.
The results of the measurements were evaluated by the method of double extrapolation according to ZIMM~O.
In case of the copolymer this method can afford only apparent values as mentioned
the measurements must be
in the theoretical part. In order to determine
performed in several (possibly good) solvents with sufficiently different refractive
indices.
The fractionation of the graft copolymer BPP-S waa described in a preceding
papere. Other samples of copolymer were fractionated by the method of precipitation chromatography in a fractionation column described inll. The amounts of
polymer for fractionation were about 8 g. The column worked within the temperature range of 25-30°C. The temperature cycle time was about 30 minutes. The solvent gradient was exponential.
aw
Table 2. Refractive index increment for polypropylene and polystyrene in different
solvents at 25"C.
Solvents
I
n-Butyl propionate
n-Butyl chloride
Toluene
Bromobenzene
*
**
***
I
I
1.3958
1.3987
1.49745 *
1.56165* *
I
0.1101
0.1082
-0.0027
-0.0599
0.2018
0.1997
0.1130
0.042***
Values quoted froml.
Calculated values.
Values quoted from4 (meaaured at 2OOC).
Results
It was shown in papefl, that the results of propylene-styrene graft copolymer
fractionation due t o the copolymer composition homogeneity do not depend on
the fractionation system used. To similar conclusions led also further fractionation of graft and block copolymers in both fractionation systems, sensitive
and not sensitive to their chemical composition, so, it seems, that low composition heterogeneity is typical of this kind of copolymers. The distributions of chemical composition of propylene-styrene copolymer samples in chloroformisopropanol system represents Fig. 1 (I,, being the cummulative weight of the
copolymer). The chemical composition of the copolymer was determined by
means of pyrolytic gas chromatography.
It is to be seen that except the region of lowest molecular weights the chemical composition of copolymer samples does not change with the order of
fraction but remains constant.
47
g. F L O R ID.
~ ,LATHand Z. MGLSEK
I
I
I
0.5
In
1
Fig. 1. Chemical composition distributions of propylene-styrenecopolymers.
m P - 5 (0
01, BPP-S (0 0 0 0 ) s CPP-S (00 0O),
DPP-S ( A A A A ), pp-sp . (
W).
Besides the fractionation results we tried to express the composition heterogeneity of the graft copolymer B P P S by means of the composition heterogeneity parameters P and Q by measuring the light scattering of copolymer in
several solvents (Table 3). A mutual comparison of the molecular weight values
Map, measured in different solvents shows that the differences are relatively
small what indicates that the composition heterogeneity parameters are small
too.
(RW= 4.08.105,
P/Rw = 1.017.10-2, Q/ZW = 8.633 * 10-3)
.
Table 3. Variation of &pp values with the refractive index of the solvent used for
the BPP-S copolymer.
Solvent
VPg
- VPP
V
n-Butyl propionate
n-Butyl chloride
Toluene
Bromobenzene
0.658
0.665
3.373
- 3.732
l -
,
M
. 10-5
3.96
4.12
4.20
4.88
These results confirm a high composition homogeneity of copolymers which
has been stated on the basis of fractionation results. Therefore it follows that
48
Propykne-Styrene Graft and Block Copolymers
the molecular weight of the copolymer may be also determined with a sufficient
accuracy by measuring light scattering in a solvent with a sufficiently high
value of the refractive index increment.
The molecular weight EWof propylene-styrene copolymer samples measured
in n-butyl chloride as well as molecular weight
and chemical composition
are given in Table 4.
an
Copolymer
-
M,
. 10-5
-
M,
. 10-5
Y
XPS
It is interesting to compare the molecular weights of graft copolymer samples
APP-S, BPP-S and CPP-S among themselves. These samples have been prepared from the same fraction of atactic polypropylene. Since no noticeable change
of molecular weight of polypropylene takes part during the grafting under given
conditions, the molecular weight of the copolymer increases proportionally with
the content of polystyrene in the copolymer. The copolymer polydispersity
decreases partly with the degree of grafting. A rather low polydispersity of the
propylene-styrene block copolymer is due to the method of its preparation
(a reduction of polypropylene heterogeneity in the course of modificationl2). By
the measurements of light scattering in n-butyl chloride were further determined the molecular weights of the fractions used for the calibration of the
relation between the limiting viscosity number and molecular weight of the
copolymers (in the region of molecular weights with constant chemical composition) (Table 5).
It is to be seen from Table 5 that the value of a in the dependences of limiting
viscosity number on the molecular weight of the copolymers in toluene is
constant within a broad range of copolymer composition.
By using of values in Table 5 the molecular weight distribution curves of
copolymer samples were plotted (Fig. 2).
The detailed fractionation results for the samples of graft as well as block
propylene-styrene copolymers are summarized in 13.
49
&.FLORIAN,D. LATHand Z. MAGLSEK
relationship of the graft and
Table 5. Constants K and a in the M~RK-HOUWINK
block propylene-styrene copolymers.
Copolymer
APP-S
BPP-S
CPP-s
DPP-S
PP-PS
K -103
Toluene
12.2
7.13
Toluene
a
n-Butyl chloride
7.24
0.747
0.796
5.98
0.783
0.798
n-Butyl chloride
4.12
13.5
0.779
0.665
3.62
10.4
0.785
0.686
0.783
0.798
7.13
5.98
I
I"
Fig. 2. Molecular weight distributions of propylene-styrene copolymers.
APP-S (0 0 0 ) s BPP-S (00 0 O), CPP-S (0 KO,
D P P d ( A A A A ) , pp-ps (a W a).
Discussion
The most conspicuous common feature of all samples of graft and block copolymers is a very narrow distribution of their prevailing part as regards chemical
composition polydispersity; this should demand the length or number of polystyrene branches to be proportional to the length of main polypropylene chain.
If we get round this hypothesis and seek the cause of a low composition
in homogeneityof copolymer in separation procedures in the course of which the
part of copolymer with different composition could dissolve, it is difficult to
explain the fact that the copolymer samples with quite different chemical composition have been isolated by practically the same separation procedure and a
higher composition inhomogeneity has appeared just a t the lowest molecular
50
Propylene-Styrene &aft and Block Copolymers
weights. Similarly, the method of successive precipitation used for the separation of the APP-S and BPP-S samplesbrought about equal results5~6. Therefore,
let us admit that a correlation between the lengths of main polypropylene and
polystyrene chains exists.
tries to explain such a correlation observed at polyisobutylene grafting by the idea that the growth of the polystyrene chain involves an
increase of “internal” incompatibility of polymer components in copolymer
what has relation to the chain lengths. Provided the polystyrene chains reach a
certain critical length, they concentrate in the microphases in which the termination rate increases.
sEBBAN-DANON14
This idea should also enable to explain the great deviations exkiting in
composition of the lowest fractions. In case of very short molecular chains of
polypropylene, it seems probable that no separation in microphases can take
place and the grafting goes forward independently of the main chain length. The
low-molecular portion of the copolymer with a rather high composition polydispersity thus formed is then separated together with homopolymers. But the
portion having a styrene content of about 50% is not to be removed by separation procedures and accounts for the deviations in composition of the lowest
copolymer fractions (see Fig. 1).
Equally, the physical influencesmay also be effectivek mse of the block copolymer though its reaction mechanism is different. Therefore, it is not surprising
that these conclusions are valid for the results of block-copolymerfractionation, too (Fig. l).
Similarly, such low composition polydispersity of the graft copolymers of
resembling type has been observed in other papers159 16.
It is to be seen from Table 4 that the polydispersity parameter y of the graft copolymer depends on its chemical composition. Let us try to explain this fact by
investigating the polydispersity parameters a and of copolymer (see equation
1).The first idea to be applied involves the assumption that the polydispersity
of polystyrene branches is much lower than that of polypropylene chain. In
this case the polydispersity parameter y of the copolymers should decrease
with increasing content of styrene. We tried to prove this hypothesis by measuring the polydispersity parameters a of BPP-S and CPP-S copolymer samples.
The numerical average molecular weight of styrene branches could be
determined on the basis of osmotic measurements and its value was found to
= 3.12 . l o 4 for BPP-S and Ezs = 2.38 . 105 for CPP-S sample.
To determine the weight average molecular weight of styrene branches, it
was possible to use the results obtained by light scattering measurements in
toluene which had approximately an equal refractive index as polypropylene.
51
s. FLORIAN, D. LATEand 2. MAGASEK
The molecular weight of styrene branches could then be calculated by using the
relationship
Thus the molecular weight of the styrene branches was found to be Bzs =
1.20 * lo5 for BPP-S and @is
= 7.10 * 105 for CPP-S sample.
By using the number as well as the weight average molecular weights, the
polydispersity of styrene branches could then be calculated, the values of which
were a = 3.50 and a = 2.98 for BPP-S and CPP-S samples, resp. It is evident
that the polydispersity of styrene branches is relatively high and, in addition,
different for particular samples.
In a similar way, the polydispersity parameter /3 of the propylene chain was
determined for a BPP-S sample. The average molecular weight of polypropylene was obtained from the results of light scattering measurements in
various solvents (Table3) by using the equations (2)-(4).The values found were:
Zn= 7.3 * lo4 and BW= 2.91 . lo5 from which followed /3 = 3.98. It is evident that the values of a and are close for this sample. On the basis of equation
(1) it can be,found that the correlation coefficient is close to one what indicates again a maximum stoichiometric linkage between polypropylene and
polystyrene chains.
It follows from results in Table 4 that the observed dependence of the polydispersity parameter y on the polymer composition is likely to be due to
multiple repetitions of the separations procedures with samples containing a
higher content of polystyrene (a higher portion of homopolymer) what could
result in a higher loss of the low-molecular part of the copolymer which has
a strong effect on the value of the polydispersity parameter.
Since the experimentalresults have confirmed a high compositionhomogeneity
of copolymer samples in a sufficient way, it remains to answer the question
whether the number or length of polystyrene branches is proportional to the
length of the polypropylene chain. It is worth saying that the answer is not
simple. Both stated alternatives have been studied for BPP-S sample on the
basis of comblike model proposed by BERRYand 0 ~ 0 ~ 1 ~ 0the
1 7 ,A-B interactions having not been taken into consideration. Since it has been found that the
branching frequency is practically independent on the molecular weight of copolymer, the second alternative seems to be more probable 18. It should mean
that every polystyrene chain grows during synthesis till it reaches a lenght
proportional to the length of polypropylene chain.
This idea is supported by the fact that the logarithmic dependence of the limiting viscosity on the molecular weight has been found linear for all samples of the
copolymer. The typical curvature due to branching has not been observed.
52
Propylene-Styrene Graft and Block Copolymers
It is difficult to explain this course of grafting. However, it is evident that the
typical physical interactions of the components (compatibility and solubility of
polymers) may also play an important role in grafting and the polymerization
may be affected by a gel-effect, the operation of which is controlled by the
viscosity of the medium and other parameters. It may thus be assumed that
the gel-effect is in operation mainly in case of the higher molecular weights of
polypropylene and results in a grafting of longer branches on it.
On the other hand, since the values of graft BPP-S copolymer molecular
weight comply with the dependence of the limiting viscosity number on the
molecular weight of the block copolymer of equal composition (Table 5 ) , their
structures should be similar, what implies such a low number of branches, that
the effect of branching does not manifest itself in the above relationship.
Unfortunately, a direct determination of the number of branches in this copolymer has not been possible.
The branching density may, however, be judged from the results involving
the polymer mixture fractionation after grafting provided the presence of
unmodified basic polymer has been proved7. Assuming that the grafting initation
proceeds randomly, the POISSON
probability distribution of this process may
be used for the determination of the branching frequency of the main polymer
chain :
1
Wn = (R/N)n e - R / N -
n!
(6)
where Wn, R and N denote the portion of macromoleculeswith n-fold branching,
the number of branches, and the number of macromolecules of basic polymer,
resp.
By using the summation values of the POISSON
probability distribution and
data of polymer grafting efficiency, the average number of branches in main
chain may be calculated.
The extract of the reaction mixture in n-heptane has been used for the determination of polypropylene which has not been consumed in grafting. Isolated
polypropylene has been cleared of the residues of graft copolymer by extraction
with propyl acetate. Three parallel determinations have been carried out.
It follows from the relationship (6) that BPP-S sample (41.8% of polypropylene which has not been consumed in grafting) and the DPP-S sample
(36.8% o f polypropylene which has not been consumed in grafting) have 0.90
and 1.03, resp., polystyrene branches in polypropylene chain on the average.
For the DPP-S sample (the highest degree of grafting) the numbers of branches
have had this distribution: 58.2% of macromolecules have had one, 29.1% haye
two, 9.6%have three and only 2.4% have four branches. For other samples of
53
g. F L O R ~ D.
N , LATHand Z. MASASEK
the propylene-styrene graft copolymer the portion of macromolecules with
single branch has been stiU higher.
A monodispersity of the basic polymer has been supposed in calculations. As
polypropylene fractions with relatively narrow distribution (p w 1.8) have been
used for grafting, the real number of branches should be close to the calculated
one provided the given assumption was fulfilled. That means that mostly one
or two branches of polystyrene may be expected for this type of graft copolymer
to be in the main polypropylene chain.
Further informations on the structure of this copolymer are t o be obtained
by the study of properties of its fractions what will be the subject of another
paper.
1
2
3
4
5
6
P. CORDIER,
J. Chim. physique 64 (1967) 423.
W. H. STOCKMAYER,
L. D. MOORE,M. FIXMAN
and B. M. EPSTEIN,
J. Polymer
Sci. 16 (1955) 517.
W. BUSEUK
and H. BENOIT,C. R. hebd. Seances Aced. Sci. 246 (1958) 3 167.
W. BUSEUKand H. BENOIT,Canad. J. Chem. 36 (1958) 1616.
V. JWRANI~OVA,s. FLORL~N
and D. BEREK,European Polymer J. 6 (1970) 57.
g. FLORIAN,D. LATHand V. ~ U R ~ O V IAngew.
~ ,
makromolekulare Chem. 10
(1970) 189.
J. PAVLINEC,
M. L A Zand
~ Z. MANASEK, J. Polymer Sci. C 16 (1967) 1113.
* A. ROMANOV,
S. J. MAUARIKand M. L A Z ~Vysokomol.
,
Soedin. 9 (1967) 292.
9 G. LANQEAMMER,
R. BERQER
and M. SEIDE,Plaste und Kautschuk 11 (1964) 472.
10 B. H. ZIMM, J. chem. Physics 16 (1948) 1093, 1099.
11 J. PoLA~EK,
L. SCHULZ
and I. KOSSLER,J. Polymer Sci. C 16 (1967) 1327.
12 A. ROMANOV,
Dissertation, Bratislava 1966.
13 8. FLORIAN,
Dissertation, Bratislava 1968.
14 J. SEBBAN-DANON,
J. Chim. physique 58 (1961) 246.
15 P. CORDIER,J. Chim. physique 64 (1967) 439.
16 A. CEUPIRO,
P. CORDIER,
J. JOSEFOWICZ
and J. SEBBAN-DANON,
J. Polymer Sci.
C 4 (1963) 491.
17 G. C. BERRY
and T. A. OROFINO,J. chem. Physics 40 (1964) 1614.
18 8. FLORIAN and D. LATE,Intern. Conference “Chemical Transformations of
Polymers”, Bratislava 1968, P 70.
7
64
Документ
Категория
Без категории
Просмотров
1
Размер файла
614 Кб
Теги
block, graf, propylene, stud, scattering, copolymers, light, polydispersity, styrene, fractionation
1/--страниц
Пожаловаться на содержимое документа