Use of a fluoroelastomer processing aid with non-olefin polymers.
Fluoroelastomers (FE) are among the fluoropolymers that have become widely used in processing polyolefins, notably linear low-density polyethylene (LLDPE). Their presence leads to the suppression of instabilities known as shark-skin and slip-stick fracture, which limit the output rates in which the polymer may be processed. Several empirical and theoretical accounts of the function of FE and of related additives conclude that they act as a die lubricant, modifying the properties of the interface between flowing polymer and die wall (1-7). As a result, the onset of instabilities is shifted to much higher output rates and the force of extrusion is reduced, signalling a lowered apparent melt viscosity. Two previous publications from this source (8, 9) have contributed to this knowledge. In one (8), it was shown that the effectiveness of FE action, expressed as a unitless coefficient, [Theta], could be related to the migration of FE through the host polymer matrix and to the mechanical strength of the interface between FE and the die wall. In a subsequent paper (9) we examined the problem of purging FE from a coated die by the extrusion of LLDPE, and a mathematical model was proposed to account for the dependence of purging on variables of the extrusion process.
If FE functions as a die lubricant, then it seems logical to ask whether its usefulness is restricted to LLDPE and other polyolefins, and whether it may not also be useful with other thermoplastics. So far this topic appears not to have been addressed. It forms the focus for the present study, where FE was used as a potential flow modifier for the extrusion of polystyrene (PS), a chlorinated polyethylene (CPE), polycarbonate (PC), and Surlyn ionomer (SN). An attempt is made to relate the performance of FE in these cases to the thermodynamic interaction between the polymers, as evaluated by methods of inverse gas chromatography (IGC).
i) Materials: The polymers and the fluoroelastomer used in this research are identified in Table 1. They were selected on the basis of their ability to act as Lewis acids and bases and were used as received from the supplier. The CPE of course qualifies as a polyolefin, but the presence of specifically interactive halide linkages differentiates it from previously studied LLDPE resins.
ii) IGC: Inverse Gas Chromatography was used to specify the surface interaction characteristics of the polymers noted in Table 1. Thus, the polymers were coated from solution onto Chromosorb G (AW-DMCS treated, 60/80 mesh). Ashing procedures showed the supported polymer to be in the 7 wt% range. The coated support was dried and packed in previously washed and degreased stainless steel columns, 6 mm in diameter and [approximately]40 cm long. Packed columns were placed in a Varian 3400 gas chromatograph equipped with both thermal conductivity and flame ionization detectors. The columns were conditioned for at least 4 h at 120 [degrees] C in a flow of He, which also served as carrier gas in IGC experiments. These involved probing the polymer surfaces with the following vapors, injected at extreme dilution: n-alkanes from hexane (n-C6) to nonane (n-C9), diethylether (Ether), tetrahydrofuran (THF), ethylacetate (EtAc), acetone, and chloroform (CH[Cl.sub.3]). All probes were analytical grade, used without further purification. Measurement temperatures for PC were 80 to 120 [degrees] C, and 40 to 70 [degrees] C for PS, well below the [T.sub.g] values of these polymers. Absorption of probes into the bulk of PC and PS could therefore be neglected, and retention characteristics were dominated by surface adsorption. For CPE, SN, and FE the measurements were made between 30 and 90 [degrees] C, above the [T.sub.g] values for the polymers. Since under these circumstances retention volumes are the sum of contributions from surface adsorption and bulk absorption (10, 11), a separation is required if surface characteristics are to be secured. We followed previously.reported procedures (10) based on increasing carrier gas flow rates. In these cases He gas flow rates varied from 20 to 70 mL/min.
iii) Flow measurements: The experimental protocols for evaluating the flow-modifying effectiveness of FE, described previously (8), were followed in this work. They made use of an Instron Rheometer connected to an on-line recording computer. A flat-entry die, 1.75 mm in diameter and with L/D = 20, was used throughout. To quantify the effectiveness of FE as a flow modifier, flow resistance measurements were made in two types of extrusion:
Type I: In every run of this type, host polymer was extruded through a freshly cleaned die.
Type II: Host polymer was extruded through the die, which had been precoated with FE.
Table 1. Polymers Used in This Study. Polymer Code Specification Supplier PS PS 204-00 Mn = 9.1 x [10.sup.4], Polysar Ltd. Mw = 1.9 x [10.sup.5] PC Lexan 130- Bisphenol-A General Electric 111 Inc. Surlyn Surlyn 9950 MI = 5.5, Cation DuPont Inc. type = Na CPE Tyrin 4213 Dow Chemicals FE Dynamar(*) Mooney Viscosity: 48 3M Inc. * Copolymer of vixylidene fluoride and hexafluoropropylene.
As before (8), the FE effectiveness coefficient [Theta] was calculated from
[Theta] = (1 - [F.sub.I]/[F.sub.II]) x 100% (1)
where F is the force on the rheometer piston.(*) At constant extrusion rates, force readings were averaged over the period of stable force readings (8). Maximum readings off were used in comparisons between Type I and II procedures when extruding at varying output rates. Details of processing conditions for each of the polymers will be stated in pertinent sections of the Results and Discussion section.
RESULTS AND DISCUSSION
i) Polymer surface properties: The route followed was based on the well established relationship (12, 13) between the net retention volume Vn of a probe and its characteristic dispersive surface energy [Mathematical Expression Omitted]
[Mathematical Expression Omitted] (2)
where [Mathematical Expression Omitted] is the dispersion surface energy of the polymer phase, N is Avogadro's number, c is an integration constant relating to a given column, R and T have their usual meaning and a is the molecular area of the adsorbed probe molecule. In this work the a values used were those reported by Schultz and coworkers (13), as given in Table 2. Also given in the table are acceptor and donor numbers A[N.sup.*], DN, describing the probes' (Lewis) acid and base properties (in kCal/mole), respectively. These numbers are drawn from Gutmann's theory of acids and bases (14), as modified by Riddle and Fowkes (15).
Equation 2 calls for a straight line to be generated when RTln Vn is plotted vs. [Mathematical Expression Omitted], for the alkane probes. From the slope of that line [Mathematical Expression Omitted] is readily obtained. An example of this is shown in Fig. 1 for PS at 40 [degrees] C. Specific interactions between polar probes and polymer surfaces are estimated by using the straight line as a reference for the position of the polar probes, also illustrated in Fig. 1. Formally, the normal distance from the placement of each point to the corresponding one on the reference line evaluates [Delta][G.sub.sp], the contribution of specific, or acid-base, interactions to the free energy of desorption:
[Mathematical Expression Omitted] (3)
The [Delta][G.sub.sp] datum is then used to obtain the acid and base interaction constants, [K.sub.a] and [K.sub.b], for the polymer. A requirement is the evaluation of [Delta][G.sub.sp] over a suitable temperature range, leading to the computation of the acid/base enthalpy parameter [Delta][H.sub.sp]. This is then used, as suggested by Papirer (12), in the expression
[Delta][H.sub.sp] = [K.sub.a] [multiplied by] DN + [K.sub.b] [multiplied by] A[N.sup.*] (4)
so that a plot of [Delta][H.sub.sp]/A[N.sup.*] vs. DN/A[N.sup.*] should yield the desired acid-base parameters. Figure 2 illustrates the expected excellent linearity for PS in the upper portion, and for FE in the lower portion. A summary of dispersion surface energies, of typical [Delta][G.sub.sp] values and of the [K.sub.a], [K.sub.b] parameters for the polymers of this work is given in Table 3.
The surface energy values are in good agreement with previously published data (16), but the datum for PC seems surprisingly low. In part the cause may be the more elevated temperature for this reading. Each of the polymers is able to act as both electron donor and acceptor in the temperature ranges pertaining to these determinations. Basicity is dominant in all but [TABULAR DATA FOR TABLE 2 OMITTED] the CPE, which is a net acid. The degree of acid-base interaction between polymer pairs can be quantified by empirical formulae, as discussed earlier (16, 17). The simplest of these, applied in this work, is:
[I.sub.sp] = [K.sub.a(X)] [multiplied by] [K.sub.b(F)] + [K.sub.b(X)] [multiplied by] [K.sub.A(F)] (5)
where [I.sub.sp] is the acid-base pair interaction parameter, X represents any one of the host polymers and F the FE flow modifier. Values of [I.sub.sp] are listed in Table 4. Also shown, for reasons to be clarified later in this section, are values of the parameter normalized to the [I.sub.sp] for CPE-FE.
Previously (9), it was argued that the effectiveness of FE in reducing extrusion force, and its capability to function over significant flow times depended on the rate of migration of FE into the flowing host polymer, and on the tendency of the host polymer to strip the FE from the die wall by mechanical, or frictional forces. Both of these factors should increase in importance with a rise in the degree of molecular contact established between matrix and FE, and thus on the [I.sub.sp] value. In other words, we would expect increasing values of [I.sub.sp] to designate interfacial states that are less suited for the performance of FE as a beneficial flow modifier. The pair interaction parameters classify acid/base interactions between FE and the present polymers in the following descending order: SN [greater than] PC [greater than] PS [greater than] CPE. (We note here that for polyolefins such as LLDPE, [K.sub.a] = [K.sub.b] = 0, so that [I.sub.sp] for FE/LLDPE also = 0.) In terms of the above supposition, we would then expect initial values of [Theta] to decrease in the order LLDPE [greater than] CPE [greater than] PS [greater than] PC [greater than] SN, and variations of [Theta] with extrusion time to change in the reverse order. The expectations will be examined in following portions of the paper.
ii) FE as modifier of flow behavior: Quantitative analyses of the effects on extrusion exerted by the FE followed the steps outlined earlier in this paper and [TABULAR DATA FOR TABLE 3 OMITTED] detailed in Ref. 8. The effectiveness coefficient [Theta] was evaluated according to Eq 1, by comparing the flow resistance generated under identical settings of melt temperature and output rate in runs of Type I and Type II. Furthermore, by reloading the rheometer barrel after an initial run without cleaning the die, Type II (and occasionally Type I) runs were repeated several times, allowing the [Theta] parameter to be followed over a number of extrusion cycles. This procedure lead to the evaluation of an average decay rate in [Theta], thus providing a measure of the rate at which FE was purged from the extrusion die. An example of the results obtained by these steps is shown in Fig. 3 for the extrusion of CPE at 160 [degrees] C and a shear rate = 71.4 [s.sup.-1]. Clearly, the substantial drop in extrusion force in Type II runs confirms the beneficial effects exerted by the FE coating. The purging rate is low, since only minor differences exist between the initial and subsequent runs of Type II.
Table 4. Specific Interaction Parameter [I.sub.sp] Between Host Polymers and FE. Polymer Couple [I.sub.sp] [I.sub.sp(X)/[I.sub.sp](CPE)] CPE/FE 1.35 1.00 PC/FE 2.32 1.72 PS/FE 1.79 1.32 Surlyn/FE 3.48 2.58
A detailed summary of the extrusion results for all of the host polymers is presented in Table 5. Necessarily, the temperatures and output rates varied somewhat from polymer to polymer, as indicated in the table. Returning to the CPE matrix, the average reduction in [Theta] of 0.5% per run is indeed low, albeit greater than the [approximately]0.09% per run reported earlier (8, 9) for LLDPE under similar extrusion conditions. By inference the CPE melt slips over the FE layer without massive diffusion across the boundary, nor does this melt cause major losses of FE due to mechanical stripping from the extrusion die. As noted above, the degree of specific interaction for this pair of polymers is low. Thus, acid-base coupling appears to provide little driving force for contact at molecular levels.
The system PC/FE is more strongly interactive, and the flow data appear to reflect this. The initial [Theta] value here is not greatly different from the preceding case, but the decay rate of 3.3% per run far exceeds corresponding values for CPE and LLDPE. The strongly interactive PC therefore may classify as a good purging agent for FE, but little lasting benefit from reduction in extrusion force should be expected. Of course, the effect may not be due entirely to acid-base interaction effects, since, at the elevated temperature of this case, the adhesion of FE to the die wall may be seriously affected and the migration kinetics into the flowing matrix polymer accelerated.
Extrusion of PS and Surlyn ionomer was under identical settings of shear rate and temperature. The initial effect of FE on the extrusion of PS is slight and the decay rate is rapid. The PS exerts a strong purging effect on the FE deposit. The data for SN are perhaps the most interesting in the set. Here the presence of FE actually raises the flow resistance, indicative of a viscosity increase. The loss rate from the die is very rapid, and the viscosity characteristic of Type I runs is soon established. The SN/FE system is the most strongly interactive in the present selection of materials, raising the possibility that these polymers may form bonds through interspecies entanglements, or via chemical interaction. It is to be noted, however, that no evidence was found for chemical bonding from DSC and infrared spectroscopic data.
iii) Acid-base interaction and FE effectiveness: The inferences, above, of possible corelations between acid-base pair interactions and the effectiveness of FE as a flow modifier warrant further examination. This is done by means of Figs. 4 and 5, the former showing the variation of the initial [Theta] value with the normalized [I.sub.sp] parameter, the latter the average rate of change in [Theta] per run with that parameter. Earlier reported data for LLDPE are included in the comparison. The existence of significant corelations in both cases is indicated. Figure 4 documents the decrease in the initial [Theta] with rising acid-base interaction and suggests that FE ceases to be a flow modifier when the normalized [I.sub.sp] value exceeds about 2 (corresponding to an absolute [I.sub.sp] of [approximately]0.65). Whether a crossing of the [Theta] = 0 boundary displayed by the ionomer is a general occurrence at high levels of acid-base interaction remains a point to be resolved by additional experimentation. The relationship in Fig. 5 is unambiguous, showing the accentuated loss in FE effectiveness with increasing acid-base coupling. As stated earlier in this article, the findings are consistent with mechanisms postulated to account for the flow modifying action of the FE (8, 9). At higher levels of acid-base interaction, the migration of FE into the polymer matrix would be enhanced. Under these conditions we may assume the rapid creation of an interphase between matrix polymer and FE, and this would also enhance the random mechanical stripping of FE from the die.
[TABULAR DATA FOR TABLE 5 OMITTED]
The inferred correlations are to be viewed with caution, however, since other factors, not taken into consideration, may also influence the effectiveness of FE flow additives. No account has been taken of the contribution made by dispersion forces to events at the FE/polymer interface. An exact calculation of interfacial surface energies under appropriate conditions of temperature and shear, currently unavailable, would be needed for this purpose. The viscoelastic properties of the polymers under the defined extrusion conditions would also affect migration and mechanical stripping. Each of the polymers in this study has a characteristic dependence of melt viscosity and elasticity on extrusion variables; a full rationalization of the variation in FE effectiveness over the range of extrusion variables pertaining to this study would call for a consideration of these dependencies. Finally, attention is drawn to the temperature differences between IGC measurements from which the interaction parameters are drawn, and the extrusion measurements. These differences, imposed by experimental limitations, are significant in all cases, but particularly so for PC. Earlier studies by Panzer and Schreiber (16) and ongoing work (18) show that the acid-base functionalities of PC and of other polymers are temperature dependent, with a tendency to decrease with rising temperature. Consequently, the relationships in Figs. 4 and 5 are indicative of trends but do not necessarily define quantitative corelations.
* The extrusion behavior of thermoplastics other than LLDPE was shown to be affected by FE coatings applied to the extrusion die. The flow modifier was effective in reducing the apparent melt viscosity of CPE, but less so in processing PC, PS, and Surlyn ionomer.
* IGC techniques have been applied to determine the acid-base functionalities of FE and the matrix polymers of this study, showing that acid-base pair interactions varied in decreasing order: Surlyn/FE [greater than] PS/FE [greater than or equal to] PC/FE [greater than] CPE/FE.
* As a consequence of the above statement, FE is considered to be an effective flow modifier when the matrix polymer is acidic, but strongly interactive, basic polymers will purge the FE rapidly from the die surface.
* While acid-base interaction is an evident factor in the performance of FE and related flow modifiers, the presented relationships only approximate true corelations, because of significant differences in the temperatures at which extrusion and IGC determinations could be carried out.
We thank the Natural Sciences and Engineering Research Council, Canada for financial support of this research through Operating and Cooperative Research and Development grants. The assistance of Dr. D. Duchesne and of other members of 3M Canada research staff, through the supply of flow modifiers and valuable discussions, is acknowledged with gratitude.
* FE benefits the extrusion process by lowering the force on the rheometer piston. Under these circumstances [Theta] [less than] 0, but we follow the precedent set in Refs. 8 and 9, by reporting a positive index of effectiveness. In contrast a reported negative [Theta] indicates that the extrusion force has actually risen, hindering the extrusion process.
1. P.S. Blatz, U.S. Patent 3,125,547 (1964).
2. C. K. Shih, Polym. Eng. Sci., 16, 742 (1976).
3. A. V. Ramamurthy, J. Rheol., 30, 337 (1986).
4. A. V. Ramamurthy, Adv. Polym. Technol., 6, 489 (1986).
5. D. S. Kalika and M. M. Denn, J. Rheol., 31,815 (1987).
6. S. G. Hatzikiriakos and J. M. Dealy, J. Rheol., 35, 497 (1991).
7. S. G. Hatzikiriakos and J. M. Dealy, J. Rheol., 36, 703 (1992).
8. K. C. Xing and H. P. Schreiber, Polym. Eng. Sci., 36, 387 (1996).
9. K. C. Xing and H. P. Schreiber, Int. J. Polym. Proc., in press.
10. P. Mukhopadhyay and H. P. Schreiber, Macromolecules, 26, 6391 (1993).
11. P. Mukhopadhyay and H. P. Schreiber, Colloids and Surfaces A: Physicochem. Eng. Aspects, 100, 47 (1995).
12. C. Saint Flour and E. Papirer, J. Colloid Interf. Sci., 91, 63 (1983).
13. J. Schultz and L. Lavielle, in ACS Symposium Series 391, ch. 14, D. R. Lloyd, T. C. Ward, and H. P. Schreiber eds., Amer. Chem. Soc., Washington, D.C. (1989).
14. V. Gutmann, The Donor-Acceptor Approach to Molecular Interactions, Plenum Press, New York (1978).
15. F. L. Riddle Jr. and F. M. Fowkes, J. Amer. Chem. Soc., 112, 3259 (1990).
16. U. Panzer and H. P. Schreiber, Macromolecules, 25, 3633 (1992).
17. J. Kloubek and H. P. Schreiber, J. Adhesion, 42, 87 (1993).
18. G. Tovar, P. J. Carreau, and H. P. Schreiber, to be published.
|Printer friendly Cite/link Email Feedback|
|Author:||Xing, K.C.; Wang, W.; Schreiber, H.P.|
|Publication:||Polymer Engineering and Science|
|Date:||Oct 1, 1996|
|Previous Article:||A calibration technique to evaluate the power-law parameters of polymer melts using a torque-rheometer.|
|Next Article:||Viscoelastic features of molten blends of poly(vinyl chloride)/poly(ethylene-vinyl acetate).|