"THE TECHNOLOGY OF PLASTICIZERS"
BY J. KERN SEARS AND JOSEPH R. DARBY
CHAPTER ON "PERMANENCE OF PLASTICIZED PVC"
…from the plasticizers occupying more of the surface area at higher concentrations. The surprisingly low volatility of DBP at 25 to 43% concentration probably results from the case hardening of the sample commonly observed with volatile plasticizers. The high and rapidly increasing volatility of the chlorinated paraffin hydrocarbon plasticizer is what would be expected of a highly inefficient plasticizer (a secondary plasticizer used as a primary plasticizer) whose volatile loss is in large measure controlled by difficulty of migration to the surface.
3. Life Expectancy
Reed and Connor (62) went beyond Equation 6.2 with data averaged for several typical monomeric plasticizers to obtain Equation 6.3 which permits estimation of weight loss at one condition from known weight loss at some other condition.
where W2, t2, d2, and T2 are respectively the weight loss, exposure time, sheet thickness, and temperature (° C) under the desired conditions; W1, t1, d1, and T1, represent the same under known conditions, and K = eb which is approximately 1.10 or 100.042, according to the average of their data in which e is the base of natural logarithms, and b is a constant, 0.096, the average slope of W versus T From this it also appears that for constant time and thickness, a 7.2° C rise in temperature doubles the volatile loss. Similarly, a reduction of 7.2° C in temperature doubles the time required for a given weight loss. Thus they estimate 29 years for a 4-mil (0.1 -mm) film plasticized with DOP to lose a fourth of its plasticizer content at room temperature.
Quackenbos (59) used a different approach for prediction and estimation of service life of plasticized PVC-one based on plasticizer vapor pressure. After assuming arbitrarily that a film may be failed when it has lost 10% of its mass (about 25 to 30% of the plasticizer content), he constructed the log-log plot of vapor pressure versus time for 1O% loss (Fig. 6.12).
Figure 6.12) Vapor pressure of plasticizers at 98° C versus time for 10% loss from a 4-mil (0.1-mm) PVC Film at 98° C. The 10% loss is based on total composition and represents about a fourth to a third of the plasticizer lost . Quackenbos (59).
Although the data were all obtained at 98° C the graph applies to any temperature as long as the PVC is flexible. Table 6.2 shows predicted service life at 98° C from measured vapor pressure and at 25° C from vapor pressure extrapolated to room temperature. The simple technique of estimating vapor pressure of plasticizer blends from weighted averages of their individual pressures is usually quite satisfactory at these temperatures. Thinius, Schroeder, and Keatner (76) have shown that at temperatures above 200° C, vapor pressures of blends may begin to depart seriously from values predicted in this way. Predictions seem to be too long for the real world at the time of this writing-hundreds of hours at 98° C and 1000 years at room temperature. Yet Graham (30), using the same method, calculated service lives for PVC plasticized with DOP ranging from 4.2 years to 40° C to 1150 years at 0° C. If DBP were used in place of DOP the service lives were only 0.058 years (21 days) at 40° C to 11 years at 0° C. Service life of a low volatile/high volatile blend is usually longer than predicted on vapor pressure. Autooxidation, photooxidation. and hydrolysis can destroy plasticizers and the PVC itself long before their predicted life-span based on volatility has elapsed. Extraction and migration may supersede volatility and shorten service fife. Yet predictions for some uses can be simple and very helpful.
The preceding figures permit rapid graphical estimates of service life. As an example, a specific request was received for estimated service life on the Arabian desert for 125-mil (3-mm) thick objects of PVC plasticized with BBP when volatility was the expected limiting factor. Figure 6.5 (or tables in the Appendix) shows the volatile loss from 40-mil PVC/BBP (100/67) to be about 3% of the total mass in one day at 87° C. Percent loss is essentially independent of plasticizer content. If we should plot this one point on Figure 6.10 and draw a line through it essentially parallel to those shown, we can adjust for temperature. With the average annual maximum temperature for Kuwait City 85° F (29.4° C) and the average maximum for August 104° F (40° C) as guides (17), we should find losses of about 0.005 and 0.02% per day for 40-mil film which would be 1.8 and 7.3% per vear for the 29.4 and 40° C average temperatures. These two points are transferred to our enlargement of Reed and Connor's graph--the lower part of Figure 6.9. Straight dashed lines drawn through them to the lower right comer permit us to read the estimated loss for 3-mm thick material as 0.5 and 2.3% loss per year for these two temperatures.
If we then use Quackenbos' criterion that 10% loss is failure, these values suggest a life as long as 20 years at 85° F (29.4° C) or as short as 4.5 years at 10.4° F (40° C). These ballpark estimates are designed to be conservative by the choice of maximum temperatures. The fact that 20-mil black pigmented sheets plasticized with 50 phr of BBP are somewhat stiff but tough after nine years continuous exposure, not only to heat but to sun and rain, in southern Florida tend to confirm the reasonableness of the predictions. The annual average maximum temperature at the Florida test site is 83° F (24.4° C) and the average maximum August temperature is 91° F (33° C).
In Figure 6.12 only one data point is seriously out of line; that is the one for polyethylene glycol di(2-ethylhexoate). The polyether structure of this plasticizer makes it very easily oxidized, and in Quackenbos’ volatility tests at 98° C it probably suffered oxidative fragmentation (see later). What he measured, therefore, was pseudovolatility for this one plasticizer. With a proper antioxidant present we would predict its data point would shift to the right until it reached the line at a service life of about 200 hr. Even this may be fictitiously short if his vapor pressure data was influenced by either current or previous oxidation. Decomposition of some plasticizers such as phenol alkylsulfates during vapor pressure measurements has proved to be a problem (76).
C. MIGRATION TO OTHER MATERIALS
The mobility of a plasticizer which enables it to soften, flexibilize, and toughen PVC also permits it to leave the PVC and go into other solid materials which are in contact with it, if they have the ability to absorb the plasticizer. In most cases the migration process is complex and difficult to describe precisely. Several controlling factors are at play.
1. Controlling Factors
As with volatility, the rate of migration may be controlled by the ease of loss from the PVC surface (surface control) or it may be controlled by the rate of diffusion of plasticizer from the interior of the PVC mass to the surface (diffusion control). Therefore the molecular size and shape of the plasticizer are highly important; small molecules migrate faster than large ones. linear molecules migrate faster than bulky, branched ones, and highly solvating ones that produce an open gel structure migrate faster than those that are "frozen in" to isolated pockets. Since, after the first few moments of contact, plasticizer cannot be lost from the surface faster than it can migrate to the surface, its rate of diffusion establishes the ultimate limit on speed of migration; it may migrate slower but it cannot migrate faster.
When a material in contact with the PVC surface is efficient enough as an absorbent to remove plasticizer as rapidly as it reaches the surface, the rate of diffusion will control plasticizer loss. In such cases, loss will theoretically equal the loss by evaporation into a vacuum at that temperature. With diffusion in control, the rate of loss will gradually decrease as the concentration of plasticizer inside the PVC "reservoir" decreases and as plasticizer concentration builds up in the receiving material, since the rate varies with the concentration gradient. Curve A of Figure 6.13 shows the effect. When this type loss is plotted against the square root of time the curve is linear, in keeping with Fick's law and as seen in Equation 6.1. Curve A represents the loss to certain grades of silica gel, such as Linde silica, which not only are highly absorbent but which provide good wicking action to lead the plasticizer from the surface. Few materials have this ability.
When a material in contact with plasticized PVC can absorb plasticizer at a rate slightly slower than the rate at which plasticizer can migrate to the surface, migration is at first linear with time as shown in Curve B, Figure 6.13. As time goes on and the concentration of plasticizer in the PVC "reservoir" decreases, the diffusion rate decreases until the receiving material can accept plasticizer as fast…
For more information call 800-OK-LINER today!
To request one of our free Product Sample Books, please email your name, phone number, and address to firstname.lastname@example.org.