Antioxidant behavior in flexible PU foam.

Antioxidant behavior in flexible PU foam

The reduction of chlorofluorocarbons (CFCs) and other halogenated blowing agents in slabstock foam has forced the use of foam formulations with higher and higher levels of water to meet density requirements. These elevated water levels can raise the maximum temperature exposure in the foam increasing the risk of scorch degradation and discoloration (polymer decomposition). In the extreme case, the foam may suffer a severe loss of physical properties and even self-ignite.

Antioxidants are typically present in commercial polyols to prevent polymer decomposition. Oxidation inhibitors are available in a wide variety of molecular classes which are attributed with various mechanisms of action (refs. 1 and 2). Regardless of the specific mechanism through which a given antioxidant may inhibit decomposition, mobility is a parameter common to all types of antioxidants. A given molecule must be adequately permeable in the polymer matrix in order to migrate to a site of action. On the other hand, a compound which migrates too freely may migrate out of the foam resulting in a loss of effective antioxidant concentration.

In addition to the solubility and diffusion parameters which affect a compound's migratory ability, thermal gradients contribute significantly to the movement of small molecules in polymers (ref. 3). With regard to polyurethane foam buns, very few data exist describing the exotherm profiles which are produced. Several factors such as formulation changes (especially water level, index and alternative blowing agents), bun dimensions, reactant temperatures and ambient conditions are likely to contribute to the temperature profiles measured in foam buns.

In this article, data are given regarding the AO concentration profiles across foam buns as a function of the corresponding temperature profile of the foams studied, three were produced on a Varimax pilot line located at Dow Chemical Company Freeport, Texas. The fourth foam used for this study was a production scale foam made at Texas Fibers, a Division of Leggett & Platt, Brenham, Texas. The foams produced on the Varimax include formulations using 4.1, 5.1 and 6.1 parts per hundred parts (pphp) water based on polyol and can be compared to the 4.1 pphp water foam made at Texas Fibers.

Temperature data, collected from a foam-in-place grid of thermocouples, gives the time-temperature profile within the foam bun as a function of location. Foam samples which have been removed from locations corresponding to the thermocouples are examined, via methylene chloride extraction and liquid chromatographic analysis, for antioxidant content. Our objective is to determine the significance of the foam environment, as a function of formulation, on the behavior of antioxidants.

Materials and methods

Foam formulations and manufacture The polyurethane foams produced for this study are described in table 1. Foams were produced at each of the three water levels (4.1, 5.1 and 6.1 pphp) using the Varimax pilot line at Dow Chemical. One foam at the 4.1 pphp water level was made at Texas Fibers. Initial reactant temperatures were maintained at 20-22 [degrees] C. Ambient temperature for Freeport runs was about 22 [degrees] C and from 25-34 [degrees] C at Texas Fibers. Any deviations in formulation components are noted in table 1.

Table : Table 1 - foam formulations

             Varimax      Varimax      Varimax     Texas
             pilot line   pilot line   pilot line   Fibers
Polyol              100          100          100      100
Water               4.1          5.1          6.1      4.1
TDI                56.6         68.2         79.9     56.6
Surfactant          1.0          1.1          1.2      1.0
Amine               .12          .11          1.0     .095
Tin                .162          .18          .23       na

Component description

Polyol - Voranol(1) 3512TB; TDI - Voranate(1) T-80 (120 index); Surfactant - Q2-5160 (pilot line) Niax(2) L-562 (Texas Fibers); Amine - D 8264 (pilot line), BL 13/Dabco(3) 33LV 1.5/1 (Texas Fibers); Tin - stannous octoate

1 - Voranol and Voranate are trademarks of the Dow Chemical Co.; 2 - Niax is a trademark of Union Carbide Corp.; 3 - Dabco is a trademark of Air Products.

The foam bun dimensions varied with the formulations used and the conveyor width of the two foam machines. The 4.1, 5.1 and 6.1 pphp water pilot line foams were 27" x 50", 36" x 51" and 40" x 50", respectively. The Texas Fibers foam was 37" x 85".

Foam physical properties Samples (15" x 15") were cut from the center of each of the foams produced for physical testing according to ASTM procedures.

Temperature data acquisition Foam bun temperature profiles were measured using a foam-in-place aluminum rack supporting a grid of 15, 20 gauge thermocouple wires which interfaced with a personal computer. The actual location of each of the thermocouples within the foam was measured and is shown in figure 1. Temperatures were recorded every 30 seconds, from pour through 20 hours.

The rack was manually placed in the rising foam a few feet from the mixhead. The rack was stabilized until the foam gel strength was adequate to support the thermocouple grid. Slight folds occurred near the top of the foam where it came in contact with the rack. Except for small lines of densification along the paths of the thermocouple wires, the internal foam cell structure appeared undisturbed. The foam buns were cut as they exited the foam machine tunnel with the rack centered in an eight foot long section.

Antioxidant extraction and analysis A vertical slice (one inch thick) of each of the foams produced was taken at approximately 20 hours after pour. Samples (about 1 gram) were taken from locations in the foam corresponding to each of the thermocouple locations. These samples were placed in methylene chloride (60 ml) for three days prior to analysis. The AO concentration of the extracts was determined by reverse phase liquid chromatography using a Novapak C18 column and water: acetonitrile (60:40) as the eluting solvent. Repeat extraction of a single foam sample with fresh solvent yielded no detectable AO. Initial AO content of the polyols used in this study were determined similarly.

In the case of the 5.1 pphp water foam produced in Freeport, additional foam samples for AO analysis were taken from vertical slices at 6.5, 33.5, 61.5 and 106 minutes. The 6.5 minute sample was collected at the cut-off saw after removing the initial three feet of foam produced. Four eight foot long buns (one containing the thermocouple grid) were cut and allowed to stand. The 33.2, 61.5 and 106 minute foam slices were taken from the centers of three of these eight foot buns. In this way, the additional foam slices were collected such that they experienced identical heat histories.

Results and discussion

The physical properties of the foams prepared for this study are presented in table 2. With regard to heat dissipation and the potential for antioxidant migration, foam density is of primary concern. The density varies, as expected with increasing water level, from 1.04 to 1.53 pcf on the pilot foam line. The two 4.1 pphp water foams produced (Freeport and Texas Fibers) have essentially identical densities (1.53 and 1.56, respectively). Additional physical property parameters are presented for inspection.

Table : Table 2 - foam physical properties

                 Varimax      Varimax      Varimax      Texas
                 pilot line   pilot line   pilot line   Fibers
Water                   4.1          5.1          6.1       4.1

Density

  (lbs./cu.ft.)        1.53         1.23         1.04      1.56
Tensile, psi           16.7         13.8         13.9      15.4
Elong., %               237          148          147       158
Tear/, lbs./in.         2.8          1.7          2.1       1.8

Resil, %

(DB)                     48           37           41        48
Comp. set 90            5.5          7.7         13.5       6.1

ILD 

  25%                  39.5         43.7         46.7      46.6
  65%                  69.9         77.1         81.4      94.1
  ret. to 25%          25.9         25.5         24.7      30.8
% hyst. retrn.         65.6         58.3         52.9      66.2
Modulus                1.77         1.77         1.74      2.02
Air flow               4.08         0.21        <0.16      2.22

The peak exotherm for each foam occurred within approximately fifty minutes from pour (figure 2). As expected, the maximum core exotherm of the three pilot line foams is proportional to the water level. Despite some scatter in the production foam temperature data, its maximum temperature (154 [degrees] C) is closely duplicated in the pilot line foam (155 [degrees] C). These peak exotherm results are in general agreement ([+ or -] 5 [degrees] C) with previous thermocouple probe observations.

These peak core temperatures were sustained over a range of about one to two hours in each of the three pilot line foams. The rate of core temperature loss was proportional to the maximum exotherm observed. That is, after 2-1/2 hours the core temperatures of these three foams were about 1, 6 and 12 [degrees] C cooler (compared to 50 minutes, figure 2) for the 4.1, 5.1 and 6.1 pphp water foams respectively (figures 2 and 3). The differences in cooling rates observed in these three foams may be due in part to their respective densities. The rapid cooling of these pilot line foams results in equalization of the 5.1 and 6.1 pphp water foam core temperatures (149 and 148 [degrees] C) at about five hours following pour (figure 4).

As mentioned above, the Varimax pilot line and the larger production size foam, both at 4.1 water, showed similar peak core exotherms. The similarity of these two foam temperature profiles continues through 2-1/2 hours (figure 3), but shortly after three hours the differences in heat loss rates become a factor. Examination of the five hour exotherm profile (figure 4) shows a 24 [degrees] C difference between these two foams. It is also apparent that the larger production foam has a different temperature profile from top to bottom with an apparently warmer upper foam area in comparison to the pilot line foams.

The differences observed in the long-term, 5 to 20 hours after pour, thermal histories of the pilot line and production foam are most easily explained by the surface area to volume differences which exist in these two cases. As expected, the smaller foam with a high surface area to volume ratio, dissipates heat more rapidly than the larger foam with its lesser surface area to volume ratio. The entire twenty hour heat profiles, as measured by thermocouples 2, 5, 8, 11 and 14, are presented for each of the four foams (figures 5 to 8). The 20 hour antioxidant concentration profiles are presented for the phenolic and amine inhibitors in tables 3 and 4, respectively. The amount of both AOs is seen to be reduced in the center (corresponding to thermocouples 5, 8 and 11) of all foams studied.

Table : Table 3 - phenolic antioxidant concentration profile (ppmp polyol)

         TC #1   TC #4   TC #7   TC #10   TC #13
         to #3   to #6   to #9   to #12   to #15

4.1 Water

Top      3,836   4,820   4,639   4,361    4,262
Center   3,705   1,754   1,393   1,639    2,689
Bottom   3,656   1,656   2,738   2,344    2,262

5.1 Water

Top      1,737   1,930   2,158   2,105    1,807
Center   1,298   1,140   1,158   1,123      965
Bottom   1,228   1,439     930   1,035    1,475

6.1 Water

Top      1,925   2,189   1,830   1,717    1,849
Center   1,887   1,340   1,377   1,396    1,830
Bottom   1,925   1,849   2,302   2,226    1,849

4.1 Water production foam

Top      2,871   5,565   7,597   4,694    2,016
Center   1,548   1,000   1,113   1,226    1,742
Bottom   1,081     919     903     919    1,081

Table : Table 4 - amine antioxidant concentration profile (ppmp polyol)

         TC #1   TC #4   TC #7   TC #10   TC #13
         to #3   to #6   to #9   to #12   to #15

4.1 Water

Top      1,721   1,574   1,803    1,639   1,738
Center   1,607     934     902      918   1,115
Bottom   1,689     934   1,344    1,213   1,295

5.1 Water

Top        544     386     439      368     754
Center     175     175     175      175     474
Bottom     632     596     456      474     858

6.1 Water

Top        396     547     245      264     264
Center     189     226     226      226     321
Bottom     981     660     491      811     868

4.1 Water production foam

Top        790   1,048   1,129    1,177     758
Center     613     823     516      661     468
Bottom     581     371     258      339     387

The 4.1 pphp water pilot line foam contains high concentrations of both phenolic and amine AO along the top row and left column of thermocouple locations. In fact, all of these values are greater than the original polyol AO content (3,308 ppm phenolic/1,564 ppm amine). The mean phenolic AO content for this cross section is found to be 3,068 ppm or 93% of the initial AO level. The average amine AO content for this sample is 1,362 ppm or 87% of the initial AO level.

The 5.1 and 6.1 pphp water pilot line foams have similar AO concentration profiles with much less AO remaining than the 4.1 pphp water pilot line foam. The mean phenolic AO content is 1,435 ppm (5.1 water) and 1,833 ppm (6.1 water) for these foams. This level of extractable phenolic AO represents 43% and 55%, respectively, of the initial polyol AO content. The average amine AO content for these two foams is 496 ppm (5.1 water) and 534 ppm (6.1 water). These values are only 32% and 34% of the initial polyol amine AO content.

The AO concentration profile of the 4.1 water production foam shows elevated values predominantly along the sides and especially the top of the foam bun. The phenolic AO values corresponding to thermocouple locations 4, 7 and 10 far exceed the 3,420 ppm initially present in the polyol used for this foam. The mean phenolic AO content for this cross section is 2,285 ppm or 67% of the initial value. The amine AO profile does not contain values which exceed the initial polyol content of 1,713 ppm and gives a cross sectional average of 661 ppm or 39% of the starting amount.

The 5.1 water pilot line foam AO concentration profile was also studied as a function of time. The phenolic AO profiles from 6.5, 33.5 and 106 minutes from pour are considerably different from the final 20 hour profile. These earlier cross sections show high concentrations of phenolic AO in the center of the foam relative to the edges. The 6.5 minute cross section gives a mean phenolic AO content of 3,068 ppm or 93% of the initial polyol value. This cross section was taken well after blow-off and indicates little loss in phenolic AO. The 33.5, 61.5 and 106 minute cross sections yield 2,238 ppm, 2,214 ppm and 2,043 ppm average phenolic AO. These values represent 68%, 67% and 62%, respectively, of the initial polyol AO content. As mentioned above, the 20 hour cross sectional average phenolic AO content is found to be 1,435 ppm or 43% of the starting value. The amine AO concentration profile shows similar results as a function of time but with much lower cross sectional averages. The 6.5, 33.5, 61.5 and 106 minute average values are 1,149 ppm, 614 ppm, 486 ppm and 532 ppm or 73%, 39%, 31% and 34% of the initial polyol amine AO content. Again, the 20 hour amine AO cross sectional average for this foam is 445 ppm or 28% of the starting amount.

Conclusions

The three water levels (4.1, 5.1 and 6.1 pphp) used to produce foams for this study represent a range of potential change for the future of polyurethane foam manufacture. The 4.1 pphp water foam is a common upper limit for an all water blown foam as is currently practiced by commercial foam producers. The 5.1 and 6.1 pphp water foams represent likely changes for the future as the use of alternative blowing agents is reduced. The core temperature and maximum exotherm values of the three pilot line foams correspond well with prior single probe data. The temperature profiles reported here define the gradient which results as a function of the low thermal conductivity of polyurethane foam. This property results in a sustained warm core with decreasing temperature toward the foam edge. The peak exotherms of these three foams occur within an hour of pour and increase proportionately with higher water level in the formulation. After about two hours, these foams begin to cool at a rate proportional to their maximum exotherm. The hotter the foam core temperature, the more rapidly it cools. Because of the different rates in cooling, the 5.1 and 6.1 water foams have similar temperature profiles at about five hours after pour.

The foam bun cooling rate is also affected by the dimensions of the foam. The 4.1 pphp water production and pilot line foams have similar peak exotherm profiles; the same amount of initial heat is produced. As these foams begin to cool, however, the higher surface area to volume (mass) of the pilot line foam allows it to cool much more rapidly. This difference becomes apparent by five hours following pour. By ten hours after pour, there is nearly a 50 [degrees] C difference in these two foams.

The impact of the different heat histories of the four foams studied can be seen in the antioxidant concentration profiles. The coolest foam, the 4.1 pphp water pilot line foam, retains the minority of its initial AO package even after 20 hours.

The other two pilot line foams, 5.1 and 6.1 pphp water, lose more than half of their starting AO content by 20 hours. When the production and pilot line foams are compared at the same water level (4.1 pphp), different AO profiles are seen. The larger production foam with a more sustained temperature gradient has less recoverable AO than the more rapidly cooling pilot line foam.

Both of the 4.1 pphp water foam AO concentration profiles support the hypothesis of AO migration. The AO concentrations detected along the side and edges of these foams exceed the initial loading of AO in the polyols used. The only way this could occur is by migration of the AO from the warm center of the foam out toward the cooler edges.

Finally, the AO concentration examined as a function of time indicates that little AO is lost due to blow off. It appears that at least through the first 1-1/2 hours substantial amounts of AO remain in the foam center. It can also be seen, however, that total AO content is diminished between 6-1/2 minutes and 1/2 hour following pour. The remaining AO loss and final AO concentration profile occurs after 1-1/2 hours up until some time prior to 20 hours following pour.

It is not known how much of the undetected AO has been consumed and how much is lost to the atmosphere via migration. Also, because these foams did not have significant discoloration, we are not able to examine the interaction of AO concentration profiles and color formation. These areas will be the topic of future work.

References

[1. ] Lutz, J.T., Jr., ed. 1989. Thermoplastic polymer additives: Theory and practice. New York: Marcel Dekker, Inc. [2. ] Ranby, B. and J.F. Rabek. 1975. Photodegradation, photooxidation and photostabilization of polymers. London: John Wiley and Sons. [3. ] Tochacek, J. and J. Sedlar. 1989. "The influence of molecular weight on the efficiency of phenolic antioxidants as stabilizers for polypropylene - a short communication," Polymer Degradation and Stability, 24: 1-6.

PHOTO : Figure 1 - thermocouple locations (view from mixhead)

PHOTO : Figure 2 - foam exotherm profile (maximum 50 minutes)

PHOTO : Figure 3 - foam exotherm profile (at 2.5 hours)

PHOTO : Figure 4 - foam exotherm profile (five hours)

PHOTO : Figure 5 - 4.1 water pilot line foam

PHOTO : Figure 6 - 5.1 water pilot line foam

PHOTO : Figure 7 - 6.1 water pilot line foam

PHOTO : Figure 8 - 4.1 water production foam