Printer Friendly

Novel high load polymer polyol technology for improved fatigue and processing.

Polyurethane foam carpet underlay has been manufactured for decades, both as prime foam and as rebond material. Over the last few years the production of prime carpet underlay has been declining, primarily due to customer complaints about unsatisfactory durability performance. It is generally known that durability is reduced by lowering the foam density, by increasing the isocyanate index and by increasing the proportion of polymer polyol in the formulation. In other words, an economically desirable move towards lower foam densities, combined with maximum hardnesses, resulted in significant deteriorations of foam durability as observed in real life use and measured by various fatigue tests. More recently the durability of prime carpet foam has been improved by producing higher density grades with hardnesses lower than before. Lower TDI indexes were the major tool towards achieving a better fatigue performance and consequently lower hardnesses.

Our research has shown complex relationships between formulation variables like water level, solid polymer content and TDI index and resulting properties like density, hardnesses, mechanical properties and fatigue (ref. 1). It became apparent that for a given density and hardness a combination of low TDI indexes and high polymer solids was desirable for obtaining improvements in durability. It is also known, however, that combinations of high polymer polyol levels and low TDI indexes (less than 110) lead to processing problems like top and internal splits. We directed further research towards utilizing the quantitative relationships between formulation variables and foam properties for the design of a new polymer polyol, with good processing characteristics in the range of TDI indexes and polymer contents where the improved fatigue performance was expected.


The laboratory foams evaluated in this study were made by the following bench mixing technique - all polyols, water, silicone and amine were premixed with a drill press and blade stirrer for one minute, then allowed to degas for 15 seconds. The tin catalyst was added, followed by 10 seconds of agitation. Finally the TDI was added over five seconds, with agitation for 10 seconds. The mixtures were poured into cardboard boxes (14" x 14" x 6"); cream time and blow off time were observed; the rise profiles were recorded with an ultrasonic device. The foams were subjected to a surface cure in an oven at 120 [degrees] C for five minutes. After ambient cure for at least three days the foams were cut into test samples (12" x 12" x 4") and tested for physical properties. Machine foams were prepared on our low pressure multi stream pilot Maxfoam machine; individual blocks representing one variable change (40" x 24" x 20") were cured at ambient for at least three days, then cut into test samples and tested like the bench made foams.

Foam physical properties were measured according to ASTM D 3574. Durability was measured by the Stomper Fatigue test, with 50,000 cycles at 90% compression, with CFD testing at 25% and 65% performed before and after the compression cycles. Fatigue losses are expressed as a percentage CFD loss at 25% and 65%, respectively.

Optimization of foam properties and processing

The most important properties considered in this study were density, hardness (IFD) and fatigue. Other properties like tear strength, tensile strength, elongation and compression sets were also recorded.

Density is primarily influenced by the water level in the formulation. A relatively small, but significant, effect of TDI index on density has also been observed. Lower indexes generally lead to higher densities. A small effect of polymer polyol concentration on density was ignored in this study since it is insignificant in the range of polymer concentrations investigated. The effect of water on density, for different TDI indexes, is shown in figure 1.

Hardnesses and fatigue are influenced by water level, polymer content and index, to different degrees. These relationships, resulting from previous studies (ref. 1) are shown in figures 2 to 5, for a given density, e.g., 2.1 pcf, equivalent to 2.7 parts water in the formulation. This density or water level represents a significant proportion of commercially produced carpet underlay foam and was therefore selected for the majority of the subsequent studies. It can be seen from these graphs that a grade of that density, at an IFD 25% of 100 lbs. can be made with 2.7 php water, at an index of 120, with 77 php Arcol HS-100 polyol, with a corresponding fatigue loss of 41 %. The same grade can be made at the same water level at an index of 110 with 80 php Arcol HS-100 polyol, with a fatigue loss of 34% (the small increase in density at the lower index is ignored for this comparison). With a further small reduction in index, to 108, the formulations for this grade would read: 2.7 php water, 108 index, 88 php Arcol HS-100 polyol and a fatigue loss of 33%. At this index and polymer level, however, top and internal splits were usually observed on our laboratory Max-foam machine. These splits would be expected to be even more pronounced on large scale production equipment.

This illustrates the benefits of reformulating such carpet underlay grades towards lower TDI indexes but highlights some associated problems. The graph of fatigue loss vs. TDI index suggests that there may be no benefit in reducing the index below 105 since the fatigue losses appeared to level out or even slightly increase again. However, only limited data points were available at indexes below 110 and virtually none at index 100; the prediction model may, therefore, not be very accurate in this range.

Therefore, it was decided to seek the answer to two questions: Can the fatigue performance be further improved by further index reductions? And is it possible to produce foam without defects or processing problems at these low indexes?

This study was focused on TDI indexes between 100 and 110, with higher indexes as control points. Work below 100 index was not attempted due to suspected problems of sub-index foam such as more rapid discoloration under light, possibly increasing compression sets, etc.

Tables 1 and 2 show selected results of a laboratory study where two formulations were run at there TDI indexes, each at different tin catalyst levels.


It can be seen from table 1 and table 2 that the IFDs with 95 php Arcol HS-100 polyol at index 108 are comparable to those with 85 php HS-100 polyol at index 118; the fatigue losses are approximately 15% lower. The IFDs with 95 php HS-100 polyol at index 100 are comparable to those with 85 approximately 10%. This means that there is a potential benefit from reducing the TDI index to 100. However, a definite tendency towards splitting was seen in these bench foams, rated slight at index 108 and severe at index 100. Increasing the tin catalyst at index 100 only led to tight foams which still showed splits.

At the next step it was decided to investigate the process-ability at low index. A series of three experimental polymer polyols was prepared, with variations in the molecular structure, which were expected to influence foam stability at medium to low index. A series of scouting laboratory experiments was run, with 100% Polymer polyol, and TDI index and tin catalyst level as variables, in order to investigate the processing latitude of these polyols. Selected results from this processing study are presented in table 3.


The following conclusions could be drawn from these experiments. Arcol HS-100 polyol, the control polyol, showed a good processing range at index 118, as expected. but could not be Processed at index 108 or 100. The experimental polyols A and B led to an acceptable processing range at index 118 and good range at indexes 108 and 100. The experimental polyol C showed a very narrow processing range at index 118, an intermediate one at index 108 and a good one at index 100. Foam physical properties for this series were within expectations; these are not included in the table.

The experimental polyols A and B were selected for further studies; these polyols gave the best compromise in processing latitude at high index and stability at low index. A series of machine runs was conducted on our pilot Maxfoam machine; selected results as well as the operable tin latitude are summarized in tables 4-6.


The following conclusion were drawn from the machine runs. The machine-made foam generally needed more tin catalyst or a more stabilizing polyol than the bench foams to achieve sufficient stability. Polyol A could not be processed on the machine at index 100 but processed well at indexes 108 and 118. Polyol B processed well at indexes 100 and 108; the processing latitude at index 118 was narrow but may be sufficient.

Only split-free foam was subjected to testing of full foam physical properties; foams with splits were only tested for density, porosity and stomper fatigue. Contrary to the earlier prediction model, there was a significant trend towards lower fatigue values with lowering the TDI index to 100. The absolute fatigue values for the machine-made foams were approximately 10-15% lower than those from the corresponding bench foams.

Figure 6 gives a graphic picture of the fatigue loss vs. TDI index, at two levels of polymer polyol, corresponding to the previous model shown in figure 3. The experimental polyols led to hardnesses similar to those obtained with HS-100 polyol, for a given TDI index and porosity. In other words, the IFD model used figures 2 and 4 earlier proved to be valid also for the experimental polyols, down to an index of 100. The reduction in index improved other properties besides fatigue, notably their strength and elongation.


It has been demonstrated that the durability of carpet underlay foam can be improved over currently achievable values by the following techniques: for a given density and IFD a combination of higher polymer polyol levels and lower TDI indexes should be sought for obtaining optimum durability performance. For a TDI index of 108 or below the use of a more stabilizing polymer polyol such as the experimental polyol B becomes necessary. The reduction in TDI index also improves some mechanical properties and may offer some environmental advantages e.g., a lower emission of TDI vapor. Further work is planned to investigate the impact of utilizing polymer polyols such as experimental polyol (B) at indexes below 100.


[1.] Lawler, L. F. 1992. "The effects of fomrmulation variables on foam durability: Water content, polymer solids, index" 34th Annual Polyurethane Technical Marketing Conference, Polyurethanes |92, October 21-24-1992. (*) This article is reprinted by permission of the Polyurethane World Congress 1993.
COPYRIGHT 1994 Lippincott & Peto, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Craig, T.A.
Publication:Rubber World
Date:Apr 1, 1994
Previous Article:Rotational casting of PU covered rolls.
Next Article:CB's role in compound curing behavior.

Related Articles
Materials and machinery advances will shine at polyurethane conference.
Fluoropolymers: the new breeds.
Antioxidant behavior in flexible PU foam.
New additives and polyols surface at urethane conference.
Polyurethane conference preview: foam formulators chase moving target.
New 'eco-logical' RIM equipment.
'Third-generation' blowing agents starred at Polyurethanes '95.
New polyols for elastomers boost performance at lower cost.
Polyurethanes Expo 2003 held.
Corn-based engineering plastics on the way.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters