Polyurethane systems for structural composites

By Dan Heberer and David Bareis, Huntsman Polyurethanes

Polyurethanes (PU) are a promising alternative to traditional vinyl ester and epoxy-based matrix systems. They offer the potential of short cycle times, high toughness and exceptional durability. They have not been widely adopted in the past because they suffer from short pot life and gel times.

But by using new catalyst packages, urethane systems can be adjusted to offer unique processing parameters, cure temperatures, cure cycles and physical properties to make it possible to challenge the dominance of incumbent materials.

This paper shows data on resin cure properties and rheology as well as the physical properties and composite production methods that are achievable with urethane materials. It shows that with correct formulation, polyurethane systems can have significant production benefits over traditional epoxy formulations in a wide number of applications.

By modifying compounding parameters and ingredient choices, it is possible to make PU resin systems with glass transition temperatures (Tg)s from 104°C to 225°C, viscosity values as low as 150 cP, and pot-lives of several hours. These urethane resins have outstanding impact strength and are considerably higher than epoxy resins with an equivalent Tg. Advances in urethane formulation and catalysis has greatly expanded the ability to use urethane chemistry for composite applications and processing methods including pultrusion, RTM, resin infusion, filament winding, and cure-in-place pipe repair.

By understanding the processing method, cure time and temperature, and the required features of the finished part, HPU can select the right polyurethane resin for many composite applications.

Delayed-action, snap-cure

With traditional formulations, the urethane system begins to react and viscosity rises immediately after mixing. These highly reactive systems have a short working time which prevents the production of large composite parts.

We have developed a breakthrough in catalysis chemistry that offers formulations with pot-lives of up to several hours at room temperature and a snap-cure during processing at elevated temperatures.

This catalyst allows polyurethane composites to be produced using common thermoset composite processing operations. This combination opens up new opportunities for material substitution. Target materials are: metals, expensive high-performance epoxies, phenolics, vinyl ester and polyester systems.

This paper shows how modified catalyst systems have the properties needed to make some of the possible substitutions.

In the work we discuss below, resins are designated by a letter and a number. The letter indicates the catalyst system: PUR is a traditional catalyst system and V resins are formulated with the advanced system. The number indicates nominal Tg. This is determined by the onset temperature from the storage modulus in dynamic mechanical analysis DMA experiments.

For example, a system called V-135 has a glass transition temperature of about 135°C and is made with the new catalyst system.

In the following work, two commercial epoxy systems with glass transition temperatures in the same range as the polyurethane materials were used as comparison materials.

Experimental

Rheological tests were carried out either with a Brookfield Viscometer or on a Dynamic Stress Rheometer using disposable parallel plates with a 25-mm diameter. To measure the onset of cure as evident by a rapid rise in viscosity, the instrument was run in a dynamic (oscillating) stress mode with a peak stress of 4 Pa, a frequency of 1 rad/sec and a 1 mm gap separation. However, due to the low initial viscosity of the polyurethane materials, it was necessary to revert to steady state (non-oscillating) tests to accurately determine the viscosity of the uncured material. All components were mixed together in a Flaktek mixer for 30 seconds before being placed in the rheometer.

Pot-life at room temperature is the time needed after mixing for the material to reach the string gel, point. This is when the material will form a string when a probe is withdrawn from the material. To find this point, 100g of system was mixed in a 125 ml cup and sampled with the probe periodically until the string time was reached. The material was not mixed during the test.

The DMA analysis on cured materials was conducted in oscillatory mode with a frequency of 1 Hz and a heating rate of 3°C/min in a torsion beam geometry. The glass transition temperature is the temperature at which the storage modulus G’ begins to rapidly fall with increasing temperature.

To make void-free, bubble-free samples the system components were individually degassed, then combined and degassed again in a second container.

Mixed components were transferred to a heated mould and cured under a pressure of 2 bar using the given cure conditions.

The tensile properties of the neat resins were evaluated in accordance to ASTM D638-03, ASTM D790 was used to determine the Flexural properties, and D256 used for the Un-notched Izod impact tests.

Results and discussion

Figure 1 is a plot of the G’ storage modulus versus temperature from DMA experiments on the fully cured polyurethane materials. This shows that PUR systems have a Tg range from 104°C to 157°C; the V systems have a range that starts at 135°C and extends to a high of 225°C. The different systems have been developed to cover the widest possible range of Tg to fit a number of different processing and physical property constraints.

Two standard epoxies are included in the table, one has a Tg of 140°C, the other 195°C.

Flow and viscosity

Table 1 gives the basic rheological data on the epoxy and polyurethane systems.

The two epoxy systems have room temperature viscosity values of 700 and 750 cP. These viscosities are at the high end of the range for the polyurethane materials. Isocyanates have a very low viscosity so the viscosity of the polyurethane systems can be as low as 150 cP immediately after mixing. The highest viscosity of one of the polyurethane systems is 950cP, shown by the formulation with a Tg of 160°C at 950cP.

This implies that the lower viscosity of polyurethane systems could more easily wet-out fibres and faster injection rates.

The pot-life, or the working time, of the polyurethane systems are shorter than the epoxy, but in all cases is 35 minutes or greater.

The PUR materials do not contain catalysts so this would be the maximum gel time available. It could be decreased if catalysts were added to give faster demould times.

This is in contrast to the V systems which are catalysed but which have extended gel times greater than 90 minutes.

The catalyst technology used in these materials gives an induction period where there is minimal viscosity growth, followed by a rapid snap-cure.

Figure 2 illustrates the change in viscosity of PUR-160 and V-225 as they are heated at a rate of 3°C/min.

The shape of the viscosity curves is determined by two competing processes: the decrease in viscosity that occurs as the resins are heated and the increase in viscosity as the materials begin to react.

At about 41°C, the PUR-160 mixture becomes clear and there is a decrease in viscosity, but by 63° the viscosity of PUR-160 is higher than it was at room temperature and viscosity rises steadily as the material begins to cure.

The V-225 system displays a minimum viscosity at 52°C but in contrast to PUR-160, viscosity remains below the viscosity at room temperature until 77°C is reached. At this temperature, the material exhibits snap-cure behaviour with rapid conversion to a cured material.

The ability to maintain a low viscosity for an extended time enables the production of large composite parts.

This experimental information is encouraging, but it is important to understand the viscosity profiles of the resin systems under conditions simulating RTM moulding operations. To do this we measured the way that viscosity changed with time under constant (isothermal) temperature conditions.

The viscosity profiles at 60°C, shown in Figure 2, show the snap-cure behaviour of the V-225 system. Initially there is a drop in viscosity for both PUR-160 and V-225 as the material heats up to 60°C. Both materials then begin a slow increase in viscosity as curing starts and progresses. The viscosity of V-225 remains very low until snap-curing after about seven minutes.

Processing advantages

The rapid, full-cure property would enable parts to be de-moulded shortly after resin injection. Because polyurethane chemistry is very flexible, it is possible to adjust the temperature at which the snap-cure starts and the time taken for the polyurethane part to gel sufficiently for a safe de-mould. By tailoring these properties to the production process it would be possible to give the lowest possible cycle time.

Table 2 gives typical mechanical property data for the two epoxy and six urethane materials for neat resin plaques prepared with the indicated cure temperatures and times. If the ultimate Tg of the material is above the initial cure temperature a post-cure was added to complete the curing process. The times used were not optimised to indicate the fastest available cure cycle, rather they were used simply to prepare the plaques for evaluation.

Toughness

The key benefits of polyurethane systems over other materials are their greater toughness. This is shown in Table 2 through the tensile yield strength and tensile break strain, the flex strain, and the un-notched Izod tests.

The tensile modulus and tensile strengths of the epoxy and urethane materials all fall in the same range with no significant differences between the materials for a given Tg.

The key differences between epoxy and polyurethane materials are the yield observed in the PUR samples and the increased tensile break strain exhibited by the PUR and V systems.

The improvement in the tensile strain of the polyurethane materials is achieved without sacrificing the tensile modulus and tensile strength.

Comparing two materials with equivalent Tg: the standard epoxy (Tg = 140°C) and the V-135 (Tg = 135°C), shows the tensile break strain of V-135 is 88% higher than the standard epoxy. V-135 fails at 8.5%, the standard epoxy at 4.5%.

The tensile strain at break for V-135 is higher and it has the same tensile strength as the standard epoxy. This shows that it has a higher toughness. Likewise, comparing the high Tg epoxy (Tg = 195°C) to the V-225 (Tg = 225°C) shows tensile strain at break is 15% higher for V-225 at 2.3% even though its Tg is 30°C higher than the epoxy which failed at 2%. In this case the higher tensile strength for an equivalent tensile break strain indicates a higher toughness for the V-225.

Similar observations are seen in the flexural property measurements where the improvements in flexural break strain occur independently of the flexural modulus and flex strength.

In flexion, PUR materials yield and do not break which shows toughness, the V systems demonstrate high flexural break strains.

The most direct measurement of toughness that was measured was the Unnotched Izod Impact Test (UIIT). The impact strength of the systems was inversely related to Tg: samples with lower Tg gave highest impact values; samples from formulations with higher Tg gave the lowest impact values.

As with the tensile break strain and flexural break strain, it is best to compare values from materials with equivalent Tg. This comparison can be found by plotting the unnotched Izod impact value versus the Tg of the material. An improvement is easily seen for all Tgs and as an example, a comparison shows that the V-135 UIIT value is 66% higher than the standard epoxy.

Applications of the new urethane chemistry

This advanced polyurethane chemistry has greatly expanded the ability to use urethane chemistry for composite applications and resulted in new resin systems offering material processing advantages over incumbent materials

The new materials can be processed using methods that were not possible with standard urethane systems. Table 3 shows the fit between the urethane systems and five different processing methods. The colour of the box indicates the suitability of a resin system with a given processing method. A green box means that the resin/processing combination has been used to prepare either glass or carbon fibre parts. The yellow box shows that a system would be expected to be compatible but it has not yet been demonstrated. The red boxes indicate that the resin would not be a candidate material for the given process method (for instance the high initial viscosity and the viscosity build versus time of PUR-160 would be problematic for vacuum infusion).

A wide range of parts can be made with processes using one of the urethane resins. For example, carbon fibre composite tubes made via filament winding that were subsequently post-cured in an oven. Carbon fibre tubes as long as eight meters have been made with V-225.

A collection of pieces of pultruded parts showing the wide variety of profiles and constructions are possible with urethane chemistry. Products have been made with either glass, basalt, or carbon fibre and with or without additional fillers. Urethanes have been demonstrated to run at high production speeds with minimal scrap or down time.

Carbon fibre composite made by vacuum infusion with perimeter injection and a central outlet and degas point. The low viscosity and long open time of the V-225 urethane produced a 110 X 120 cm part with very good infusion of the resin into the fibres.

This is an edited version of a paper presented at the CPI conference in Phoenix, Arizona in September 2013.

References

1. Connolly M., King J.P., Shidaker T.A., Duncan A.C., “Pultruding Polyurethane Composite Profiles: Practical Guidelines for Injection Box Design, Component Metering Equipment and Processing,” Proceedings of Composites 2005, American Composites Manufacturers Association, Sept. 2005.

2. Connolly M., King J.P., Shidaker T.A., Duncan A.C., “Processing and Characterization of Pultruded Polyurethane Composites,” Proceedings of the 8th World Pultrusion Conference, European Pultruders Technical Association, March 2006.

3. Connolly M., King J.P., Shidaker T.A., Duncan A.C., “Characterization of Pultruded Polyurethane Composites: Environmental Exposure and Component Assembly Testing,” Proceedings of Composites 2006, American Composites Manufacturers Association, Oct. 2006.

4. Connolly, M., van Boxel, R. and Brennan, M., “Rheo-Kinetics and Cure Modeling for the Pultrusion of

Polyurethane Composites,” Proceedings of the 1st Global Pultrusion Conference, Baltimore, MD, June 2009.

5. Brennan M., Connolly M., Shidaker T.A., “CFD Modeling of the Closed Injection Wet-Out Process For Pultrusion,” Proceedings of the 9th World Pultrusion Conference, European Pultruders Technical Association, March 2008.

6. Joshi, R.R., Varas, L.L., Padsalgikar, A.D., “Polyurethanes in Pultrusion: Styrene Free Alternative Systems,” International Composites Expo 1999-SPI, (Society for the Plastics Industry), April 1999.

Biographies

David Bareis is a market development manager for Urethane Composites in Huntsman Polyurethanes Business Development. For over 20 years at Hunstman David has been involved in the development and commercialisation of polyurethane systems for automotive and composite applications.

Dan Heberer, Ph.D. is a graduate of the Polymer Science Department at the University of Akron. His career has primarily focused on the development and commercialisation of adhesives for automotive applications. Currently he is a technical manager at Huntsman Polyurethanes.