Thermosets in the '90s: plenty of them yet.

Thermosets in the '90s

Trends in thermoset resin development in the next decade are likely to mirror those in thermoplastics (see preceding stories). Though a limited number of new polymers may be developed to fill specific processing and performance niches, suppliers will probably rely more on blending, alloying, and interpenetrating polymer networks (IPNs) of existing thermosets (sometimes together with thermoplastics) to answer future market demands.

The current regulatory climate has created special demands on polyester resins. They must meet the challenge of lower styrene emissions and provide better smoke/flame performance without sacrificing their low cost, processability, and corrosion resistance. New intermediates and monomer diluents could present some opportunities here. Otherwise, polyesters could face increased competition from phenolics, which some believe are poised for a renaissance.

In the next 10 years, epoxy materials will provide improved toughness in finished parts and lower viscosity during molding without a corresponding loss in high-temperature performance. These improvements will come from development of new multi-polymer hybrids or multi-phase materials, or brand-new epoxy backbones.

A simultaneous improvement in processing and properties will be much more difficult to achieve with polyimide materials. To promote the use of these materials in non-military commercial markets, suppliers may compromise certain properties such as temperature performance to facilitate easier processing and lower cost. Meanwhile, newer thermosets such as cyanate esters and BCBs will be competing for niches in the high-performance market sector.

POLYESTERS: THE STYRENE CHALLENGE

EPA and OSHA restrictions on volatile organic compounds (VOCs) and styrene volatiles in the workplace, respectively, will force unsaturated polyester suppliers to find innovative ways to reduce the level of styrene in their materials and support development of "cleaner" processing technologies.

Suppliers agree that although there will be some improvements in the use of additives to reduce styrene emissions, that approach will probably be a temporary "quick fix." Use of alternative monomers will be explored somewhat, although that could lead to polyesters that cost as much as three times more, according to Charles Dudgeon, director of technology and commercial development of the Composite Polymers Div. of Ashland Chemical Inc., Columbus, Ohio. Understandably, none of the material suppliers interviewed foresee wholesale replacement of styrene. Barring any drastic regulatory changes, styrene will continue to be the dominant monomer diluent for polyesters throughout the '90s because of its availability.

The drive will be to reduce the use of styrene as much as feasible. One of the approaches to complying with Rule 1162 in Southern California is to use a resin containing less than 35% styrene. Less-volatile monomers such as vinyl toluene, paramethylstyrene (PMS), or acrylates, figure to be used increasingly as partial substitutions to bring the styrene content down. Since the substitute monomers are more costly than styrene, their use will probably be limited to what is needed to meet environmental criteria. For reasons of cost, PMS may be a more likely substitution than vinyl toluene.

PMS is made only by Deltech Corp. in Baton Rouge, La. Besides lower volatility, PMS reportedly has greater compatibility with low-shrink and low-profile modifiers, and also higher reactivity for faster and more thorough cures. The latter can help reduce shrinkage and improve hot strength and heat-distortion resistance. The main advocate of vinyl toluene in polyesters is Dow Chemical Co., Midland, Mich. Vinyl toluene is actually 30% PMS and 70% metamethyl isomer, and can be used interchangeably with pure PMS, according to Dow.

Perhaps the main thrust toward styrene suppression will be in changing polymer structures so less monomer is required in resin formulations. Efforts are focusing on reducing polymer molecular weight to provide a workable viscosity at low monomer levels, while retaining good mechanical properties. However, more studies are needed to reveal what reducing the molecular weight of a polymer does to its water and chemical resistance.

"Right now, the data are not there to say that a resin with 35% styrene content will have the same corrosion resistance as one with 45% styrene," according to Terry Sprow, technical manager for the Resins and Coatings Div. of Owens-Corning Fiberglas Corp., Toledo, Ohio. "Even by mid-'92, when regulations are scheduled to become stricter in California, there will probably not be enough data for everyone to be comfortable."

Ashland's Dudgeon says that in the next few years, his firm will be able to maintain the viscosity performance of a 45%-styrene resin for spraying, rolling, and glass saturation with only 35% styrene content. However, he also stresses that maintaining current corrosion resistance at lower styrene contents will be a separate challenge.

NEW INTERMEDIATES & HYBRIDS

At least one supplier (who did not wish to be identified) is working with a new intermediate that could make polyesters more competitive with vinyl esters in the future. The new material, 2-methyl-1,3-propanediol from ARCO Chemical Co., Newtown Square, Pa., is being evaluated for its ability to provide corrosion and uv resistance, higher tensile strength and greater toughness. The new material was discussed at this year's SPI Composites Institute Conference in Washington, D.C.

Another key R&D push is toward increased toughness for polyesters. Ashland, for one, is continuing to develop tougher SMC resin systems with new thermoplastic low-profile additives. But what appears to be the most popular area of toughening research involves combining polyester and urethane chemistries (see PT, March '87, p. 58; April '89, p. 37).

This was the basis of Chicago-based Amoco Chemical Co.'s first venture into polyester resin systems in 1988 (PT, March '88, p. 44). Owens-Corning is developing a polyester/urethane hybrid system for increased strength, heat resistance, and toughness in RIM parts. Cook Composites and Polymers Co., Kansas City, Mo. (formerly Freeman Chemical, PT, April '90, p. 163), is also working on low-styrene polyester/urethane IPNs for liquid molding. The resulting materials are expected to provide at least 15% elongation, according to James Purdy, Cook's director of international services.

Similar research at Cook has been expanded into developing vinyl ester/urethane IPNs with an eye toward increased toughness, high solids content, low viscosity, and low styrene emissions. Also developing vinyl ester/urethane hybrids is Interplastic Corp., Minneapolis. Says Terry McCabe, technical marketing director, "If we can achieve greater toughness with corrosion resistance, we'll have a new grade of engineering material." And Reichhold Chemicals, Inc., Durham, N.C., is continuing its work on toughened vinyl ester alloys, according to James E. Maass, director of marketing for the Reactive Polymers Div., even though Ford Motor Co. has delayed mass production of an SMC pickup truck bed based on Reichhold's isocyanurate-modified vinyl ester (PT, March '89, p. 15).

UV CURE IS COMING

Ashland is currently working to commercialize new technology licensed from BASF AG of West Germany for uv-curing of polyesters using photoinitiator catalysts. Previous photoinitiator technology has been applicable mainly to curing very thin layers--adequate for coating applications but not for laminates.

The technology allows curing of up to 3/4-in.-thick glass-reinforced laminates using low-powered uv lamps. Dudgeon says that a part with a room-temperature curing time of 6-8 hr might cure in 15 min. with uv. "But the big advantage will be for people who currently cure at elevated temperatures and will then be able to get the same productivity without putting parts in ovens," he adds.

The photoinitiating material being developed by Ashland is reportedly sensitive to a narrow band (350-400 nm) of uv light that is reportedly not harmful to humans. This allows visible light (with wavelengths above 400 nm) to be used for working without curing the material. Unlike other photoinitiators that work in the visible light range, the Ashland material does not impart a color to the part, according to Dudgeon.

Ashland plans to sell the catalyst premixed with polyester resin, saving a step for the processor. It will be possible to make a rolled prepreg of reinforcement, resin, and protective film in advance, and temperature for several weeks or months with no change in viscosity or cure. Due to the lack of air exposure, styrene emissions will be greatly reduced.

UV curing will also be suitable for novel techniques such as vacuum-forming thermoset resin prepregs, since the photoinitiator will not be activated by heat (a concept advocated by BASF for several years--see PT, Jan. '84, p. 70). Curing will begin only when forming is complete and the mold is exposed to uv light.

Cook Composites and Polymers, which has developed uv-curable binder resins for glass preforms, may also develop uv-cure molding resins, according to Jim Purdy.

OTHER TRENDS

A trend expected by R.L. Holtzendorf, vice-chairman, Alpha Corp., Collierville, Tenn., will be toward improved foaming processes that will allow greater use of unsaturated polyesters in cellular foams to save weight and cost.

Among other polyester R&D goals is higher heat resistance. "We're shooting for higher temperature performance for our materials as long as the economics stay within reason," according to Thomas George, executive director of P.D. George Co., St. Louis.

And some suppliers are working to reduce the need for finishing operations. Owens-Corning is trying to increase the pigmentability of low-shrink polyester systems for applications such as power tool and electrical appliance housings. The trick is to retain low shrinkage while increasing the pigmentability. Current low-profile/low-shrink resins can be pigmented in whites and pastels, but Owens-Corning hopes to take advantage of the potential market for stronger shades.

Addressing the growing quality concerns of manufacturers and their ever-growing emphasis on statistical process control (SPC), Owens-Corning also plans to offer more data to processors about each batch of polyester material shipped. The data will be supplied as requested by the processor, in forms such as SPC charts.

BURNING ISSUES

The need to meet stricter regulations for flame spread, visible smoke and smoke toxicity will make it increasingly difficult for polyesters to be used in the aircraft/aerospace, transportation, and building/construction industries. Currently, fire-retardant polyester resins typically incorporate chlorine or bromine. Because of their alleged smoke toxicity, these halogens may face greater restrictions or customer resistance in the future. Research is focusing on finding effective ways to use lower halogen levels to get the same fire performance, or to replace halogens with something else.

Interplastic's McCabe says current polyesters and vinyl esters can be made to produce 60-80% less smoke than current levels. "We're on the edge of obtaining those types of results while maintaining the handling characteristics of our resins," he said.

But some suppliers are considering other thermoset resin technologies--most notably phenolics. "Polyester resins can be made to produce less smoke," says Ashland's Dudgeon, "and I think they will come close to the performance of phenolics. Therefore, a lot of the applications that would go to phenolics today will eventually go back to polyesters. But there will always be a market segment where phenolics will be the material of choice," he said. Although it has not yet committed itself fully to the program, Ashland is in the early stages of developing a phenolic with low free formaldehyde, room-temperature cure, and processability similar to a polyester.

RENAISSANCE FOR PHENOLICS?

Given this demand for improved fire and smoke performance in many applications, the high-temperature, chemical and environmental resistance, and good mechanical properties of phenolics may cause a resurgence in their use in the 1990s, especially in fiber-reinforced composites. There are very few other materials that can match the performance profile of phenolics at comparably low cost. Suppliers of other resins, such as Allied-Signal, Ashland, and Reichhold, are entering or considering to enter the market with novel materials that offer improvements over traditional phenolics. Key trends in phenolic development will be in improving toughness and impact strength without sacrificing dimensional stability under heat and load. This may be accomplished through alloying with epoxies, thermoplastics and elastomers.

However, since commercial phenolics have been around since 1909, many people wonder what more can be done to improve them. "It's a good basic material, and there's a limit to how far you can go with exotic alloying systems and reinforcements and still maintain an acceptable cost, particularly when using it for metal replacement," according to Walter Hayes, v.p. of Rogers Corp.'s Molding Materials Div., Manchester, Conn.

One key challenge for the '90s will be in the further manipulation of phenolic cure rates to allow processors to minimize the amount of post-cure required. Alkaline curing systems that are less corrosive to machinery and glass reinforcements, and that are safer for operators, represent one direction of development.

Here are some of the signs of activity in phenolics:

* Indspec Chemical Corp., Pittsburgh, is a relatively new firm spun off from the now-defunct Koppers Co. It is continuing development of a family of resorcinol resins that cure by a non-acid mechanism (PT, Dec. '88, p. 19; April '90, p. 151). Resorcinol is a phenol with a second hydroxyl group, reportedly making it more reactive, faster-curing, and less toxic than straight phenol. While most of its materials are currently used in hand lay-up and spray-up, Indspec is developing special grades for filament winding and pultrusion.

Indspec also plans to offer resorcinol derivatives that will behave like engineering plastics at higher temperatures. To process these materials, two powders will be mixed together and melted during processing.

* A new patented process for phenolic pultrusion is available for licensing from Weyerhauser Co., the building-materials giant in Tacoma, Wash., using a non-acid-curing resin from Plastics Engineering Co. (Plenco), Sheboygan, Wis. (PT, April '89, p. 25).

* New resins for phenolic pultrusion, filament winding, spray-up, and SMC are being introduced in North America by Norold Composites, Mississauga, Ontario, a new joint venture of Reichhold Ltd. of Canada and Norsolor S.A. of France (PT, Feb. '90, p. 14).

* Durez Molding Compound Div. of Occidental Chemical Corp., N. Tonawanda, N.Y., has also been pursuing phenolic pultrusion and SMC with new non-acid-cure formulations for several years, but so far with limited success (PT, March '86, p. 75; March '87, p. 90). Use of phenolic SMC in aircraft/aerospace remains small-scale, owing to lack of a satisfactory surface finish for interior applications.

Meanwhile, Occidental/Durez expects to develop granular phenolic compounds with glass loadings higher than 60% in the near future. The company is especially hopeful about the prospects for its high-temperature molding (HTM) or "hot-cone" process, in which a molding compound is injected over a heated cone, producing material temperatures as high as 500 F as the material enters the mold cavity (PT, May '81, p. 67; Sept. '85, p. 19; June '86, p. 12). This reportedly increases flow, drives out volatile gases, and accelerates cure to the point where phenolics can beat theirmoplastic molding cycles for thick parts.

Thomas Quealy, Durez business manager, says the HTM process is currently being used by about 40 processors and has reduced cycle times and scrap rates by as much as a third compared with regular injection molding. He also says that HTM leads to improved physical properties such as tensile and flex strength, low porosity, and good dimensional stability.

* Soon-to-be available mineral-and-glass reinforced phenolic molding compounds from ICI Fiberite Molding Materials, Winona, Minn., use a proprietary thermoplastic alloying agent to increase toughness (PT, April '90, p. 23).

* Allied-Signal Inc., Morristown, N.J., is developing new phenolic-triazine (PT) thermoset resins which have the basic structure of phenolics, but produce no volatiles during cure, and require no catalyst (PT, March '90, p. 105). Expected to compete with polyimides for aircraft interior and structure applications, PT resins boast the flame retardance of phenolics and the performance of polyimides. As PT resins are developed, they are expected to cost less and process easier than polyimides.

In the meantime, says Harvey Dailey, group leader of Indspec's High Performance Resins Dept., "The real improvement in properties of phenolic laminates in the 1990s isn't as much as resin question as it is a glass reinforcement question." Sizings used on glass were developed mainly for polyesters, epoxies, and vinyl esters. Phenolic producers such as Indspec have cooperative programs with glass manufacturers for developing more phenolic-compatible binders. Nearly all the major glass-fiber companies are involved in such efforts, especially for filament winding and pultrusion.

"When they are fully commercial, which I would anticipate within a three- to five-year period, I think you'll see a major improvement in phenolic properties because of the new surfacing agents," Dailey said. Possibilities include using aminosilane as opposed to current binders made with bisphenol-A, starch, polyester, or urea.

MORE PROCESSABLE EPOXIES

A key development during the 1980s was the increased availability of epoxy resins in narrow viscosity ranges. Over the last 10 years, this has led to a maturation of the epoxy market, and since registration of brand-new chamistries under EPA's ToSCA rules can be extremely difficult and costly, suppliers may continue to experiment with current chemistries that have yet to be exploited fully. so, while we will probably not see as many new polymer backbones in the next decade as we saw in the last, you can expect greater fracture toughness and higher elongations in finished parts. Because of health concerns, suppliers also expect a continuing move away from hardeners that contain MDA.

"Rather than seeing brand-new epoxy backbones, we are going to see more modifiers that provide increased toughness, lower water absorption, and lower viscosities," predicts Michale Vallance, director of matrix and electrical power research & development for the Plastics Div. of Ciba-Geigy Corp., Ardsley, N.Y. Vallance notes the possibility of using two or more different epoxies in one material system--such as one to provide toughness while the other provides low viscosily.

Ciba-Geigy will commercialize a new epoxy for the commercial aircraft market later this year. The material has lower viscosity than any of their current high-performance epoxy products. It is the first tetra-functional resin to provide room-temperature viscosity comparable to lower-performance bisphenol-A epoxies, according to Vallance.

As for toughness, Vallance says to look for innovations such as rubber-coated carbon fibers and impact-modified thermoplastics, all within a thermoset matrix.

Research on alloying epoxies with urethanes and thermoplastics is underway at Dow Chemical. According to Fred Corson, v.p. of research and development, the aim of the research is to develop a technology base for toughened thermoset materials. As part of a NASA program to commercialize advanced materials, Dow is also developing toughened epoxies for RTM of commercial aircraft parts.

Dow is also experimenting with unconventional rubber modification of its thermoset materials, according to Corson. although he declined to give details, the work involves tying various elastomers into a thermoset matrix without diminishing thermal properties, unlike traditional rubber blends, which have led to reduced heat resistance.

Anthony Schroer, business manager for resins at Shell Chemical Co., Houston, expects improved flexibility and toughness to result from Shell's epoxy/elastomer hybrid products (PT, July '89, p. 27). He says Shell is also trying to develop new epoxy backbones to gain more toughness without using hybrids or multi-phase approaches.

POLYIMIDE DIRECTIONS

As with many other materials, suppliers are hesitant to predict revolutionary changes in polyimides. New catalysts may bring about shorter curing times, but it appears that working temperatures will remain high in order to produce materials with high-temperature performance. On the other hand, some suppliers are willing to trim back some of the performance properties of their resins to reduce cost, and others are finding ways to make them usable with lower-cost processing methods.

One of the more novel developments in these resins of late has been the appearance of chopped-fiber molding compounds that can be processed by fairly conventional compression and transfer molding. Dexter Composites, Cleveland, is the first to have introduced compounds of PMR-15 resin in granular form, preform slugs, and even SMC (PT, Feb. '87, p. 37; Sept. '89, p. 13; March '90, p. 53).

Dexter is now working on a new PMR-15 SMC that's expected to cost half as much ($30-35/lb) as the current product ($60-65/lb) as use increases. PMR-15 products with increased continuous-temperature service up to 700 F may also become available, but would cost as much as $500/lb, according to Bernard Nowak, sales manager.

Ciba-Geigy recently unveiled a new class of polyimides, called allylnadicimides, that eases processability by being soluble in common solvents and curing at epoxy-like conditions, while still providing cured [T.sub.g.'s] over 572 F and good mechanical proerties (PT, June '87, p. 48). The company expects its conventional condensation polyimides to remain in use as a toughening additive for other resins, as they do not see great prospects for decreasing the high melt viscosity of these polyimides.

Because of the need for good dielectric performance, advanced circuit board applications are the focus of development in addition-type polyimides at National Starch and Chemical Corp., Bridgewater, N.J. According to Robert Rossi, technical director for electronic materials and adhesives, multilayer, multi-chip modules are a particular target. He notes a trend toward using photoimageable polyimides, which would eliminate the need for a photo-resist material and etching process in manufacturing these modules. Development work seeks to build photo-reactivity into the polyimide backbone, rather than relying on polymer groups that disappear during cure. According to Rossi, a workable photoimageable polyimide could reduce the number of processing steps for multi-chip modules from 11 to five and bring a typical production time of four weeks down to 1-1/2 weeks. "We should have photoimageable polyimides in under five years," Rossi says.

Other targets of development at National include:

* A low-stress polyimide to match a circuit-board substrate's low coefficient of thermal expansion. The matching will avoid residual stresses between the substrate and layers of polyimide, which can cause cracking or delamination. This material will be available later this year, according to Rossi.

* Stress-absorbing modifiers that may be mixed in with the polyimides should be available within 5 years. This type of blending would build a low-modulus segment into polyimides, which are typically brittle.

* A lower-cost material with a dielectric constant in the range of 2-2.5. Currently, this level of dielectic performance can be achieved only with expensive monomers, but National feels the price must be driven down to develop the market. Polyfluorinated polyimides with dielectric constants as low as 2.4 currently could cost as much as $2500/lb dry weight.

BMI MATERIALS MATURING

When they were first introduced, bismaleimides (BMIs) were somewhat difficult to process, since the resin had to react with itself to produce a cured part. Since then, two-component systems that use curing agents such as diallyl bisphenol have allowed easier processing. As processors and end users have become more familiar with these materials, they also have more clearly indicated what they like and don't like about BMIs.

For instance, Ciba-Geigy says it may have built too much temperature capability into its BMIs for some applications. "We may be able to take some properties back and exchange them for others, such as providing better toughness in exchange for higher viscosity, or providing both better toughness and low viscosity if the customer can tolerate a lower glass-transition temperature," says Michael Vallance. "As the market begins to have a clearer opinion of what's needed, it becomes easier for us to supply what our customers want."

Likewise, Herman Mihalich, business manager for Advanced Materials at Rhone-Poulenc Inc., Princeton, N.J., says that "customers are telling us, 'We don't really need 250 C [heat resistance], how about 200-220 C?' And that goes not only for aerospace but also for electronics applications. By taking some temperature performance out of the BMI, we can provide improved toughness, lower viscosities, shorter curing times, or a reduced price."

Trends in BMI development will be very similar to those in epoxies, but with more emphasis on processing improvements, according to Ciba-Geigy's Vallance. One challenge will be to maintain the resins' temperature capability and simultaneously improve toughness. Ciba-Geigy has been experimenting with a range of modifiers to provide higher glass-transition temperatures, lower water absorption, greater thermo-oxidative stability, extended working time or pot life at standard processing temperatures, and optimized melt viscosity and reactivity.

Ciba-Geigy is using reactive diluents to develop a BMI system for RTM and filament winding, and is adding BMIs to its allylnadic-imide system to produce a tougher material for making aerospace prepregs and structural adhesives. The company says that the new materials may be commercialized by the end of the year.

Rhone-Poulenc has just begun offering developmental quantities of a melt-processable, non-MDA-curing BMI material. Previously, all of the company's BMI materials required the use of a solvent. Another developmental melt-processable product is a toughened BMI that contains a second phase of high-temperature silicone rubber. R-P plans to offer a range of new melt-processable products, including low-viscosity materials for RTM, filament winding, and pultrusion.

CYANATE ESTERS & BCB RESINS

The dielectric performance of cyanate ester resins figures to drive their use in applications such as multi-chip modules and radomes in the next decade. They may also see increased use as an additive, to work as a curing agent and provide dielectric performance to a material such as an epoxy, while maintaining the epoxy's corrosion resistance.

"We, our customers, and other suppliers have only begun to scratch the surface of cyanate esters' potential," according to Jack Christenson, business manager for Electronics and Composites at Hi-Tek Polymers, Inc., Louisville, Ky., which was recently acquired by Rhone-Poulenc. "Rhone-Poulenc is interested in combinations of cyanate esters and BMIs, and Hi-Tek has already looked at combinations of cyanate esters with epoxies. We also have a patent on blending thermoplastics with cyanate esters for toughness, a route that many structural composite prepreggers are now taking." New versions of polysulfone, polyethersulfone, polyetherimide, and polyarylate, are being developed by some thermoplastic suppliers with the aim of making them more compatible with cyanate esters.

Hi-Tek is developing new cyanate esters for processes such as RTM, filament winding, and pultrusion. Desired properties of future materials include better dielectric performance, low melt viscosity, and low moisture absorption. New catalysts will provide greater tailoring of gel time, pot-life latency, and curing rates, says Christenson. He notes that Hi-Tek cyanate esters are already available with room-temperature viscosities as low as 100 cp. (See also accompanying SAMPE news report).

Like BMIs, the temperature performance of cyanate esters may not be pushed as much in the next decade as it was in the last. "We've had just as many requests for materials with slightly lower glass-transition temperatures as for higher ones," according to Christenson. Pricing of the materials is expected to decrease in the next decade as production increases.

Dow is working on various cyanate ester resins for NASA's Langley Research Center (PT, March '87, p. 62; June '87, p. 50; Aug. '88, p. 33; May '89, p. 35; July '89, p. 25; April '90, p. 45). Dow is developing grades for RTM of commercial aircraft parts with hot/wet service temperatures in the range of 200-250 F, and for prepregging aimed at higher performance applications with hot/wet service temperatures significantly above 350 F. "We are shooting for high performance in terms of both temperature and damage tolerance," says Ed Woo, senior associate scientist for Dow's Central Research Group.

Another arm of the NASA program seeks to develop polymers based on the benzocyclobutene (BCB) monomer and represents Dow's newest thrust into the high-performance thermoset area (PT, Jan. '90, p. 13; April '90, p. 45). BCB materials are said to have high glass-transition temperatures, long processing latency, and low moisture absorption. Woo hopes that the program will produce a competitive material within the next couple of years for evaluation by the aircraft industry.