Three suppliers share the latest in coldbox binder systems: at this year's AFS Casting Congress, foundry binder suppliers discussed their most recent coldbox binder innovations, detailing the results of casting trials with the new systems.

The term "coldbox" originally was used only for the phenolic urethane process, but now it designates other gas-catalyzed binder systems such as C[O.sub.2]-cured sodium silicate, furan and S[O.sub.2], acrylic and acrylic-epoxy S[O.sub.2] systems, phenolic ester-cured and C[O.sub.2]-cured alkaline phenolic technologies.

In all coldbox systems, a binder-sand mixture is cured at room temperature, and that cure is accelerated or initiated by a vapor, gas catalyst or co-reactant passed through the sand mix. Some examples of catalysts include the triethyl or dimethylethylamines for phenolic urethanes, S[O.sub.2] for furan and acrylic-epoxy resins, methyl formate for ester-cured alkaline phenolics, and C[O.sub.2] for silicate and alkaline phenolic binders.

Over the years, coldbox technologies have benefited from a number of advancements since the first phenolic urethane coldbox (PUCB) process was introduced to the foundry industry by Ashland Chemical at the 1968 Casting Congress held in Cleveland. Coldbox systems represented a great leap forward for foundry-men, who were now able to boost production of cores with good tensile strength, hot strength and dimensional accuracy.

However, for all the benefits of coldbox systems, there are a few potential drawbacks, including environmental and productivity issues. Coldbox systems may exhibit a short mixed bench life, the tendency [TABULAR DATA FOR TABLE 2 OMITTED] to stick to tooling surfaces and the emission of unacceptable levels of VOCs or HAPs.

Table 1. Methyl Ester Physical Properties

Characteristic                Typical Value

Appearance                    Yellow liquid
Moisture                        0.1% max
Viscosity @ 40C, cSt              4.6
Flash Point                      190C
Density (lb/gal)                 7.30

As suppliers have used their experience and technical know-how to formulate optimal coldbox systems that overcome these problems, foundries have used these new compositions to produce better quality castings. The following article was compiled from presentations made at the 1999 AFS Casting Congress and details some of the latest advancements in coldbox systems.

In their presentation on PUCB binder technology, D.M. Trinowski, Delta-HA, Inc., Detroit, Michigan, and G. Ladegourdie and K. Lochte, Huttenes-Albertus Chemische Werke Gmbh, Dusseldorf, Germany, discuss how the Technikure Z system offers better bench life, greater humidity resistance, resistance to degradation from aqueous core coatings and improved release from foundry tooling.

Also in the PUCB arena, R.A. Laitar and S.R. Iyer, Borden Chemical, Inc., Forest Park, Illinois, shared the results of trials with Borden's newest PUCB system for aluminum permanent mold applications. This system offers improved shakeout (lower hot strength), reduced smoke and VOC emissions and higher tensile strength.

Ashland Specialty Chemical Co., Columbus, Ohio, took the same opportunity to offer information on its newest coldbox system, an amine-cured resin. D.G. Hendershot, E.G. Toplikar, M.A. Neu, B.L. Semler, J.J. Archibald, Ashland Specialty Chemical Co., Columbus, Ohio and K.R. Crawford, Tonkawa Foundry, Inc., Tonkawa, Oklahoma, discussed how this system, which compares favorably with PUCB from strength and productivity standpoints, offers easier equipment clean-up and a substantial improvement to mixed sand benchlife.

This article only skims the surface of what is available or achievable, but by taking a glimpse at the newest systems, foundrymen can get a taste of what binder technology may offer in the years to come.

PHENOLIC URETHANE SYSTEMS

All phenolic urethanes are three-part systems consisting of a modified phenolic resin Part I, polymeric isocyanate Part II, and tertiary amine catalyst Part III. The "active" ingredients in a PUCB resin system consist of a polyol and a polyisocyanate. Both the phenolic resin component and the polymeric isocyanate are traditionally dissolved in organic solvents, due to the very high viscosity of both materials. However, the difference in polarity of the polyisocyanate and phenolic resin limits the choice of appropriate solvents that are compatible with both components. This compatibility is necessary to achieve complete reaction and curing of the binder, as well as the speed of cure.

In its infancy, PUCB suffered from limitations, such as short mixed sand bench life, sticking of cores to tooling surfaces, sensitivity to different sand types and process sensitivity due to foundry environmental variations.

Today, phenolic urethanes account for the largest market share of any coldbox (more than 85%) system in use. PUCB sand is reclaimable and associated with high productivity.

                          Benzene lb/ton iron     Voc lb/ton iron

Conventional cold-box           0.055                  0.27
New cold-box                    0.52                   0.22
% reduction                     5.45                  18.5

Delta's Technikure Z System

The aromatic solvents in PUCB binders have created disadvantages, due to emissions in production and storage of cores and molds. The non-polar aromatic solvents are rich in both VOCs and HAPs, both of which are carefully regulated. At the high temperatures encountered during the casting operation, the binder components are subjected to a pyrolysis process involving creation of new, stable compounds. In the presence of aromatic hydrocarbons, this pyrolysis process generates benzene, toluene and xylenes, which exhibit particularly great thermal stability.

Delta-HA's Technikure Z uses plant-based solvents (methyl esters of vegetable oils) for the resins and co-reactants instead of high-boiling aromatic hydrocarbons. Aside from the ecological advantages of these environmentally friendly and C[O.sub.2]-neutral natural products, the new solvents meet all physical requirements for PUCB systems. They are high-boiling, sufficiently low in viscosity, odorless and are classified as innocuous. They are furthermore nonflammable, a property that considerably simplifies transportation and storage of the resin and co-reactant solutions prepared from them. Table 1 (see p. 45) shows typical physical properties of a methyl ester solvent.

VOCs and Emissions

Typical VOC content for traditional coldbox systems range from 40-50% in the Part 1 and from 20-30% in Part II. Table 2 (see p. 45) shows a VOC content comparison between a conventional PUCB system and the new methyl ester technology. The methyl ester technology results in a VOC reduction of more than 50%.

In tests conducted in Europe using German PUCB formulations, remarkable differences have been seen in benzene emissions. The workplace exposure of the employee at the shakeout station is lower when the new system is in use.

Further, the environmental burden is reduced. This has been demonstrated by measurements made in the hoodstack of a foundry. The levels at the shakeout station and in the hoodstack, through which the exhaust air leaves the foundry, are listed. These emission measurements led to identical results in a number of foundries: the benzene, toluene and xylene emissions were decreased by 25-50%, depending on the operating conditions, with the new methyl ester-based formulas.

Reduction in benzene emissions has not yet been as well documented in the U.S. Limited hoodstack testing conducted at a Midwestern ductile iron foundry has shown a modest 5% reduction in benzene and an 18% reduction in VOCs, as shown in Table 3.

Strength Development

Figure 1 shows the tensile strength development provided by the new coldbox system in comparison to those in a standard system. In general, out-of-box strengths are similar to those of standard systems, with the exception of immediate and early strengths. However, the somewhat lower initial strengths developed with the methyl ester systems are compensated for by easier release from the tooling. No deleterious effects from this phenomenon have been observed in actual foundry practice. Furthermore, improvements in resistance to degradation due to exposure to high humidity and aqueous core coatings are seen with the new system.

Improved Release from Tooling

During the mixing operation, the sand grains are coated with a film of binder consisting of resin and co-reactant. During the blow cycle, when the corebox is filled, the sand grain strikes the surface of the corebox at high speed. This impingement against the surface of the corebox causes the binder layers to spall off the grain and deposit on the corebox. In conventional coldbox binders, attempts have been made to diminish this effect by including release agents in the resin and activator formulations. Vegetable oil methyl esters provide for excellent release of the core from the tooling. A lower fraction of core breakage, longer runs and reduced consumption of external release agents result.

Improved Cure Efficiency

An unexpected benefit of the new solvent technology was improved cure efficiency. Methyl esters act to increase the transport of amine gas throughout the core faster than aromatic solvents, a phenomenon referred to as bulk cure. Foundry results consistently have shown cycle time reductions of 5% or more due to this effect, resulting in greater core production.

Thermal Characteristics

PUCB systems exhibit a markedly stronger tendency to soften during pouring than do hotbox and shell-molded components. The results are erosion, increased finning, scabbing and inaccurate casting dimensions. In the new coldbox systems, the vegetable oil methyl esters polymerize when enough heat is present. This polymerization of the solvent stabilizes the polyurethane links and increases the strength levels in the high temperature range. The tendency for finning is markedly reduced, and the dimensional accuracy of the castings is good.

Gas Development

Gas in core and mold sands can lead to defective castings, for example at high binder levels when gas evolution occurs at an early stage or when the silica sand exhibits low gas permeability. In the new coldbox systems, the powerful initial gas burst is weakened and its time of occurrence shifted slightly.

A conventional system also was compared to the new coldbox system where the test cores were exposed to the dipping test in an uncoated condition. The differences are even greater in the case of coated cores. A coating with a high gas permeability permits the gases developed in the core to escape in front of the column of molten metal. Typical defects (gas bubbles) are minimized by this coating behavior.

The results of an otherwise identical test using a coating exhibiting low gas permeability show that the conventional coldbox system quickly develops a high gas pressure in the core. The new coldbox system features a slower rise in the gas pressure, which reaches its peak at a time when the initial casting skin already has formed.

Borden's Sigma Cure for AI

In discussing the evolution of urethane coldbox binder systems for aluminum applications, Borden Chemical detailed its newest system for permanent mold aluminum applications, Sigma Cure PM14 and PM25. The company also introduced its 7121 Phenolic Urethane Part 1 Resin and 7516 Isocyanate Part 2 Resin for ferrous applications at this year's CastExpo. Both binder products offer less smoke and odor, higher tensile strength, better core storage life, and better resistance to degradation of strength when using water-based core coatings.

Aluminum Systems

The first coldbox system introduced for aluminum applications employed modified Part 2 resins, designed to breakdown faster than those products used for ferrous applications. A lower molecular weight, Part 2 resin was developed with the objective of producing a polyurethane binder with lower hot strength. The cores produced using the modified Part 2 resin show shorter collapse times than those made using a system designed for ferrous applications. Shorter failure times indicate lower hot strength and suggest improved shakeout. The lower hot strength was not obtained at the expense of tensile strength or bench life properties of the system. These properties are comparable for both systems.

Reduced Smoke

The production of aluminum castings in permanent molds has a unique set of binder requirements. Because the castings cool much faster, the cores see less heat to promote breakdown of the binder system. Also, permanent molds are typically run with more operator interaction in core setting, casting removal and, occasionally, metal pouring. In addition to causing problems for personnel operating the permanent mold, high levels of smoke are associated with the buildup of carbonaceous residues on the permanent mold. These residues can plug vents and affect casting finish.

Table 4. Typical Lab Tensile Performance Ranges of Isomax vs.
Conventional System

Tensile Measurement (time)          New System (psi)      PUCB (psi)

Immediate (30 sec)                       150-230           140-240
5 min                                    230-400           220-350
1 hr                                     260-450           250-380
24 hr                                    300-550           300-450
24 hr @ 90% RH                           130-290           100-300
Hot core wash                             80-150            50-120
Cold core wash                           200-500           200-450

Sand type: 55 AFS silica sand
Catalyst: Dimethylethylamine (DMEA)
Binder level: 1.0-1.5% based on sand

This problem was studied in the laboratory to determine the factors contributing to smoke generation. The test conditions consist of preparing tensile dog bones from the binder system being studied and applying heat in the form of a Bunsen burner flame. The heat is applied for 30 sec and then removed. The cores are then observed for intensity and duration of smoke evolution. The permanent mold core produced lower smoke intensity, slowing faster and ending sooner than the conventional system. Productivity is improved by reducing the downtime encountered to clean residues from the permanent molds. Tensile and bench life properties are comparable to conventionally formulated systems. A particular advantage of the permanent mold binder system is a reduction in the gas evolution of the core [ILLUSTRATION FOR FIGURE 2 OMITTED].

Reduced VOC Emissions

The LOI and resultant emissions of a low-VOC system were compared to a conventional system over a 24-hr period. The low-VOC systems lost less solvent, over time, than did conventional systems, resulting in the low emissions observed. Again, the reduction of emissions was accomplished with minimal loss of sand performance. The immediate strengths of the low-VOC aluminum coldbox system tended to be higher than those of conventional systems, while the humidity resistance tends to be somewhat lower.

High Tensile/Better Shakeout

When compared at the same binder level, the new system develops higher tensile strengths than conventional systems [ILLUSTRATION FOR FIGURE 3 OMITTED]. The bench life properties also are improved [ILLUSTRATION FOR FIGURE 4 OMITTED]. The hot strength of the high-tensile system was found to be lower than conventional systems using both the dilatometer and BCIRA hot distortion. While the lab results suggested the shakeout properties would be improved, confirmation was carried out using a test casting.

The new system shows greater amounts of sand removed at each of the test times. The improved shakeout was obtained at the same binder level. This new binder system is particularly suited for foundries that use a core pack without a mold and vacuum and/or pressure-assisted pouring. In these applications, core integrity is critical and higher core strengths are needed for handling and pouring purposes.

AMINE-CURED SYSTEM

Ashland's Isomax System

As the inventor of PUCB binders, Ashland Chemical continues to explore improved binder formulations including PUCB products for high productivity, environmental awareness and quality in the ferrous casting market. Other binder products reduce core removal efforts and provide environmental advantages for the aluminum casting market. Ashland also has developed an amine catalyst recovery program.

In the interest of offering new opportunities to improve productivity and reduce emissions in the coldbox core-making arena, Ashland introduced a new amine-cured binder technology at CastExpo '99. Tested in production at Oklahoma's Tonkawa Foundry, Isomax offers extended mixed sand bench life (2-20 times that of PUCB), consistent sand flowability as a function of mix age, no hardening or crusting of mixed sand to hoppers, blow heads and mixing equipment, reduced VOCs on mixing, and reduced susceptibility to veining.

Chemistry and Performance

This new binder is comprised of numerous components packaged as a two-part system. The number of constituents within the binder system contributes to multiple curing mechanisms that occur upon catalysis. The cure chemistry of the new system involves the reaction of a specialty polymer with an initiator in the presence of a tertiary amine catalyst, such as triethylamine (TEA), dimethylethylamine (DMEA) or dimethylpropylamine (DMPA).

Testing has been performed on a variety of sand types with AFS grain fineness numbers ranging from 40-90. Results indicate that the new system is suitable for use with a variety of sands, including silica, lake, specialty, and thermally and mechanically reclaimed. Typical sand tensile performance for the system on silica sand is listed in Table 4 (see p. 47).

The tensile performance/profile can be tailored to meet specific applications. Furthermore, the system has built-in humidity resistance, which offers outstanding core shelf-life with little loss of tensile or scratch hardness.

In addition, the new system affords excellent dip-redry performance where the use of a water-based refractory coating is required. Typical binder levels for iron applications are in the 1-1.7% range, based on sand. Table 5 shows laboratory tensile data, with respect to mixed sand bench life over 24 hr. The new resin actually improves in performance within the first 8-24 hr of mix age. This is indicated by the substantial increase in immediate tensiles for aged sand versus those obtained for freshly mixed sand. Along with maintaining acceptable tensile performance for extended periods of time, cores produced with aged sand show little weight loss or loss of definition versus those produced with fresh sand.

The consistent core density, as a function of mix age, is indicative of excellent flowability and extended bench life. Additionally, the new resin system is tolerant of high sand temperatures ranging from 120-150F (49-66C) without detriment to cure or core quality. Bench life of several hours under these harsh conditions has been observed. Even when a sand mix lacks sufficient strength development, the mix remains highly flowable and takes on a "dried-out" appearance.

Resin Mixing, Process Considerations

The new resin employs standard process equipment and catalysts currently used in the foundry industry for use with PUCB technology. Acceptable sand mixes can be produced with any style of mixer. Because the new system operates at Part I:Part II ratios of 15:85-35:65, proper distribution of the minor component (Part I) is crucial for achieving a suitable mix that yields high-quality, cured specimens. Both Part I and Part II are low-viscosity liquids [(less than 200 centipoise at 70F (21C)] and are capable of being delivered with conventional pumping technology.

In terms of curing aspects, production cycle times with the new resin are equivalent to those achievable with PUCB. The major difference in curing the new system involves an extended retention time for the tertiary amine catalyst. This is required to properly activate the curing mechanism, which is dependent upon catalyst concentration. Early process testing indicated that excessive amounts of catalyst were required to cure the system. However, through the use of restricted venting, improved tooling seals and reduction of gassing/purge pressures, amine consumption is only about 1.5 times that of current PUCB technology. In light of the increased amine usage, one would expect increased amine odors retained in cured NRS specimens. However, the retention of the amine catalyst odor [TABULAR DATA FOR TABLE 6 OMITTED] in cured specimens is virtually undetectable within minutes.

Table 5.24-hr Lab Bench Life Tensile Performance Ranges of Isomax

Tensile Measurement (time)        New System (psi)        PUCB(psi)

Immediate (30 sec)                     150-290              0-80
5 min                                  170-310              0-120
1 hr                                   190-340              0-140
24 hr                                  240-380              0-160
24 hr @ 90% RH                         100-230              0-80
Hot core wash                           60-110              0-40
Cold core wash                         200-350              0-130

Sand type: 55 AFS silica sand
Catalyst: Dimethylethylamine (DMEA)
Blinder level: 1.0-1.5% based on sand

Casting Performance

The casting attributes of the new resin have been examined through gray iron test castings (stepcone, 2x2 penetration and erosion wedges). Table 6 lists the results against standard PUCB technology, with and without a water-based graphite refractory coating. The new system offers casting attributes similar to PUCB when a refractory coating is employed with a reduced susceptibility toward veining. But, because the new system offers improved thermal breakdown, it can be susceptible to erosion in ferrous alloy applications where severe gating and/or no refractory coating or sand additive is employed.

Field Applications

For in-field testing, the binder was taken to Tonkawa Foundry, a small Ni-Resist facility. With resin splits ranging between 15:85-35:65 (Part I to Part II) and resin levels at 0.6% to 0.8%, based on sand, batch mixing was found to be most efficient. However, with accurate pumps and controls, continuous mixing was able to accommodate the process. Existing core blowers and gas generation units were used with no modifications.

Additional findings in the curing mechanics revealed that little or no purge air was required. Further, the use of heated purge air was of no real benefit for the system. The only advantage heated air provided was to ensure delivery of a true amine vapor when using an injector-style generator. Venting modifications to ensure a plenum back pressure of 5-8 psi during gassing provided for optimum curing. Additionally, a short dwell time after gassing allowed for a thorough reaction of the amine with the resin. Even though gas consumption was somewhat higher, the overall curing cycle was either the same or reduced as in PUCB systems. Specifically, cores/molds, ranging from 0.5-75 lb, employed cure times from 2-10 sec.

With the flowability and additional immediate strength generated with the new system, core and/or mold release may be adversely affected. A silicone release agent, under standard operating conditions, resolves this issue. In addition, the new resin is similar to other PUCB processes when comparing required binder level and cure time of chunky cores versus thin, lacy cores. The new system will not react with air or moisture in an open environment, which reduces mixer core blower and tooling cleaning. At the pouring line, the system exhibited a reduction in smoke, gas and VOC emission. Casting validation was evaluated on several different jobs, ranging from thin-walled impellers to deep draw sweeped-vane diffusers used in submersible pump applications for the petrochemical industry. Castings poured were produced in Ni-Resist Type I (a high-Ni alloy) poured between 2675-2800F (1468-1538C) with sand-to-metal ratios between 4:1-6:1 with both cores and molds made with the new resin. However, because of the reduced hot strength associated with the present chemistry of this technology, erosion is a possibility. Care should be given in gating practices so as to avoid severe impingement of high metal velocity on cored/mold surfaces. Additional casting validation also was obtained in a high-chrome, ferrous, wear-resistant alloy. In all applications (cores and molds), typical casting finish as well as dimensional stability is good.