Final Report | Carbon Dioxide Soluble Binders for Environmentally Benign Metal Forming Operations| Research Project Database | NCER | ORD

Final Report: Carbon Dioxide Soluble Binders for Environmentally Benign Metal Forming Operations

EPA Grant Number: R831504
Title: Carbon Dioxide Soluble Binders for Environmentally Benign Metal Forming Operations
Investigators: Manke, Charles W. , Enick, Robert , da Rocha, Sandro R.P.
Institution: Wayne State University , University of Pittsburgh - Main Campus
EPA Project Officer: Bauer, Diana
Project Period: January 1, 2004 through December 31, 2006
Project Amount: $349,967
RFA: Technology for a Sustainable Environment (2003)
Research Category: Pollution Prevention/Sustainable Development

Description:

Objective:

This project investigates the use of two classes of low molecular weight hydrocarbons—sugar acetates and tert-butylated aromatics—as carbon dioxide-soluble binder materials for metal forming operations, such as sand casting and powder injection molding. The use of carbon dioxide-soluble binders can potentially facilitate new, environmentally benign processes for extraction and recovery of binder materials and casting sand, thereby alleviating serious environmental problems associated with conventional debinding processes, including emission of hazardous air pollutants and volatile organic compounds, and production of waste materials. Two fundamental requirements for carbon dioxide-extractable binder materials are that they must be highly soluble in carbon dioxide at reasonable temperatures and pressures, and that the rate of dissolution of the binder in carbon dioxide must be rapid. Our study performed pressure-volume-temperature (PVT) measurements, and sand plug dissolution experiments on candidate binder materials, and we have identified three materials that meet these basic requirements, namely β-D galactose pentaacetate, tri-tert- butylbenzene, and tri-tert-butylphenol, whose structures are shown below, in Figure 1. We also found that a low-cost commercial polymer, polyvinyl acetate, was highly soluble in dense carbon dioxide, but would require much higher pressures for CO₂ extraction than the compounds listed above.

Figure 1. Structural Formula of a) b-D Galactose Pentaacetate, b) 1,3,5-Tri-tert-butylbenzene, c) 2,4,6-Tri-tert-butylphenol

Introduction

The three main objectives of the project were:

Determine the Equilibrium and Transport Properties of the Sugar Acetate Binder–CO₂ System. Detailed PVT measurements were performed on the β-D galactose pentaacetate–CO₂ binary system, the tri-tert-butylbenzene–CO₂ binary system, and the tri-tert-butylphenol–CO₂ binary system. These measurements were performed to evaluate the solubility of these binder candidates in dense carbon dioxide, and to elucidate the phase behavior of the systems at temperatures and pressures representative of binder extraction processes. Dissolution experiments were performed by binding sand into small (1 cm³) plugs with each of the binder materials listed above, and then exposing the sand plugs to dense carbon dioxide in a visible cell to determine the time required for complete extraction of binder from the sand. These experiments were performed in the laboratory of Professor Manke at Wayne State University and in the laboratory of Professor Enick at the University of Pittsburgh.
Determine the Wetting and Interfacial Properties of the Sugar-Based Additive–Solid–CO₂ Interface. The sand plug dissolution experiments described above show that drainage of CO₂ swollen binder liquids from the sand plugs is an important mechanism for binder extraction. Interfacial and wetting properties of the CO₂ swollen binder liquid play an important role in this process. These properties were measured in the laboratory of Professor da Rocha at Wayne State University.
Develop Polymer and Surfactant Additives Required to Improve Mechanical Properties of the Mold, Template, or Green Part. The solubility of polyvinyl acetate and other polymers in carbon dioxide, pure and in mixtures with β-D galactose pentaacetate, were investigated in Professor Enick’s laboratory at the University of Pittsburgh.

Summary/Accomplishments (Outputs/Outcomes):

Solubility and Phase Behavior of the β-D Galactose Pentaacetate–CO₂ Binary System

Figure 2 shows the phase envelope of the b-D galactose pentaacetate–carbon dioxide binary system measured by dew and bubble point experiments at various temperatures between 308 K and 323 K (Dilek, et al., 2006). The curves represent phase boundaries at constant temperature separating a two-phase system from a single-phase system. The region below each boundary is the two-phase region, where the system exists as a CO₂-rich upper fluid phase, and a sugar acetate-rich lower fluid phase in equilibrium. In the region above the phase boundary, the system exits as a single fluid phase. The maximum pressure required for the b-D galactose pentaacetate–carbon dioxide binary system to exist as a single phase is 10.3 MPa at 308 K, and this pressure increases to 15.7 MPa at 323 K.

Figure 2. Pressure-Concentration Plot of b-D Galactose Pentaacetate–CO₂ System Phase Envelopes at (◊) 308 K; (Δ) 313 K; (□) 318 K; (o) 323 K. Data are taken from (Dilek, et al., 2006).

These data show that the solubility of b-D galactose pentaacetate in carbon dioxide reaches 25–30 wt % in a single upper phase at pressures of temperatures of 308–323 K (95–122°F), and corresponding pressures of 10–16 MPa (1450–2300 PSI). These conditions are well within the range of processing equipment for supercritical fluid extraction.

Solubility and Phase Behavior of the Tri-Tert Butylbenzene (TTBB)–CO₂ Binary System

The phase behavior of the 1,3,5-tri-tert-butylbenzene–carbon dioxide binary system was also studied with dew and bubble point measurements in a high-pressure variable volume sapphire cell in our laboratories (Dilek, et al., in press, 2007). The phase envelope of the binary system at 328 K is shown in Figure 3. 1,3,5-tri-tert-butylbenzene liquefies in the presence of carbon dioxide at 2 MPa at 328 K, considerably below the 343 K melting point of pure TTBB. This melting point is represented with a solid–fluid–vapor equilibrium horizontal line, which also forms the lower boundary of the fluid–vapor binary phase equilibrium. The fluid–vapor region of the binary system exits between the fluid–vapor equilibrium boundary curve formed by dew and bubble points, and the solid–liquid–vapor equilibrium line. In this region, the binary system exists as a TTBB-rich lower phase and a carbon dioxide-rich upper phase in equilibrium. Above the fluid–vapor boundary, the binary system exits as a single phase. The phase envelope in Figure 3 shows that extraction of TTBB from a solid matrix can be conducted at pressures over 2 MPa (290 PSI) at a temperature of 328 K (131°F). Thus TTBB offers the potential for binder extraction at relatively low pressure, compared to b-D galactose pentaacetate. At present, however, TTBB is only available as a high-cost laboratory reagent.

Figure 3. Pressure-Concentration Plot of 1,3,5-Tri-tert-butylbenzene–CO₂ Binary System Phase Diagram at 328 K. (□) Dew points on the liquid–vapor equilibrium boundary; ( Symbol. ) near-critical points on the liquid–vapor equilibrium measured by the turbidity method; (■) bubble points on the liquid–vapor equilibrium boundary; (▲) solid–liquid equilibrium boundary; (●) solid–liquid–vapor equilibrium line. Data are taken from (Dilek, et al., in press, 2007).

Tri-tert-butyl Phenol (TTBP)–CO₂ System

The phase behavior of the TTBP–CO₂ system, measured at 328 K, is shown below in Figure 4. The phase behavior of this system is very similar to the TTBB–CO₂ phase diagram shown above. Although less CO₂-philic than TTBB, TTBP is still quite CO₂-soluble. Further, TTBP has been commercially available in extremely large volumes (e.g., railroad cars) at a cost of roughly $3/lb. An extensive study of the phase behavior of this compound in CO₂ has been completed at the University of Pittsburgh. Pressures slightly greater than 1000 psia and mild heating to 50–60oC are required to dissolve the TTBP-bound sand in several minutes.

Figure 4. Phase Envelope of 2,4,6-Tri-tert-butylphenol–Carbon Dioxide Binary System at 328 K.

Dissolution Rates of Binders in Carbon Dioxide

To investigate the disassociation of sand mold in carbon dioxide, cylindrical sand plugs were cast by melting the binder and mixing with sand. This molten sand–binder mixture was poured into a glass tube and the mixture was allowed to cool and solidify at ambient conditions. The solid sand plug was then removed from the glass mold and placed into the high-pressure, variable-volume visible cell. The cell was loaded with carbon dioxide and brought up to a temperature and pressure that were determined previously from preliminary binder solubility tests. Figure 5 shows the dissolution of b-D galactose pentaacetate from a sand plug in the high-pressure, variable-volume sapphire vessel. The composition of the sand plug was 50% sand and 50% b-D-galactose pentaacetate by weight and the cylinder was approximately 1 cm³. The disassociation took place at 314 K and 15.3 MPa with the observation of sand particles falling away under gravity. During the debinding, dripping of a low-viscous liquid from the cylinder was observed, which contributed to the removal of binder from the disintegrating sand plug. The complete disintegration of the bound sand cylinder took approximately 5 minutes, leaving a mass of unbound sand on the surface of the piston at the end of the experiment. This experiment shows that debinding of bound sand could be performed at reasonable temperatures and pressures with rapid dissolution of sugar acetate in supercritical CO₂. When setting a similar b-D-galactose pentaacetate-bound sand mold in chloroform, it was observed that approximately 20 minutes are required to completely disassociate the sand mold.

Figure 5. Pictures of Dissolution of Binder b-D Galactose Pentaacetate From 1-cm³ Sand Plug With Supercritical Carbon Dioxide at 313 K and 15.3 MPa.

The same procedure was followed for extraction of other binders from sand plugs. All the cylindrical sand plugs were approximately 1 cm³. The binders were extracted with high-pressure carbon dioxide and organic solvents separately for method comparison. Disassociation data are presented in Table 1.

Table 1. Comparison of Sand Plug Debinding Experiments

Binder	Extraction substance	Temperature	Pressure	Sand plug composition (wt % Binder)	Time of dissolution
b-D galactose pentaacetate	Supercritical CO₂	313 K	15.3 MPa	50 wt %	≈ 5 min
b-D galactose pentaacetate	Chloroform	298 K	Atmospheric	50 wt %	≈ 20 min
1,3,5-tri-tert-butylbenzene	Near-critical CO₂	301 K	7.7 MPa	20 wt %	≈ 40 sec
1,3,5-tri-tert-butylbenzene	Acetone	298 K	Atmospheric	20 wt %	≈ 7.5 min
2,4,6-tri-tert-butylphenol	Supercritical CO₂	338 K	17.3 MPa	18 wt %	≈ 4 min
2,4,6-tri-tert-butylphenol	Acetone	298 K	Atmospheric	20 wt %	≈ 7.5 min

When comparing the disassociation in supercritical or near-critical carbon dioxide with that in conventional organic solvents, such as acetone and chloroform, it was found that the disassociation in supercritical and near-critical carbon dioxide dissolution was much faster than that in organic solvent debinding. The rapid disassociation in supercritical carbon dioxide is due to the enhanced transport properties and low surface tension of carbon dioxide–binder systems at conditions near the critical point of pure carbon dioxide. Usually the diffusivity of supercritical fluid is approximately two to three orders of magnitude larger than that of an organic liquid in ambient conditions. Low surface tension and viscosity of carbon dioxide close to critical points enhances the penetration of the fluid into the pores of the sand–binder matrix, where it easily dissolves the binder and breaks down the sand mold.

Although quantitative measurements of the strength of these sand plugs were not made, it was readily apparent that the TTBP sand plug was much stronger than either the sugar acetate-bound plug or the TTBB plug. It was easy to handle, retained its shape, and was very difficult to break. The sand plug formed with TTBB, however, was much more brittle and could be easily broken. Although the sugar acetate-bound plug had more integrity than the TTBB, it was not nearly as rigid as the TTBP-bound plug. It is not surprising, therefore, that the TTBB plug was the easiest to dissolve in CO₂, and the TTBP was the most difficult.

Interfacial Properties of the CO₂–Sugar Acetate–Substrate Interface

Experimental Setup. The interfacial properties of the sugar acetate–CO₂ mixture were determined using a similar setup to that described above, with a few changes. The system is shown in Figure 6. The modifications were necessary since sugar acetate is solid at ambient conditions and also because is has very high CO₂ solubility. The sugar acetate phase, therefore, had to be pre-saturated with CO₂ in another variable-volume, high-pressure cell, called the drop phase cell, which is used to generate the pendant drop/drop for contact angle measurements.

Figure 6. Schematic Diagram of the Tandem High-Pressure Pendant Drop Tensiometer and Contact Angle Goniometer for the Sugar Acetate System.

Interfacial Tension. The tension/contact angle experiment consisted of adding a known amount of sugar acetate and CO₂ into the high-pressure vessel, and then bringing the system to a specified density. Based on the phase behavior of the CO₂–sugar acetate mixture, an excess amount of sugar acetate was added to the cell in an attempt to prevent the pendant drops formed at the tip of the capillary to be immediately solubilized by the CO₂ continuous phase. Even in that case, the formation and stabilization of a droplet at the capillary for the tension measurements was very difficult. CO₂ dissolved in the droplet phase would tend to leave the droplet when entering the measurement cell, thus causing an instantaneous solidification of the drop. The viscosity of the CO₂-saturated sugar acetate phase was also very low and thus hard to flow slowly into the measurement cell and to be contained at the tip of the capillary. The tension of the interface was also very low, which is a positive aspect for the proposed technology. For the few drops that actually stayed at the capillary (for a short time), results indicate a tension lower than 0.5 mN·m^-1 at 308 K.

Contact Angle. The contact angle of the saturated sugar acetate solution was easier to measure since drops of sugar acetate on a silane substrate (glass cover slide) were easier to form than pendant drops. The contact angle results at 308 K are shown in Figure 7 as a function of the CO₂ pressure. Three independent runs are shown. We observed that (i) the contact angle is fairly low, even at low pressures (i.e., sugar acetate wets the silica-based substrate under CO₂ atmosphere), and (ii) similar to the C|W|substrate system, an increasing pressure also causes an increase in contact angle (dewetting). We have also attempted to measure the contact angle on a hydrophobic substrate (C₈TS-modified glass slide). Sugar acetate is seen to completely wet the C₈TS-modified substrate under compressed CO₂ atmosphere.

Figure 7. Contact Angle of the CO₂–Sugar Acetate–Glass Substrate at 308 K. Three independent runs were conducted. Inset: snapshot of the sugar acetate droplet under CO₂ atmosphere.

Overall, the results show that sugar acetate is a very promising binder candidate, with very low interfacial tension against CO₂, and with good wetting properties on silicate materials even at very low CO₂ pressures.

Solubility of Polymer Binders and Binder Additives in Carbon Dioxide

Polyvinyl acetate (PVAc)—This is the most CO₂-soluble, non-fluorous, commodity polymer (based solely on C, H and O) that has yet been identified. PVAc costs about $3/lb. Unfortunately, CO₂ pressures ranging from 6,000–10,000 psia are required to dissolve the PVAc.
Polylactic acid (PLA)—We have recently found that PLA, a biodegradable polymer, is soluble in CO₂ over a very wide range of molecular weights. Although the pressure required to dissolve PLA in CO₂ does not increase substantially with pressure, the pressure range of 18,000–21,000 psia is roughly twice as high as that required to dissolve PLA.
CTA oligomers (Oligo-CTA)—Very small sugar acetates are highly CO₂ soluble (see #1 above). Cellulose triacetate (CTA), a high molecular weight commodity polymer with a backbone of sugar acetates, is so crystalline, however, that it is completely CO₂-insoluble. We have found that, by “chopping” the CTA into smaller polymers using a technique known as pivaloylysis, very small polymers (oligomers) of cellulose triacetate were formed. Oligo-CTA with roughly 5 or fewer sugar acetate groups in the backbone, is soluble in CO₂ at several thousand psia.
Poly (1-O-(vinyloxy)ethyl-2,3,4,6-tetra-O-acetyl-β-D-glucopyranoside)–Poly(AcGIcVE)—It was desired to exploit the solubility of sugar acetates while avoiding the crystallinity associated with a polymeric backbone of sugar acetates. Therefore, a polymer was synthesized with side groups composed of sugar acetate. This polymer, poly(AcGIcVE) was found to be slightly less soluble in CO₂ than PVAc. As such, it is the second-most CO₂-soluble compound that has been identified. Nonetheless, given the price of its starting materials and the difficulty in synthesis, this polymer provides no clear advantage over PVAc for this application.

Conclusions:

Tri-tert-butylbenzene (TTBB) exhibits extremely high solubility in dense carbon dioxide at relatively low pressures. It also can be extracted very rapidly into dense CO₂. In a CO₂ binder extraction process, TTBB binders could be extracted efficiently at pressures of 1000 psi, or less, and at temperatures only slightly higher than room temperature (~35°C). Because of these excellent properties, TTBB should be investigated further for metal forming binder applications. However, it is not presently available as a commercial chemical, so its availability and cost for industrial use are not known. Its present cost as a reagent grade chemical would be much too high for any industrial application.
β-D galactose pentaacetate and tri-tert-butylphenol (TTBP) also exhibit high solubility in dense carbon dioxide, and rapid extraction rates in dissolution experiments. β-D galactose pentaacetate may be problematic for industrial use because it does not have good long-term stability, and could not be stored for long times. TTBP is available commercially, has excellent stability, and relatively low cost. Consequently, among the compounds evaluated, TTBP is the most promising candidate for use as a metal forming binder material. TTBP can be extracted efficiently into dense carbon dioxide at pressures ranging from 1000–2500 psi, and temperatures of 50–60°C.
Among low-cost commercial polymers, polyvinyl acetate (PVA) exhibits the highest solubility in dense carbon dioxide. Thus PVA is also a viable candidate material for CO₂-extractable metal forming binders, and a possible strength additive for such binders. However, the pressure required to extract PVA into carbon dioxide is 6000–10000 PSI, which is much higher than the extraction pressures for the materials listed above.

References:

Dilek C, Manke CW, Gulari E. Phase behavior of 1,3,5-tri-tert-butylbenzene-carbon dioxide binary system. Journal of Supercritical Fluids 2007 Jul 27 [Epub ahead of print] doi:10.1016/j.supflu.2007.07.009.

Dilek C, Manke CW, Gulari E. Phase behavior of β-D galactose pentaacetate–carbon dioxide binary system. Fluid Phase Equilibria 2006;239(2):172-177.

Journal Articles on this Report: 1 Displayed | Download in RIS Format

Other project views: All 4 publications 1 publications in selected types All 1 journal articles

Type Citation Project Document Sources

Journal Article Dilek C, Manke CW, Gulari E. Phase behavior of β-D galactose pentaacetate–carbon dioxide binary system. Fluid Phase Equilibria 2006;239(2):172-177. R831504 (Final)

Abstract: Science Direct Abstract

Other: Science Direct PDF

Supplemental Keywords:

Type	Citation	Project	Document Sources
Journal Article	Dilek C, Manke CW, Gulari E. Phase behavior of β-D galactose pentaacetate–carbon dioxide binary system. Fluid Phase Equilibria 2006;239(2):172-177.	R831504 (Final)	Abstract: Science Direct Abstract Other: Science Direct PDF

supercritical fluid technology, compressible solvents, interfacial phenomena, interfacial tension, contact angle, viscosity, swelling, phase behavior, metal casting, lost foam, debinding, green chemistry, alternatives, sustainable development, clean technologies, innovative technology, renewable, waste reduction, waste minimization, environmentally conscious manufacturing; environmental chemistry, engineering, measurement methods, manufacturing industry, VOC's, , Sustainable Industry/Business, Scientific Discipline, RFA, Technology for Sustainable Environment, Sustainable Environment, Environmental Engineering, Environmental Chemistry, foundry industry, foundry mold and core processing, metal casting industry, carbon dioxide soluble binder, environmentally conscious manufacturing, alternative materials, air pollution control, emissions control, binder removal, environmentally benign alternative

Progress and Final Reports:
Original Abstract

Top of page

The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.

Research Project Search

Final Report: Carbon Dioxide Soluble Binders for Environmentally Benign Metal Forming Operations

Description:

Local Navigation