Although the overall consumption of coffee in the United States is flat or declining slightly, the demand for decaffeinate coffee is growing. The original technology for coffee decaffeination involved the extraction of the caffeine from the beans with methylene chloride. However, methylene chloride has many detrimental effects on both humans and the environment and has been targeted by the EPA for emissions reduction. In addition, public concerns about potential residual methylene chloride in the decaffeinated beans makes this option unacceptable. Another process for coffee decaffeination has been developed that uses water to extract the caffeine from the beans (the "conventional" or "water" process). Although originally motivated by consumer opposition to food processing with chlorinated solvents, this process aims at pollution prevention through solvent substitution; i.e. substituting water for methylene chloride. Unfortunately, the water also removes some of the aromas and flavors, which need to be recovered and added back to the beans to maintain a high quality product. Another drawback is that methylene chloride is still used in the process to remove the caffeine from the water extract; however, it does not ever contact the beans. The caffeine/methylene chloride mixture can be sold, primarily to pharmaceutical companies. A final alternative that has been developed in the last 15-20 years is the use of supercritical carbon dioxide to extract the caffeine. This process involves extracting the caffeine from the beans at moderate pressures (~4350 psi). To avoid recompression costs the caffeine can be washed from the CO2 (which is recycled) at high pressure with water. The aromas and flavors are not extracted from the beans with CO2 so the water wash contains essential pure caffeine and can be sold in that form. The main disadvantage of the supercritical CO2 process is the high pressures needed for operation, which will increase capital equipment costs and require appropriate safety procedures. In the United States, Kraft General Foods makes use of CO2 for coffee decaffeination in a large (estimated at about 50 million lb/yr) supercritical extraction unit in Houston, TX.
Groups at both (i) West Virginia University and (ii) Notre Dame have looked at the CO2 coffee decaffeination process. The main difference between the two projects is that the West Virginia project only does the design for the CO2 extraction process but uses reverse osmosis to concentrate the caffeine/water stream and focuses on the mass transfer modeling of the caffeine extraction from the green coffee beans. The Notre Dame project develops designs for both the CO2 extraction process and the conventional water process so comparisons can be made between the two processes. However, the Notre Dame CO2 extraction just uses nitrogen stripping to concentrate the water/caffeine mixture and does a superficial job of the mass transfer modeling of the caffeine extraction. The two projects are listed below separately.
Caffeine from batches of green coffee beans is extracted using supercritical carbon dioxide. The caffeine is recovered by scrubbing with water (at supercritical conditions for carbon dioxide) followed by reverse osmosis. Some key issues in this design are mass transfer modeling of the caffeine extraction from the green coffee beans, sizing the reverse osmosis unit, and design of the absorber for fluids of roughly equal densities.
This process would be an interesting one-semester design project. Material available on this project includes a project report, PFDs, economic optimization results and P&IDs. Instructors can obtain this information by contacting Professor Shaeiwitz.
CLICK HERE to obtain an Adobe Acrobat file about this project.
In this project, designs were developed for both the conventional "water" process and the supercritical carbon dioxide process for the extraction of caffeine from coffee beans. In the water process, the beans are contacted countercurrently with the water and the caffeine extracted from the water with pressures (~12,000 psi) are required to achieve sufficiently high oil solubility. Finally, the transport of soybean flakes at high pressure would be quite difficult and the actual implementation of this process would hinge on the successful scale-up of a high pressure screw conveyor.
This project actually involves two designs: the conventional hexane process and the supercritical CO2 process so that comparisons could be made between the two technologies. Either design would be an appropriate project for a one-semester senior design project. Brief reports, PFDs, economic evaluations and full AspenPlus designs are available for both of these processes. Instructors can contact Professor Brennecke for this information.
