Case Study: Grinding Corn Germ for the Catalytic Extraction of Oil

A thermal analysis of unprocessed germ was performed to identify whether ambient or cryogenic grinding would be recommended.

Figure 1. A thermal analysis of raw corn germ as measured with a DSC Q1000 showed the T_G and T_M to be consistent with literature values for similar materials.

Corn oil typically is recovered from the corn germ through a solvent-based process using hexane. Due to capital equipment requirements and special handling and disposal considerations, the use of hexane is considered a rather expensive process. Unfortunately, water-based methods for oil recovery have not been successful because the yields of corn oil are much less than those obtained using hexane extraction.

Figure 2. Samples were tested in a LIN pilot-scale process that incorporated a helical screw conveyor.

Recently, researchers at the USDA’s Agricultural Research Services in Philadelphia investigated the use of a combination of enzymes and an aqueous system to extract oil from corn germ. The corn oil located within the germ is contained in a starch-and-protein matrix in small inclusions with a nominal 1-μm diameter. Because the primary function of the enzyme in the new process is to depolymerize starch, grinding prior to extraction should accelerate the extraction process and increase the oil yield.

A thermal analysis of the unprocessed germ was performed to identify whether ambient or cryogenic grinding would be recommended. The transition temperatures were measured on a differential scanning calorimeter (DSC). The DSC output shown in figure 1 identified a glass transition temperature (T_G) for the starch component at around -72°F (-58°C) and a melting temperature (T_M) for the oil component at around -4°F (-20°C). These results were consistent with literature values for similar materials. Therefore, based on the rather low T_G, the material was considered appropriate for using cryogenic conditions in grinding to expose sufficient surface area to maximize the oil extractability.

Figure 3. The cylindrical tube at the top of the helical screw conveyor was fitted with a LIN injection manifold and appropriate controls to introduce the refrigerant into the process. The body of the conveyor was insulated to mitigate the cryogenic burn hazard and maximize cooling efficiency.

Samples were tested in a pilot-scale process that incorporated a helical screw conveyor in which LIN was added to the process (figures 2 and 3). The corn germ then was passed through the grinding mill, and the powder subsequently was separated from the resulting nitrogen gas in the cyclone separator. Using nitrogen in this particular process provided a secondary benefit of minimizing the oxidation of the oil fraction during processing.

Figure 4. The particle size distribution for the cryogenically ground corn germ (as measured with a Horiba LA-910) demonstrated that LIN was an effective and efficient method of controlling in-feed temperature.

The process conditions for the trial are summarized in the table. While the LIN ratio often can be determined experimentally, the LIN requirement in this case was calculated from the specific heat as a function of temperature (as determined from the data in figure 1) and the measured temperature change through the conveyor. The corn germ was cooled to the desired inlet temperature of the grinding mill using LIN spray in a helical screw conveyor with a retention time of five minutes. The mill inlet temperature shown in the table is much lower than the material T_Gto compensate for the heat input from the mechanical grinding process.

This table shows the average settings and run-time calculations for several Microtec UTM-200 corn germ trials conducted using LIN.

A representative particle size distribution is given in figure 4. The resulting corn germ D50 value was 17.4 μm for this sample, as taken downstream of the cyclone separator. The smallest measured particle was 2.3 μm. Nonetheless, the injection of LIN into the feed conveyor was an effective and efficient method of controlling the in-feed temperature of the corn germ so that the desired temperature conditions could be maintained throughout the grinding process. Furthermore, as the power input to the sample was varied, it was possible to use dynamic control of the LIN addition to maintain the desired processing conditions within the range tested. This example illustrates the flexibility that can be achieved using a LIN-based temperature control system.