Carbon dioxide mobility

Achiral-chiral chromatography has also been accomplished using subcritical fluid chromatography (Phinney et al., 1998). In this work, the structurally related [3-blockers, 1,4-benzodiazepines, and two cold medicines were separated using methanol or ethanol modified carbon dioxide mobile phases. The (3-blockers were separated using cyanopropyl and Chiracel OD columns connected in series. Likewise, an amino bonded phase and Chiracel OD column were used for the separation of the 1,4-benzodiazepines. Guaifenesin and phenylpropanolamine from cough syrup were separated on cyanopropyl and Chiralpak AD columns in series. 

The second approach was taken by practicing liquid chromatographers. They routinely dealt with thermally labile, highly polar molecules and frequently sacrificed resolution, and speed in their separations because of the aqueous mobile phases that were required. With the enhanced diffusion and decreased viscosity of supercritical fluids over liquids, chromatographic run-time and resolution could be improved when supercritical fluids were used. But solubility in pure carbon dioxide mobile phases, which has the solvating powers from hexane to methylene chloride under normal density ranges, was a problem for these polar molecules. To compensate for this, experimentalists started working with mixed mobile phases. These mixed phases were based on... 

Beyond the density changes that can be used to control method modifications in SFC, the mobile phase composition can also be adjusted. Typical LC solvents are the first choice, most likely because of their availability, but also because of their compatibility with analytical detectors. The most common mobile phase modifiers, which have been used, are methanol, acetonitrile and tetrahydrofuran (THF). Additives, defined as solutes added to the mobile phase in addition to the modifier to counteract any specific analyte-column interactions, are frequently included also to overcome the low polarity of the carbon dioxide mobile phase. Amines are among the most common additives. 

Actual operating conditions can be found in the figure caption. A given limitation of SFC, relative to HPLC, as described is the ability to dissolve samples in a solvent system compatible with the methanol/ carbon dioxide mobile phase. For this particular mobile phase, other compatible sample diluents that worked effectively are pure methanol,... 

FIGURE 4.2 The van Deemter curves for HPLC and SFC. Conditions (a) LC, 55 cm x 250 xm ID fused-silica capillary columns packed with 5 pm porous particles, nitromethane test solute, 25°C, UV detector (214 nm) (b) SFC, 75 cm x 250 pm ID fused-silica capillary columns packed with 5-pm porous particles, 45°C, 230 atm, carbon dioxide mobile phase, methane test solute, FID. (Adapted from Wu, N. et al. Anal. Chem. 71 5084-5092. With permission.)... 

The SFC analysis of barbiturates has been reported by Smith and Sanagi [16]. The method utilzed packed columns of ODS bonded silica (200 x 3 mm ID) or polystyrene-divinylbenzene polymer (150 x 4.6 mm ID) with a carbon dioxide mobile phase and FID detection. The barbiturates were strongly adsorbed on the ODS column and were not observed to elute. The drugs did elute with reasonable retention times on the polymer column. Unfortunately, there was peak tailing, and adsorption remained a problem with all the barbiturates studied. The authors were prevented from adding methanol as a polar modifier to the mobile phase because the detection was by FID. 

Edder et al. reported the capillary supercritical fluid chromatography of basic drugs of abuse, namely nicotine, caffeine, methadone, cocaine, imipramine, codeine, diazepam, morphine, benzoylecgonine, papverine, narcotine, and strychnine [25]. They compared the separation of these drugs on DBS and DB wax columns. The chromatographic conditions included a carbon dioxide mobile phase and a flame-ionization detector. It was noted that on the DBS column, all peaks other than methadone and cocaine were separated. With the exception of benzoylecgonine and papaverine, all other peaks were separated on a DB wax column. A reproducibility of less than 5% was obtained with an internal standard method. The detection limits obtained were within 10-50 ppm on both the columns. A linearity of >0.99 was obtained for methadone, codeine, and morphine in the concentration range 10-1000 ppm. 

Li and co-workers demonstrated the use of SFC in the analysis of panaxadiol and panaxatriol in ginseng, a famous traditional Chinese medicine [28 J. A capillary SB-cyanopropyl-50 column with a carbon dioxide mobile phase and flame-ionization detector was used for the analysis. Methyltestosterone was used as an internal standard for the quantitation. Figure 7.17 shows the SFC separation. The method was found to be linear (r > 0.999) in the range studied and the precision obtained was in the range 2.2-5.7%. 

Either the asphaltene precipitation or the multiple phase behavior of the process under "miscible" conditions appears to cause a certain extent of reduction of the mobility of the carbon dioxide, as compared to the mobility which would be calculated from the relative permeability of the carbon dioxide divided its viscosity (14-16). However, it is usually considered that this is not sufficient to resultinafavorable mobility ratio (a ratio of carbon dioxide mobility to crude oil mobility less than one). 

The tests cannot be extrapolated directly to field reservoir performance because the spatial geometry is different. The core tests have a linear, 1-dimensional flow geometry while the actual reservoir has radial, 3-dimensional flow. In 3-dimensional flow the displacement efficiency is typically less than that measured in linear displacement studies. Chilton (1987) showed in his computer simulation studies that, as compared with the linear flow case, the predicted oil produced was 10% less for the two-dimensional model and 27% less for the three-dimensional model. However, when mobility control was used with a tenfold decrease in carbon dioxide mobility, the calculated improvement in displacement efficiency was much less for the linear case than the three-dimensional case. This result indicates that the increase in displacement efficiency under field conditions should be greater than that recorded in these linear laboratory tests. 

Fig. 1 Use of polymer-coated silica stationary phase with a neat carbon dioxide mobile phase. Column Deltabond SFC Methyl (150 X 2.0 mm) mobile phase CO2 pressure gradient 75 bar initial for 2 min, then 50 bar/min ramp to 180 bar, then 15 bar/min ramp to 300 bar flow 0.5 mL/min temperature 100°C injection 5 jjlL detection flame ionization detector sample solvent methylene chloride. Peak key 1 BHT 2 dimethyl azelate 3 triethyl citrate 4 tributyl phosphate 5 methyl palmitate 6 methyl stearate 7 diethylhexyl phthalate 8 Tinuvin 327 9 Spectra-Sorb UV531 10 tri(2- ethylhexyl) trimellitate 11 dilauryl thiodipropionate ...

Saturated solutions in methanol can actually be injected into methanol/ carbon dioxide mobile phases (although not recommended) without the solute falling out of solution. This is probably due to the extensive adsorption of the modifier onto the stationary phase (38), creating a form of reservoir of solvent to receive the injected sample. 

The aromatic hydrocarbon content of diesel fuel affects the cetane number and exhaust emissions. One test method (ASTM D-5186) is applicable to diesel fuel and is unaffected by fuel coloration. Aromatics concentration in the range 1-75 mass% and polynuclear aromatic hydrocarbons in the range 0.5-50 mass% can be determined by this test method. In the method, a small aliquot of the fuel sample is injected onto a packed silica adsorption column and eluted with supercritical carbon dioxide mobile phase. Mono- and polynuclear aromatics in the sample are separated from nonaromatics and detected with a flame ionization detector. The detector response to hydrocarbons is recorded throughout the analysis time. The chromatographic areas corresponding to the mononuclear aromatic constituents, polynuclear aromatic constituents, and nonaromatic constituents are determined, and the mass-percent content of each of these groups is calculated by area normalization. 

Figure 1 SFC chromatogram of soybean oil sample. Carbon dioxide mobile phase at 200 C, 10 m x 50 pm i.d. SB-Methyl-100 column, linear density programmed at 0.005 g/mL/min, FID at 375 C. Fatty acid group identification P-Palmitic, O-Oleic, S-Stearic, L-Linoleic, Ln-Linolenic.

S. Rokushika, K. P. Naikwadi, A. L. Jadhav, and H. Hatano. Polyacrylate liquid crystalline stationary phases in supercritical fluid chromatography with carbon dioxide mobile phase. Chromatographia, 22 209-212,1986. 

Figure 7.2 Supercritical fluid chromatogram of polymer additives listed in Tables 7.1 and 7.2. Conditions 10 m x 50 pm id fused-silica capillary column, crosslinked methylpolysiloxane stationary phase (0.25 pm film thickness) carbon dioxide mobile phase at 140 "C 15 - 35 MPa at 0.3 MPa/min after an initial 12 minutes isobaric period ...

Willian and Lilly [27] also used SFC carbon dioxide mobile phase with FID to monitor the separation of MMA oligomers and polymer additives. 

Biicherl and co-workers [48] has described an integral restrictive interface with jet separation for coupling capillary column SFC with carbon dioxide mobile phase with high resolution mass spectrometry. 

Polymeric materials that are thermally sensitive require alternative analytical approaches. Again, processing and analysis of the representative sample "as is" can be achieved with varied SF instrumentation and methods. Figure 7 gives the off-line SFE microreflectance-FTIR spectrum of a contaminant removed from a rubber composite product by an SFE in-process treatment at 340 atm (5000 psi) and 100 C over a several hour period. The SFE method used both static (no flow of mobile fluid through the pressurized vessel) and dynamic exposure of the product to a flow of 4-5 mL liquid carbon dioxide mobile fluid at 340 atm (5000 psi). The extracted contaminant was identified as an oligomeric fluorocopolymer based on FTIR spectral library comparisons. Presence of the contaminant was shown to be detrimental to in-Tield performance of the product and efficient removal was needed with an environmentally acceptable process. 

---