Anal. Chem. 1999, 71, 3808-3813
Retention Behavior of Phenols, Anilines, and Alkylbenzenes in Liquid Chromatographic Separations Using Subcritical Water as the Mobile Phase Yu Yang,* Aaryn D. Jones, and Clinton D. Eaton
Department of Chemistry, East Carolina University, 205 Flanagan, Greenville, North Carolina 27858
The unique characteristic of subcritical water is its widely tunable physical properties. For example, the polarity (measured by dielectric constant) of water is significantly decreased by raising water temperature. At temperatures of 200-250 °C (under moderate pressure to keep water in the liquid state), the polarity of pure water is similar to that of pure methanol or acetonitrile at ambient conditions. Therefore, pure subcritical water may be able to serve as the mobile phase for reversed-phase separations. To investigate the retention behavior in subcritical water separation, the retention factors of BTEX (benzene, toluene, ethylbenzene, and m-xylene), phenol, aniline, and their derivatives have been determined using subcritical water, methanol/water, and acetonitrile/water systems. Subcritical water separations were also performed using alumina, silica-bonded C18, and poly(styrene-divinylbenzene) columns to study the influence of the stationary phase on analyte retention under subcritical water conditions. Reversed-phase liquid chromatography (RPLC) is a very popular separation and analytical technique employed in almost every chemistry laboratory worldwide. However, a mixture of relatively polar organic solvent and water (e.g., methanol/water or acetonitrile/water) has to be used as the mobile phase in RPLC. Therefore, the major disadvantage of RPLC is the need for organic solvents that are expensive, are potentially harmful to the operator and the laboratory environment, and need to be disposed of. Water is nonhazardous, but it is too polar to dissolve many organic species at ambient conditions. Therefore, ambient water cannot elute most organic analytes when it is used as the eluent in LC. However, the solubility of organic compounds is dramatically enhanced by increasing water temperature under moderate pressure to keep water in the liquid state. For example, the solubility of benzo[e]pyrene at 350 °C has been reported to be 10 wt %, an increase of 25 million-fold over its solubility at ambient conditions.1,2 Recent studies demonstrate that increases in solubility of 104-105-fold are typical for pesticides and polycyclic aromatic * Corresponding author: (tel) 252-328-1647; (fax) 252-328-6210; (e-mail)
[email protected]. (1) Sanders, N. D. Ind. Eng. Chem. Fundam. 1986, 25, 171. (2) Mackay, D.; Shiu, W. Y. J. Chem. Eng. Data 1977, 22, 399.
3808 Analytical Chemistry, Vol. 71, No. 17, September 1, 1999
hydrocarbons by raising water temperature from 25 to 200 °C.3 Furthermore, the solubilities of more water soluble hydrocarbons such as benzene and alkylbenzenes are also enhanced 10-30 times by raising temperature from ambient to 200 °C.4,5 The organic solubility increase in water at elevated temperatures is partially caused by the decrease in the polarity of water. For example, the dielectric constant of water is decreased from 80 to ∼30 by raising water temperature from ambient to 250 °C (under moderate pressures to maintain water in liquid).6 The polarity of water at 200-250 °C is similar to that of methanol or acetonitrile at ambient temperature. Therefore, high-temperature water behaves like an organic solvent, and reversed-phase separations may be achieved by using pure water at elevated temperatures. Although reversed-phase separations have been performed at elevated temperatures, the focus in these studies7-15 has been either to observe the changes in retention factors of the analytes or to examine the thermal stability of the stationary phases. Efforts of eliminating the use of organic modifiers in the mobile phase were not reported in these works. In addition, the temperature employed in these studies was generally lower than 120 °C.7-15 Pure water separations using nonporous particles at ambient conditions have been reported. Reasonable retention factors have been achieved for some aromatic analytes by substantially decreasing the ratio of the stationary phase relative to the mobilephase volume.16 Very recently, initial studies of using subcritical water as an eluent for potential LC separations have been reported.17-19 Miller (3) Miller, D. J.; Hawthorne, S. B. Anal. Chem. 1998, 70, 1618. (4) Mathis, J. A.; Yang, Y. The 8th International Symposium on Supercritical Fluid Chromatography and Extraction, St. Louis, MO, July 1998. (5) Yang, Y.; Miller, D. J.; Hawthorne, S. B. J. Chem. Eng. Data 1997, 42, 908. (6) Haar, L.; Gallagher, J. S.; Kell, G. S. National Bureau of Standards/National Research Council Steam Tables; Hemisphere Publishing Corp.: Bristol, 1984. (7) Cole, L. A.; Dorsey, J. G. Anal. Chem. 1992, 64, 1317. (8) Cole, L. A.; Dorsey, J. G. Anal. Chem. 1992, 64, 1324. (9) Tchapla, A.; Heron, S.; Colin, H.; Guiochon, G. Anal. Chem. 1988, 60, 1443. (10) Sander, L. C.; Field, L. R. Anal. Chem. 1980, 52, 2009. (11) Li, J.; Carr, P. W. Anal. Chem. 1997, 69, 837. (12) Chen, H.; Horvath, Cs. J. Chromatogr., A 1995, 705, 3. (13) Kalghatgi, K.; Horvath, Cs. J. Chromatogr. 1988, 443, 343. (14) Hancock, W. S.; Chloupek, R. C.; Kirkland, J. J.; Snyder, L. R. J. Chromatogr., A 1994, 686, 31. (15) Chloupek, R. C.; Hancock, W. S.; Marchylo, B. A.; Kirkland, J. J.; Boyes, B. E.; Snyder, L. R. J. Chromatogr., A 1994, 686, 45. (16) Foster, M. D.; Synovec, R. E. Anal. Chem. 1996, 68, 2838. 10.1021/ac981349w CCC: $18.00
© 1999 American Chemical Society Published on Web 07/23/1999
and Hawthorne investigated subcritical water separations of alcohols, polyhydroxylbenzenes, and amino acids with flame ionization detection.17 Smith and Burgess demonstrated separations of other polar analytes using superheated water as the mobile phase and UV detection.18 Yang et al. studied the elution ability of subcritical water from different sorbents.19 In the present study, the retention factors of BTEX (benzene, toluene, ethylbenzene, and m-xylene), phenol, aniline, and their derivatives have been determined using subcritical water, methanol/ water, and acetonitrile/water systems to study the retention behavior in subcritical water separations. To evaluate the effect of stationary phases on the retention of the solutes, subcritical water separations have also been performed using alumina, silicabased C18, and poly(styrene-divinylbenzene) columns at three different water temperatures. EXPERIMENTAL SECTION Reagents. Benzene, toluene, ethylbenzene, and m-xylene were obtained from Fisher Scientific (Fair Lawn, NJ). Phenol, o-cresol, 4-chlorophenol, 2,4-dichlorophenol, 2,4,5-trichlorophenol, aniline, 3-chloroaniline, 2,3-dichloroaniline, and 2,4,6-trichloroaniline were purchased from Aldrich (Milwaukee, WI). Among the mobile phases used in this study, HPLC grade methanol and acetonitrile were obtained from Fisher Scientific, while the deionized water (18 MΩ) was prepared in our laboratory using a Sybron/ Barnstead system (Sybron/Barnstead, Boston, MA). All mobile phases were purged with helium before each use. Columns. Both normal- and reversed-phase columns were used in this study. An Alumina column (150 × 2 mm i.d.) and a Nucleosil C18 AB column (150 × 2 mm i.d.) were purchased from Keystone Scientific (Bellefonte, PA). A Poly(styrene-divinylbenzene) (PRP-1) column (150 × 2.1 mm i.d.) was obtained from Hamilton (Reno, NV). The particle size is 5 µm for all three columns. The pore size is 100 Å for the Nucleosil C18 AB and the poly(styrene-divinylbenzene) column. The reason for using columns with a smaller internal diameter is that faster thermal equilibrium can be achieved with these “thinner” columns at elevated temperatures. The reversed-phase columns were stable at elevated temperatures during the course of this study since no significant change in retention was observed at the beginning and the end of this work. However, the potential hydrolysis or degradation of the packing material at higher temperatures over long-term use is currently under investigation. Subcritical Water Separation. A homemade system was used for both subcritical water and the traditional RP separations in this work. A Hewlett-Packard pump (series 1050, Avondale, PA) was used to deliver the mobile phase at a flow rate of 0.25 mL/ min. This low flow rate was used for two reasons. First, the pressure drop of the analytical column would be very high with higher flow rates since columns with smaller internal diameter were used in this study. Second, the stability and performance of the UV detector might be influenced by the flow rate of the hightemperature water. A lower flow rate is more desirable since less heat will reach the UV flow cell compared with higher flow rates. The outlet of the pump was connected to a Valco six-port injector (17) Miller, D. J.; Hawthorne, S. B. Anal. Chem. 1997, 69, 623. (18) Smith, R. M.; Burgess, R. J. J. Chromatogr., A 1997, 785, 49. (19) Yang, Y.; Belghazi, M.; Lagadec, A.; Hawthorne, S. B.; Miller, D. J. J. Chromatogr., A 1998, 810, 149.
fitted with a 2-µL sample loop (purchased from Keystone Scientific). The injector was located just outside the oven, while the analytical column was placed inside a Fisher Isotemp oven. A Knauer UV detector was placed next to the oven, and the inlet of the UV detector was connected to the outlet of the analytical column. The outlet of the UV flow cell was connected with a short packed column (Keystone, 20 × 4.6 mm with a particle size of 5 µm) to prevent water from boiling at temperatures above 100 °C. The back pressure provided by this short column was typically a few bars. Please note that the total pressure in the system is the sum of the pressure drops of the separation column and the restrictor column and it is higher than the back pressure. If this back pressure is not applied, not only would the stationary phase be exposed to steam but also the UV signal would fluctuate very quickly (the detector would be unstable). Any packed columns or back-pressure regulators with pressures higher than a few bars are not recommended since the high pressure may blow up the expensive UV cell. When the above precautions are taken properly, a nonmodified UV detector can be used in subcritical water separation. The detector used in our studies was stable and functioned normally during the entire study. A wavelength of 254 nm was used for all of the experiments. However, a shorter wavelength can be used with subcritical water as the mobile phase. A HP 3396 series II integrator was used for data acquisition. The stock solutions of the analytes were prepared in methanol (∼1000 ppm). However, the dilution of these analytes was prepared in pure deionized water (∼100 ppm). After the purge of the deionized water with helium, water was continuously pumped through the column at a flow rate of 0.25 mL/min. The oven was set to the desired temperatures. To ensure that the separations were carried out at the desired temperatures, the first injection of each temperature was not made until ∼20 min after the desired oven temperature was reached. Please note that the temperature of the stationary phase and the mobile phase inside the column lagged behind the oven temperature by approximately 5-15 min, depending on the temperature employed. During separations at elevated temperatures (e.g., 150-200 °C), be aware that the outside of the oven and the tubing between the column and the detector can be hot. Traditional Reversed-Phase Separations Using Organic Solvent/Water. Separations were also performed using methanol/ water or acetonitrile/water as the mobile phase. Please note that the analytical column was not heated in these separations. RESULTS AND DISCUSSION Temperature Effect on Retention in Subcritical Water Separations. Separations of polar analytes such as phenol, aniline, and their derivatives were performed using the C18 AB column at three different temperatures. As shown in Table 1, the retention factors are decreased by increasing the water temperature (decreased dielectric constant). For example, 4-chlorophenol was not eluted by subcritical water at 100 °C until 19 min (k′ ) 6.92) as shown in Table 1 and Figure 1. However, its retention was reduced to 6.4 min (k′ ) 1.71) using water at 150 °C, as demonstrated in Table 1 and Figure 2. Even 2,4,5-trichlorophenol was eluted by pure water at higher temperature (150 °C) within 32 min (Figure 2). The trend of decreased retention by raising the temperature of water was also found in separations of aniline Analytical Chemistry, Vol. 71, No. 17, September 1, 1999
3809
Table 1. Temperature Effect on the Retention Factors of the Selected Solutes on a Nucleosil C18 AB Column Using Subcritical Water at a Flow Rate of 0.25 mL/min temperature (°C) dielectric const phenol o-cresol 4-chlorophenol 2,4-dichorophenol aniline 3-chloroaniline 2,3-dichloroaniline benzene toluene ethylbenzene m-xylene
100 56
150 44
Retention Factor (k′) 2.12 5.50 6.92 22.4 2.35 6.12 20.1 9.64 27.0 >36 >36
200 35
0.516 1.47 1.71 5.31
0.035 0.170 0.179 0.856
0.561 1.71 5.09
0.026 0.279 1.07
3.48 8.92 19.9 21.6
1.64 4.84 12.1 12.3
Figure 2. Subcritical water separation of phenol and its derivatives from the Nucleosil C18 AB column at 150 °C.
Figure 1. Subcritical water separation of phenol and its derivatives from the Nucleosil C18 AB column at 100 °C.
and its derivatives as demonstrated in Table 1 and Figures 3 and 4. Since good separations of polar analytes were achieved by subcritical water as shown in Table 1, we then tried subcritical water separations of nonpolar compounds such as BTEX. The retention factors of BTEX obtained by subcritical water are also listed in Table 1. As expected, the retention times of BTEX are generally longer than those of phenol, aniline, and their derivatives for separations at the same temperature. However, like the polar analytes, the BTEX retention factors were also reduced by raising the water temperature. For example, toluene’s retention factor was 27 at 100 °C, but it was decreased to 4.84 when the temperature was raised to 200 °C, as shown in Table 1. Figures 5 and 6 show the separations of benzene, toluene, and ethylbenzene obtained by subcritical water at 150 (Figure 5) and 200 °C (Figure 6). Since the retention factors of BTEX are generally greater than those of the polar solutes, subcritical water is a weaker eluent for BTEX than for phenol, aniline, and their derivatives. The following factors might contribute to the temperature effect on the retention of subcritical water separations. As discussed earlier, the polarity of water is decreased by raising the water temperature since the dielectric constant drops significantly. The lowered polarity can strengthen the eluting power of water for reversed-phase separations.20-22 It was reported that a lower 3810 Analytical Chemistry, Vol. 71, No. 17, September 1, 1999
Figure 3. Subcritical water separation of aniline and its derivatives from the Nucleosil C18 AB column at 100 °C.
solubility of the solute corresponds to a longer retention of this solute in RPLC.7 The same trend was found in our study. For example, phenol has the highest solubility in the phenol group and thus the shortest retention as demonstrated in Table 1. The same phenomenon was also found for the aniline group and the BTEX group, where the retention is increased with decreasing solubility (Table 1). Since organic solubility is dramatically enhanced by heating the water, as discussed earlier, the increased solubility can help to reduce the retention in high-temperature water separations. The temperature effect on retention in RPLC using organic solvent/water mixtures was previously investigated by other researchers.7-10 The van’t Hoff equation (eq 1) was used to study (20) Melander, W. R.; Horvath. C. In High Performance Liquid Chromatography: Advances and Perspectives; Horvath, C. Ed.; Academic Press: New York, 1980; Vol. 2, pp 113-319. (21) Snyder, L. R., Kirkland, J. J., Glajch, J. L., Eds. Practical HPLC Method Development, 2nd ed.; Wiley: New York, 1997. (22) Jo ¨nsson, J. Å. Chromatographic Theory and Basic Principles; Marcel Dekker: New York, 1987.
Table 2. Retention Factors of the Selected Solutes on a Nucleosil C18 AB Column Using Methanol/Water at a Flow Rate of 0.25 mL/min methanol percentage in water dielectric constant phenol o-cresol 4-chlorophenol 2,4-dichorophenol
Figure 4. Subcritical water separation of aniline and its derivatives from the Nucleosil C18 AB column at 150 °C.
43 56
Retention Factor (k′) 0.77 0.98 1.15 1.85
68 44
90 35
0.087 0.17 0.23 0.60
