High-Resolution Gas Chromatography Retention Data as Basis for the

Apr 10, 2003 - We estimated Kow values that are usually derived by liquid chromatography solely from gas chromatographic (GC) data by selective correl...
0 downloads 0 Views 81KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 2274-2279

High-Resolution Gas Chromatography Retention Data as Basis for the Estimation of Kow Values Using PCB Congeners as Secondary Standards RUDOLF HACKENBERG, ANTONIA SCHU ¨ TZ, AND KARLHEINZ BALLSCHMITER* Department of Analytical and Environmental Chemistry, University of Ulm, Albert-Einstein-Allee 11, D 89069 Ulm, Germany

We estimated Kow values that are usually derived by liquid chromatography solely from gas chromatographic (GC) data by selective correlation with PCB congeners as secondary standards. The GC method was established and validated with literature-known values obtained with other methods. Twenty-seven chlorinated diphenyl ethers (PCDE), 19 chlorinated naphthalenes (PCN), 4,4′-DDE, and three brominated diphenyl ethers (PBDE) were used for method validation. The advantages of our method are that only amounts in the nanogram range or less are needed, complex mixtures can be analyzed, Kow values of isomers can be determined, and even the exact structure of compounds does not have to be known. The quality of the Kow values obtained by the GC method mainly depends on the accuracy of the data of the compounds used as standards for the correlation. These data should be based on reliable experimental methods. Our semi-experimental approach in approximating physicochemical data relevant to the environmental distributionsvapor pressure of subcooled liquid and log Kowscan be extended to further classes of compounds because normalized GC retention data are easily available. We exemplified our approach with a bioaccumulating naturally occurring heptachlorinated 1-methyl-1′,2-bipyrrole, which is highly abundant in fish from the South Atlantic among others.

Introduction Deriving physicochemical data from capillary gas chromatography (HRGC) retention data has been successfully applied in the past for the vapor pressure of subcooled liquids p°L of PCBs (1-6). For polychlorinated diphenyl ethers (PCDEs), the vapor pressure of subcooled liquids p°L was also determined by HRGC; water solubility (Sw) and log Kow were determined by RP-HPLC (7). The same was done for halogenated phenyl methyl ethers (anisoles) (8, 9) and alkyl dinitrates (10, 11). Miller et al. could show that the log Kow values of PCBs determined by using the generator column method are linearly related to the logarithms of their relative retention times on packed GC analysis with the nonpolar C-87 * Corresponding author phone: +49-731-50-22751; fax: +49-73150-22763; e-mail: [email protected]. 2274

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

stationary phase (12). The linear relationship between the logarithm of the isothermal relative retention of 13 PCB congeners on a C-87 packed column and their log Kow values determined by the generator column technique was confirmed by the work of Hawker and Connell (13). The latter also found that the total surface areas (TSA) of PCBs were highly correlated with log Kow. The GC approach has been tested for the correlation of Kow values of PCBs obtained by RP-HPLC (14) with their temperature-programmed capillary gas chromatography (GC) retention data (6). But this approach has never been applied since. We extended the capillary GC approach for the determination of the basic molecular property Kow here in a semi-experimental way. We derived physicochemical properties of persistent organic pollutants (POPs) usually obtained by liquid chromatography (7, 14) by adequate correlation of GC data. For this purpose, we used PCBs as secondary standards. We ran polychlorinated diphenyl ethers (PCDE), polychlorinated naphthalenes (PCN), 4,4′-DDE, HCB, and three congeners of brominated diphenyl ether (PBDE) as an accuracy check for our capillary GC approach for the determination of log Kow values. Pontolillo and Eganhouse recently published an extensive study of approximately 700 publications to obtain reliable data for the aqueous solubility and the octanol-water partition coefficient for 4,4′-DDT and 4,4′-DDE as a case study for hydrophobic organic compounds (15). They documented nature and extent of errors in determining critical environmental parameters. Such incorrect data may lead to incorrect environmental risk assessments. The problem was even termed the “Kow controversy” (16). We exemplified our approach with a heptachlorinated 1′,2-bipyrrole that has been identified in the marine environment. This compound can be described as a natural POP, a biogenic persistent organic pollutant. This 1′,2-bipyrrole that occurs intensively in marine biota has long remained unidentified in its structure as it could not be matched with any of the known anthropogenic bioaccumulating substances (17-19). Vetter reported its occurrence naming it Q1 (20) in biota from the Antarctic, in human milk from the Faroer Islands (21), and in Antarctic air (22) and suggested its natural origin. Recently, its structure has been proven by synthesis to be a heptachlorinated 1-methyl-1′,2-bipyrrole (HM1,2BP/ U3/Q1) (23). Mixed halogenated bipyrroles of the 2,2′-type (e.g., C10H6N2Br4Cl2, a 1,1′-dimethyl-tetrabromodichloro-2,2′bipyrrole) have also been found to bioaccumulate, for example, in fish, birds, and marine mammals such as the PCBs (24-27). A global spreading throughout the marine biosphere as it is observed for the PCBs seems likely for the bioaccumulative heptachlorinated 1′,2-bipyrrole (21, 28). The compound is most prominent, however, in fish from the South Atlantic and in deep-sea fish of the North Atlantic (28). Vetter made a calculation of the octanol-water partition coefficient of HM1,2BP/U3/Q1 varying between log Kow 5.9 and 6.2 using a program provided by Syracuse Research Corporation (29, 30). Our GC approach gives characteristic environmental distribution constantssparticularly the essential parameters p°L and the octanol/water partition coefficient Kowssolely from the gas chromatographic retention times.

Experimental Section Sample preparation for gas chromatography starting with fish tissue was done as described in detail in a previous paper (31). Group separation by NP-HPLC was done on aminopropyl silica. The heptachlorinated 1′,2-bipyrrol (HM1,2BP/ 10.1021/es0201294 CCC: $25.00

 2003 American Chemical Society Published on Web 04/10/2003

FIGURE 1. HRGC (CP Sil-5/C18)-ECD chromatogram of the PCB fraction (LC1) isolated from Snoek fillet (Thyrsites atun) caught in the South Atlantic. Peaks with plain figures are PCBs, numbered according to Ballschmiter-Zell (61); peaks labeled with DE and number are polybromodiphenyl ethers. HM1,2BP, heptachloro-1-methyl-1′,2-bipyrrole; Br5-PPE, 2,3-dibromopropyl-2,4,6-tribromophenyl ether. U3/Q1) elutes together with PCBs; 4,4′-DDE and the PBDE elute with the hexane eluate. The gas chromatographic separation with electron capture detection (ECD) of the hexane eluate (LC1) obtained by NP-HPLC was performed on an Varian 3800 GC with capillary column A and H2 as the carrier gas (see Figure 1). Sample injection was done using the cold on-column technique. Capillary colums and temperature programs (for all colums: 2 m retention gap, carrier gas H2): (A) Chrompack CP Sil-5/C18 (105 m × 0.32 mm, 0.1 µm film thickness); 80 °C (2 min) at 20 °C/min f 140 °C at 2 °C/min f 250 °C (15 min); (B) Varian CP Sil-8 (60 m × 0.32 mm, 0.25 µm film thickness); 80 °C (2 min) at 20 °C/min f 180 °C at 2 °C/min f 250 °C (15 min); J&W DB 1701 (60 m × 0.25 mm, 0.25 µm film thickness); 80 °C (2 min) at 20 °C/min f 200 °C at 1.5 °C/min f 265 °C (10 min).

Determination and Calculation of Partition Coefficients Determination of the Vapor Pressure by TemperatureProgrammed Capillary Gas Chromatography. The determination of the vapor pressure by gas chromatography can be performed by using the Herington equation (32, 33):

log(t′R2/t′R1) ) log(p°1/p°2) + log(γ°1,SP/γ°2,SP)

(1.1)

The following terms are used in this equation: t′R1,2: Adjusted retention times of two compounds; one with a known vapor pressure. The adjusted retention times should be within a narrow range for both compounds. p°1,2: Vapor pressure of the subcooled liquid or liquid at standard conditions. γ°1,2: Activity coefficients in the interaction with the stationary phase. Equation 1.1 can be simplified if the gas chromatographic separation is done on a nonpolar stationary phase. In this case, the activity coefficients γ°1,2 tend to approach 1, and the value of the logarithm becomes negligible:

log(t′R2/t′R1) ) log(p°1/p°2)

(1.2)

For practical reasons, linear temperature-programmed capillary GC was used. This approach has been used for PCBs and other groups in the past (1-6, 34, 35). N-Alkanes have been used as standard compounds. The relative retention time (RRT) is a simple and robust way of standardizing the GC retention. Standardizing the retention of the PCB congeners to the sum of the retention times of PCB 52 and PCB 180 has been accepted (36, 37). We adopted this approach in this paper. Determination of Log Kow from Temperature-Programmed Capillary GC Data. A widely used method is to calculate the log Kow value from the known value of a structurally similar substance by adding the fragment constants that were published by Hansch and Leo (38). Kow values of single compounds as well as mixtures of compounds can be determined by reversed-phase liquid chromatography (RP-HPLC) (39). The Kow partition constant of compounds available only in nanograms or less and detectable after high-resolution capillary gas chromatography with ECD or MSD as the detection modes can be calculated from GC retention data by choosing PCB congeners as secondary correlation standards as it is presented here. Using the Collander equation (40, 41):

log Kd(A) ) a log Kd(B) + b

(2)

it is possible to derive the distribution coefficient Kd in one solvent system (A) from the constants measured in a different system (B) if experimentally determined values for the constants a and b are available. An equation for the determination of log Kow values by GC as suggested by Miller et al. (12) and Hawker and Connell (13) and reviewed by James (42) is given by

log Kow ) a log(tR/tM - 1) + C

(3)

with tR being the retention time of the compound and tM being the dead time in the chromatographic system. The VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2275

expression (tR/tM - 1) is called the retention factor k. Connell points out the necessity of the use of reference compounds with known log Kow values for calibration (43). The elevated temperature in the GC measurement, which exceeds the boiling point of water, and the obvious differences between the octanol/water and the gas/liquid system in gas chromatography make this step necessary. Equation 3 is deduced as shown below: The retention factor k is defined by

k ) nS/nM ) cSVS/cMVM ) (tR - tM)/tM

(4)

In this equation, nS and nM are the numbers of molecules, cS and cM the concentrations of the solutes in the two phases, and VS and VM are the volumes of the stationary and mobile phases. With cS/cM ) Kgc and VS/VM ) q, k can be expressed as

k ) Kgcq ) (tR - tM)/tM

(5)

Rearrangement leads to an expression for the distribution coefficient Kgc in the gas-liquid system, which can be substituted in eq 2 to give

log Kow ) a log(tR/tM - 1) + a log 1/q + b

(6)

Combining the term [a log 1/q + b] to a new constant C, an equation of the type y ) mx + z is obtained (see eq 3). By using reference compounds with known log Kow values (e.g., specific PCB congeners), the constants a and C can be derived from a linear correlation. Deriving Further Physicochemical Properties from Log Kow. Taking the vapor pressure and the log Kow value as starting points, the water solubility and other essential partition constants can be calculated (44, 45). Several equations describing the correlation between log Kow and the water solubility were compiled and reviewed by Lyman et al. (44). These correlations are based on different sets of compounds, thus leading to linear equations with different slopes and intercepts (eqs 7-9):

log Kow ) x log Sw + y log Sw )

log Kow - y x

unit of Sw is mol/L (46) (7)

for PCB and 4,4′-DDT: x ) -0.518; y ) 2.222 (46) log Sw ) -0.922 log Kow + 4.184 unit of Sw is mg/L (47) (8) log Sw ) -1.49 log Kow + 7.46 unit of Sw is µmol/L (48) (9) Equation 7 is based on PCB and 4,4′-DDT; eqs 8 and 9 are both based on a broad range of compounds including among others chlorobenzenes and aromatic hydrocarbons. So these equations appear to be suited for the estimation of the water solubility of environmental contaminants.

Results and Discussion Comparison of Log Kow (GC) to Log Kow (HPLC). To check whether the application of the correlation (eq 3) presented above would lead to acceptable results, we calculated the partition constants by high-resolution capillary gas chromatography and compared them to HPLC-based values obtained from the literature. The determination of log Kow values by HPLC is a method of established reliability. However, the results will show some 2276

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

TABLE 1. Constants a and C of Equation 3 As Obtained by Correlation of Log(tR/tM - 1) with Log Kow (HPLC) (14) for a 40-Congener PCB Standard Solutiona capillary column A)a

Sil-5/C18 (PCB set Sil-5/C18 (PCB set B)b Sil-8 (PCB set A)a DB 1701 (PCB set A)a

constant a

constant C

correlation

4.05 3.89 4.27 3.73

3.30 3.55 2.16 2.63

0.90 0.93 0.91 0.90

a PCB set A: log K ow values were determined by HPLC on a Nucleosil 5/C18, eluent methanol/water (80:20). b PCB set B: log Kow were determined by HPLC on a Nucleosil 5/C18, eluent acetonitrile/water (70:30).

deviation if there are changes in the experimental setup (e.g., the use of different eluents). We used two sets of HPLCbased log Kow values for PCBs to indicate how this would affect the GC correlation of log Kow values. One set of log Kow values (PCB set A) was determined on a Nucleosil C18 RP column with MeOH/H2O (80:20) as an eluent; for the measurement of the other set (PCB set B), the same column with CH3CN/H2O (70:30) as an eluent was used (14). For both sets 2-ketoalkanes were used to determine the retention indices in the liquid chromatographic system. We used a standard solution of PCB that contained 40 congeners with degrees of chlorination from Cl1 to Cl10 to obtain the constants a and C required by eq 3. Data were obtained with three capillary colums of different polarity, a Sil-5/C18 (Chrompack), a Sil-8 (Varian), and a J&W DB 1701 capillary column All these capillaries showed a good correlation of the term log(tR/tM - 1) to the log Kow based on HPLC (Table 1). The standard compounds we used for the comparison of log Kow (GC) and log Kow (HPLC) values were 27 polychlorinated diphenyl ethers (PCDE) (7) and 19 polychlorinated naphthalenes (PCN) (49-52). The values for log Kow obtained by correlation of GC-based retention times (eq 3) are listed in Tables 2 (PCDE) and 3 (PCN). Published values of log Kow are only available for selected PCN congeners. We observed an acceptable conformity of our calculated log Kow (GC) values to published results for both groups of organohalogens. For the PCDE, calculated values deviate not more than half an order of magnitude for all but three congeners (Table 2). As to the PCN, this benchmark is also exceeded by three congeners (Table 3). The calculated values for the particular congeners and thus their deviation from the literature value are very similar for the three GC phases. A comparison of the log Kow values for PCDE obtained from the two HPLC PCB sets A and B shows that the results differ from about 0.10.2 units. Results calculated from the HPLC PCB set B are slightly on the higher side. Our results indicate that the method presented here for the calculation of log Kow values using GC retention data allows a good approximation and can be applied to nonpolar compounds of the POP type. Evaluation of the Log Kow (GC) Approach for 4,4′-DDE and Other POPs. We used the retention times as determined in the chromatogram given in Figure 1 to calculate the vapor pressure, the log Kow value, and the water solubility of the heptachloro-1-methyl-1′,2-bipyrrole (HM1,2BP/U3/Q1) with HCB, 4,4′-DDE, and PBDE as an internal check. To calulate the vapor pressure (p°) of the above-mentioned compounds according to eq 1.2, we used the retention time and vapor pressure of closely eluting PCB congeners. The results are given in Table 4, and retention times are given in Table 5. Using late eluting PCBs as the reference results in an increase of the calculated vapor pressure (Table 4). This emphasizes the choice of the correct reference compound. Considering the discrepancy between the values published

TABLE 2. Log Kow Values of Tri- to Octachlorinated Diphenyl Ethers (PCDEs) As Determined by HPLC (7) and by Capillary Gas Chromatography (HRGC)a

no. 17 25 31 32 39 41 63 64 68 79 82 90 91 119 126 130 137 156 163 166 170 177 180 181 190 195 203

PCDE congeners chlorine substitution 2,2′,4 2,3′,4 2,4′,5 2,4′,6 3,4′,5 2,2′,3,4 2,3,4′,5 2,3,4′,6 2,3′,4,5′ 3,3′,4,5′ 2,2′,3,3′,4 2,2′,3,4′,5 2,2′,3,4′,6 2,3′,4,4′,6 3,3′,4,4′,5 2,2′,3,3′,4,5 2,2′,3,4,4′,5 2,3,3′,4,4′,5 2,3,3′,4′,5,6 2,3,4,4′,5,6 2,2′,3,3′,4,4′,5 2,2′,3,3′,4′,5,6 2,2′,3,4,4′,5,5′ 2,2′,3,4,4′,5,6 2,3,3′,4,4′,5,6 2,2′,3,3′,4,4′,5,6 2,2′,3,4,4′,5,5′,6

log Kow (HPLC)b

log Kow (GC) mean

log Kow (Sil 5/C18) PCB set A

log Kow (Sil 5/C18) PCB set B

log Kow (Sil 8) PCB set A

log Kow (DB 1701) PCB set A

4.96 5.65 5.66 5.3 5.77 5.72 6.21 5.64 6.13 6.22 6.3 6.54 6.06 6.44 6.83 7.01 6.72 7.07 6.78 6.95 7.28 7.14 7.46 7.31 7.31 7.84 7.81

5.69 ( 0.06 5.67 ( 0.06 5.70 ( 0.05 5.62 ( 0.06 5.70 ( 0.07 6.37 ( 0.06 6.23 ( 0.06 6.17 ( 0.05 6.10 ( 0.07 6.27 ( 0.07 6.79 ( 0.14 6.67 ( 0.14 6.61 ( 0.04 6.53 ( 0.06 6.88 ( 0.05 7.13 ( 0.04 7.11 ( 0.04 7.20 ( 0.04 6.99 ( 0.04 7.06 ( 0.04 7.53 ( 0.04 7.40 ( 0.03 7.39 ( 0.04 7.36 ( 0.03 7.40 ( 0.03 7.74 ( 0.03 7.56 ( 0.03

5.62 5.62 5.65 5.54 5.68 6.31 6.21 6.12 6.09 6.27 6.84 6.59 6.56 6.52 6.86 7.09 7.08 7.17 6.96 7.03 7.47 7.35 7.35 7.32 7.36 7.69 7.52

5.78 5.78 5.81 5.71 5.84 6.45 6.35 6.26 6.23 6.40 6.51 6.96 6.68 6.65 6.97 7.19 7.18 7.27 7.07 7.14 7.56 7.44 7.45 7.41 7.45 7.77 7.60

5.63 5.62 5.66 5.58 5.65 6.32 6.18 6.13 6.04 6.20 6.86 6.56 6.56 6.48 6.82 7.09 7.08 7.16 6.96 7.03 7.51 7.39 7.35 7.33 7.38 7.73 7.54

5.71 5.65 5.69 5.64 5.63 6.40 6.18 6.16 6.02 6.19 6.96 6.57 6.62 6.47 6.85 7.14 7.10 7.20 6.97 7.04 7.57 7.43 7.39 7.36 7.40 7.77 7.57

a Log K ow (GC) mean ( mean deviation and single results as obtained by correlation of GC retention data given by three capillary colums of different polarity is presented. The structure-related numbering of PCDE follows ref 7. b Ref 7.

TABLE 3. Log Kow Values for Polychlorinated Naphthalenes (PCN) Calculated from GC Retention Timesa PCN no.

chlorine substitution

log Kow ref values

log Kow (GC) mean

log Kow (GC) Sil 5/C18

log Kow (GC) Sil 8

log Kow (GC) DB 1701

1 3 5 6 8 9 10 21 27 28 42 43 47 53 66 69 71 73 75

1 1,2 1,4 1,5 1,7 1,8 2,3 1,3,7 1,2,3,4 1,2,3,5 1,3,5,7 1,3,5,8 1,4,6,7 1,2,3,5,8 1,2,3,4,6,7 1,2,3,5,7,8 1,2,4,5,6,8 1,2,3,4,5,6,7 1,2,3,4,5,6,7,8

3.9b 4.42b 4.66b 4.67b 4.56b 4.41a 4.51b; 4.71b 5.59a; 5.35b 5.94a; 5.75b; 5.50b 5.77b 6.38a; 6.19b 5.96a; 5.76b 5.81b 6.13d 6.79d 6.69d 6.98c 7.18d; 7.69c 8.4a

4.06 ( 0.14 4.69 ( 0.07 4.56 ( 0.08 4.62 ( 0.07 4.65 ( 0.07 4.85 ( 0.08 4.78 5.08 ( 0.02 5.91 5.78 ( 0.06 5.54 ( 0.06 5.78 ( 0.06 5.69 ( 0.06 6.46 ( 0.05 6.77 ( 0.05 6.87 ( 0.06 6.90 ( 0.06 7.33 ( 0.03 7.70 ( 0.02

3.86

4.07 4.62 4.55 4.56 4.58 4.80 5.06

4.26 4.75 4.65 4.67 4.72 4.96 4.78 5.09

5.74 5.45 5.74 5.65 6.42 6.74 6.84 6.87 7.31 7.69

5.73 5.57 5.73 5.64 6.42 6.70 6.82 6.85 7.30 7.68

4.49 4.78 5.10 5.91 5.87 5.61 5.87 5.77 6.54 6.87 6.96 6.99 7.38 7.74

a Log K ow mean ( mean deviation. Coefficients a and C of eq 3 are based on PCB set A. Reference values of log Kow PCN: a, ref 49; b, ref 50; c, ref 51; and d, ref 52. The structure-correlated systematic numbering of PCN follows ref 62.

by different authors (Table 4), the calculated results for the vapor pressure of HCB, 4,4′-DDE, and three brominated diphenyl ethers correspond with published data. To obtain the constants a and C required for the calculation of the log Kow values by eq 3, the log Kow values of the seven PCB indicator congeners covering the degree of chlorination from tri to hepta (PCB 28, 52, 101, 118, 138, 153, 180) detectable in nearly every type of sample were used as reference. We included the late eluting Cl8-PCB 194 to extend

the frame of calibration. For the log Kow values of the PCB congeners, three independent sets of data were used for the correlation (Table 5). A correlation of log(tR/tM - 1) with the set 1 of log Kow value (RP-HPLC) of the PCB leads to a linear regression of the form:

y ) 4.87x + 1.91

R 2 ) 0.971

VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

(10.1) 9

2277

TABLE 4. Vapor Pressure of Subcooled Liquids at Standard Conditions - Log p°L and Its Variation Using Equation 1.2 with Various PCBs as Reference Compoundsa ref values -log p°L (Pa) PCB reference HCB HM1,2BP 4,4′-DDE Br4-DE 47 (2,2′,4,4′) Br5-DE 99 (2,2′,4,4′,5) Br5-DE 100 (2,2′,4,4′,6)

PCB 28 C12H5Cl5 -log p°L (Pa)

PCB 101 C12H5Cl5 -log p°L (Pa)

PCB 110 C12H5Cl5 -log p°L (Pa)

PCB 180 C12H3Cl7 -log p°L (Pa)

PCB 194 C12H5Cl5 -log p°L (Pa)

1.59 1.51 1.72

2.37 2.58 2.61 2.75

2.85 2.60 2.81 2.84 2.98

4.09

4.89

3.95 4.09 4.17 4.15

4.83 4.91 4.89

1.72a 2.56b 3.50c; 3.73d 4.17c; 4.75d 4.54d

a p° of the respective PCB is given in the first line (6). The numbering of the PCB and the PBDE congeners follows Ballschmiter and Zell (61) L as summarized by the U.S. EPA (36). The result based on a closely eluting PCB is boldfaced. Reference values of -log p°L: a, ref 53; b, ref 54; c, ref 55; and d, ref 56.

TABLE 5. Calculated Values of Log Kow (GC) Based on Temperature-Programmed GC Retention Data (Figure 1; Chrompack Sil-5/C18) Using Three Independent Sets of Log Kow Values of PCB Congeners Used as Secondary Standardsa PCB

tR (min) GC

RRT (52 + 180)

log(tR/tM - 1) GC

log Kow (RP-HPLC) (14)

log Kow (TSA) (13)

log Kow (selected values) (57)

28 52 101 118 138 153 180 194

26.28 28.57 36.48 42.58 46.24 44.41 52.66 60.56

0.324 0.352 0.449 0.524 0.569 0.547 0.648 0.746

0.749 0.791 0.912 0.987 1.026 1.007 1.088 1.153

5.5 5.8 6.5 6.6 6.7 6.9 7.2 7.6

5.67 5.84 6.38 6.74 6.83 6.92 7.36 7.80

5.80 ( 0.20 6.10 ( 0.20 6.40 ( 0.50 6.40 ( 0.30 7.00 ( 0.50 6.90 ( 0.20 not listed 7.10 ( 0.50

HCB HM1,2BP 4,4′-DDE Br4-DE 47 Br5-DE 99 Br5-DE 100

tR (min) GC

RRT (52 + 180)

log(tR/tM - 1) GC

(1)

22.06 35.83 38.42 52.92 63.30 60.63

0.272 0.441 0.473 0.652 0.779 0.747

0.658 0.904 0.938 1.090 1.174 1.154

5.1 6.3 6.5 7.2 7.6 7.5

calcd values for log Kow (GC) (2) (3) 5.1 6.4 6.6 7.3 7.8 7.7

5.6 6.4 6.5 7.0 7.3 7.2

ref values 5.47a 6.956b 5.9-6.2c 6.6-7.0c 6.6-7.0c

a Independent sets: (1) determined by RP-HPLC (14), (2) calculated from total surface area (TSA) (13), and (3) compiled as selected values by Mackay (57). tM in the HRGC System was 3.977 min. Reference log Kow values: a, ref 58; b, ref 15; and c, 59.

A correlation of log(tR/tM - 1) with the set 2 of log Kow value based on the TSA of the PCB leads to a linear regression of the form:

y ) 5.10x + 1.77

R 2 ) 0.981

(10.2)

A correlation of log(tR/tM - 1) with the set 3 of log Kow value (selected values published by Shiu and Mackay) (57) leads to a linear regression of the form:

y ) 3.27x + 3.44

R 2 ) 0.884

(10.3)

With values for a ) 4.87 and C ) 1.91 (set 1), a ) 5.10 and C ) 1.77 (set 2), and a ) 3.27 and C ) 3.44 (set 3), respectively, the log Kow values of compounds detected in the chromatogram (Figure 1) can be calculated using eq 3. The results are given in Table 5. The value obtained for 4.4′-DDE (log Kow (GC) ) 6.5) corresponds fairly well to the reference value of log Kow ) 6.956 (highest rating) (15). The compilation of Mackay reports a “selected value” of log Kow ) 5.7 for 4,4′-DDE (60). The log Kow (GC) value of 6.3-6.4 given in Table 5 for the heptachlorinated 1′,2-bipyrrole (HM1,2BP/U3/Q1) corresponds well to the calculation made by Vetter (Q1: log Kow 5.9-6.2) (30). For the log Kow (GC) of tetra-brominated diphenyl ether, a range of 5.9-6.2 is given by the WHO (World Health Organization) (59). In the case of tetra-BDE 47, the calculated 2278

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

value of log Kow ) 7.2 exceeds the published value by 1 order of magnitude. The results for the later eluting PBDE congeners show a slightly better conformity with the published range. Our semi-experimental approach to derive approximated physicochemical properties relevant to the environmental distribution presented here can be generally applied to other compounds because GC retention data (e.g., of pollutants of the POP type)seven if their structure is not knownsare easily available.

Acknowledgments We greatly appreciate the gift of a sample extract containing Q1 by Dr. W. Vetter, University of Jena, Germany. We would also like to thank two anonymous reviewers for their helpful comments.

Literature Cited (1) Westcott, J. W.; Bidleman, T. F. J. Chromatogr. 1981, 210, 331336. (2) Mackay, D.; Bobra, A. M.; Chan, D. W.; Shiu, W. Y. Environ. Sci. Technol. 1982, 16, 645-649. (3) Kim, Y.-H.; Woodrow, J. E.; Seiber, J. N. J. Chromatogr. 1984, 314, 37-53. (4) Bidleman, T. F. Anal. Chem. 1984, 56, 2490-2496. (5) Fischer, R. C.; Ballschmiter, K. Fresenius Z. Anal. Chem. 1989, 333 731-732. (6) Fischer, R. C.; Wittlinger, R.; Ballschmiter, K. Fresenius J. Anal. Chem 1992, 342, 421-425.

(7) Kurz, J.; Ballschmiter, K. Chemosphere 1999, 38, 573-586. (8) Fu ¨ hrer, U.; Deissler, A.; Schreitmu ¨ ller, J.; Ballschmiter, K. Chromatographia 1997, 45, 414-427. (9) Pfeifer, O.; Lohmann, U.; Ballschmiter, K. Fresenius J. Anal. Chem. 2001, 371, 598-606. (10) Fischer, R. G.; Ballschmiter, K. Chemosphere 1998, 36, 28912901. (11) Fischer, R. G.; Ballschmiter, K. Fresenius J. Anal. Chem. 1998, 360, 769-776. (12) Miller, M. M.; Ghodbane, S.; Wasik, S. P.; Tewari, Y. E.; Martire, D. E. J. Chem. Eng. Data 1984, 29, 184-90. (13) Hawker, D. W.; Connell, D. W. Environ. Sci. Technol. 1988, 22, 382-387. (14) Brodsky, J.; Ballschmiter, K. Fresenius J. Anal. Chem. 1988, 331, 259-301. (15) Pontolillo, J.; Eganhouse R. P. The Search for Reliable Aqueous Solubility (Sw) and Octanol-Water Partition Coefficient (Kow) Data for Hydrophobic Organic Compounds: DDT and DDE as a Case Study; USGS Water-Resources Investigation Report 014201; U.S. Geological Survey: Reston, VA, 2001. (16) Renner, R. Environ Sci. Technol. 2002, 36, 411A-413A. (17) Weber, H.; Goerke, K. Chemosphere 1996, 33, 377-392. (18) Froescheis, O. Ph.D. Thesis, University of Ulm, 1997. (19) Looser, R. Ph.D. Thesis, University of Ulm, 1999. (20) Vetter, W. Organohalogen Compd. 1999, 43, 433-436. (21) Vetter, W.; Alder, L.; Kalleenborn, R.; Schlabach, M. Environ. Pollut. 2000, 110, 401-409. (22) Vetter, W.; Schlabach, M.; Kallenborn, R. Fresenius Environ. Bull. 2002, 11, 170-175. (23) Wu, J.; Vetter, W.; Gribble, G. W.; Schneekloth, J. S., Jr.; Blank, D. H.; Gorls, H. Angew. Chem., Int. Ed. 2002, 41, 1740-1743. (24) Tittlemier, S. A.; Simon, M.; Jarman, W. M.; Elliott, J. E.; Norstrom, R. J. Environ. Sci. Technol. 1999, 33, 26-33. (25) Gribble, G. W.; Blank, D. H.; Jasinski, J. P. Chem. Commun. 1999, 21, 2195. (26) Tittlemier, S. A.; Blank, D. H.; Gribble, G. W.; Norstrom, R. J. Chemosphere 2002, 46, 511-517. (27) Tittlemier, S. A.; Fisk, A. T.; Hobson, K. A.; Norstrom, R. J. Environ. Pollution 2002, 116, 85-93. (28) Hackenberg, R.; Looser, R.; Froescheis, O.; Beck, H.; Niessner, W.; Jarman, W. M.; Ballschmiter, K. Prepr. Ext. Abstr. ACS Ntl. Meet., Am. Chem. Soc., Div. Environ. Chem. 2001, 41, 128-135. (29) Syracuse Research Corporation. On-line log P calculation; http:// esc.syrres.com/interkow/kowdemo.htm (accessed June 2002). (30) Vetter, W. Organohalogen Compd. 2000, 46, 244-247. (31) Froescheis, O.; Looser, R.; Caillet, G. M.; Jarman, W. M.; Ballschmiter, K. Chemosphere 2000, 40, 651-660. (32) Herington, E. F. G. In Vapor Phase Chromatography; Desty, D. H., Ed.; Butterworth: London, 1957. (33) Westcott, J. W.; Bidleman, T. F. J. Chromatogr. 1981 210, 331336. (34) Hinckley, D. A.; Bidleman, T. F.; Foreman, W. T. J. Chem. Eng. Data 1990, 35, 232-237. (35) Donovan, S. F. J. Chromatogr. A 1996, 749, 123-129. (36) Frame, G. M. Fresenius J. Anal. Chem. 1997, 357, 701-713. (37) U.S. Environmental Protection Agency. PCB ID Homepage; http://www.epa.gov/toxteam/pcbid/ (accessed June 2002). (38) Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistry and Biology; Wiley: New York, 1979. (39) Collander, R. Acta Physiol. Scand. 1947, 13, 363-381. (40) Collander, R. Acta Chem. Scand. 1950, 4, 1085-1098.

(41) Lyman, W. J. Estimation of physical properties. In Environmental Exposure from Chemicals, Vol. 1; Brook, N. W., Blau, G. E., Eds.; CRC Press: Boca Raton, FL, 1985. (42) James, K. C. Solubility and Related Properties; Dekker: New York, 1986. (43) Connell, D. W. Evaluation of the bioconcentration factor, biomagnification factor, and related physicochemical properties of organic compounds. In Bioaccumulation of Xenobiotic Compounds; Connell, D. W., Ed.; CRC Press: Boca Raton, FL, 1990; pp 10-45. (44) Lyman, W. J.; Reeh, W. F.; Rosenblatt, D. H. Handbook of Chemical Property Estimation Methods; American Chemical Society: Washington, DC, 1990. (45) Isnard, P.; Lambert, S. Chemosphere 1989, 18, 1837-1853. (46) Chiou, C. T.; Schmedding, D. W.; Manes, M. Environ. Sci. Technol 1982, 16, 4-10. (47) Kenaga, E. E.; Goring, C. A. I. Relationship between Water Solubility, Soil Sorption, Octanol Water Partitioning, and Bioconcentration of Chemicals in Biota. Presented at the American Society for Testing and Materials, Third Aquatic Toxicology Symposium, October 17-18, 1978, New Orleans, LA. (48) Chiou, C. T.; Freed, V. H.; Schmedding, D. W.; Kohnert, R. L. Environ. Sci. Technol. 1977, 11, 475-478. (49) Opperhuizen, A.; van der Velde, E. W.; Gobas, F. A. P. C.; Liem, D. A. K.; van der Steen., J. M. D. Chemosphere 1985, 14, 18711896. (50) International Programme on Chemical Safety. Concise International Chemical Assessment Document 34: Chlorinated Naphthalenes; WHO: Geneva, 2001. (51) Burreau, S.; Axelman, J.; Broman, D.; Jakobsson, E. Environ. Toxicol. Chem. 1997, 16, 2508-2513. (52) Lei, Y. D.; Wania, F.; Shiu, W. Y.; Boocock, D. G. B. J. Chem. Eng. Data 2000, 45, 738-742. (53) Farmer, W. J.; Yang, M. S.; Letey, J.; Spencer, W. F. Soil Sci. Soc. Am. J. 1980, 44, 676-680. (54) Paasivirta, J.; Sinkkonen, S.; Mikkelson, P.; Wania, F. Chemosphere 1999, 39, 811-832. (55) Wong, A.; Lei, Y. D.; Alaee, M.; Wania, F. J. Chem. Eng. Data 2001, 46, 239-242. (56) Tittlemier, S.; Stern, G.; Tomy, G. Environ. Toxicol. Chem. 2002, 21, 1804-1810. (57) Shiu, W. Y.; Mackay, D. J. Phys. Chem. Ref. Data 1986, 15, 911929. (58) Chessels, M.; Hawker, D. W.; Connel, D. W. Chemosphere 1991, 22, 1175-1190. (59) International Program on Chemical Safety. Polybrominated Diphenyl Ethers; Environmental Health Criteria 162; WHO: Geneva, 1994. (60) Mackay, D.; Shiu, W. Y.; Ma, K. C. Illustrated Handbook of Physical-Chemical Properties and Environmental Fate for Organic Compounds. Vol. 5. Pesticide Chemicals; Lewis Publishers: Boca Raton, FL, 1997. (61) Ballschmiter, K.; Zell, M. Fresenius Z. Anal. Chem. 1980, 302, 20-31. (62) Ballschmiter, K.; Bacher, R.; Mennel, A.; Fischer, R. C.; Riehle, U.; Swerev, M. J. High Resolut. Chromatogr. 1992, 15, 260-270.

Received for review July 4, 2002. Revised manuscript received February 24, 2003. Accepted February 26, 2003. ES0201294

VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2279