Sampling and Laboratory Analysis for Solvent Stabilizers - Environmental Investigation and Remediation - page 217

TABLE 4.9
Laboratory Performance for 1,4-Dioxane Analyses by EPA Method 8261, Vacuum
Distillation (VD/GC/MS)

Spike
Concentration
(

Average
Recovery
(%)

Relative
Standard
Deviation a

Sample
Size

Number
of Samples

Method Detection
Limit (Estimated) Lab Matrix μ g/L)

1 5 g Soil 8 10 186 16.7 9.4 μg/kg

2 5 g Soil 7 10 87 20.9 5.7 μg/kg

1 5 mL Water 8 20 79 24.9 11.6 μg/L

2

5 mL Water 7 10 113 17.9 6.3 μg/L

2 25 mL Water 7 4 113 17.7 2.5 μg/L

Sample Matrix and
1,4-Dioxane
Concentration (ppb)

Average Recovery at Different Concentrations Relative

Standard
Deviation a

Number of
Samples Lab Low (%) Medium (%) High (%) Overall (%)

1a 9 Water: 1, 10, and 50 128 103 114 116 6.4

2b 12 Water: 1, 3, and 5 327 145 178 217 13.9

1c 9 Soil: 1, 10, and 50 137 103 122 121 21.0

2d 12 Soil: 1, 3, and 5 156 127 166 150 21.0

Matrix of Five Blind
Samples (5 mL)

Average Relative Standard
Deviation a Spike Level ( μ g/L) Average Recovery (%)

375 95 11.4

Water

Water 125 111 7.8

Water 55 102 13.6

Water 7 196 24.2

Water 1 660 24.0

Source: Strout, K., et al., 2004a, Vacuum distillation unit interlaboratory study evaluation. Paper presented at the 20th Annual
National Environmental Monitoring Conference, July 19-22, Independent Laboratory Institute, Washington, DC.
http://nemc.us/proceedings/2004proceedings.pdf (accessed January 2005). With permission.
a Lower relative standard deviation (RSD) values rel ect increased precision. Recovery of volatilized 1,4-dioxane may be
diminished by condensation in the apparatus (Hiatt, 2007).

In vacuum distillation, the effect of boiling point and relative volatility on the internal standards
must be addressed by correcting the GC-MS response. Surrogate recovery is determined to predict
the recoveries of analytes within surrogate groups on the basis of relative volatility * and boiling
point. Once the recoveries of the internal standards have been corrected for boiling-point effects,
recoveries then rel ect only the matrix effects owing to relative volatility. Relative volatility is also
corrected by using a software algorithm for processing the data generated from EPA Method 8261A
(USEPA, 2006d). Method 8261A does not list a lower limit of quantitation for 1,4-dioxane. An
estimated MDL for 1,4-dioxane on a 25 mL aliquot of water is 2.5 μg/L, whereas the routine injec-
tion of 5 mL of sample produces an estimated MDL or 6.3 μg/L (Strout et al., 2004a). Table 4.9
presents data from laboratory performance studies for EPA Method 8261A.

* “Relative volatility” describes the distillation of one compound relative to that of another compound as a ratio of their
respective partition coefi cients.