Determination of Blood Alcohol
“Dry Lab”
Introduction
Quantitative determination of blood alcohol (BAC) is one of the most common analyses performed in the forensic toxicology laboratory. GC with FIR (flame ionization detection) is the preferred technique.
Blood samples must be drawn by medical personnel and transported to the lab, where they are stored in a refrigerator. The blood tubes used for collection generally contain sodium fluoride, an anti-glycolytic, which inhibits enzyme reactions with glucose (recall that fermentation of glucose can produce alcohol!). Sample preparation is minimal and will vary by laboratory. Nearly all labs, however, will require use of an internal standard (discussed further below).
Quantitative analysis by GC is typically done using an autosampler. Even with this, the very small sample injection volumes and potential small changes in instrumental conditions, such as gas flow, might introduce other variations. One method commonly used to compensate for these difficulties is the use of internal standards.
The internal standard method involves spiking an exactly known quantity of a substance into every sample and standard. The area of the internal standard and the area of the analyte are determined, and then a ratio of these two is calculated by dividing the area of the analyte peak by the area of the internal standard peak. The result is called a peak area ratio (PAR). The idea is that even though the peak areas for a given sample may vary from one test to the next due to injection differences or instrumental variations, the ratio of the two peaks will be constant, since the variations will affect both substances equally. Then, when preparing the calibration curve, the PAR is plotted on the y-axis rather than the simple peak area.
A good internal standard has the following characteristics:
ü Yields a peak that is well resolved from other peaks
ü Has a retention time close to the analyte’s retention time
ü Normally, some structural similarity between the IS and analyte is desirable
ü Is a compound not readily available to the public and this not typically ingested
o This can be confirmed by using two Internal Standards (e.g., N-Propanol and Isobutanol)
§ The area ratio between the two internal standards should be a constant
There are two ways of introducing an internal standard into the analysis. One way is to dissolve the internal standard into the solvent used to dilute both samples and standards. A second way is to add an accurate and precise volume of concentrated internal standard solution to the samples and standards. The Internal Standard solution may contain salt (i.e., sodium chloride) to enhance the headspace analyses. Addition of salt reduces the solubility of alcohol in an aqueous solution.
For blood alcohol quantitative determinations, n-propanol is a commonly used internal standard. In the data set below, you are provided with peak areas for both ethanol and propanol. When you make the calibration curve, plot the PAR vs the concentration.
Quality control samples (QC) are a critical component of a forensic BAC. QC samples typically include:
Negative control (contains no ethanol)
Positive control (contains a known amount of ethanol)
Quality control requirements typically include:
Agreement between the calculated value found for a control and its true, assigned value
Agreement between the calculated ethanol concentrations in duplicate samples
Agreement between the retention times of calibrator and sample ethanol peaks and internal standard peaks
For this “dry lab” you will be given a set of data from a BAC run. You will need to determine the PAR, plot a calibration curve, determine the concentrations of the unknowns, and calculate the QC results to be sure they are within the pre-established limits.
In a real world forensic BAC run, you would have two separate columns running these samples simultaneously. You would also determine acetone, isopropanol, and methanol. For this dry lab, however, we will only do calculations for ethanol on one column.
Concentration (g ethanol / 100 mL)
Peak area for ethanol
Retention time ethanol, min
Peak area for n-propanol (IS spiked at 0.05g/dL)
Retention time n-propanol min
Blank
0
1259
1.141
25649
1.991
Calibrator 1
0.0100
3975
1.139
24507
1.898
Calibrator 2
0.0500
21876
1.137
25565
2.010
Calibrator 3
0.0800
37561
1.140
24610
1.995
Calibrator 4
0.100
46003
1.142
24368
1.997
Calibrator 5
0.300
133987
1.138
25117
2.001
Calibrator 6
0.500
221397
1.140
24947
1.993
Negative control 1
0
3165
1.145
25387
2.011
Positive control 1
0.0800
35587
1.135
24991
1.993
Blood sample 1
?
67590
1.141
25619
1.990
Blood sample 1 duplicate
?
70345
1.138
25116
1.891
Blood sample 2
?
98171
1.144
24819
1.993
Blood sample 2 duplicate
?
110786
1.137
25038
2.011
Negative control 2
0
1590
1.143
25437
1.995
Positive control 2
0.300
146723
1.144
24751
1.996
Report Requirements
Calculations
For each question, show the calculations for at least one example. You may write-out rather than type the calculations, if you wish.
1. Use the calibrator data to make a calibration curve using Excel. You should be able to copy and paste that data table into Excel. Remember to use the PAR, not just the ethanol peak area. Plot PAR on the on the y-axis and ethanol concentration on the x-axis. Use Excel to get a trendline (linear least squared fit to the data) and an r2 value for the calibration curve. Attach the graph to the report. (2 pts)
2. The standard operating procedure (SOP) that you are using says that the r2 value for the calibration curve must be greater than 0.995. Does your curve meet that requirement? (0.5 pts)
3. Use the trendline equation to calculate the ethanol concentrations for negative control 1 and 2. (1 pt)
4. The SOP in your lab states that the QC limit for the negative control is less than 0.0025%. (1 pt)
a. Are both of the negative controls within QC limits?
b. Suppose that your SOP says that you can’t report data if the limit is exceeded. Would you be able to report data from this run?
5. Use the trendline to calculate the ethanol concentrations for positive control 1 and 2. (1.5 pts)
6. Calculate the % error for both positive controls. (See data table for the “true” concentrations). (1.5 pts)
7. The SOP states that the positive controls must be within 5% of the “true” concentration. If they aren’t, you should re-run the batch. Can you report the data from this batch? (0.5 pt)
8. Calculate the ethanol concentrations for the four injections of Blood samples. (1.5 pts)
9. Calculate the agreement between each Blood sample and its duplicate as relative percent difference from the average (RPD). (1.5 pts)
10. Suppose your SOP requires that the RPD must be +/- 5%. Do your samples meet this requirement, and could you report this data? (0.5 pts)
11. Calculate and list the average value of the retention time for the ethanol and n-propanol for the 6 calibrators. (1pt)
12. Suppose your SOP requires that the retention times for the ethanol and n-propanol in the samples is within ±3% of the average retention time for those substances in the calibrators. Do your 4 Blood sample injections meet that requirement? Briefly justify your answer. (1.5 pts)
Questions
13. The per se limit for DUI is 0.08%. Explain what the percent unit actually means in blood alcohol testing. Is this on a weight/weight basis, a volume/volume basis, a volume/weight basis or a weight/volume basis? Would it make a difference if the limit were on a different basis? (1 pt)
14. Is the unit g ethanol/dL blood the same unit as 0.08%? What volume in is 1 dL in units of milliliters? (1 pt)
15. Were any of the samples in this dry lab data above the legal limit? If so, which ones? (1 pt)
16. The actual columns often used for BAC are proprietary. Manufacturers provide examples of operating conditions that demonstrate how well the columns separate mixtures, but disclose little information on the composition of the columns. Even though you don’t know the exact makeup of the stationary phase in the columns, would you expect them to be polar or non-polar? Explain your answer. (1 pt)
17. Chromatographic conditions for BAC are generally isothermal. Briefly explain the term isothermal in the context of GC-FID. What step in a temperature gradient GC-FID run can be omitted in an isothermal GC-FID run? Why are isothermal methods preferred for BAC? (2 pts)
Report: Title, Name, Questions answered in order and email of spreadsheet to instructor. (1 pt)
