The authors refer to "the assumed 2100:1 ratio of breath to blood alcohol" (emphasis added). In fact, this should be the "2100:1 ratio of blood- to breath–alcohol", commonly referred to as the blood/breath ratio, or BBR (2). This distinction is significant because the authors' wording indicates that the concentration of alcohol in the breath is 2100 times greater than the concentration of alcohol in the blood. In fact, the 2100:1 ratio—which is the basis for the operation of all breath–alcohol analyzers used by law enforcement agencies in the U.S. regardless of whether test results are reported in terms of blood–alcohol or breath–alcohol concentrations (3)—means that the blood–alcohol concentration estimated from a breath–alcohol analysis (BAC_Est.) is 2100 times greater than the corresponding measured breath–alcohol concentration (BrAC_Meas.). Equation 1 reflects this fact, and from it we obtain the expression, 2100 = BAC_Est./BrAC_Meas., which confirms the argument presented above.

I agree with the authors' statement concerning breath–alcohol tests that, based on Henry's law, "the concentration of ethanol in the blood is assumed to be directly proportional to the partial pressure of ethanol in the air above the blood". They provide eq 2 in this regard, specify k as the Henry's law constant, and add that, "if a person's breath alcohol concentration is measured, and the Henry's law constant is known at breath temperature (34 °C), then the concentration of ethanol in a person's blood can be calculated".

The latter argument is misleading. It is not the Henry's law constant that is used to convert the results of breath–alcohol analyses (BrAC_Meas., or the authors' [ethanol]_air) into corresponding values of BAC_Est. (or the authors' [ethanol]_blood) at an average alveolar air temperature of 34 °C, but rather the 2100:1 BBR, in accord with eq 1. It can be demonstrated via the following derivation, consistent with Thompson's analysis (4), that the BBR, which has units of "mL_breath/mL_blood" (2), is equal to k_EtOHRT, where k_EtOH is the Henry's law constant.

In the general Henry's law equation, C = kP_gas, C is the concentration of the gaseous solute in solution, P_gas is the partial pressure of the gas above the solution, and k is the Henry's law constant. In fact, this equation is consistent with the authors' statement cited above concerning Henry's law. Within the context of breath–alcohol analysis, C is BAC_Est., or "[ethanol]_blood"; k is k_EtOH; and P_gas is P_EtOH, which is given by eq 3.

(3)

Since n_EtOH/V_EtOH = BrAC_Meas., or "[ethanol]_air", eq 3 can be expressed in the form of eq 4.

The general Henry's law equation can now be written in the form of eq 5, which incorporates the information provided by eq 4 and mirrors eq 1. It should also be noted that, with k_EtOH having units of "mol_EtOH/mL_blood atm", R having units of "mL_breath atm/mol_EtOH K", and T having units of "K", the expected BBR units of "mL_breath/mL_blood" are obtained.

(5)

Thus, the Henry's law constant, k_EtOH, is not the BBR, as implied by Kniesel and Bellamy (1) via their eq 2 and as they state in their Supplemental Material, but rather a contributor to the BBR factor.

The analytical wavelength used by Kniesel and Bellamy (1) in their experiments, namely 1053 cm^–1 (9.5 μm), is not ethanol-specific, as I am sure the authors would agree. In this regard, it should be noted that this wavelength is characteristic of the carbon–oxygen stretching vibration of ethanol. Moreover, it lies within the 1111 cm^–1–1000 cm^–1 (9–10μm) IR wavelength region where other common, volatile organic compounds having carbon–oxygen functionality absorb, including other alcohols, esters, and ethers that occur, for example, in solvents, perfumes, and food (5). Given concerns over this lack of specificity for ethanol, Draeger Safety (6) developed the Alcotest 7110 MK III breath–alcohol analyzer which employs a dual analytical protocol. That is, it relies not only on IR analysis at 1053 cm^–1 (9.5 μm) but also on electrochemical fuel cell analysis to minimize the specificity problem.

A final point of clarification concerns the predicted time of 6.1 minutes determined by the authors that would typically elapse before a person's apparent blood–alcohol concentration stemming from mouth–alcohol contamination would drop to the "legal limit" they cited, namely 0.08%, which exists in many states. It is also important to recognize that more time is required for apparent BACs attributable to mouth-alcohol contamination to drop to zero (7–9). Thus, many jurisdictions, such as New York, require a 20-minute observation period before a breath test is administered to a driving-while-intoxicated (DWI) arrestee. This is a significant consideration because mouth–alcohol contamination not only impacts on test results involving DWI, but also on those results that fall within the driving-while-alcohol-impaired (DWAI) range. In New York, that range is 0.06%–0.07%.

Literature Cited

1. Kniesel, A.; Bellamy, M. K. J. Chem. Educ. 2003, 80, 1448–1450.
2. Labianca, D. A. J. Chem. Educ. 2002, 79, 1237–1240.
3. Labianca, D. A.; Simpson, G. Eur. J. Clin. Chem. Clin. Biochem. 1995, 33, 919–925.
4. Thompson, R. Q. J. Chem. Educ. 1997, 74, 532–536.
5. Labianca, D. A. J. Anal. Toxicol. 1992, 16, 404–405.
6. Draeger Safety, Inc., Breathalyzer Division, 185 Suttle Street, Durango, CO 81301-7911.
7. Spector, N. H. Science 1971, 172, 57–59.
8. Harding, P. M.; McMurray, M. C.; Laessig, R. H.; Simley, D. O., II; Correll, P. J.; Tsunehiro, J. K. J. Forensic Sci. 1992, 37, 999–1007.
9. Swope, R. S. DWI J.: Law & Sci. 1995, 10 (9), 5–8.