Environmental Fate and Transport of Solvent-Stabilizer Compounds - Environmental Investigation and Remediation - page 153

TABLE 3.15
Index of Biodegradability According to BOD 5 /COD Ratio
BOD 5 /COD Rating Example Compounds

<0.01 Not degradable 1,4-Dioxane, propylene oxide, morpholine, triethanolamine

0.01-0.1 Moderately degradable n -Butyl acetate, acetonitrile

>0.1 Relatively degradable n -Propyl alcohol, n -butyl alcohol, pyridine, n -amyl alcohol, phenol

Source: Scow, K.M., 1982 Handbook of Chemical Property Estimation Methods , pp. 9-1-9-85. Washington, DC: American
Chemical Society.

i rst-order kinetics follows a linear relationship between concentration reduction by microbial degra-
dation and time, whereas the second-order kinetics follows a hyperbolic relationship (Scow, 1982).
A screening-level index of the biodegradability of organic compounds uses the ratio of biochemi-
cal oxygen demand from a 5-day test (BOD 5 ) to chemical oxygen demand (COD) according to the
classii cation scheme shown in Table 3.15.

3.3.4.1 Predicting Biodegradability
Most of the thousands of organic compounds introduced to the marketplace and subsequently released
to the environment are subjected to only screening-level tests for biodegradability. Reliable and
consistently generated biodegradation data are not widely available. A goal in the assessment of the
environmental persistence of organic compounds is the ability to use biodegradation models to predict
whether the compounds will be mineralized by ambient soil microbes. Databases of consistently
developed biodegradation data—such as the Japanese Ministry of International Trade and Industry
(MITI) Test database, which includes more than 800 compounds—are gradually overcoming this
limitation to biodegradation modeling, and predictions of biodegradability are improving (Pavan and
Worth, 2006). One such biodegradation model is BIOWIN, a commonly used model developed by
Syracuse Research Corporation and available as part of the EPIWIN Suite (USEPA, 2007a).
BIOWIN is an algorithm that predicts the potential for aerobic biodegradation from a compound's
molecular structure and molecular weight. High-molecular-weight compounds are less prone to aero-
bic biodegradation. The BIOWIN classii cation of chemical compounds is based on the presence of
structural fragments in a molecule that are also present in compounds whose biodegradability is
known. A training set of 200 compounds whose biodegradation fates are well studied is used to pre-
dict the biodegradation potential for compounds that include the same structural fragments (Boethling
et al., 1994). For example, ester, alcohol, and carboxylic acid groups usually enhance biodegradabil-
ity, whereas the presence of halogens and nitro groups generally increases resistance to biodegrada-
tion (Scow, 1982). Alcohols, aldehydes, acids, esters, and amides are more susceptible to biodegradation
than are the corresponding alkanes, olei ns, ketones, dicarboxylic acids, nitriles, amines, and chloro-
alkanes. Ether functions are particularly resistant to biodegradation because of the high dissociation
energy of the ether linkage (about 360 kJ/mol) (Kim and Engesser, 2004). Highly branched com-
pounds are also more resistant to biodegradation than are straight chains, and short chains are not as
quickly degraded as long chains. Halogenated compounds that resist aerobic degradation may be
more rapidly degraded under anaerobic conditions (Aronson et al., 1999). Unsaturated compounds
(alkenes or alkynes) are more susceptible to biodegradation than saturated compounds (alkanes).
Table 3.16 lists output from the BIOWIN model for predicting biodegradability of solvent stabilizers
and chlorinated solvents, and Table 3.17 gives ratings for biodegradability interpreted from literature
studies and collated in the Syracuse Research Corporation's BIODEG database (SRC, 2007a).
Because of imperfections in the prediction algorithm, estimates of biodegradability predicted by
using BIOWIN must be used conservatively. An evaluation of the reliability of BIOWIN predictions
determined that the prediction “ not readily degradable ” is highly accurate (correctly predicted for more