Modeling the role of metabolic intermediates in kinetics of
phenolbiodegradation
Si-Jing Wang, Kai-Chee Loh*Department of Chemical and
Environmental Engineering, National University of Singapore, 10
Kent Ridge Crescent, S119260 Singapore, Singapore
Received 23 September 1998; received in revised form 29 January
1999; accepted 3 February 1999
Abstract
The kinetics of phenol biodegradation by Pseudomonas putida ATCC
49451 in batch cultures were investigated over a wide range
ofinitial phenol concentrations (25800 mg/l). Although the Haldane
equation could model specific growth rate as a function of initial
phenolconcentrations very well, it was found inadequate to describe
phenol degradation profiles, especially for cultures containing
high initialphenol concentrations (e.g. 800 mg/l). This was
attributed to the inhibition of metabolic intermediates of phenol
degradation and the variablecell mass yield. Consequently, a new
phenol degradation model was proposed. By incorporating the
inhibition effects of metabolicintermediates, the new model
successfully simulated phenol degradation profiles in the entire
range of initial phenol concentrations studiedby using only one set
of model parameters. Based on a comparison of the new model with
the conventionally used Haldane equation, it isconcluded that the
inhibition of metabolic intermediates plays a crucial role in
phenol degradation modeling, especially over a wideconcentration
range of phenol. 1999 Elsevier Science Inc. All rights
reserved.
Keywords: Phenol biodegradation; Cell growth; Cell mass yield;
Kinetic model; Metabolic intermediates; Inhibition; Haldane
equation
1. Introduction
Phenol and phenolic compounds are of widespread use inmany
industries such as polymeric resin production and oilrefining. As a
result, these compounds are commonly en-countered in industrial
effluents and surface water. Thesepollutants are usually treated in
activated sludge processesbecause many aerobic bacteria and fungi
are able to usephenol as a source of carbon and energy.
Biodegradation ofphenol, therefore, has long been the subject of
numerousinvestigations [16].
Cell growth on phenol has been observed to displaysubstrate
inhibition phenomena at high phenol concentra-tions, and the
Haldane equation is often used to describe cellgrowth on phenol
either by pure [2,3] or mixed cultures[1,7]. The maximum specific
growth rate obtained has beenreported to be in the range of
0.130.36/h for mixed cul-tures [1,7] and 0.290.90/h for pure
cultures [3,8]. The Ksand Ki parameters are also distributed over a
wide range,depending on cell type and culture environments [9].
Thespecific phenol degradation rate has also often been mod-
eled by using the Haldane equation [4,10]. In some cases,the
substrate removal rate was coupled to the cell growthrate with a
constant yield coefficient [3,8]. This assumptionis clearly valid
only within a very narrow range of initialphenol concentrations.
Allsop et al. [5] investigated thedynamics of phenol degradation in
continuous cultures sub-ject to step increases in phenol feed
concentration. Theyfound that the yield of biomass on phenol varied
from 0.73to 0.14 g/g during the transient stage. Variations in cell
massyield can also be expected when cells grow very slowlyduring
substrate inhibition [11]. These suggest that the as-sumption of a
constant cell yield coefficient should be usedwith caution when the
specific substrate degradation rate ismodeled as being directly
related to the specific cell growthrate.
Besides substrate inhibition, variation in the cell massyield
can also be attributed to the accumulation of
metabolicintermediates and their consequential inhibition on
substrateconsumption [5,11]. The typical metabolic pathway for
phe-nol degradation occurs via a catechol derivative before
ringcleavage through an ortho- or meta-oxidation [2]. In thecase of
P. putida, phenol is mineralized through the metapathway [2,12].
The production and accumulation of meta-bolic intermediates during
phenol degradation has been* Corresponding author. Tel.:
165-874-2174; fax: 165-779-1936.
Enzyme and Microbial Technology 25 (1999) 177184
0141-0229/99/$ see front matter 1999 Elsevier Science Inc. All
rights reserved.PII: S0141-0229(99)00060-5
commonly observed [5,1213]. Among these, 2-hy-droxymuconic acid
semialdehyde (2-HMAS), the first prod-uct of catechol ring opening
in the meta-pathway degrada-tion of phenol, has been reported to be
responsible for thecolor change of the medium (from colorless to
greenishyellow) during phenol degradation [12,13]. Morsen andRehm
[12] monitored the concentration changes of the in-termediate in
relation to phenol removal, and found that theconcentration of
2-HMAS reached a maximum when phe-nol was just about completely
depleted. Allsop et al. [5] alsoobserved accumulation of
intermediates during degradationof phenol and investigated their
inhibition effect on phenoldegradation. They suggested that the
Haldane equation wasinadequate for describing the dynamics of the
phenol de-grading system because the effect of metabolic
intermedi-ates was not accounted for in the Haldane equation, but
analternate quantitative model has not been proposed. As aresult,
the Haldane equation could simulate phenol degra-dation profiles
when applied to a wide range of phenolconcentrations only when
different sets of model parameterswere used. For example, Saez and
Rittmann [10] obtainedquite different sets of substrate inhibition
parameters fordifferent initial phenol concentrations.
In this article, a clarification and quantitative discussionof
the relationship between specific growth rate and sub-strate
consumption rate, a consideration of variable cellmass yield as
well as the role of metabolic intermediates ofphenol will be
presented. Based on these, a new kineticmodel is proposed. The
model will be validated experimen-tally over a wide range of
initial phenol concentrations (25; 800 mg/l). In addition,
comparison of the model with theconventional Haldane model for
phenol degradation will bepresented and discussed.
2. Kinetic model for degradation of phenol
2.1. Relationship between specific growth rate andsubstrate
consumption rate
Based on a material balance for substrate in a batchcultivation
(ignoring substrate consumption for productionsynthesis and
maintenance), the specific substrate consump-tion rate can be
expressed as:
qs 5 2dSXdt 5
m
Y(1)
Eq. (1) is often used to develop the specific
substrateconsumption rate by the substitution of an
establishedmodel for the specific growth rate (m) [3,78,1416].
In the case of cell growth modeled by the Haldaneequation (as in
the case of phenol) [13,78] where
m 5mmS
KS 1 S 1 S2/Ki(2)
Nomenclature
Kh Inhibition coefficient of metabolic intermediates(mg/l)
Ki Inhibition constant for cell growth (mg/l)Ki9 Inhibition
constant for substrate consumption
(mg/l)Kp Proportionality constant (mg/l)KS Saturation constant
for cell growth (mg/l)KS9 Saturation constant for substrate
consumption
(mg/l)ks Saturation constant for substrate consumption
(mg/l)qs Specific utilization rate of substrate (mg/
(mg 3 h))Rm Maximum specific consumption rate of substrate
(mg/(mg 3 h))Rm9 Maximum specific consumption rate of
substrate
(mg/(mg 3 h))S Substrate concentration (mg/l)S0 Initial
substrate concentration (mg/l)t Time (h)X Biomass concentration
(mg/l)X0 Initial biomass concentration (mg/l)Y Observed cell mass
yield (g/g)YC Theoretical cell mass yield on phenol (g/g)YE Yield
of cell mass on phenol for energy (g/g)m Specific growth rate (per
h)
if the initial substrate concentration is well above thecritical
substrate concentration where m is maximum, i.e.S0 .. =KSKi, as
substrate is being consumed (decreasingS), m is expected to
increase in a batch culture! However,cells in batch culture are
often observed to grow at a con-stant specific growth rate in the
exponential phase followedby deceleration growth and stationary
phase [5,17]. This isalso supported by our experimental data of
cell growthprofiles over a wide range of initial phenol
concentrations inthis work (data not shown). This appears to be
contrary tothe observations reported by DAdamo et al. [7] with
aheterogeneous culture on phenol. In their case, growth rateon
phenol was found to accelerate with culture time. This,however, may
be due to the fact that the microbial popula-tion was heterogeneous
and complex. So in Eq. (2), thesubstrate concentration S should
really be the initial sub-strate concentration S0 for batch
cultures [3]. Otherwise, themere substitution of the growth model
of Eq. (2) into Eq.(1), particularly for S0 .. =KSKi, will
erroneously modelsubstrate consumption, especially when the
substrate hasbeen consumed to a significant extent. Furthermore, in
usingEq. (1), it has been customary to assume that the yield
coeffi-cient Y is an average constant. This, again, is not always
true,especially at extremes of specific growth rate [11].
178 S.-J. Wang, K.-C. Loh / Enzyme and Microbial Technology 25
(1999) 177184