Final Report

College of Engineering - Department of Chemical Engineering

CHE 491: Senior Design 2

Spring 2015

EFFECT OF KINETIC INTERACTIONS ON BIOREACTOR PERFORMANCE

IN WASTE TREATMENT APPLICATIONS

(FINAL REPORT)

Group Members

Name ID#

Mahren Masud 38975

Azka Vicar Mir 39786

Fahmida Anwar 39982

Reshma Sulthana Sherif 42501

Anisul Karim 44246

Sonam Ludhani 49634

Submission Date: 26th May 2015

Advisor: Dr. Zarook Shareefdeen

Coordinator: Dr. Hussain Ahmed

1

Executive Summary

An inlet stream to a waste treatment plant has various pollutants. In a biodegradation method of

treatment microorganisms use these pollutants as their source of carbon. This project was

designed to investigate the effects of these pollutants interacting with each other. Three systems

have been studied, namely benzene-toluene, toluene-phenol and benzene-phenol. Furthermore

the effect of these kinetic interactions on the overall bioreactor size and economics have also

been determined. For the benzene-toluene system the competitive interaction in the binary

mixture lead to an overall 80% increase in the reactor volume and costs increased by up to 30%

which are deemed as significant.

2

Table of Contents Executive Summary ...................................................................................................................................... 1

List of Figures ............................................................................................................................................... 4

Introduction ................................................................................................................................................... 5

Background ................................................................................................................................................. 8

Wastewater Treatment .............................................................................................................................. 8

Bioreactors ................................................................................................................................................ 9

Bacterial Strains ...................................................................................................................................... 10

Bioreactor and Kinetics ........................................................................................................................... 10

Pollutants ................................................................................................................................................ 10

Kinetics ................................................................................................................................................... 12

Monod Kinetics ................................................................................................................................... 12

Andrews Kinetics ................................................................................................................................ 13

SKIP Model ........................................................................................................................................ 14

Methodology ............................................................................................................................................... 15

MATLAB ................................................................................................................................................ 15

MATLAB for modeling substrate and biomass behavior without interaction .................................... 15

MATLAB for modeling substrate interaction with biomass and each other for interaction ............... 16

PolyMath ................................................................................................................................................. 17

Literature Review .................................................................................................................................... 17

Results ......................................................................................................................................................... 20

Benzene and Toluene .............................................................................................................................. 20

PPO1 ................................................................................................................................................... 20

Consortium .......................................................................................................................................... 22

Toluene and Phenol ................................................................................................................................. 24

Benzene and Phenol ................................................................................................................................ 26

Design ......................................................................................................................................................... 28

Equipment Sizing and Material of Construction ..................................................................................... 28

Tank 1 (Holding/Storage Vessel) – (T-101) ....................................................................................... 28

Tank 2 (Neutralization Tank) – (T-102) ............................................................................................. 29

Reactor- (R-101) ................................................................................................................................. 30

Settling Tank- (T-103) ........................................................................................................................ 33

3

Air Stripper-(T-104) ............................................................................................................................ 37

Benzene Data .......................................................................................................................................... 37

Adsorber-(T-105) ................................................................................................................................ 41

Pumps .................................................................................................................................................. 43

Piping .................................................................................................................................................. 44

Control ........................................................................................................................................................ 45

Costing ........................................................................................................................................................ 47

Reactor Size ................................................................................................................................................ 48

Data ............................................................................................................................................................ 48

Flow rate .................................................................................................................................................... 48

1000 gal/day ................................................................................................................................................ 48

Biomass ...................................................................................................................................................... 48

2000 mg/L ................................................................................................................................................... 48

Inlet Concentration of pollutants............................................................................................................. 48

500 mg/L ..................................................................................................................................................... 48

HSE- Health, Safety and Hazard ................................................................................................................. 49

HAZOP ................................................................................................................................................... 49

UAE Regulations .................................................................................................................................... 53

Economic Analysis ................................................................................................................................. 54

Conclusion .................................................................................................................................................. 56

Future Work ............................................................................................................................................ 56

References ................................................................................................................................................... 58

Appendix ..................................................................................................................................................... 61

4

List of Figures

Figure 1: Plant Layout .................................................................................................................................. 6

Figure 2: Initial Substrate Conc. vs RE% (B-T) using PPO1 ..................................................................... 20

Figure 3: Graph of initial substrate conc. vs. removal efficiency using PPO1 – interaction ...................... 21

Figure 4: Graph of initial substrate conc. vs. removal efficiency using consortium – no interaction ......... 22

Figure 5: Graph of initial substrate concentration vs. removal efficiency using consortium – interaction 23

Figure 6: Graph of initial substrate concentration vs. removal efficiency– interaction .............................. 25

Figure 7: Graph of initial substrate concentration vs. removal efficiency– interaction .............................. 27

Figure 8: An ideal rectangular sedimentation tank illustrating the settling of discrete particles. [1] ......... 33

Figure 9 – Pressure Drop Correlation Curve .............................................................................................. 41

Figure 10: Plant layout with Control and Instrumentation.......................................................................... 46

Figure 11: Bare Module Cost of Plant using CAPCOST ............................................................................ 47

5

Introduction

This project aims to study the effect of kinetic interaction on reactor sizing, economics

and overall performance of waste treatment. Feed streams with multiple pollutants are fed to bi-

reactors for treatment. These bio-reactors contain microorganisms in order to feed on organic

matter, in this case the organic pollutants in industrial wastewater. When two pollutants enter a

bioreactor, bacteria may degrade one pollutant more over the other. These kinetic interactions

were studied in order to size the bioreactor. Monod, Andrews and SKIP models were employed

and parameters from literature [5] were obtained in order to design a continuous stirred tank

reactor (CSTR) model. A general plant layout was constructed and each equipment was sized to

estimate the overall plant cost. HAZOP was also performed on each equipment. Moreover,

comparisons of removal efficiencies and volume for interaction and non-interaction were

established. Initially, a benzene-toluene system was thoroughly examined where removal

efficiencies in interaction and non-interaction system was analyzed. Subsequently, different

pollutant systems were studied like toluene- phenol and benzene-phenol. Figure 1 below

illustrates a simple plant layout which was used in preliminary equipment sizing using heuristics.

6

Figure 1: Plant Layout

Note: Not drawn to scale

Components

T101: Holding Tank

P101A/B: Centrifugal Pump

P102A/B: Dosing Pump

T102: Flow Equalization Tank/Neutralization Tank

R101: Bioreactor

T103: Settling Tank

P103A/B: Centrifugal Pump

T104: Air Stripper

T105: Adsorber

T106: Adsorber

7

In the schematic above the daily influx of wastewater is initially stored in the holding

tank (T-101). Then it is pumped (via P-101A/B) to an equalization tank (T-102) where the

composition of the water and other parameters such as pH and temperature are measured. If the

pollutant concentration is higher than what the bioreactor can process then the wastewater may

be diluted before being pumped to the bioreactor (R-101). The effluent from the CSTR is fed to a

settling tank (T-103) wherein the solid biomass is allowed to settle. The exit stream from the

tank is pumped (via P-103A/B) to a stripping tower which the feeds to the air stripper (T-104). It

has been determined that the post CSTR treatment should be performed using an adsorber (T-

105/T-106) [1] and not via liquid/liquid extraction or distillation. Liquid/liquid extraction will

require solvents which in turn will generate more chemical waste and distillation is not be

feasible since the outlet concentrations of the pollutants will be extremely low thereby making it

an exceptionally expensive method.

The adsorber will be packed with activated carbon. Recent research into “high

performance activated carbon for benzene/toluene adsorption from industrial wastewater” [1] has

shown promising results. Moreover, it has been cited by the US Environmental Protection

Agency as “one of the best available environmental control technologies” [1].

8

Background

Wastewater Treatment

Wastewater can be treated either physically, chemically or biologically. Physical means

of treatment is generally best applicable for wastewaters with high solid content and can be

usually carried out through sedimentation, filtration and screening. Examples of chemical

treatment are chemisorption and chemical oxidation.

Physical means such as filtration, sedimentation and screening aims to remove or separate solids

from a liquid stream. However, this project deals with studying industrial waste waters which is

high in organic pollutants not solids. [1]

Adsorption can be a physical or chemical surface phenomena where pollutants can be

removed using absorbers; in other words, physisorption and chemisorption. In physisorption, the

absorbate physically sticks on to the surface of absorbent molecules. The larger the surface area

of the absorbent, the stronger the absorption. In chemisorption, reaction takes place on the

absorbent surface during absorption. Adsorption using activated carbon as an absorbent is quite

popular due its high surface area. However, it can be very costly since activated carbon would

need to be replaced frequently. [1]

Chemical oxidation uses chemical oxidants in order to transform pollutants to less

harmful products. Common oxidizing agents are potassium dichromate, permanganate and

hydrogen peroxide. Although this method is effective in removing contaminants from waste

waters, it also results in toxic chemicals and more byproducts. Removing these chemicals would

be highly costly as additional separation units would be required. [2]

9

This project, on the other hand, focuses on biodegradation which is a type of biological

treatment. Basically, biodegradation uses micro-organisms to remove organic pollutants from

wastewaters and treatment can be either aerobic or anaerobic. Anaerobic treatment is carried out

in absence of oxygen whereas aerobic treatment uses oxygen to maximize the growth and

efficiency of biomass. Thus, this project was narrowed down to focusing on aerobic treatment.

Bioreactors

In general, bioreactors can simply be classified as either CSTR (continuous flow reactor),

PFR (plug flow reactor), batch or semi-batch. Based on the kinetics, the reactor can be sized.

Currently only a simple CSTR system has been sized and analyzed; however, in the future, more

reactors of this type will be looked at such as the ones mentioned below.

There are different types of reactors that can be used in this process:

1. Fixed-film options bio-tower – These towers consist of a layer of media and a rotary

distributor arm that sprays pretreated wastewater over the surface of the media. This

water moves downwards as air is circulated upwards. As the water moves downwards, a

“biological slime of microbes” [3] gets cultivated on the surface.

2. Rotating biological contactors (RBCs) – These consist of biomass coated media

arranged vertically on a horizontal rotating shaft. The media is 40% submerged in

wastewater and is exposed to atmospheric oxygen. As the surface area is high, biomass

population increases steadily and excess growth is continuously shed and removed.

3. Submerged biological contactors (SBCs) – This system is similar to RBCs except that

these contain 90% submerged wastewater and provide three times the surface area.

10

4. Membrane bioreactors (MBR) – These reactors combine both anoxic and anaerobic in

one system. They have submerged membranes that maintain an active biomass

production, thus reducing sludge-settling issues.

5. Jet aeration – This system works either by aerating wastewater “mechanical surface

aeration” or by injecting pure oxygen through “submerged diffusers” [4]

6. Surface aeration – This system uses a propeller that sprays wastewater into the air.

Bacterial Strains

As mentioned earlier, biodegradation requires microorganisms or bacterial strains to

remove pollutants. Such bacterial strains require a carbon source to grow which can degrade

hydrocarbons through bio-oxidation reactions. These microorganisms should also be able to

adapt to different pollutants.

So far, the bacterial strains which have been worked with are P. Putida, Rhodococcus,

Gliomastix indicus and consortium, where consortium is a mixture of different bacterial cultures.

Based on these bacterial strains, parameters for calculations were obtained from research.

Bioreactor and Kinetics

Bioreactor refers to an engineered device in which chemical processes are carried out

involving biological organisms. Within the bioreactor a reaction between the pollutants and

biological organism (i.e. bacteria) takes place. This biological reaction results in further growth

of the organism along with the production of one or many byproducts.

Pollutants

A pollutant is a substance presented into the environment that may have undesired effects

on the usefulness of a resource. There are two main groups of pollutants; Inorganic pollutants

11

and organic pollutants .This research focuses on organic pollutants since it deals with industrial

wastewater which is rich in organic pollutants.

12

Kinetics

The amount of bacteria present within the bioreactor is termed as biomass. The growth of

biomass (i.e. X) present within the bioreactor is proportional to the rate of change in biomass

(i.e.𝒅𝑿

𝒅𝒕).

The specific growth rate of biomass can be obtained by:

𝑑𝑋

𝑑𝑡= µ 𝑋 (1)

Equation 1: Growth rate of Biomass

Where rate of change of biomass is equal to proportionality (i.e. µ) times the growth of biomass.

µ is also known as the specific growth rate of biomass. Several kinetic models can be used to

define or mathematically describe µ. The kinetic models used in this research project are as

follows.

Monod Kinetics

Monod model is a growth model that introduces the concept of limiting nutrients

(substrate). A nutrient is considered to be limiting when a relationship between its exhaustion

and end of its growth is established. This model describes the relation between the growth rate

and the concentration of the limiting nutrient:

𝑑𝑋

𝑑𝑡= 𝑋

µ𝑚𝑆

𝐾+𝑆 (2)

Equation 2: Growth rate using Monod model

The µ in the previous equation has been replaced another term. Here X is the concentration of

biomass at time t, S is the substrate concentration at t, µm is the maximum specific growth rate and

K is the kinetic constant that supports half-maximum specific growth rate.

13

Andrews Kinetics

Andrews kinetic model evaluate the growth kinetics of inhibitory substrates, when the

cells grow on single substrate. This model defines µ as follows:

µ =µ𝑚𝑆

𝐾+𝑆+𝑆2

𝐾𝑖

(3)

Equation 3: Growth rate using Andrew’s model

Where S is the substrate concentration at t, µm is the maximum specific growth rate and K is the

kinetic constant that supports half-maximum specific growth rate and Ki is the inhibition

constant.

A binary mixture with competitive interaction containing (j, q) as pollutants can use the

following equations to define µ.

µ𝑗 =µ 𝑚𝑗∗ 𝑆𝑗

𝐾𝑠𝑗+𝑆𝑗+𝐾𝑗𝑞𝑆𝑞 (4)

Equation 4: Maximum specific growth rate for pollutant j

µ𝑞 =µ 𝑚𝑞∗ 𝑆𝑞

𝐾𝑠,𝑞+𝑆𝑞+𝐾𝑞𝑗𝑆𝑗 (5)

Equation 5: Maximum specific growth rate for pollutant q

Where µmj the maximum specific growth is rate of pollutant j and µmq is the maximum specific

growth rate of pollutant q. Sj and Sq. is the substrate concentration of pollutants j and q, and Ks,j is

the kinetic constant for pollutant j and Ks,q is the kinetic constant for pollutant q and K qj is the

interaction parameter describing the effect of q on j and similarly Kjq is an interaction parameter

describing the effect of j on q .

14

SKIP Model

The SKIP model most closely determines the biodegradation of binary mixtures. It can be

used to fit unspecified inhibition types. Where µ𝑚𝑎𝑥 , 𝐾𝑆 and 𝑌𝑋/𝑆 were determined from single-

substrate experiments.

µ =µ𝑚𝑎𝑥,1 𝑆1

𝐾𝑆,1 + 𝑆1 + (𝐾𝑆,1

𝐾𝑆,2) 𝑆2

+µ𝑚𝑎𝑥,2𝑆2

𝐾𝑆,2 + 𝑆2 + (𝐾𝑆,2

𝐾𝑆,1) 𝑆1

Equation 6: Growth rate using sum kinetics model

15

Methodology

In order to meet the objective of sizing a bioreactor that is acting with and without substrate

interactions, it was necessary to utilize intensive modeling and simulation programs. The results

of the model was evaluated in terms of removal efficiency. This was accomplished through the

use of MATLAB and PolyMath.

MATLAB

MATLAB codes were written for two purposes. MATLAB was first used for modeling substrate

behavior with biomass without the any other substrate interactions in the system. Then

MATLAB was used for modeling substrate interactions with biomass and each other when the

substrates are interacting with each other.

MATLAB for modeling substrate and biomass behavior without interaction

Please refer to Appendix for the specific MATLAB code.

In order to examine the change in substrate over time, it was assumed that biomass was constant

at 2000 mg/L and residence time of 30 minutes. The constants for the model such as specific

growth rate, yield and kinetic parameter were inputted into the program as fixed values. Based

on this, the kinetic model was programmed as a differential equation of change of substrate with

respect to time on MATLAB using the ode45 tool. A list of initial concentrations was developed

in intervals of 50 mg/L. The differential model was run for all values of initial concentration to

give the final concentration of the substrate. This value was used by the MATLAB program to

find the removal efficiency. Initial substrate concentration was plotted against final removal

efficiency as the output of the program.

16

The program required a guess for the final substrate concentration. However, considering that the

final value was a function of initial value and time, it was impossible to provide a constant value

for the guess that would be applicable for points of all concentration. Thus, the guess was

provided as a percentage of the initial concentration. For example, if the initial substrate

concentration is S0, the guess was stated as RS0 where R was usually 0.05 representing a guess

that the final concentration of the substrate would be approximately 5% of the initial substrate

concentration. This was used as a starting point by the MATLAB program to find the final

substrate concentration.

MATLAB for modeling substrate interaction with biomass and each other for interaction

Please refer to Appendix for the specific MATLAB code.

Similar to the first MATLAB model, it was assumed that biomass was constant at 2000 mg/L

and residence time of 30 minutes and a list of initial substrate concentrations were provided as an

input in steps of 50 mg/L. The constants for the model such as specific growth rates, yields,

kinetic parameters and interaction parameters were inputted into the program as fixed values.

This was used to develop the kinetic models for both substrates as differential equations using

the ode45 function.

However, in this case, the final substrate concentration was a function of time, initial

concentration and another substrate’s degradation as well. The interaction of one substrate upon

another was accounted for in the differential model by developing two equations each

representing the degradation of a substrate respectively while accounting for the interaction of

the substrates. The two differential equations were solved simultaneously to yield the final

concentrations for both substrates. The program used the final concentrations to calculate

17

removal efficiencies for both substrates. These values were plotted against the initial

concentrations to examine the effect of interaction on the removal efficiency.

The graphs for the two MATLAB programs were compared to see the difference that substrate

interaction made on the system.

PolyMath

Please refer to Appendix for PolyMath program.

PolyMath was used to confirm the results from the MATLAB simulations. The same data was

tested on PolyMath. A similar program was developed on PolyMath and the results were

analyzed. The analysis showed the same results as the MATLAB simulation and thus confirmed

the values.

Literature Review

Paper 1[6]

The paper published by Reardon, Mosteller & Bull Rogers, (2000) focuses on the bioremediation

of three aromatic compounds namely, benzene, toluene and phenol which are common pollutants

in the industrial and municipal wastewaters. The biodegradation was performed by a culture of

Pseudomonas putida F1 in batch cultivations separately on each component and then their

mixtures. The sum kinetics with interaction parameters (SKIP) model helped describe the binary

substrate results. This model, coupled with various parameters obtained from single and binary

substrate experiments aided in the prediction of the kinetics of the tertiary substrate mixture.

These kinetic interaction parameters have been extracted in order to be used as data for this

project’s MATLAB modelling.

18

Paper 2[7]

This journal by Reardon, Mosteller, Bull Rogers, DuTeau & Kim (2002) deals with the same set

of aromatic compounds, benzene, toluene and phenol. The microorganisms involved are

Pseudomonas putida F1 and Burkholderia sp. strain JS150. The interaction parameters obtained

from this paper can be used to confirm the data obtained from the previous source and thereby be

used in modelling.

Paper 3[11]

This paper by Oh, Shareefdeen, Baltzis, & Bartha (1994) was actually the one that was used to

lay the groundwork for the research. The cultures used in this experiment was Pseudomonas

putida O1 and Consortium and the parameters have been obtained as documented below.

Paper 4 [12]

The paper by Zarook, Shaikh and Ansar (1996) helps lay some groundwork for conducting the

shock loading effects. It includes data for pseudo steady state and random variations to the feed.

A transient bio filtration model developed in this paper can be used for solving transient models

in a CSTR. It also depicts various transient responses of exit concentrations to the random

perturbations.

Paper 5 [13]

The paper published by Rafiei, Naeimpoor and Mohammadi (2014) deals with the performance

of hybrid bioreactors under inlet loading shocks. Conventional membrane bioreactors were

compared against hybrid reactors and it was found that the hybrid reactors had better resistance

19

to fouling effects and also were efficient. This article will be used to perform trials on membrane

bioreactors that are widely used in the chemical industry.

Paper 6 [14]

The paper published by Yao, Ren, Deng, & Wei (2010) discusses the substrate interactions of m-

cresol and pyridine as single and dual substrates in petrochemical and coking wastewater. The

phenol-biodegradation was done using the bacterium Lysinibacillus cresolivorans and the cell

growth and substrate biodegradation kinetics were studied. Results concluded that the Haldane

model worked well for the single substrate kinetics and dual substrates was found to exhibit

inhibitory effects. The kinetic parameters and results were extracted from the paper and included

into the pollutants database.

Safety book

This book provided a set of heuristics to follow in order to do sizing for most of the unit

operations illustrated in the plant layout.

Hazardous Waste Management

This book was used as a guide to size the air stripper and the adsorbers.

20

Results

Benzene and Toluene

The graphs obtained below are the results tabulated from the non-interaction and interaction models

between benzene and toluene using the bacterial strain PPO1 and Consortium.

PPO1

NO INTERACTION

Figure 2: Initial Substrate Conc. vs RE% (B-T) using PPO1

Figure 1: Graph of initial substrate concentration vs. removal efficiency using PPO1 – no

interaction

0

20

40

60

80

100

120

0 200 400 600 800 1000 1200

Re

mo

val E

ffic

ien

y, R

E (%

)

Initial Substrate Conc., S0 (mg/L)

Initial Substrate Conc. (S0) vs. Removal Efficiency (RE%)

Benzene

Toluene

21

INTERACTION

Figure 3: Graph of initial substrate concentration vs. removal efficiency using PPO1 –

interaction

Results for PPO1 (@ S0 = 500 mg/L)

Pollutant Removal Efficiency (%)

No Interaction

Benzene 98.2

Toluene 96.9

Interaction

Benzene 15.5

Toluene 22.1

Table 4: Results for PPO1 @ S0 = 500 mg/L

0

20

40

60

80

100

120

0 200 400 600 800 1000 1200

Rem

ova

l Eff

icie

ny,

RE

(%)



Benzene

Toluene

22

Consortium

NO INTERACTION

Figure 4: Graph of initial substrate concentration vs. removal efficiency using consortium – no

interaction

0

20

40

60

80

100

120

0 200 400 600 800 1000 1200

Rem

ova

l Eff

icie

ny,

RE

(%)



Benzene

Toluene

23

INTERACTION

Figure 5: Graph of initial substrate concentration vs. removal efficiency using consortium –

interaction

0

20

40

60

80

100

120

0 200 400 600 800 1000 1200

Rem

ova

l Eff

icie

ny,

RE

(%)



Benzene

Toluene

24

Results for Consortium (@ S0 = 500 mg/L)

Pollutant Removal Efficiency (%)

No Interaction

Benzene 97.5

Toluene 98.4

Interaction

Benzene 82.6

Toluene 95.5

Table 5: Results for Consortium @ S0 = 500 mg/L

Toluene and Phenol

1. Toluene non interaction was again simulated in MATLAB using parameters from a different

source [5]. Toluene was modelled using Monod kinetics in this study.

25

2. Phenol results illustrated using the Monod model.

Figure 6: Graph of initial substrate concentration vs. removal efficiency– interaction

0

10

20

30

40

50

60

70

80

90

100

0 2000 4000 6000 8000 10000 12000

Re

mo

val E

ffic

ien

cy,

RE

(%)

Initial Substrate Concentration, S0 (mg/L)


Toluene

Phenol

26

Benzene and Phenol

1. Benzene results illustrated using the Monod model.

1. Phenol results illustrated using the Monod model.

27

Figure 7: Graph of initial substrate concentration vs. removal efficiency– interaction

0

10

20

30

40

50

60

70

80

90

100

0 2000 4000 6000 8000 10000 12000

Re

mo

val E

ffic

ien

cy,R

E (%

)

Initial Substrate Concentraton,S0 (mg/L)


Benzene

Phenol

28

Design

The sizing of the plant (Figure 1) has been performed as follows:

Equipment Sizing and Material of Construction

Tank 1 (Holding/Storage Vessel) – (T-101)

Purpose of the equipment

Storage of the entering waste water before any processing begins.

Assumed parameters

𝜏 = 30 𝑚𝑖𝑛

𝑄 = 0.9 𝑚3

𝑚𝑖𝑛= 342367

𝑔𝑎𝑙

𝑑𝑎𝑦

Calculations for Sizing

Using the flow rate and time constant, volume can be calculated:

𝜏 × 𝑄 = 𝑉

30𝑚𝑖𝑛 ×1 ℎ𝑟

60 𝑚𝑖𝑛×

1 𝑑𝑎𝑦

24 ℎ𝑟×

342367 𝑔𝑎𝑙

𝑑𝑎𝑦= 7132.65 𝑔𝑎𝑙

7132.65 𝑔𝑎𝑙 = 27.00 𝑚3

Safety Factor of 1.5 to be added to volume estimation

𝑉𝑠𝑎𝑓𝑒𝑡𝑦 = 27.00 × 1.5 = 40.50 𝑚3

Heuristic: For less than 3.8 m3 use vertical tank on legs

Heuristic: Liquids subject to breathing losses may be stored in tanks with floating or expansion roofs for

conservation

∴ Horizontal tank with floating roof will be designed

Heuristic: L/D=2-5 with 3.0 as the optimal ratio

Assume L/D=3.0

𝑉 = 𝜋 ×𝐷2

4× 𝐿

𝐿 = 3𝐷

𝑉 = 𝜋 ×𝐷2

4× 3𝐷

𝜋 ×𝐷2

4× 3𝐷 = 40.50

𝑫 = 𝟐. 𝟓𝟖𝟎 𝒎

𝑳 = 𝟑𝑫 = 𝟕. 𝟕𝟒𝟐 𝒎

Therefore, Volume=40.50 m3, Diameter=2.580 m, Length=7.742 m

29

Tank 2 (Neutralization Tank) – (T-102)


To hold the waste water while it is being brought to operating conditions. In this tank, parameters such as

concentration, pH and temperature will be adjusted before it enters the bioreactor.

Assumed parameters

𝜏 = 30 𝑚𝑖𝑛

𝑄 = 0.9 𝑚3

𝑚𝑖𝑛= 342367

𝑔𝑎𝑙

𝑑𝑎𝑦


Using the flow rate and time constant, volume can be calculated. This holding tank follows the same

heuristics as the first one. However, for the second holding tank, the volume is assumed to be 50% more

than the first one to account for the addition of diluting fluid and pH neutralizers.

𝑉 = 𝑉𝐻𝑜𝑙𝑑𝑖𝑛𝑔𝑇𝑎𝑛𝑘1 ∗ 1.50 = 60.75 𝑚3

Assume L/D=3.0

𝑉 = 𝜋 ×𝐷2

4× 𝐿

𝐿 = 3𝐷

𝑉 = 𝜋 ×𝐷2

4× 3𝐷

𝜋 ×𝐷2

4× 3𝐷 = 60.75

𝐷 = 2.954 𝑚

𝐿 = 3𝐷 = 8.863 𝑚

Horizontal tank with floating roof, Volume=60.75 m3, Diameter=2.954 m, Length=8.863 m

Therefore, Volume=40.50 m3, Diameter=2.580 m, Length=7.742 m

30

Material of Construction

Considering all the available options and their associated pros and cons, three options for

material of construction have been finalized. Ideally, hasteloy, monel-nickel or stainless steel

would be used to construct the tank due to their high resistance to corrosion. However, they are

extremely costly to implement. This poses a major issue considering that petroleum industry

waste water typically has a high concentration of suspended particles which leads to severe

corrosion.

Considering this, mild steel will be used. To account for its low resistance to corrosion, a strong

polyphoshate inhibitor will be added. This protects against corrosion and prevents scaling from

occurring if the wastewater is left standing for a long period of time.

Reactor- (R-101)

Heuristics:

L/D = 1

Power input in homogeneous reaction stirred tank = 0.5 – 0.15 HP per 1000 gallons

Ideal CSTR is when 500 – 2000 revolutions of a well- designed stirrer

Benzene (no interaction)

𝑣0 = 𝑣 = 0.9 𝑚3

𝑚𝑖𝑛

𝑣0 = 0.9 𝑚3

𝑚𝑖𝑛 ×

1000 𝐿𝑖𝑡𝑟𝑒

1 𝑚3 = 900

𝐿

𝑚𝑖𝑛

𝐶𝐵0 = 400 𝑚𝑔/𝐿

𝐶𝐵 = S = 20 𝑚𝑔

𝐿 @ 95% conversion

31

𝜇𝑚𝑎𝑥 = 0.44 ℎ𝑟−1

𝐾 = 3.36 𝑚𝑔

𝐿

𝜇 =𝜇𝑚𝑎𝑥 × 𝑆

𝐾 + 𝑆=

0.44 ℎ𝑟−1 × 20 𝑚𝑔

𝐿

3.36 𝑚𝑔

𝐿 + 20 𝑚𝑔

𝐿

= 0.3767 ℎ𝑟−1

−𝑟𝐵 = 1

𝑌× 𝜇𝑋 =

1

0.65× 0.3767

1

ℎ× 2000

𝑚𝑔

𝐿= 1159.1

𝑚𝑔

𝐿.ℎ= 19.32

𝑚𝑔

𝐿.𝑚𝑖𝑛

𝑉 =𝐹𝐵0 − 𝐹𝑩

−𝑟𝐵=

𝑣0 (𝐶𝐵0 − 𝐶𝐵)

−𝑟𝐵

𝑉 =900

𝐿𝑚𝑖𝑛 (400 − 20)

𝑚𝑔𝐿

19.32𝑚𝑔

𝐿. 𝑚𝑖𝑛

= 17701.9 𝐿𝑖𝑡𝑟𝑒𝑠 = 𝟏𝟕. 𝟕 𝒎𝟑

𝐷 = √𝑉 × 4

1 × 𝜋

3

= 𝟐. 𝟖𝟐𝟓 𝒎

𝐿 = 𝟐. 𝟖𝟐𝟓 𝒎

𝜏𝐵 =𝑉

𝑣0 = 20 𝑚𝑖𝑛

Toluene (no interaction)

𝑣0 = 𝑣 = 0.9 𝑚3

𝑚𝑖𝑛

𝑣0 = 0.9 𝑚3

𝑚𝑖𝑛 ×

1000 𝐿𝑖𝑡𝑟𝑒

1 𝑚3 = 900

𝐿

𝑚𝑖𝑛

𝐶𝑇0 = 400 𝑚𝑔/𝐿

𝐶𝑇 = S = 20 𝑚𝑔

𝐿 @ 95% conversion

32

𝜇𝑚𝑎𝑥 = 0.72 ℎ𝑟−1

𝐾 = 15.07 𝑚𝑔

𝐿

𝜇 =𝜇𝑚𝑎𝑥 × 𝑆

𝐾 + 𝑆=

0.72ℎ𝑟−1 × 20 𝑚𝑔

𝐿

15.07 𝑚𝑔

𝐿 + 20 𝑚𝑔

𝐿

= 0.4106 ℎ𝑟−1

−𝑟𝑇 =1

𝑌× 𝜇𝑋 =

1

0.64× 0.4106

1

ℎ× 2000

𝑚𝑔

𝐿= 1283.1

𝑚𝑔

𝐿.ℎ= 21.385

𝑚𝑔

𝐿.𝑚𝑖𝑛

𝑉 =𝐹𝑇0 − 𝐹𝑇

−𝑟𝑇=

𝑣0 (𝐶𝑇0 − 𝐶𝑇)

−𝑟𝑇

𝑉 =900

𝐿𝑚𝑖𝑛 (400 − 20)

𝑚𝑔𝐿

21.385 𝑚𝑔

𝐿. 𝑚𝑖𝑛

= 15992.52 𝐿𝑖𝑡𝑟𝑒𝑠 = 𝟏𝟓. 𝟗𝟗 𝒎𝟑

𝐷 = √𝑉 × 4

1 × 𝜋

3

= 𝟐. 𝟕𝟑 𝒎

𝐿 = 𝟐. 𝟕𝟑𝒎

𝜏𝑇 =𝑉

𝑣0 = 17.7 𝑚𝑖𝑛

Material of Construction for CSTR

Options:

1. Carbon steel – inexpensive, it is capable of resisting abrasion and alkali. Moreover, it is

readily available. However, not resistant to acids and strong alkali and can cause

contamination.

2. Monel-Nickel – Minimal discoloration and contamination. It is also resistant to chlorides.

However, can corrode in oxidizing environments and can be costly.

33

Settling Tank- (T-103)

Purpose of the equipment:

Continuously removing the biomass that is being deposited from the reactor using the process of

sedimentation.

Assumed parameter

Volumetric flow rate (Q) = 0.9 m3/min

Calculations for sizing of the settling tank:

Assume a Detention Time (DT) of 1 hour:

DT = 𝑉𝑜𝑙𝑢𝑚𝑒

𝑄

𝑉𝑜𝑙𝑢𝑚𝑒 = 𝐷𝑇 𝑥 𝑄 = 1ℎ𝑟 𝑥 𝑜. 9 𝑚3

𝑚𝑖𝑛 𝑥

60𝑚𝑖𝑛

1ℎ𝑟= 54 𝑚3

Figure 8: An ideal rectangular sedimentation tank illustrating the settling of discrete particles. [1]

34

Table 1 Typical design values for a sedimentation basin. [1]

Using Table 1 the typical values are:

Depth of the settling zone (H) = 3.5m; and the ranges are:

Length of the settling zone (L) = 15-25 m

Width of the settling zone (W) = 3 -24 m

The lowest value of the length is too large for the purposes of this project. Therefore, a typical value

for depth is assumed at 3.5 m and a length to width ratio of 1:4 is also assumed [1]. These assumptions

are then utilized to find the length of the clarifier.

Volume = L x W x H

𝐿 = 4𝑊

Volume = 4𝑊2 x H

𝑊 = √𝑉𝑜𝑙𝑢𝑚𝑒

4 𝑥 𝐻

2

= √54𝑚3

4𝑥 3.5𝑚

2

= 1.96𝑚

𝐿 = 4𝑥 𝑊 = 4𝑥 1.963 = 7.86𝑚

Therefore, using this length area can be calculate:

𝐴𝑟𝑒𝑎 = 𝐿𝑥𝑊

𝐴𝑟𝑒𝑎 = 7.86 𝑥 1.96 = 15.42 𝑚2

35

Therefore, the over flow rate is [1]:

𝑣𝑜 =𝑄

𝐴

Where:

vo is the overflow rate (m/min)

Q is the volumetric flowrate (m3/min)

A is the area (m2)

𝑣𝑜 =0.9

𝑚3

𝑚𝑖𝑛

15.42𝑚2 = 0.058𝑚

𝑚𝑖𝑛≅ 0.06

𝑚

𝑚𝑖𝑛

Assuming an efficiency (η) of 95 %:

Using the formula:

𝜂 =ℎ

𝐻

Where:

η is the efficiency.

H is the depth of the settling zone.

h is the vertical fall over length.

ℎ = 𝜂 𝑥 𝐻 = 0.95 𝑥 3.5 = 3.33 𝑚

Using the h, the settling velocity (vs) can be calculate by:

𝑣𝑠 =ℎ𝑥𝑄

𝐻𝑥𝐴

36

𝑣𝑠 = 3.33 𝑚

3.5𝑚𝑥

0.9 𝑚3

𝑚𝑖𝑛15.42 𝑚2

= 0.055𝑚

𝑚𝑖𝑛≈ 0.06

𝑚

𝑚𝑖𝑛

To assure a good sedimentation tank design settling velocity (vs) must be greater than or equal to the

overflow rate (vo) [1].

It can be seen above that vs is equal to vo, hence it can be concluded that the design of the sedimentation

tank is reliable.

The Weir overflow rate (WOR):

𝑊𝑂𝑅 =𝑄

𝐿𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑡ℎ𝑒 𝑤𝑒𝑖𝑟

𝐿𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑡ℎ𝑒 𝑤𝑒𝑖𝑟 = 2 𝑋 𝑤𝑖𝑑𝑡ℎ 𝑜𝑓 𝑡ℎ𝑒 𝑠𝑒𝑡𝑡𝑙𝑖𝑛𝑔 𝑡𝑎𝑛𝑘

𝐿 𝑤𝑒𝑖𝑟 = 2 𝑥 1.96𝑚 = 3.92𝑚

Therefore,

𝑊𝑂𝑅 =0.9

𝑚3

𝑚𝑖𝑛3.92 𝑚

= 0.229𝑚3

𝑚 𝑚𝑖𝑛


Taking all available options into consideration and weighing their pros and cons, two material such as

stainless steel and carbon steel have been finalized. They are extremely costly however, they have high

resistance against corrosion and prove to be cost efficient in the long run.

37

Air Stripper-(T-104)

An air stripper was selected to remove the pollutants from the outlet of the settling tank as it is cost-

effective for removing volatile organic compounds and is very effective for low concentrations.

Sizing

Benzene Data

Benzene Data

Benzene

Concentration (mg/L) 1.4

Flowrate (gal/day) 10,000

Viscosity of Liquid (kg/m.s) (Engineering

Toolbox, n.d.)

7.98E-04

Viscosity of Gas (kg/m.s) (Engineering Toolbox,

n.d.)

1.90E-05

Density of liquid (kg/m3) (Engineering Toolbox,

n.d.)

995.7

Density of gas (kg/m3) (Engineering Toolbox,

n.d.)

1.165

Density of benzene (kg/m3) (Engineering

Toolbox, n.d.)

868

Critical surface tension (N/m) (Engineering

Toolbox, n.d.)

0.04

Surface Tension (N/m) (Engineering Toolbox,

n.d.)

0.0712

Temperature (K) 303

Diffusivity of Benzene in water, DL (cm2/s)

(HyperTextBookShop, n.d.)

1.02E0-5

Molar Mass of Benzene (g/mol) 78.11

𝑀𝑜𝑙𝑎𝑟 𝑉𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝐵𝑒𝑛𝑧𝑒𝑛𝑒, (𝑐𝑚3

𝑚𝑜𝑙) =

𝑀𝑜𝑙𝑎𝑟 𝑀𝑎𝑠𝑠

𝐷𝑒𝑛𝑠𝑖𝑡𝑦

𝑀𝑜𝑙𝑎𝑟 𝑉𝑜𝑙𝑢𝑚𝑒 =78.11 ∗ 1003

868 ∗ 1000= 89.99

𝑐𝑚3

𝑚𝑜𝑙

𝐻𝑒𝑛𝑟𝑦′𝑠 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡, 𝐻 = exp (𝐴

𝑇+ 𝐵) = exp (

−3.19 ∗ 103

303+ 5.53) = 6.75 ∗ 10−3𝑎𝑡𝑚.

𝑚3

𝑚𝑜𝑙

𝐻𝑒𝑛𝑟𝑦′𝑠 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡 (𝐷𝑖𝑚𝑒𝑛𝑠𝑖𝑜𝑛𝑙𝑒𝑠𝑠) =6.75 ∗ 10−3

8.25 ∗ 10−5 ∗ 303= 0.270

Henry’s Constant (LaGrega,

Buckingham, & Evans, 2001)

A -3.19*103

B 5.53

Packing (Tripack (PVC))

Size (m) 2.50E-02

F 28

at (m-1) 279

38

𝐴𝑠𝑠𝑢𝑚𝑖𝑛𝑔,

𝐴𝑖𝑟 𝑡𝑜 𝑊𝑎𝑡𝑒𝑟 𝑅𝑎𝑡𝑖𝑜 (𝑄𝐴

𝑄𝑊) = 20

𝐴𝑠𝑠𝑢𝑚𝑖𝑛𝑔 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟 𝑜𝑓 𝑐𝑜𝑙𝑢𝑚𝑛 = 0.5𝑚,

𝐴𝑟𝑒𝑎 𝑜𝑓 𝑐𝑜𝑙𝑢𝑚𝑛 = 0.20𝑚2

Onda Correlations (LaGrega, Buckingham, & Evans, 2001)

𝐿𝑖𝑞𝑢𝑖𝑑 𝑚𝑎𝑠𝑠 𝑙𝑜𝑎𝑑𝑖𝑛𝑔 𝑟𝑎𝑡𝑒, 𝐿 =10,000

𝑔𝑎𝑙𝑑𝑎𝑦

∗ 8.31𝑙𝑏

𝑔𝑎𝑙

0.20𝑚3 ∗ 24ℎ𝑟 ∗ 3600𝑠 ∗ 2.204𝑙𝑏= 2.18

𝑘𝑔

𝑚3. 𝑠

𝐷𝑖𝑓𝑓𝑢𝑠𝑖𝑜𝑛 𝐶𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 𝑖𝑛 𝐴𝑖𝑟, 𝐷𝐺 = 0.09234𝑐𝑚2

𝑠 (LaGrega, Buckingham, & Evans, 2001)

𝑄𝑊 =10,000

𝑔𝑎𝑙𝑑𝑎𝑦

∗ 35.3147 𝑓𝑡3

264.172 𝑔𝑎𝑙 ∗ 24ℎ𝑟 ∗ 3600 𝑠= 0.015

𝑓𝑡3

𝑠

𝑄𝐴 = 0.015 ∗ 20 = 0.30𝑓𝑡3

𝑠

𝐺𝑎𝑠 𝑚𝑎𝑠𝑠 𝑙𝑜𝑎𝑑𝑖𝑛𝑔 𝑟𝑎𝑡𝑒, 𝐺 =0.30

𝑓𝑡3

𝑠 ∗ 1.165𝑘𝑔𝑚3 ∗ 0.02832

𝑚3

𝑓𝑡3

0.20𝑚2= 0.049

𝑘𝑔

𝑚2. 𝑠

𝑅𝑒𝑦𝑛𝑜𝑙𝑑𝑠 𝑁𝑢𝑚𝑏𝑒𝑟, 𝑅𝑒 =𝐿

𝑎𝑡𝜇𝐿=

2.18

279 ∗ 0.798 ∗ 10−3= 9.79

𝐹𝑟𝑜𝑢𝑑𝑒 𝑁𝑢𝑚𝑏𝑒𝑟,𝐿2𝑎𝑡

𝜌𝐿2𝑔

=2.182 ∗ 279

995.72 ∗ 9.81= 1.36 ∗ 10−4

𝑊𝑒𝑏𝑒𝑟 𝑁𝑢𝑚𝑏𝑒𝑟,𝐿2

𝜌𝐿𝜎𝑎𝑡=

2.182

995.7 ∗ 0.0172 ∗ 279= 2.40 ∗ 10−4

Equation 6- aw equation [1]

𝑎𝑤

𝑎𝑡= 1 − exp [−1.45 (

𝜎𝑐

𝜎)

0.75

(𝐿

𝑎𝑡𝜇𝐿)

0.1

(𝐿2𝑎𝑡

𝜌𝐿2𝑔

)

−0.05

(𝐿2

𝜌𝐿𝜎𝑎𝑡)

0.2

𝑎𝑤 = 279 ∗ [1 − 𝑒(−1.45 ∗ 0.560.75 ∗ 9.790.1 ∗ (1.36 ∗ 10−4)−0.05 ∗ (2.40 ∗ 10−4)0.2)] = 81.90𝑚−1

Equation 7-kL equation [1]

𝑘𝐿 (𝜌𝐿

𝜇𝐿𝑔)

13

= 0.0051 (𝐿

𝑎𝑤𝜇𝐿)

23

(µ𝐿

𝜌𝐿𝐷𝐿)

−0.5

(𝑎𝑡𝑑𝑝)0.4

39

𝑘𝐿 [995.7

(0.798 ∗ 10−3) ∗ 9.81]

13

= 0.0051 ∗ [2.18

81.90 ∗ (0.798 ∗ 10−3)]

23

∗ [0.798 ∗ 10−3

995.7 ∗ (1.02 ∗ 10−9)]

−0.5

∗ [279 ∗ 0.025]0.4

𝑘𝐿 = 8.15 ∗ 10−5𝑚

𝑠

Equation 8 - kG equation [1]

𝑘𝐿

𝑎𝑡𝐷𝐺= 5.23 (

𝐺

𝑎𝑡𝜇𝐺)

0.7

(µ𝐺

𝜌𝐺𝐷𝐺)

13

(𝑎𝑡𝑑𝑝)−2

𝑘𝐺 = 279 ∗ 9.234 ∗ 10−6 ∗ [5.23 ∗ (0.049

279 ∗ 1.9 ∗ 10−5)

0.7

∗ (1.9 ∗ 10−5

1.165 ∗ 9.234 ∗ 10−6)

13

∗ (279 ∗ 0.025)−2

= 1.59 ∗ 10−3𝑚

𝑠

Equation 9 - Kla equation [1]

1

𝐾𝐿𝑎=

1

𝐻′𝑘𝐺𝑎+

1

𝑘𝐿𝑎

1

𝐾𝐿𝑎=

1

0.270 ∗ 1.59 ∗ 10−3 ∗ 81.90+

1

8.15 ∗ 10−5 ∗ 81.90

𝐾𝐿𝑎 = 0.0056𝑠−1

Preliminary Design

𝐿𝑖𝑞𝑢𝑖𝑑 𝑚𝑜𝑙𝑎𝑟 𝑙𝑜𝑎𝑑𝑖𝑛𝑔 𝑟𝑎𝑡𝑒, 𝐿 = 2.18𝑘𝑔

𝑚2. 𝑠∗

103𝑔

𝑘𝑔∗

𝑚𝑜𝑙

18𝑔= 121.11

𝑚𝑜𝑙

𝑚2. 𝑠

𝑆𝑡𝑟𝑖𝑝𝑝𝑖𝑛𝑔 𝐹𝑎𝑐𝑡𝑜𝑟, 𝑅 = 𝐻′ (𝑄𝐴

𝑄𝑊) = 0.270 ∗ 20 = 5.4

𝐻𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑇𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑈𝑛𝑖𝑡, 𝐻𝑇𝑈 =121.11

55,600 ∗ 0.0056= 0.40𝑚

𝐶𝑖𝑛 = 1400𝜇𝑔

𝐿 𝐶𝑜𝑢𝑡(99.5% 𝑟𝑒𝑚𝑜𝑣𝑎𝑙) = 0.005 ∗ 𝐶𝑖𝑛 = 7

𝜇𝑔

𝐿

40

Equation 10- NTU equation [1]

𝑁𝑇𝑈 =𝑅

𝑅 − 1∗ ln [

𝐶𝑖𝑛𝐶𝑜𝑢𝑡

∗ (𝑅 − 1) + 1

𝑅]

𝑁𝑇𝑈 =5.4

5.4 − 1∗ ln [

140070

∗ (5.4 − 1) + 1

5.4] = 6.3 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟 𝑢𝑛𝑖𝑡𝑠

𝐻𝑒𝑖𝑔ℎ𝑡, 𝑍 = 𝐻𝑇𝑈 ∗ 𝑁𝑇𝑈 = 6.3 ∗ 0.4 = 2.5𝑚 = 8.3𝑓𝑡

𝑇𝑎𝑘𝑖𝑛𝑔 𝑎 𝑠𝑎𝑓𝑒𝑡𝑦 𝑓𝑎𝑐𝑡𝑜𝑟 𝑜𝑓 20%,

𝑍 = 8.3 ∗ 1.20 = 10𝑓𝑡

Pressure Drop Calculations

𝐿 = 2.18𝑘𝑔

𝑚2. 𝑠∗ 0.2048 = 0.450

𝑙𝑏

𝑓𝑡2. 𝑠

𝑄𝐴 = 0.3𝑓𝑡3

𝑠∗

𝑚3

35.3147𝑓𝑡3= 0.0085

𝑚3

𝑠

𝐺 = 0.049𝑘𝑔

𝑚2. 𝑠∗ 0.2048 = 0.010

𝑙𝑏

𝑓𝑡3. 𝑠

𝐿

𝐺(

𝛿𝐴

𝛿𝑊)

0.5

=0.450

0.010∗ (

0.073

62.2)

0.5

= 1.5

𝐺2𝐹

𝛿𝐴𝛿𝑊𝑔=

0.012 ∗ 28

0.073 ∗ 62.2 ∗ 32.17= 1.9 ∗ 10−5

Using the following chart, we can find the pressure drop in the tower. However, the calculated value for

the Ordinate is lower than the starting point in the chart. Hence, the lowest curve was assumed to be the

pressure drop.

41

Figure 9 – Pressure Drop Correlation Curve [http://www.tankonyvtar.hu]

𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 𝑑𝑟𝑜𝑝 = 10𝑓𝑡 ∗ 0.0015𝑖𝑛

𝑓𝑡= 0.015 𝑖𝑛𝐻2𝑂 = 3.7 𝑃𝑎

Adsorber-(T-105)

Powdered activated carbon was used for adsorption process as it is widely used in biological treatment

processes.

𝐶𝑖𝑛 = 70𝜇𝑔

𝐿 𝐶𝑜𝑢𝑡 = 3.5

𝜇𝑔

𝐿 (95% 𝑟𝑒𝑚𝑜𝑣𝑎𝑙)

𝐹𝑙𝑜𝑤𝑟𝑎𝑡𝑒 𝑜𝑓 𝑎𝑖𝑟 = 0.0085𝑚3

𝑠

𝑀𝑜𝑙𝑎𝑟 𝑓𝑙𝑜𝑤𝑟𝑎𝑡𝑒 𝑜𝑓 𝑎𝑖𝑟 = 0.0085𝑚3

𝑠∗

𝑚3

1.165𝑘𝑔∗

𝑚𝑜𝑙

29𝑔∗

1000𝑔

𝑘𝑔= 0.25

𝑚𝑜𝑙

𝑠

42

𝐹𝑟𝑒𝑢𝑛𝑙𝑖𝑐ℎ 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛, 𝑞 = 𝑘𝐶𝑓

1𝑛

𝑘 = 47.9𝐿

𝑚𝑔, 𝑛 = 0.533 [4]

𝑞 = 47.9 ∗ 0.00350.533 = 2.4𝑚𝑔 𝐵𝑒𝑛𝑧𝑒𝑛𝑒

𝑔 𝑐𝑎𝑟𝑏𝑜𝑛

𝐶𝑎𝑟𝑏𝑜𝑛 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 = 0.0024𝑔 𝐵

𝑔 𝐶

𝐴𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝐵𝑒𝑛𝑧𝑒𝑛𝑒 𝑅𝑒𝑚𝑜𝑣𝑒𝑑

= 0.0085𝑚3

𝑠∗ (0.07 − 0.0035)

𝑚𝑔

𝐿∗ 103

𝐿

𝑚3∗ 3600

𝑠

ℎ𝑟∗ 24

ℎ𝑟

𝑑𝑎𝑦∗ 10−3

𝑔

𝑚𝑔

= 48.8 𝑔𝐵𝑒𝑛𝑧𝑒𝑛𝑒𝑟𝑒𝑚𝑜𝑣𝑒𝑑

𝑑𝑎𝑦

𝐶𝑎𝑟𝑏𝑜𝑛 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 = 48.8𝑔𝐵

𝑑𝑎𝑦∗

1

0.0024

𝑔𝐶

𝑔𝐵= 20,333

𝑔𝐶

𝑑𝑎𝑦

𝐶𝑎𝑟𝑏𝑜𝑛 𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑 𝑓𝑜𝑟 6 𝑚𝑜𝑛𝑡ℎ𝑠 = 20,333𝑔𝐶

𝑑𝑎𝑦∗ 30𝑑𝑎𝑦𝑠 ∗ 6 𝑚𝑜𝑛𝑡ℎ𝑠 ∗

1𝑘𝑔

1000𝑔= 3,660𝑘𝑔 𝐶𝑎𝑟𝑏𝑜𝑛

According to Hutchins [3],

Height of one column, d = 2.3m

Assuming,

Height of adsorption zone, AZ = 2m

𝑁𝑢𝑚𝑏𝑒𝑟 𝑎𝑛𝑑 𝑠𝑖𝑧𝑒 𝑜𝑓 𝑎𝑑𝑠𝑜𝑟𝑝𝑡𝑖𝑜𝑛 𝑐𝑜𝑙𝑢𝑚𝑛𝑠, 𝑛 = (𝐴𝑍

𝑑) + 1 = (

2

2.3) + 1 = 1.9 = 2 𝑢𝑛𝑖𝑡𝑠

Assuming L/D ratio of 2,

Diameter of adsorption unit = 1.15m

43

Pumps


Pump the waste from the holding tank to the bioreactor

Pump the water from the settling tank to the air stripper

Assumptions

𝑄 = 0.9 𝑚3/𝑚𝑖𝑛

𝐷𝑒𝑛𝑠𝑖𝑡𝑦 = 𝜌 = 1000 𝑘𝑔/𝑚3

A centrifugal single stage pump will be used


Based on Heuristics:

Power for pumping liquids: kW = (1.67)*[Flow (m3/min)]*[ΔP (bar)]/ε

Efficiency is selected as 60%

Pump 1-(P-101A/B)

ΔP (bar) = 𝜌 ∗ 𝑔 ∗ ∆ℎ = 1000𝑘𝑔

𝑚3∗ 9.81

𝑚

𝑠2∗ 8 𝑚 = 74.48 𝑘𝑃𝑎 = 0.7848 𝑏𝑎𝑟

Power = 1.67∗0.9

𝑚3

ℎ𝑟∗0.7848 𝑏𝑎𝑟

0.6= 2.0 𝑘𝑊

P1 (Pressure at suction) = Vapor Pressure inside tank (using Raoult’s law)

𝑃1 = 𝑃°𝑏𝑒𝑛 ∗ 𝑥𝑏𝑒𝑛 + 𝑃°𝑡𝑜𝑙 ∗ 𝑥𝑡𝑜𝑙 = 94.6 𝑡𝑜𝑟𝑟 ∗ 0.5 + 29.1 ∗ 0.5 = 61.8 𝑡𝑜𝑟𝑟 = 0.082 𝑏𝑎𝑟

Assuming Benzene and Toluene are present in equal proportion

Pressure at discharge (P2) = P1 + ∆P = 0.082 + 0.7848 = 0.8668 𝑏𝑎𝑟

44

Pump 2-(P-103A/B)

ΔP (bar) = 𝜌 ∗ 𝑔 ∗ ∆ℎ = 1000𝑘𝑔

𝑚3 ∗ 9.81𝑚

𝑠2 ∗ 5.4 𝑚 = 530 𝑘𝑃𝑎 = 0.530 𝑏𝑎𝑟

Power = 1.67∗0.9

𝑚3

ℎ𝑟∗0.530 𝑏𝑎𝑟

0.6= 1.33 𝑘𝑊

P1 (Pressure at suction) = Pressure inside previous unit operation

Pressure at discharge (P2) = P1 + ∆P = 0.8668 + 0.530 = 1.40 𝑏𝑎𝑟


Steel and stainless steel alloys provide protection against chemical and rust corrosion and have higher tensile

strengths than plastics, corresponding to higher pressure ratings.

Piping

Typical fluid velocities and allowable pressure drops, which can be used to estimate pipe sizes, are as

follows

Table 2:Typical pipe properties (Towler & Sinnott, 2008)

Total Pipe length= 43.8 m

Q=0.9 m3/min=0.015m3/s

45

From the above table taking the typical properties of non-viscous pumped liquid

𝐷 = √4𝑄

𝜋𝑢

𝐷 = √4 ∗

0.015𝑚3

𝑠

𝜋 ∗3𝑚

𝑠

𝑡𝑜 √4 ∗

0.015𝑚3

𝑠

𝜋 ∗1𝑚

𝑠

𝐷 = 0.0798 𝑚 𝑡𝑜 0.1382 𝑚

However for economical pipe the following equation is given

𝑑𝑖 , 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 = 𝑄0.5 (Towler & Sinnott, 2008)

Q in 𝑚3/s

𝑑𝑖 = 0.90.5

𝑑𝑖 = 0.122 𝑚

Control

The control and instrumentation implemented in this design have been outlined in the plant

layout below.

The level gauge monitors the amount of liquid in the systems they are implemeted in.

Furthermore the TI (temperature indicator), TT (temperature transmitter) and TC (temperature

controller) work to control and maintain the temperature in the system since microorganisms are

very sensitive to any change to temperature and pH (pH indicators and controllers are also part of

the system).

46

The fluid flow is a very important factor in the adsorber and hence there is a FT (flow

transmitter) and FC (flow controller) in place to monitor this.

Finally a pressure relief valve is placed on the reactor in case of an excessive pressure build-up.

Figure 10: Plant layout with Control and Instrumentation

47

Costing

Figure 11: Bare Module Cost of Plant using CAPCOST

48

Reactor Size

The results illustrated below are the difference in reactor volume and thereby prices due to

kinetic interaction. This was performed using MATLAB. As tabulated below the same

parameters were used. The only variable was residence time (τ), which was manipulated to

obtain the same removal efficiency when there is no interaction.

Data

Flow rate 1000 gal/day

Biomass 2000 mg/L

Inlet Concentration of pollutants 500 mg/L

Pollutant Sinlet Soutlet RE%

Benzene 500 12.6599 97.46802

Toluene 500 8.1119 98.37762

τ (h) Pollutants SB0 ST0 SB ST RE%Benzene RE%Toluene

0.5 Benzene+Toluene 500 500 86.8235 22.7142 82.6353 95.45716

0.9 Benzene+Toluene 500 500

11.9554

4.0123

97.60892

99.19754

Volume(m3) 0.07885 0.1420

Cost ($) 11,600 15,000

Due to the effect of interaction the volume and the cost of the reactor increase by 80% and 30%

respectively.

49

HSE- Health, Safety and Hazard

HAZOP Tank 1

Deviation Possible Cause Consequence Action Required

No Flow Line Fracture

Loss of feed to tank and

reactor. Reduced output

Regular maintenance and

inspection of lines. Set up

maintenance schedule

Wastewater discharged to

adjacent area

Estimate quantity

released and develop

management plan

Damage to pump due to

air flow

LSL to turn pump off if

level not high on enough

More Flow Error in FCV Tank level too high.

Reactor overfill

HSL to turn on pump and

turn off inlet flow.

More Pressure Buildup of vapor VOC release into air Install floating roof

More acidic or basic

compounds present Inflow conditions

Increased rate of

corrosion. Damage to

equipment

Add corrosion inhibitor to

protect tank

More hardness Inflow conditions

Increased scaling. Loss of

efficiency and equipment

damage

Add resin to entering

flow to soften water

Tank 2

Deviation Possible Cause Consequence Action Required

No Dosage

Damage of dosing pump Equalization will not

occur. Biomass threat.

Switch to standby dosing

pump

Empty equalization supply

tanks

Control will not be

accomplished

Low level alarm on the

supply tanks

Error in indicators and

transmitters

Control will not be

accomplished.

Regular maintenance and

re-calibration of

transmitters.

50

No mixing Corrosion of mixer

paddles

Uneven mixture.

Control not accurate. Inhibitor added to tank

More Temperature Industrial discharge at

high temp Threat to biomass

Jacketed vessel cascade

control system

Less Flow

Pump malfunction Delay to process Switch to standby pump

Line Blockage Fouling and pressure

build-up in pipes

Regular checking of pipes

and maintenance.

Acidic or basic

compounds Inflow conditions Threat to biomass

pH control in equalization

tank

Reactor

Deviation Causes Consequences Action

High pressure

High volatile compounds

and increased CO2

production in the reactor

Temperature increase in

reactor

Install high temperature

alarm (TAH)

No/ very less degradation

of pollutants

Excessive organic

loading

Inefficient removal of

pollutants

Reduce inlet flow or add

additional aeration basins

or increase biomass in

reactor.

Low dissolved oxygen Inadequate air supply Removal efficiency

decreases Increase air supply

More death rate than

growth rate

Improper environmental

conditions

No degradation of

pollutants

Use appropriate control

in order to improve

conditions.

High pH Incoming effluent with a

larger pH

Decrease in biomass

growth

Adjust pH using a pH

indicator controller and

dosing pumps.

51

Settling Tank


High flow Issues with the FCV

(flow control valve)

The settling tank

washouts

HSL (high level switch)

to close FCV at the inlet.

No flow Accumulation of

activated sludge

Overflow of untreated

water

HSL to open the FCV at

outlet

Low flow Issues with the FCV The detention time

increases

The FCV to be opened at

the inlet

Air Stripper


More liquid flow in the

inlet

Increased pumping power

due to blockage in the

pipe

Flooding in the column Flow control with Air to

Water set point

More pressure drop Blockage in pipe Flooding in the column Regular maintenance

Less liquid flow Less liquid pumped from

the settling tank

Liquid entering the

adsorber Increase pump flowrate

Less liquid pressure Pipe ruptured Flooding Regular maintenance

No liquid flow pump failure No stripping Activate backup pump;

regular maintenance

52

Adsorber


No flow Pipe leak Leaking of contaminated air Regular maintenance

No adsorption Saturated Activated Carbon No adsorption Regular maintenance,

Activate backup column

More concentration of

pollutants Malfunction of Air stripper Saturation of carbon

Activate backup adsorption

column

Other contaminants in the

inlet

Improper settling in the

settling tank clogging of activated carbon

Maintenance of the settling

tank

Pumps


High Flow High switch valve not

functioning

Impingement of water on

the impeller causes

erosion corrosion

Activate standby pump

Regular maintenance

No flow Gas Locking

No fluid enters or leaves

the pump; the pumping

system is useless

Activate the standby

pump and deviate flow

Less flow Cavitation

Reduction in pump

efficiency; damage to

pump components

Low switch level (LSL)

on the preceding tank

should go off

Reverse Flow Pump Failure Pipe rupture due to liquid

build up

Activate standby pump

Regular maintenance

53

Piping


No liquid flow Blockage in pipe and

pipe Leak Pump cavitation

If it occurs right before the primary pump, shut

primary pump and activate backup pump. If it

occurs in the common pipeline connecting both

the pump, activate alarm to shut down plant for

maintenance.

Less liquid flow Partial pipe blocks and

Pipe leakage

Less flow in the

reactor and air

stripper

Regular maintenance, Increase pumping power

to raise flow, activate back up pump

Liquid flow in the

reverse direction

Pump cavitation or

pipeline block

No liquid flow to

the equipment’s

and Pipe rupture

Use one-way valve and regular maintenance

UAE Regulations

Organic compounds such as Benzene and Toluene are classified as carcinogens, and exposure to

high concentration is hazardous. According to Regulation & Supervision Bureau, the water

leaving any waste treatment can only have a maximum concentration of 10 µg/L (Regulation,

2013). Therefore the calculated value of 7 µg/L at a conversion of 99.5 % is with in this limit of

regulation and can be deemed safe. In addition, the Air quality standard by the UAE EHS

(Environment, Health and Safety) requires that the maximum amount of Benzene leaving any

plant cannot exceed the limit of 5 µg/L (Department, 2013). This plant releases Benzene with a

concentration of 3.5 µg/L at 95% conversion, which is lower than the allowable limit. Therefore

is it safe to release it into the atmosphere.

54

Economic Analysis

Wastewater treatment plants are categorized as public projects. The criteria that defines a public

project are: large investment, long life span of 30-50 years, no profit, funding from taxes and

subsidiaries, provides an essential service to the public on behalf of the government and is

politically inclined. Thus the bioreactor waste treatment facility can be considered a public

project.

While profitability analysis is generally adopted for a private sector analysis, public sector

projects do not have profitability as the main goal. In these cases, cost-benefit analysis and ratio

are used to evaluate the feasibility of completing a project. The cost-benefit ratio can be

described by the following equation:

𝑀𝑜𝑑𝑖𝑓𝑖𝑒𝑑𝐵

𝐶𝑅𝑎𝑡𝑖𝑜 =

(𝐵𝑒𝑛𝑒𝑓𝑖𝑡𝑠 − 𝐷𝑖𝑠𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠) − 𝐶𝑜𝑠𝑡

𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡

Benefits are the advantages to the public quantified in monetary terms. In this treatment plant,

the benefit would be the savings of the water treatment plant that would have otherwise occurred

using more expensive treatment methods.

Disbenefits are the disadvantages to the public quantified in monetary terms. For the case of this

project, the disbenefit would be the loss of income by using the land that could have otherwise

been relegated to another purpose in an industrial district.

The costs are the expenditures made by the government and include the daily operating costs and

maintenance expenditures. For example, transport and pump and pipe maintenance would be a

cost to this project.

55

Capital Investment is the investment in building the plant and the process. This includes land

costs and all installation costs of the equipment.

If the B/C analysis results in a value greater than 1, the project is considered economically

justified for the specified lifetime.

56

Conclusion From the results obtained in this project one may conclude that the effect of kinetic interactions on the

bioreactor sizing is quite significant. As illustrated earlier the kinetic interactions cause an 80% increase

in reactor volume and a 30% increase in the overall cost.

Future Work

The system has been costed for two systems: a benzene and toluene system without interaction

and a benzene and toluene system with interaction. In the future, this work will be expanded for

different substrate systems such as toluene-phenol and benzene-phenol substrate interaction

systems.

Following this, the existing results and sizing parameters will be simulated on SuperPro where

the substrate systems will be tested in the sized equipment. This will give a good indication of

the overall efficiency of the process and its practical applicability.

To improve the general model and simulation process, the biomass mass balance will be taken

into account. The biomass is currently treated as a constant at 2000 mg/L. However, the biomass

has its own mass balance because it has its own growth and death rate while producing carbon

dioxide and water after the oxidation occurs. Therefore, for future work, the biomass mass

balance will be incorporated into the model and the product carbon dioxide will be accounted

for.

Furthermore, current costing does not take into account anything except for equipment purchase

and installation costs. In the future, costs for compressed air, corrosion inhibitor, hardness

reducing resin, utilities such as steam, biomass, acid and base for the neutralization process and

water for concentration equalization will be accounted for. In addition to this, because it is a

governmental project, there is no profitability analysis. However, an economic analysis can be

57

conducted based on a cost benefit ratio. This requires information such as benefits and

disbenefits to the public. In the future, this will be incorporated into the economic analysis to

provide a clear picture of how sustainable this project is in the long run and if the benefit of the

project is more than its various associated costs.

58

References [1] T. E. Schultz, "Biological Wastewater Treatment," Chemical Engineering, vol. 112, no. 10,

pp. 44-46, 48-51, October 2005.

[2] S. A. Bolles, "Modeling wastewater aeration systems to discover energy savings

opportunities," n.d..

[3] ICON, "Pollutants in urban wastewater and sewage sludge," Office for Official Publications

of the European Communities, Luxembourg, 2001.

[4] Z. Shareefdeen, Effects of Kinetic Interactions on Bioreactor Performance in Wastewater

Treatment Applications, 2014.

[5] Zerajic, S., & Stevanovic, J. (2007). The Kinetic Models of the Bioprocess with Free and

Immobilized Cells.CI&CEQ, 13(4), 216-226. Retrieved October 20, 2014, from

http://www.doiserbia.nb.rs/img/doi/1451-9372/2007/1451-93720704216Z.pdf

[6] Reardon, K. F., Mosteller, D. C., & Bull Rogers, J. D. (2000). Biodegradation Kinetics of

Benzene, Toluene, and Phenol as Single and Mixed Substrates for Pseudomonas putida F1.

Biotechnology and Bioengineering, vol. 69, no. 4, pp. 387-400. [Online]. Available:

http://www.chemeng.queensu.ca/courses/CHEE884/documents/Reardonetal2000Biotechnol

Bioeng69-385.pdf. [Accessed: Sept. 29, 2014].

[7] Reardon, K. F., Mosteller, D. C., Bull Rogers, J. D., DuTeau, N. M., & Kim, K. (2002).

Biodegradation Kinetics of Aromatic Hydrocarbon Mixtures by Pure and Mixed Bacterial

Cultures. Environmental Health Perspectives, vol. 110, no. 6, pp. 1005-1011. [Online].

Available: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1241285/ [Accessed: Sept. 29,

2014].

59

[8] Chang, Y., Cheng, H., Lai, S., & Ning, H., Biodegradation of naphthalene in the oil refinery

wastewater by enriched activated sludge, International Biodeterioration & Biodegradation,

Volume 86, Part C, January 2014, Pages 272-277, ISSN 0964-8305,

http://dx.doi.org/10.1016/j.ibiod.2013.09.018. [Accessed: Sept. 29, 2014]

[9] Yao, H., Ren, Y., Deng, X., & Wei, C., (2011) Dual substrates biodegradation kinetics of m-

cresol and pyridine by Lysinibacillus cresolivorans, Journal of Hazardous Materials,

Volume 186, Issues 2–3, 28 February 2011, Pages 1136-1140, ISSN 0304-3894,

http://dx.doi.org/10.1016/j.jhazmat.2010.11.118. [Accessed: Sept. 29, 2014]

[10] Pierre Le-Clech, Vicki Chen, Tony A.G. Fane, Fouling in membrane bioreactors used in

wastewater treatment, Journal of Membrane Science, Volume 284, Issues 1–2, 1 November

2006, Pages 17-53, ISSN 0376-7388, http://dx.doi.org/10.1016/j.memsci.2006.08.019.

[Accessed: Sept. 29, 2014]

[11] Oh, Y. S., Shareefdeen, Z., Baltzis, B. C., & Bartha, R. (1994). Interactions between

benzene, toluene, and p-xylene (BTX) during their biodegradation. Biotechnology and

Bioengineering, 44(4), 533-538. [Accessed: Sept. 29, 2014]

[12] Zarook, S. M., Shaikh, A. A., & Ansar, Z. (1997). Development, experimental validation

and dynamic analysis of a general transient biofilter model. Chemical Engineering Science,

52(5), 759-773. doi:10.1016/S0009-2509(96)00428-9 [Accessed: Dec. 16, 2014]

[13] Rafiei, B., Naeimpoor, F., & Mohammadi, T. (2014). Bio-film and bio-entrapped hybrid

membrane bioreactors in wastewater treatment: Comparison of membrane fouling and

removal efficiency. Desalination, 337, 16-22. doi:10.1016/j.desal.2013.12.025 [Accessed:

Nov. 22, 2014]

60

[14] Yao, H., Ren, Y., Deng, X., & Wei, C. (2011). Dual substrates biodegradation kinetics of m-

cresol and pyridine by lysinibacillus cresolivorans. Journal of Hazardous Materials, 186(2),

1136-1140. doi:10.1016/j.jhazmat.2010.11.118 [Accessed: Sept. 29, 2014]

[15] Department, E. (2013, 01). Trakhees. Retrieved 05 25, 2015, from

http://www.ehss.ae/forms/guidelineno.en-001airenvironmentjanuary2013.pdf

[16] Engineering Toolbox. (n.d.). The Engineering ToolBox. Retrieved from

http://www.engineeringtoolbox.com/

[17] Hutchins, R. (1974). New Method Simplifies Design of Activated Carbon Systems. Chem.

Engr. , 80, 133-138.

[18] HyperTextBookShop. (n.d.). Retrieved from

http://www.hypertextbookshop.com/biofilmbook/working_version/artifacts/tables/Modul

e_004/Table4-1_DiffCoeffH2O.htm

[19] LaGrega, M. D., Buckingham, P. L., & Evans, J. C. (2001). Hazardous Waste Management

and Environmental Resources Management. Thomas Casson.

[20] Regulation, W. Q. (2013, 04 10). Regulation & Supervision Bureau. Retrieved 05 25, 2015,

from http://www.rsb.gov.ae

[21] Speth, T., & Miltner, R. (1990). Benzene/GAC Isotherm. Retrieved from

http://iaspub.epa.gov/tdb/pages/contaminantProcess/contaminantProcessReferences.do

[22] Towler, G., & Sinnott, R. (2008). Chemical Engineering Design: Principles, Practice, and

Economics of Plant and Process Design. Oxford: Elsevier.

61

Appendix

Meeting Minutes Sample

Senior Design 2 (CHE 491)

Meeting Minutes

Date: 23rd May 2015 Location: Library, IC1

Objective: Final Report and Presentation

Start Time Item Responsible 17:30 0:02 Calls Meeting to order

Opening Remarks to welcome participants

Reviews Agenda

Time Keeper Coordinator

17:32 0:01

Greetings and Welcome

Leader

17:33 17:36 18:00 18:25 18:28 18:30

0:03 0:24 0:25 0:03 0:02

Reviews Last Meeting’s Minutes Discussion about Report

MATLAB

Sizing

Interaction sets

Plant layout Discussion about Presentation

Prepare plant layout with instrumentation

Individual sizing slides

Economic analysis

Finalize what should be done by next meeting

Recorder reviews assignments

Coordinator proposes when next meeting should be: time, date, location

Meeting Adjourned

Leader All Members All Members Recorder Meeting Coordinator

62

Gantt Chart

63

MATLAB

Sample Code (Toluene Non-interaction)

%tau=1800 and X=2000 mg/L for all

%columns: umax(1/s), Ks, Ki, Kbt, Ktb, Y, S0, S, RE (all other units in

%mg/L or mg/mg or %)

tau=1800; %s

x=2000; %mg/L

load parametersoftoluene.txt %load file with all the parameters

umax=parametersoftoluene(1,1); %identify the parameter according to matrix position

ks=parametersoftoluene(1,2);

ki=parametersoftoluene(1,3);

Y=parametersoftoluene(1,6);

r=size(parametersoftoluene,1);%find number of rows

c=size(parametersoftoluene,2); %find number of columns

for m=1:r

B=@(s)parametersoftoluene(m,7)-s-(((1/Y)*umax*s*tau*x)/(ks+s));

if parametersoftoluene(m,7)>296

x0=0.3*parametersoftoluene(m,7);

else x0=0;

end

parametersoftoluene(m,c+1) = fzero(B,x0);

parametersoftoluene(m,c+2) = ((parametersoftoluene(m,7)-

parametersoftoluene(m,c+1))/parametersoftoluene(m,7))*100;

parametersoftoluene(m,c+3)=x0;

end

64

d=size(parametersoftoluene,2);

plot(parametersoftoluene(:,7),parametersoftoluene(:,d-1))

title('Initial Substrate Conc. (S0) vs. Removal Efficiency (RE%) toluene')

xlabel('Initial Substrate Conc., S0 (mg/L)')

ylabel('Removal Efficieny, RE (%)')

grid on

% fid = fopen('toluenesolvedputida.txt','wt');

% for ii = 1:size(parametersoftoluene,1)

% fprintf(fid,'%g\t',parametersoftoluene(ii,:));

% fprintf(fid,'\n');

% end

% fclose(fid);

65

Polymath

Non- Interaction p.putida 01 for Benzene (monode): tau=30 #mins #Fo= 3785.41 lit/day Fo=157.7255 #L/hr V=(tau/60)*Fo #L F=.95*Fo mu=.44 #1/hr ks=3.36 #mg/L y=0.65 #g/g X=2000 #mg/L f(So)=Fo*So+F*S-(1/y)*((mu*S)/(ks+S))*X*V So(0)=500 #mg/L f(S)= Fo*So+F*S-(1/y)*((mu*S)/(ks+S))*X*V S(0)=8.8812 #mg/L

Re= (So-S)/(So) *100

Results :

p.putida 01 for toluene (andrews): tau=30 #mins #Fo= 3785.41 L/day Fo=157.7255 #L/hr V=(tau/60)*Fo #L F=.95*Fo mu=.72 #1/hr

66

ks=15.07 #mg/L ki= 44.43 #mg/L y=0.64 #g/g X=2000 #mg/L f(So)=Fo*So+F*S-(1/y)*((mu*S)/(ks+S+ (S^2)/ki))*X*V So(0)=500 #mg/L f(S)= Fo*So+F*S-(1/y)*((mu*S)/(ks+S+(S^2)/ki))*X*V S(0)=15.4809#mg/L

Re= (So-S)/(So) *100

Results :

#consortium for toluene (andrews): tau=30 #mins #Fo= 3785.41 L/day Fo=157.7255 #L/hr V=(tau/60)*Fo #L F=.95*Fo

67

mu=.86 #1/hr ks=11.03 #mg/L ki= 78.94 #mg/L y=0.71 #g/g X=2000 #mg/L f(So)=Fo*So+F*S-(1/y)*((mu*S)/(ks+S+ (S^2)/ki))*X*V So(0)=500#mg/L f(S)= Fo*So+F*S-(1/y)*((mu*S)/(ks+S+(S^2)/ki))*X*V S(0)=8.1119#mg/L

Re= (So-S)/(So) *100

Results :

consortium for benzene (monode): tau=30 #mins #Fo= 3785.41 lit/day Fo=157.7255 #L/hr V=(tau/60)*Fo #L F=.95*Fo mu=.68 #1/hr ks=12.22 #mg/L y=0.71 #g/g X=2000 #mg/L f(So)=Fo*So+F*S-(1/y)*((mu*S)/(ks+S))*X*V

68

So(0)=500 #mg/L f(S)= Fo*So+F*S-(1/y)*((mu*S)/(ks+S))*X*V S(0)=8.8812 #mg/L

Re= (So-S)/(So) *100

Results:

Environmental Health Perspectives • VOLUME 110 | SUPPLEMENT 6 | DECEMBER 2002 1005

Organic chemical mixtures are prevalent inwastewater from industrial and municipalsources as well as in contaminated groundwa-ter. Common examples of chemical mixturesthat often become pollutants include gasolineand other petroleum fuels, pesticides, andwood-treating substances. Landfill leachates arecomplex mixtures that contaminate groundwa-ter supplies around the world. Pollutant mix-tures may contain only organic chemicals ormay also include inorganics, heavy metals, orradionuclides. The occurrence of contaminantsin mixtures is an important problem becausethe removal or degradation of one componentcan be inhibited by other compounds in themixture and because different conditions maybe required to treat different compoundswithin the mixture. The work reported herewas motivated by the first of these issues as itapplies to pollutant biodegradation.

Researchers have noted that microbialdegradation (metabolism) of a compound ina mixture can be strongly affected by othersubstituents of the mixture (1–4). This hasbeen observed not only for mixtures of toxicchemicals (bioremediation) but also for mix-tures of pollutants and readily degraded com-pounds (wastewater treatment) and mixturesof sugars (fermentation). To understand mix-ture effects, one must consider the metabolicrole each compound plays for the microor-ganisms. The terms “homologous” and “het-erologous” have been proposed by Harderand Dijkhuizen (5) for compounds that servethe same or different roles, respectively.

The effects of other compounds in amixture of homologous carbon and energysubstrates on the biodegradation of a chemi-cal can be positive, as in the case of increasedgrowth at low substrate concentrations (6,7)or induction of required degradativeenzymes (8). More commonly, negativeinteractions are reported. Reasons fordecreased biodegradation rates include com-petitive inhibition (9–11), toxicity (12), andthe formation of toxic intermediates bynonspecific enzymes (13,14).

Although mathematical models of mixedhomologous substrate consumption andmicrobial growth have been proposed [e.g.,(2,11,15–19)], this body of literature ismuch smaller than that for the modeling ofsingle-substrate growth kinetics. Most mod-els have been tested with only two sub-strates, and their applicability to largermixtures has been assumed without valida-tion. More recently, models have been pro-posed and tested for larger mixtures.Examples include the growth of Escherichiacoli on six sugars (16), the growth of a mixedculture on benzene, toluene, ethylbenzene,and o- and p-xylene (BTEX compounds)(11), and the biodegradation of threepolycyclic aromatic hydrocarbons (20).

In addition to the interactions amongchemical components of a mixture undergoingbiodegradation, the interactions among micro-bial species in a mixed culture may be impor-tant. For example, Lewandowski andco-workers (21) studied the biodegradation of

phenol by several two-species mixed cultures.Excellent agreement between pure-and-simplecompetition theory and experimental dataoccurred when the two species in an experi-ment were both isolated from the same envi-ronment. However, when a mixture wascomposed of two organisms from differentenvironments, there was no agreement withthe pure-and-simple competition model.Conversely, in research by Murakami andAlexander (22), interspecies interactionsbeyond pure-and-simple competition, includ-ing interactions harmful to one species whilethe other was unaffected, occurred betweenmembers of a binary culture isolated from thesame sewage treatment plant.

At Colorado State University, our grouphas been studying the biodegradation kineticsof chemical mixtures for several years. Thelong-term goal of this research is to under-stand (and model mathematically) thebiodegradation of complex chemical mixturesby microbial communities. Our strategy is tofirst learn from simpler (but representative)systems: pure cultures degrading mixturesand mixed cultures degrading single chemi-cals (Figure 1). In this report, we review ourresults at this intermediate level and thendescribe the results from a simple chemicalmixture–microbial mixture experiment.Finally, we discuss some preliminary resultsthat may provide insights on the observationsmade previously.

This work has focused on the biodegrada-tion of benzene, phenol, and toluene. Thesemonoaromatic compounds are ideal represen-tatives of chemicals found in pollutant mix-tures. They are produced in very largequantities for use as fuels, solvents, and starting

This article is part of the monograph Application ofTechnology to Chemical Mixture Research.

Address correspondence to K.F. Reardon, Dept. ofChemical Engineering, 200 W. Lake St., ColoradoState University, Ft. Collins, CO 80523-1370 USA.Telephone: (970) 491-6505. Fax: (970) 491-7369.E-mail: [email protected]

This work was supported by grant 5 P42ES05949-05 from the National Institute ofEnvironmental Health Sciences and by the ColoradoAdvanced Technology Institute through a fellowshipfor D.C.M. received from the Colorado Institute forResearch in Biotechnology. P. putida F1 was pro-vided by D. Gibson, and Burkholderia sp. strainJS150 was provided by J. Spain. T. Keefe providedstatistical analysis and assistance.

Received 18 December 2001; accepted 13 August2002.

Chemical Mixtures

Microbial growth on pollutant mixtures is an important aspect of bioremediation and wastewatertreatment. However, efforts to develop mathematical models for mixed substrate kinetics havebeen limited. Nearly all models group either the microbial population (as “biomass”) or the chem-ical species (e.g., as biological oxygen demand). When individual chemical species are considered,most models assume either no interaction or that the nature of the interaction is competition forthe same rate-limiting enzyme. And when individual microbial species are considered, simple com-petition for the growth substrate is the only interaction included. Here, we present results usingPseudomonas putida F1 and Burkholderia sp. strain JS150 growing individually and together onbenzene, toluene, phenol, and their mixtures and compare mathematical models to describe theseresults. We demonstrate that the simple models do not accurately predict the outcome of thesebiodegradation experiments, and we describe the development of a new model for substrate mix-tures, the sum kinetics with interaction parameters (SKIP) model. In mixed-culture experiments,the interactions between species were substrate dependent and could not be predicted by simplecompetition models. Together, this set of experimental and modeling results presents our currentstate of work in this area and identifies challenges for future modeling efforts. Key words: benzene,biodegradation kinetics, mixed growth substrates, phenol, Pseudomonas putida F1, toluene.Environ Health Perspect 110(suppl 6):1005–1011 (2002).http://ehpnet1.niehs.nih.gov/docs/2002/suppl-6/1005-1011reardon/abstract.html

Biodegradation Kinetics of Aromatic Hydrocarbon Mixtures by Pure and Mixed Bacterial Cultures

Kenneth F. Reardon,1,2 Douglas C. Mosteller,1 Julia Bull Rogers,1 Nancy M. DuTeau,2 and Kee-Hong Kim1

1Department of Chemical Engineering and 2Department of Microbiology, Immunology, and Pathology, Colorado State University, Fort Collins, Colorado, USA

1006 VOLUME 110 | SUPPLEMENT 6 | DECEMBER 2002 • Environmental Health Perspectives

materials for chemical syntheses (23). As anoutcome of this prevalent use, monoaromaticsare widespread environmental contaminants,usually in mixtures. Thirty monoaromatics arelisted in the U.S. Environmental ProtectionAgency’s Priority Pollutants (24), and 11 ofthese compounds are in the top 100 chemicalson the Agency for Toxic Substances andDisease Registry’s Priority List of HazardousSubstances (25). Two bacterial strains wereused in the work presented here: Pseudomonasputida F1 and Burkholderia sp. JS150.P. putida F1 uses toluene dioxygenase (TDO)to initiate the metabolism of toluene, benzene,phenol, and other aromatics (26). In contrast,Burkholderia sp. JS150 can express at leastthree initial dioxygenases (12). These strainsare thus interesting and distinct model systemsfor the study of mixture biodegradationkinetics.

Materials and Methods

Microorganisms. P. putida F1 is a well-characterized aromatic hydrocarbon–degrad-ing bacterium that can use toluene, benzene,ethylbenzene, phenol, and other aromatics assole carbon and energy sources (26). Thebiodegradation of toluene by P. putida F1begins with the oxidation of the aromatic ringby TDO to form cis-toluene dihydrodiol(26–28), which is then dehydrogenated toform 3-methylcatechol. This molecule is thencleaved at the meta position and then con-verted in three steps to acetaldehyde and pyru-vate (29–31). TDO also catalyzes theoxidation of benzene (28,32,33) and phenol(34,35). In both cases, catechol is formed afterdehydrogenation and is then further degradedby meta ring cleavage and other reactions totricarboxylic acid cycle intermediates. Thus,P. putida F1 uses the same metabolic pathwayto metabolize toluene, benzene, and phenol.

Burkholderia sp. JS150 is a nonencapsu-lated mutant of Burkholderia sp. JS1obtained after ethyl methane sulfonate muta-genesis of strain JS1 (12). In addition totoluene, benzene, and phenol, this species isable to degrade a wide range of substitutedaromatic compounds. Strain JS150 has amuch greater metabolic capability thanP. putida F1, with the ability to synthesize at

least four ring-fission pathways and use threeseparate initial dioxygenases (including anonspecific TDO) when grown on varioussubstrates (12).

Media. For all experiments, a modifiedHutner’s mineral base was used as the carbon-free medium (36), and toluene, benzene,and/or phenol was added. Phenol was addedbefore autoclaving, but toluene and benzenewere added after autoclaving to minimizelosses from volatilization (37). For strainmaintenance, cultures of both bacteria weregrown on toluene vapors and stored at –70°Cin 10% glycerol.

Chemicals. Benzene (Sigma, St. Louis,MO, USA; HPLC grade), toluene (Baker,Phillipsburg, NJ, USA; HPLC grade), andphenol (Sigma, >99.5% pure) were used as thecarbon sources. Chloroform and p-xylene(both from Baker; GC grade) were used toprepare samples for gas chromatography (GC).All chemicals used for media preparation werereagent grade.

Analytical methods. Cell concentrationswere measured as optical density at 600 nm(OD600) with a Bausch & Lomb Spectronic21 spectrophotometer (Bausch & Lomb,Rochester, NY, USA) and correlated to bio-mass concentration (37,38). To quantify thetwo cell populations in the mixed cultureexperiments, a fluorescence in situ hybridiza-tion (FISH) method was developed (39). Inthis procedure, 30 µL of a culture sample wasapplied to slides, which were then dried,fixed, and dehydrated. The samples were thenexposed to species-specific oligonucleotideprobes that were 5´-end labeled with fluores-cein isothiocyanate (40), rinsed, and preparedfor counting under a Leitz epifluorescencemicroscope (Leica Microsystems AG, Wetzlar,Germany) equipped with a BioQuant imageanalysis system (R&M Biometrics Inc.,Nashville, TN, USA)(39). For speciesBurkholderia sp. JS150, correlations were alsodeveloped between cell concentration(cells/mL or cfu/mL) and biomass concentra-tion (mg/L): 3.5 × 106 cells/mL equivalent to1.0 mg/L biomass for cells grown on toluene,and 2.4 × 106 cells/mL equivalent to1.0 mg/L biomass for cells grown on phenol.

Benzene, toluene, and phenol concentra-tions were measured by GC. Aqueous sampleswere extracted with chloroform, and p-xylenewas used as an internal standard (37). Sampleswere stored at 4°C in 2-mL screw-cap vialswith Teflon-lined rubber septa, until analysis.Benzene, toluene, and phenol standards wereprepared as aqueous solutions and extractedwith chloroform/p-xylene. The detection limitof this method for each of the threecompounds was 5 µM.

Protocol for batch biodegradationexperiments. All data for biodegradationkinetics modeling were obtained from batch

bioreactor cultivations inoculated from shakeflask cultures grown on the same carbonsource(s) used in the bioreactor. Two 3-LApplikon batch bioreactors (Applikon,Foster City, CA, USA) were used for thebiodegradation kinetic experiments. Thetotal initial substrate concentration wasapproximately 0.5 mM in the liquid phase,regardless of the number of substratesinvolved. In mixture experiments, the sub-strates were added in approximately equi-molar amounts. Henry’s law was used tocalculate the amount of toluene or benzeneto be added. All experiments were run at theoperating and initial conditions found toprovide intrinsic biodegradation kinetics,including a low inoculum size (expressed asthe ratio of substrate to cell mass; a value of300 was used) (37). The bioreactor was runas a closed system with no air sparging toeliminate the substrate loss due to volatility.The system operated aerobically (dissolvedoxygen levels remained above 5 mg/L),30°C, and without pH control (although thepH remained in the range of 6.7–6.9). Lessthan 1% of toluene and phenol was lost insterile control experiments. Biodegradationexperiments were performed in duplicate,and replicates were not performed simultane-ously. Additional experimental details can befound in Reardon et al. (37).

Determination of biodegradation kineticsmodel parameters. Several mathematicalmodels were compared for their ability to fitor predict the experimental biodegradationkinetics data. The values of all requiredmodel parameters were determined by per-forming nonlinear curve fitting to the exper-imental data using SimuSolv, a modelingand simulation package (Dow ChemicalCompany, Midland, MI, USA). Simusolvemployed a Gear method to solve the differ-ential equations and maximized the log of thelikelihood function (LLF) to optimize theunknown parameters and discriminatebetween models (37). The model with themaximum LLF value and most homogeneouserrors residual plots was chosen. For eachfinal model, the percent variation explained(PVE; similar to r 2 value for linear regression)was calculated using the LLF. The averagevalue for each of the parameters was found byseparately determining the values for each ofthe duplicate experiments and then averagingthese two values.

The tested models were those for cellgrowth kinetics (as a function of growth sub-strate consumption). An equation was alsoneeded to model substrate depletion. For therelatively nonvolatile substrate phenol, therate of consumption was described as

[1]dSdt

XY X / S

= −µ

,

Chemical Mixtures • Reardon et al.

Single chemical

Single microbialspecies

Single chemical

Multiple microbialspecies

Chemical mixture

Single microbialspecies

Chemical mixture

Multiple microbialspecies

Figure 1. Levels of complexity in biodegradationkinetics research.


where S is the substrate concentration, t istime, µ is specific growth rate, YX/S is thebiomass yield, and X is biomass concentration.To determine the yield, YX/S, the concentrationof cells produced (cells/mL) was divided by theconcentration of substrate consumed (mM).

Because toluene and benzene are volatile,Equation 1 required modification to accountfor the presence of toluene in both the gasand liquid phases in the bioreactor. Microbialgrowth rates depend on the liquid-phase sub-strate concentration only, whereas the bio-mass yield is a function of the change in totalmass of substrate. Because the cultivationconditions were chosen to ensure that masstransfer rates (from gas to liquid phase) werealways faster than biodegradation rates (37),the masses of toluene in the liquid and gasphases could be related using Henry’s law,yielding

[2]

Here, m refers to the mass of toluene in thegas phase (subscript G), liquid phase (L), orthe entire system (TOT). H is the Henry’slaw constant, R is the gas constant, T is thetemperature, and VG and VL are the gas andliquid phase volumes. Henry’s law constantsof 8.08 × 10–3 atm·m3/mol for toluene and7.31 × 10–3 atm·m3/mol for benzene at 30°Cwere used (41). The temperature and volumeof liquid remained essentially constantduring an experiment, and therefore the rateof substrate consumption can be written as

[3]

In most experiments, a certain amount oflag time was observed before any measurabledepletion of substrate or growth of organismsoccurred. Because the models do not accountfor this lag time, time zero for modeling wasdefined as the time when 2% of the substratehad been consumed.

ResultsBiodegradation of chemical mixtures bypure cultures of P. putida F1. The first setof experiments involved the use of P. putidaF1 to biodegrade benzene, toluene, phenol,and their binary and tertiary mixtures. Inthe single-substrate experiments, the growthkinetics were well fit by the Monod model,

[4]

in which µmax is the maximum specificgrowth rate and KS is the Monod half-satura-tion constant. Equation (3) was used tomodel the consumption (biodegradation) oftoluene and benzene. The Monod modelparameter values for each of the three sub-strates are listed in Table 1. The Monodmodel provided the best fit for the biodegra-dation of toluene and benzene by P. putidaF1, although substrate inhibition has beenreported for growth on toluene by othermicroorganisms (10). However, in the case ofgrowth on phenol, well known as aninhibitory substrate, the fit to the experimen-tal data was slightly improved by use of theAndrews model (37):

[5]

where Ki is an inhibition parameter. Inaddition, the growth pattern with phenol as asubstrate was different than that for toluenein that biomass production continued forapproximately 10 hr after phenol wasdepleted. This is indicative of the transientproduction of an intermediate that was thanconsumed for growth, and we therefore testedvarious models that included such an inter-mediate. However, none of these yielded animproved fit to the data (37). We chose touse the Monod model rather than theAndrews model because the differencesbetween the model fits were small andbecause use of the Andrews model with itsadditional parameter did not improve theprediction of mixture experiments.

The results of a biodegradation experi-ment with toluene and phenol are shown inFigure 2. Toluene was consumed beforephenol, and phenol biodegradation did notbegin until toluene was nearly depleted.Although this sequential substrate consump-tion is reminiscent of diauxic growth, theclassic definition of that phenomenon (induc-tion or derepression of catabolic enzymes)does not apply here because P. putida F1 usesthe same enzymes to metabolize both sub-strates (35). Similarly, when this species wasgrown on a 50:50 mixture of benzene andphenol, benzene was degraded first, and phe-nol consumption did not begin until benzeneconcentrations were near zero (Figure 3). Inthe case of the toluene–benzene mixture, P.putida F1 consumed both of these substratessimultaneously during most of the cultiva-tion, but toluene biodegradation began beforethat of benzene, and toluene was depletedfirst (Figure 4).

A common model for cell growth onhomologous substrate mixtures is a no-inter-action sum kinetics model, in which the spe-cific growth rate is the sum of the specificgrowth rates on each substrate i (µi). The rateof consumption for substrate i can be

µ = =

µ

+ +

12X

dXdt

S

K S S Kmax L

S L L i/,

µ = =

µ+

1X

dXdt

SK S

max L

S L

,

α

dSdt

S X

YL L

X / S

= −µ( )

.

m m m

mHRT

VV

m

L G

LG

LL

TOT = +

= +

=1 α .

Chemical Mixtures • Bacterial biodegradation kinetics of mixtures

Table 1. Parameters for Monod and SKIP models of biodegradation of mixtures.a

Growth µm KS YX/S I1,2 I2,1Microorganism substrate (per hour) (mg/L) (g/g) (-) (-) PVE

P. putida F1 Toluene 0.86 ± 0.01 13.8 ± 0.9 1.28 ± 0.13 N/A N/A 98.4Benzene 0.73 ± 0.03 0.12 ± 0.02 1.20 ± 0.05 N/A N/A 86.6Phenol 0.11 ± 0.01 32.0 ± 2.4 0.80 ± 0.07 N/A N/A 93.9Toluene–phenol * * * 55 ± 5 0.01 ± 0.002 98.1Toluene–benzene * * * 5 ± 0.3 0.01 ± 0.003 95.7Benzene–phenol * * * 18.5 ± 1.5 0.01 ± 0.002 94.2Toluene–benzene–phenol * * * * * 96.7

Burkholderia sp. JS150 Toluene 0.39 ± 0.01 1.01 ± 0.28 1.03 ± 0.09 N/A N/A 96.3Phenol 0.31 ± 0.03 0.51 ± 0.38 0.88 ± 0.005 N/A N/A 99.1Toluene–phenol * * * 80.6 ± 6 0.6 ± 0.03 97.3

aFor the parameters I1,2 and I2,1, subscript 1 refers to the first chemical in the pair. The notation “N/A” is shown when a parameter was not used to model growth on the substrate indicated;*indicates that previously determined values of that parameter (from single-substrate experiments) were used.

Time (hr)

Cell

con

cent

ratio

n (m

g/L)

Subs

trat

e co

ncen

trat

ion

(mg/

L)

60

50

40

30

20

10

00 5 10 15 20 25 30 35 40

30

25

20

15

10

5

0 ◆◆◆◆◆◆◆◆◆◆◆◆

◆◆

◆◆◆◆

◆◆ ◆◆◆◆

◆◆

◆◆ ◆◆◆◆ ◆◆

●●

▲▲▲▲▲▲▲▲▲▲

▲

▲

▲

▲

▲▲ ▲

●

●

●

●●●●

●◆◆

▲

PhenolTolueneBiomassSumCompetitiveSKIP

▲

●◆◆

◆◆

▲

Figure 2. Experimental data and model output forbatch biodegradation of a toluene–phenol mixtureby P. putida F1. Symbols indicate measurements ofliquid-phase toluene (●), phenol (▲), and biomassconcentrations (♦♦). Lines are predictions from thesum kinetics, no-interaction model (dotted lines),competitive inhibition model (dashed lines), andSKIP model (solid lines). Adapted from Reardonet al. (37).

modeled using Equation 1 or Equation 3, asappropriate. Because the Monod model wasfound to be suitable for biodegradation ofeach of the three monoaromatics individually,the no-interaction sum kinetics model is

[6]

where the subscripts 1 and 2 refer to each ofthe two substrates. The predictions of thismodel for the toluene–phenol mixture areshown in Figure 2. Comparison of these pre-dictions with the toluene–phenol data clearlyreveals mixture effects because phenolbiodegradation occurred later and at a lowerspecific (per cell) rate than predicted by themodel. Thus, the presence of toluene inhib-ited phenol biodegradation. However, phenolhad little effect on toluene consumption.Benzene also inhibited phenol biodegrada-tion, although phenol did not have a signifi-cant impact on the rate of benzenemetabolism (37). Finally, when the modelwas applied to toluene–benzene mixtures, thebiodegradation of benzene was predicted tobe earlier and faster than was actually mea-sured, suggesting that the presence of tolueneinhibited the degradation of benzene. In con-trast, the presence of benzene had little effecton toluene consumption (37). Thus, mixtureeffects (i.e., nonadditivity) were found withall three pairwise combinations of these threemonoaromatics.

Because P. putida F1 uses TDO toinitiate catabolism of all three chemicals, onemight expect that these mixture effects aredue to competitive inhibition of this enzyme.A sum kinetics model incorporating purelycompetitive substrate kinetics (18) is

[7]

Predictions from this model are shown inFigures 2–4 for each of the binary mixtures.In the case of the toluene–phenol mixture,the model prediction for phenol degradationrepresented the data better than did the no-interaction model, but the agreement withthe toluene data was worse. Thus, the one-sidedness of the mixture effect was not wellpredicted. Similar phenomena occurredwhen Equation 7 was used to predict thebiodegradation of benzene–phenol andtoluene–benzene mixtures. Models incorpo-rating noncompetitive and uncompetitiveinteractions have also been tested, but nonegave satisfactory results (37).

To account for these mixture effects, analternative model was formulated by incorpo-rating an interaction parameter Ii,j into thesum kinetics framework (37):

[8]

Here, Ii,j indicates the degree to whichsubstrate i affects the biodegradation of sub-strate j, with larger values corresponding tostronger inhibition. Yoon et al. (18) were thefirst to propose a model of this type, which

we call sum kinetics with interaction parame-ters (SKIP). To obtain the values of the inter-action parameters (Table 1), the SKIP modelwas fitted to each set of binary mixture datasets using values of µm, KS, and YX/S deter-mined from the single-substrate experiments.The fitted SKIP model accurately describesthe biodegradation data for all three binarymixtures (Figures 2–4), demonstrating thatthe SKIP model can be used to fit unspecifiedtypes of inhibition between two substrates.

The ability of the SKIP model to predictthe outcome of the 3-substrate mixture wasalso examined. As shown in Figure 5, the con-sumption of toluene began first, followed bybenzene, and these two chemicals were thendegraded simultaneously. Significant phenolconsumption did not begin until the tolueneconcentration was nearly zero and the benzeneconcentration was low. A three-term versionof Equation 8 successfully predicted this pat-tern using parameters determined indepen-dently from the one- and two-substratemixture experiments (37).

Biodegradation of chemical mixtures bypure cultures of Burkholderia sp. strainJS150. A second study, using Burkholderia sp.strain JS150, was performed to investigatemixed-substrate biodegradation by a bac-terium that employs different catabolic path-ways to degrade the mixture components. TheMonod model was found to fit the biodegra-dation data well for both toluene and phenol(Table 1) (38). During growth of strain JS150on phenol, the release of several metabolitesinto the medium was noted, and one wasidentified as 2-hydroxymuconic semialdehyde(38). However, these metabolites did notinhibit phenol consumption.

When strain JS150 was grown on anequimolar solution of toluene and phenol,

µ =µ

+ +

+µ

+ +

max,

S

max,

S

S

K S I S

S

K S I S

1 1

1 1 2 1 2

2 2

2 2 1 2 1

, ,

, ,

µ =µ

+ +

+µ

+ +

max,

S,S,

S,2

max,

S,S,

S,

S

K SKK

S

S

K SKK

S

1 1

1 11

2

2 2

2 22

11

.

µ =µ

++

µ

+max,

S

max,2

S

S

K S

S

K S1 1

1 1

2

2 2, ,

,



PhenolBenzeneBiomassCompetitiveSKIP

▲

◆◆■

◆◆

Subs

trat

e co

ncen

trat

ion

(mg/

L)

Cell

conc

entr

atio

n (m

g/L)

Time (hr)

60

50

40

30

20

10

0

35

30

25

20

15

10

5

0 ▲ ▲ ▲ ▲ ▲

▲▲ ▲▲▲▲▲▲

▲▲

▲

▲

◆◆

◆◆

◆◆◆◆

◆◆ ◆◆ ◆◆ ◆

◆◆ ◆◆

◆◆

◆◆◆◆

■ ■■■■

■

■■■

0 10 20 30 40 50

▲

■

◆◆

◆◆ ◆◆◆◆◆◆

▲

Figure 3. Experimental data and model output forbatch biodegradation of a benzene–phenol mixtureby P. putida F1. Symbols indicate measurements ofliquid-phase benzene (■), phenol (▲), and biomassconcentrations (♦♦). Dashed lines are predictionsfrom the competitive inhibition model, and solidlines are curve fits for the SKIP model. Reprintedfrom Reardon et al. (37) with permission from JohnWiley & Sons, Inc.

25

20

15

10

5

0

70

60

50

40

30

20

10

0

■ ■ ■■ ■

■

■

■

■◆◆

◆◆

◆◆

◆◆

◆◆

◆◆◆◆ ◆◆ ◆◆

● ●

●●

●

●●

Subs

trat

e co

ncen

trat

ion

(mg/

L)

Cell

conc

entr

atio

n (m

g/L)

0 2 4 6 8 10 12 14 16

Time (hr)

◆◆■

●

TolueneBenzeneBiomassCompetitiveSKIP

◆◆■●

◆◆

●●

◆◆

Figure 4. Experimental data and model output forbatch biodegradation of a toluene–benzene mix-ture by P. putida F1. Symbols indicate measure-ments of liquid-phase toluene (●), benzene (■),and biomass concentrations (♦♦). Dashed lines arepredictions from the competitive inhibition model,and solid lines are curve fits for the SKIP model.Reprinted from Reardon et al. (37) with permissionfrom John Wiley & Sons, Inc.

PhenolTolueneBenzeneBiomass

▲

◆◆

●

■

◆◆

◆◆

25

20

15

10

5

0

70

60

50

40

30

20

10

0

Subs

trat

e co

ncen

trat

ion

(mg/

L)

0 5 10 15 20 25 30

Cell

conc

entr

atio

n (m

g/L)

Time (hr)

▲▲ ▲ ▲▲▲▲▲▲▲

▲

▲

▲

▲

▲▲ ▲ ▲ ▲ ▲

●● ●●

●●

●

●

●●

◆◆

■■ ■ ■■ ■■

■

■

■

■

■◆◆ ◆◆ ◆◆◆◆◆◆◆◆

◆◆

◆◆◆◆

◆◆◆◆

◆◆ ◆◆ ◆◆ ◆◆ ◆◆◆◆

Figure 5. Experimental data and SKIP modelpredictions for batch biodegradation of atoluene–benzene–phenol mixture by P. putida F1.Symbols indicate measurements of liquid-phasetoluene (●), benzene (■), phenol (▲), and biomassconcentrations (♦♦); lines are model predictions.Reprinted from Reardon et al. (37) with permissionfrom John Wiley & Sons, Inc.

toluene consumption began first. However,phenol was degraded while toluene waspresent in the medium, in contrast to theexperiments with P. putida F1. The use of theno-interaction mixtures model (Equation 6)revealed that the presence of each substratehad an inhibitory effect on the biodegrada-tion of the other. Competitive (Equation 7),noncompetitive, and uncompetitive inhibi-tion models were also tested, although notmechanistically supported because strainJS150 uses multiple biodegradation pathways(12). The predictions from all three modelswere poor. Finally, the SKIP model wasapplied, with IT,P and IP,T values obtained byfitting Equation 8 to the data. The model fitswere very good (PVE = 97.3%), and thevalues of the interaction parameters indicatethat toluene inhibited phenol degradationmuch more than the reverse.

Biodegradation of single chemicals bymixed cultures. Biodegradation models formixed cultures often treat the microorganismsas a single lumped quantity (e.g., total bio-mass). However, this was shown to be inade-quate in the case of a 1:10 mixture of strainJS150 and P. putida F1 growing on phenol(39), suggesting that interactions betweenthese two species were important. Further evi-dence for complex interactions was obtainedby cultivating 1:1 mixtures of these species onphenol and measuring the sizes of the twopopulations using the FISH protocol (Figure6). The resulting kinetics did not follow amodel derived from the concept of pure-and-simple competition, in which the only interac-tion is competition for a growth-limitingsubstrate. Instead, P. putida F1 grew muchmore than the model predicted, and strainJS150 grew less than predicted by this simple

competition model. Further investigationdemonstrated that strain JS150 released ametabolite, probably 2-hydroxymuconic semi-aldehyde, which strain F1 was able to use as agrowth substrate, and thus the interactionbetween the two species included commensal-ism in addition to competition (39). Furthercomplexity was added by including Bacillussubtilis American Type Culture Collection7003, a species unable to grow on phenol, inthe mixed culture. When medium containingphenol was inoculated with a 1:1:1 ratio of thethree microorganisms, B. subtilis grew to agreater extent than did species JS150,presumably by competing for metabolicintermediates (40).

Purely competitive interactions were alsoinsufficient to describe the dynamics betweenstrains JS150 and F1 when similar experimentswere conducted with toluene (Figure 7). Inthis case, P. putida F1 grew more slowly and toa lesser extent than predicted by the pure-and-simple model. Using spent medium tests, thiswas determined to be the result of inhibitionby an unidentified chemical released by speciesJS150 (39). Thus, amensalism occurred alongwith competition when these species grewtogether on toluene.

Biodegradation of a chemical mixture by amixed culture. Finally, the biodegradation andgrowth kinetics of the 1:1 mixed culture ofstrains JS150 and F1 were examined for anequimolar mixture of toluene and phenol. Theexperimental results for the aromatic hydrocar-bons and both microbial populations areshown in Figure 8, along with the predictionsof a model based on the SKIP representation ofsubstrate consumption (Equation 8) and thepure-and-simple kinetics representation ofmicrobial growth. As was the case when thismixed culture was grown on either toluene orphenol alone, the model predictions were poor.

P. putida F1 grew faster and to a greater extentthan predicted by the model, and the growth ofstrain JS150 was less than predicted. Given theconflicting impacts on strain F1 in the phenol-only and toluene-only cultivations noted above,it is interesting to note that the phenol patterndominated in this mixed substrate experiment.In addition the concentrations of both sub-strates reached nondetect levels sooner thanpredicted by the model, indicating that themixed culture is able to degrade the mixturefaster than either pure culture alone.

Discussion

The results presented here clearly illustratethat the biodegradation kinetics of chemicalmixtures can be complex and difficult todescribe mathematically, even when thechemicals serve as homologous substrates forpure cultures of microorganisms. Althoughthese kinetics can in some cases be describedby relatively simple no-interaction (16) orcompetitive inhibition (9,18,20) models, wehave demonstrated that such models areinadequate for P. putida F1 growing on mix-tures of toluene, benzene, and phenol and forBurkholderia sp. JS150 growing on mixturesof toluene and phenol. Furthermore, thebiodegradation kinetics of a mixed culturegrowing on 1-butanol, 2-butoxyethanol, andN,N-dimethylethanolamine also were notwell predicted by competitive inhibition(42). These findings led us to develop theSKIP model, in which a fitting parameter, Ii,jwas introduced to describe the influence ofchemical i on the rate of biodegradation ofchemical j. Using Ii,j values obtained fromthe two-chemical experiments, we demon-strated the ability of the model to predict theoutcome of the three-chemical biodegrada-tion experiments. The SKIP framework hasalso been used as the basis of a model in



●●

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

20

15

10

5

0▲▲ ▲ ▲ ▲

Phen

ol c

once

ntra

tion

(mM

)

Cell

conc

entr

atio

n (1

07 cel

ls/m

L)

●●◆◆ ●●◆◆ ●●◆◆●●●●◆◆ ◆◆ ◆◆ ◆◆

◆◆◆◆

●●

●●

▲▲

▲

▲

▲

◆◆

●●●●

▲ ▲ ▲

▲

▲

●●

●●●●

●●

◆◆ ◆◆

◆◆◆◆◆◆

Time (hr)

0 5 10 15 20 25 30 35

▲Phenol

JS150

P. putida F1

Figure 6. Experimental data and model predictionsfor batch biodegradation of phenol inoculated witha 1:1 mixture of strains JS150:P. putida F1. Symbolsrepresent the experimental data values for phenol(▲), strain JS150 (●●), and strain F1 (♦♦). Lines depictthe pure-and-simple competition model output.Error bars represent one standard deviation basedon replicate analyses of each sample. Reprintedfrom Bull Rogers et al. (39) with permission fromJohn Wiley & Sons, Inc.

◆◆ ◆ ◆ ◆◆

0.40

0.35

0.30

0.25

0.20

0.15

0.10

0.05

0

12

10

8

6

4

2

0

Tolu

ene

conc

entr

atio

n (m

M)

Cell

conc

entr

atio

n (1

07 cel

ls/m

L)● ● ●●

●●

●

● ● ● ● ●● ● ● ● ●● ● ●

●

●●●●

◆

◆

◆

◆◆

◆◆

●

Time (hr)0 2 4 6 8 10 12 14 16

Toluene

JS150

P. putida F1

Figure 7. Experimental data and model predictionsfor batch biodegradation of toluene inoculatedwith a 1:1 mixture of strains JS150:P. putida F1.Symbols represent the experimental data valuesfor toluene (●), strain JS150 (●●), and strain F1 (♦♦).Lines are the pure-and-simple competition modeloutput. Reprinted from Bull Rogers et al. (39) withpermission from John Wiley & Sons, Inc.

◆◆

●●

0.30

0.25

0.20

0.15

0.10

0.05

0

15

10

5

0

Time (hr)

Subs

trat

e co

ncen

trat

ion

(mM

)

Cell

conc

entr

atio

n (1

07 cel

ls/m

L)

●

▲ ▲▲Phenol

Toluene

JS150

▲

▲

▲

▲

▲

▲

▲▲▲

● ●

●

●

▲

●●

●

●●◆◆

◆◆

◆◆●●●▲

●

◆◆

◆◆

◆◆◆◆

◆◆

P. putida F1

0 5 10 15 20 25

Figure 8. Experimental data and model predictionsfor batch biodegradation of a toluene–phenol mix-ture by the 1:1 binary mixture of strains JS150:P.putida F1. Symbols represent the experimental datavalues for toluene (●), phenol (▲), strain JS150(●●), and strain F1 (♦♦). Lines are predictions from apure-and-simple competition model between thespecies and the SKIP model for substrate interac-tions. Reprinted from Bull Rogers et al. (39) withpermission from John Wiley & Sons, Inc.

which a 13-chemical mixture was dividedinto four groups and Ii,j values determinedfor interactions between groups. In caseswithout substrate inhibition, this modifiedSKIP model accurately predicted theexperimental outcomes (43).

Although the differences between thepredictions of the SKIP and other models arehighly significant in a statistical sense, theirimpacts do not necessarily appear large in thebatch experiments presented here. The nov-elty of the SKIP model is the inclusion ofinhibition terms that are different than thosein purely competitive, uncompetitive, or non-competitive inhibition. In the case of thetoluene–phenol mixture, toluene inhibitsphenol consumption to a much greater extentthan predicted by the other models, andphenol inhibition of toluene degradation ismuch less. In a batch experiment, the mainoutcomes of this inhibition are a prolongedlag phase before phenol consumption beginsand a faster toluene degradation rate. Becausetoluene is rapidly consumed in this batch cul-tivation, the impacts on phenol degradationare relatively small. However, in a continu-ous-flow bioreactor, the differences amongthese various models would be much morenoticeable. Because phenol is not consumeduntil toluene concentrations fall below somelow level (only accurately represented by theSKIP model), the hydraulic residence timesand sizes of continuous bioreactors treatingtoluene–phenol mixtures would be substan-tially underpredicted unless the SKIP modelwere used. Because aquifers can also be repre-sented as continuous-flow bioreactors, theresult of using the SKIP versus another modelwould be similar but expressed in terms of thesize of the contaminant plume and the lengthof time required for remediation.

Despite the success of the SKIP model inthe cases presented here, the need to includethe fitting parameter Ii,j with no clear mecha-nistic basis is unsatisfying. This is particularlytrue in the case of P. putida F1, where thesame set of enzymes appears to be involved inthe biodegradation of toluene, benzene, andphenol. We have investigated this phenome-non further using two-dimensional polyacry-lamide electrophoresis of soluble proteins (44).Although this proteomic study has not beencompleted, our current evidence points to dif-ferences in the cell membrane composition asone of factors involved in these unexpectedkinetics. Based on the identification of acylcarrier protein as one of the proteins with tran-sient synthesis during biodegradation oftoluene–phenol mixtures, we performed analy-ses of the phospholipid fatty acid content ofP. putida F1 cells. The predominant phospho-lipid fatty acid of cells growing on toluene wascis-7-hexadecenoic acid (16:1w7c), whereascells growing on phenol had high levels of

cyclopropylheptadecanoic acid (cy17:0) intheir membranes. The membranes of cellsgrowing on toluene–phenol mixtures shiftedfrom 16:1w7c to cy17:0 after degradation oftoluene in the medium was complete (45).Based on these findings, we have developed thehypothesis that the inhibition of phenolbiodegradation in the presence of toluene iscaused by very slow transport of phenol intothe cell when the membrane has adapted tothe more hydrophobic environment. Then,when toluene is depleted from the medium,the membrane composition shifts to a formthrough which phenol can more readily dif-fuse. A model based on this hypothesis hasbeen shown to predict toluene–phenol mixtureresults very well using only data from single-substrate experiments. We are continuing ourinvestigations into this hypothesis and will alsoconsider the implications of other proteins thatare differentially expressed by cells growing ontoluene versus phenol.

We have also shown here that theinteractions between microbial species in amixed culture are both significant for thebiodegradation kinetics and difficult to pre-dict. In particular, we noted a large effect ofthe carbon source on the nature of the micro-bial interactions, with commensalism occur-ring when the cells grew on phenol andamensalism observed when toluene was thegrowth substrate. We also noted that thepresence of a secondary degrader (B. subtilis)had an additional impact on the biodegrada-tions by introducing a new type of interac-tion. Although the mechanism of theseinteractions could be determined after theywere observed, it seems unlikely that theycould be predicted from pure culture experi-ments without prior knowledge of all possiblemetabolites produced by each species. Finally,it is interesting to consider the question ofwhether microbial species interactionsbecome less important as the mixed culturesbecome more diverse. For example, althoughthe SKIP model alone did not describe thekinetics of the binary P. putida F1/Burkholderia sp. JS150 culture when “totalbiomass” was used in the model, it was veryaccurate in describing the biodegradationkinetics of a larger (estimated 10–20 species)mixed culture growing on a mixture of 13organic chemicals (43).

Conclusions. Although the biodegradationkinetics of mixed microbial cultures growingon mixtures of organic contaminants areoften assumed to be simple extensions ofpure-culture/single-substrate kinetics, we havedemonstrated that they are not. In the case ofpure cultures growing on aromatic chemicalmixtures, neither a no-interaction nor a com-petitive inhibition model accurately predictedthe mixture kinetics. To overcome this diffi-culty, we developed the SKIP model, which

used model parameters from single- and dual-substrate mixture experiments to accuratelypredict the outcome of the 3-substrate mix-ture experiment. When we conducted similarexperiments with a binary mixed culturerather than pure cultures, we found thatinteractions between the species had a signifi-cant impact on the biodegradation kinetics,and that the nature of these interactionsdepended on the growth substrate(s).

These findings reveal the significantchallenges that face efforts to model real-world biodegradation kinetics, in whichmixed substrates and mixed cultures are therule. Predictive modeling of these systems willbe difficult and time-consuming if one mustdetermine all pairwise chemical interactions(e.g., as required by the SKIP model) and allspecies interactions (with corresponding con-centrations of inhibitors and metabolic inter-mediates). Options to these traditionalapproaches may be developed through a fun-damental understanding of the effectsinvolved [e.g., as hinted at by the proteomicresults in (44,45)] and by alternative model-ing approaches such as that presented by Liaoet al. in this volume (46).

REFERENCES AND NOTES

1. Egli T. The ecological and physiological significance ofthe growth of heterotrophic microorganisms with mixturesof substrates. In: Advances in Microbial Ecology, Vol 14(Jones JG, ed). New York:Plenum Press, 1995;305–386.

2. Klecka GM, Maier WJ. Kinetics of microbial growth onmixtures of pentachlorophenol and chlorinated aromaticcompounds. Biotechnol Bioeng 31:328–335 (1988).

3. Meyer JS, Marcus MD, Bergman HL. Inhibitory interactionsof aromatic organics during microbial degradation. EnvironToxicol Chem 3:583-587 (1984).

4. Saéz PB, Rittmann BE. Biodegradation kinetics of a mixturecontaining a primary substrate (phenol) and an inhibitory co-metabolite (4-chlorophenol). Biodegradation 4:3–21 (1993).

5. Harder W, Dijkhuizen L. Strategies of mixed substrateutilization in microorganisms. Philos Trans R Soc Lond B297:459–480 (1982).

6. Schmidt SK, Alexander M. Effects of dissolved organiccarbon and second substrates on the biodegradation oforganic compounds at low concentrations. Appl EnvironMicrobiol 49:822–827 (1985).

7. McCarty PL, Rittman BE, Bouwer EJ. Microbial processesaffecting chemical transformations in groundwater. In:Groundwater Pollution Microbiology (Bitton G, Gerba CP,eds). New York:John Wiley & Sons, 1984;89–115.

8. Alvarez PJJ, Vogel TM. Substrate interactions of benzene,toluene, and para-xylene during microbial degradation bypure cultures and mixed culture aquifer slurries. ApplEnviron Microbiol 57:2981–2985 (1991).

9. Chang M-K, Voice TC, Criddle CS. Kinetics of competitiveinhibition and cometabolism in the biodegradation of ben-zene, toluene, and p-xylene by two Pseudomonas isolates.Biotechnol Bioeng 41:1057–1065 (1993).

10. Oh Y-S, Shareefdeen Z, Baltzis BC, Bartha R. Interactionsbetween benzene, toluene, and p-xylene (BTX) duringtheir biodegradation. Biotechnol Bioeng 44:533–538 (1994).

11. Bielefeldt A, Stensel HD. Modeling competitive inhibitioneffects during biodegradation of BTEX mixtures. Wat Res33:707–714 (1999).

12. Haigler BE, Pettigrew CA, Spain JC. Biodegradation ofmixtures of substituted benzenes by Pseudomonas sp.strain JS150. Appl Environ Microbiol 58:2237–2244 (1992).

13. Bartels I, Knackmuss H-J, Reineke W. Suicide inactivationof catechol 2,3-dioxygenase from Pseudomonas putidamt-2 by 3-halocatechols. Appl Environ Microbiol47:500–505 (1984).



14. Klecka GM, Gibson DT. Inhibition of catechol 2,3-dioxyge-nase from Pseudomonas putida by 3-chlorocatechol. ApplEnviron Microbiol 41:1159–1165 (1981).

15. Nikolajsen K, Nielsen J, Villadsen J. Structured modelingof a microbial system. III: Growth on mixed substrates.Biotechnol Bioeng 38:24–29 (1991).

16. Lendenmann U, Snozzi M, Egli T. Kinetics of thesimultaneous utilization of sugar mixtures by escherichiacoli in continuous culture. Appl Environ Microbiol62:1493–1499 (1996).

17. Tsao GT, Hanson TP. Extended Monod equation for batchcultures with multiple exponential phases. BiotechnolBioeng 17:1591–1598 (1975).

18. Yoon H, Klinzing G, Blanch HW. Competition for mixedsubstrates by microbial populations. Biotechnol Bioeng19:1193–1210 (1977).

19. Kompala DS, Ramkrishna D, Jansen NB, Tsao GT.Investigation of bacterial growth on mixed substrates:experimental evaluation of cybernetic models. BiotechnolBioeng 28:1044–1055 (1986).

20. Guha S, Peters C, Jaffé P. Multisubstrate biodegradationkinetics of naphthalene, phenanthrene, and pyrene mix-tures. Biotechnol Bioeng 65:491–499 (1999).

21. Lewandowski GA, Baltzis BC, Kung C-M, Frank ME. Anapproach to biocatalyst modeling of mixed populationsusing pure culture kinetic data. In: BiotechnologyApplications in Hazardous Waste Treatment(Lewandowski GA, Armenante P, Baltzis BC, eds). NewYork:Engineering Foundation, 1989;95–110.

22. Murakami Y, Alexander M. Destruction and formation oftoxins by one bacterial species affect biodegradation by asecond species. Biotechnol Bioeng 33:832–838 (1989).

23. Budavari S, ed. The Merck Index: An Encyclopedia ofChemicals, Drugs, and Biologicals. Whitehouse Station,NJ:Merck, 1996.

24. U.S. EPA. Priority pollutants. In: Code of Federal Regulations,Title 40, Chapter 1, Part 423, Appendix A. Washington,DC:U.S. Environmental Protection Agency, 1996.

25. ATSDR. Priority List of Hazardous Substances. Atlanta,GA:Agency for Toxic Substances and Disease Registry,1997.

26. Gibson DT, Zylstra GJ, Chauhan S. Biotransformationscatalyzed by toluene dioxygenase from Pseudomonasputida F1. In: Pseudomonas: Biotransformations,Pathogenesis, and Evolving Biotechnology (Silver S,Chakrabarty AM, Iglewski B, Kaplan S, eds). Washington,DC:American Society for Microbiology, 1990;121–132.

27. Finette BA, Subramanian V, Gibson DT. Isolation andcharacterization of Pseudomonas putida PpF1 mutantsdefective in the toluene dioxygenase enzyme system.J Bacteriol 160:1003–1009 (1984).

28. Yeh WK, Gibson DT, Liu T-N. Toluene dioxygenase: amulticomponent enzyme system. Biochem Biophys ResCommun 78:401–410 (1977).

29. Menn F-M, Zylstra GJ, Gibson DT. Location and sequenceof the todF gene encoding 2-hydroxy-6-oxohepta-2,4-dienoate hydrolase in Pseudomonas putida F1. Gene104:91–94 (1991).

30. Lau PCK, Bergeron H, Labbé D, Wang Y, Brousseau R,Gibson DT. Sequence and expression of the todGIH genesinvolved in the last three steps of toluene degradation byPseudomonas putida F1. Gene 146:7–13 (1994).

31. Zylstra GJ, Gibson DT. Toluene degradation byPseudomonas putida F1. J Biol Chem 264:14940–14946(1989).

32. Irie S, Doi S, Yorifuji T, Takagi M, Yano K. Nucleotidesequencing and characterization of the genes encodingbenzene oxidation enzymes of Pseudomonas putida.J Bacteriol 169:5174–5179 (1987).

33. Subramanian V, Liu T-N, Yeh W-K, Serdar CM, WackettLP, Gibson DT. Purification and properties offerredoxinTOL. J Biol Chem 260:2355–2363 (1985).

34. Spain JC, Gibson DT. Oxidation of substituted phenols byPseudomonas putida F1 and Pseudomonas sp. strain JS6.Appl Environ Microbiol 54:1399–1404 (1988).

35. Spain JC, Zylstra GJ, Blake CK, Gibson DT.Monohydroxylation of phenol and 2,5-dichlorophenol bytoluene dioxygenase in Pseudomonas putida F1. ApplEnviron Microbiol 55:2648–2652 (1989).

36. Cohen-Bazire G, Sistrom WR, Stanier RY. Kinetic studiesof pigment synthesis by non-sulfur purple bacteria. J CellComp Physiol 49:25–68 (1957).

37. Reardon KF, Mosteller DC, Rogers JD. Biodegradationkinetics of benzene, toluene, and phenol as single andmixed substrates for Pseudomonas putida F1. BiotechnolBioeng 69:385–400 (2000).

38. Bull Rogers J, Reardon KF. Modeling substrateinteractions during the biodegradation of mixtures oftoluene and phenol by Burkholderia sp. JS150. BiotechnolBioeng 70:428–435 (2000).

39. Bull Rogers J, DuTeau NM, Reardon KF. Use of 16S-rRNAto investigate microbial population dynamics duringbiodegradation of toluene and phenol by a binary culture.Biotechnol Bioeng 70:436–445 (2000).

40. DuTeau NM, Rogers JD, Reardon KF. Species-specificoligonucleotides for enumeration of Pseudomonas putidaF1, Burkholderia sp. strain JS150, and Bacillus subtilisATCC 7003 in biodegradation experiments. Appl EnvironMicrobiol 64:4994–4999 (1998).

41. Montgomery JH. Groundwater Chemicals DeskReference. Boca Raton, FL:Lewis Publishers, 1996.

42. Arroyo AY. Biofiltration of Off-Gases from Aluminum CanProduction [MS Thesis]. Ft Collins, CO:Colorado StateUniversity, 1996.

43. Brown DE. Biodegradation of Organic Pollutant Mixtures:Grouping Compounds to Simplify Kinetics Modeling [MSThesis]. Ft Collins, CO:Colorado State University, 1998.

44. Kim K-H, Reardon KF. Two-dimensional electrophoresisanalysis of protein production during growth ofPseudomonas putida F1 on toluene, phenol, and theirmixture. Electrophoresis 23:2233–2241 (2002).

45. Kim K-H, Bull Rogers J, Reardon KF. Membrane lipidchange in Pseudomonas putida F1 during biodegradationof toluene-phenol mixtures. In: Abstracts of the 101stGeneral Meeting of the American Society forMicrobiology, 20–24 May 2001, Washington, DC.

46. Liao K-H, Dobrev ID, Dennison JE Jr, Andersen ME,Reisfeld B, Reardon KF, Campain JA, Wei W, Klein MT,Quann RJ, Yang RSH. Application of biologically basedcomputer modeling to simple or complex mixtures.Environ Health Perspect 110(suppl 6):957–963 (2002).



Biochemical Engineering Journal 67 (2012) 156– 166

Contents lists available at SciVerse ScienceDirect

Biochemical Engineering Journal

journa l h o me pa ge: www.elsev ier .com/ locate /be j

Regular article

Substrate interactions and kinetics study of phenolic compounds biodegradationby Pseudomonas sp. cbp1-3

Jiao Liu, Xiaoqiang Jia ∗, Jianping Wen, Zhengxi ZhouDepartment of Biological Engineering and Key Laboratory of Systems Bioengineering of the Ministry of Education, School of Chemical Engineering and Technology, Tianjin University,Tianjin 300072, PR China

a r t i c l e i n f o

Article history:Received 9 February 2012Received in revised form 7 June 2012Accepted 16 June 2012Available online 26 June 2012

Keywords:BiodegradationGrowth kineticsModelingSubstrate inhibitionPseudomonas sp.Phenolic compounds

a b s t r a c t

A new strain cbp1-3 was isolated from activated sludge of a coking plant in Tianjin and identified asPseudomonas sp. based on physiological and 16S rRNA gene sequence analysis, which could completelydegrade 1400 mg/L phenol, 800 mg/L m-cresol and 150 mg/L 4-chlorophenol (4-CP) as sole carbon andenergy source within 60 h, 40 h and 60 h, respectively. Investigation on substrate interactions in thebiodegradation of mixed phenols indicated that phenol and m-cresol exhibited a mutual inhibition to eachother, and they (only below 300 mg/L) both promoted the 4-CP biodegradation. However, 4-CP stronglyinhibited the biodegradation of other phenols in their mixtures despite of the low concentration. Furtheranalysis showed that the newly proposed kinetic model for cell growth on ternary substrates fitted theexperimental data very well. And sensitivity analysis suggested that substrate interaction coefficients fi(i = 12, 13 and 23) of the models were the most sensitive parameters. Finally, yield coefficient, calculatedfor all conditions, taken as a whole, followed an allometric decline pattern with the increase of substrateconcentration.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

The widespread phenolic compounds are typical organic pollu-tants from pesticides, oil refineries, coking plants, pharmaceuticalsand so on. So far, it has still been difficult to dispose of these com-pounds safely owing to their chemical complexity [1]. Since thebiological treatment of phenolic compounds pollutants exhibitsmany advantages such as the environmental friendly and cost effec-tive, it has gained much interest as an attractive progress in thisfield recently [2–6]. In the past few decades, the related researchesprimarily focused on the single-substrate biodegradation progress.However, the biodegradation of multiple pollutants by differentmicroorganisms, which possessed diverse behaviors with complexand unspecified interactions of the substrates, was considered tobe more applicable [7]. Moreover, the kinetics studies played greatimportant role in understanding the detailed behaviors of multi-ple pollutants biodegradation, in which all of these biodegradationinteractions studies had to be performed in model wastewater thatcontained certain phenols with fixed concentrations.

Several investigations on multiple pollutants biodegradationhave gained deep insight into mathematical models upon differ-ent interactions between substrates. Yoon et al. [8] developed a

∗ Corresponding author. Tel.: +86 22 27890492; fax: +86 22 27403389.E-mail address: [email protected] (X. Jia).

generalized additive Monod model for cell growth on two sub-strates with pure competitive inhibition of them, whereas it failedto predict the complex interactions of phenol, benzene and toluenesatisfactorily [9]. Reardon et al. [9] proposed a sum of kinetics withinteraction parameters model (SKIP model), similar to the additiveMonod model, to figure out the unspecified interactions of thesearomatic hydrocarbon mixtures. Additionally, the SKIP model wasalso proved to be effective to describe the biodegradation of BTEXcompounds (benzene, toluene, ethylbenzene and xylene isomers)successfully [10].

Substrate interactions have been considered in describing phe-nolic compounds biodegradation progresses in many researches[11–13] by using kinds of microorganisms such as Pseudomonasspecies [6,14–17]. In these researches, the additive specific growthrate model, derived from the Andrew model and the additiveMonod model, was applied to describe the degradation of phenolicmixtures with strong substrate inhibition. Especially, to accountfor more complicated interactions between the correspondingsubstrates, an alternative model was developed based on thehypothesis that sorts of inhibitions had been taken into considera-tion in cell growth process to predict the microorganisms growthon phenolic compounds mixtures at high concentrations [12,17].Although most models stated above were capable of describing thedual substrates biodegradation, they might not address the behav-iors of the multiple substrates (more than two) biodegradationappropriately.

1369-703X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.bej.2012.06.008

dx.doi.org/10.1016/j.bej.2012.06.008

http://www.sciencedirect.com/science/journal/1369703X

http://www.elsevier.com/locate/bej

mailto:[email protected]

dx.doi.org/10.1016/j.bej.2012.06.008

J. Liu et al. / Biochemical Engineering Journal 67 (2012) 156– 166 157

Nomenclature

ai, a′i, a′′

i, a′′′

istoichiometric coefficients (dimensionless) (i = 1,

2, 3)fij substrate interaction coefficient (L/mg)Ii substrate interaction coefficient (dimensionless)ki, k′

i, k′′

ireaction rate constants (i = +1, −1, +2, +3, −3, +4, −4,+5, −5) (unites: k+1, k+3, k+4, k+5, k′

+1, k′+3, k′

+4, k′+5,

k′′+1, k′′

+3, k′′+4, k′′

+5 (L/(mg h)); k–1, k+2, k–3, k–4, k–5,k′

−1, k′+2, k′

−3, k′−4, k′

−5, k′′−1, k′′

+2, k′′−3, k′′

−4, k′′−5 (h−1))

Kii self-inhibition constant (mg/L) (i = 1, 2, 3)Ksi saturation constant (mg/L) (i = 1, 2, 3)pi parameters in the cell growth modelS initial substrate concentration (mg/L)S global substrate concentration (mg/L)�SC amount of carbon consumed (mg/L)t time (h)X biomass concentration (mg/L)YX/C observed overall yield coefficient (mg dry cell

formed/mg carbon consumed)

Greek symbols�mi

maximum specific growth rate on single substrate(h−1) (i = 1, 2, 3)

�X overall specific growth rate (h−1)�Xi

specific growth rate on single substrate (h−1) (i = 1,2, 3)

Subscripts1 substrate, phenol2 substrate, m-cresol3 substrate, 4-CP

In the present work, we focused on the biodegradation processesof three typical high toxic, carcinogenic, mutagenic and teratogenicphenols (phenol, m-cresol and 4-chlorophenol (4-CP) [18–20]) bythe newly isolated Pseudomonas sp. cbp1-3. New intrinsic kineticmodels of cell growth on ternary substrates were also proposedbased on the hypothesis of cell growth processes similar to enzy-matic reactions and used to evaluate the complicated interactionsof the phenols.

2. Materials and methods

2.1. Isolation and identification of microorganism

Strains were isolated as described by Jiang et al. [21] with somemodifications. Activated sludge was collected from a coking plantin Tianjin, and the coking wastewater contained kinds of phenols inthe range of 400–2500 mg/L. Then the activated sludge was selec-tively enriched by five steps for four weeks. The five repeatedexperiments were performed using phenol as the sole carbonsource in the mineral medium (MM) with the phenol concentra-tion increasing from 200 to 1000 mg/L (interval of 200 mg/L). Thefinal cultures were diluted and plated onto agar Luria–Bertani (LB)agar plates containing 1000 mg/L phenol, and then incubated after-wards at 30 ◦C for 36–48 h. Several microorganisms were obtained.Following that, the degradation capacity (for phenol, m-cresol and4-CP) of the obtained strains was further detected for the second-round selection of the expected strains.

The selected strain was identified by 16S rRNA gene sequenceanalysis. The genome extraction and 16S rRNA gene amplificationof Pseudomonas sp. 1–3 were performed as described by Qiu et al.[22]. DNA sequencing was served by the Beijing Genomics Institute

(Beijing, China). Gene alignment was implemented by the BLASTprogram of NCBI (www.ncbi.nlm.nih.gov).

2.2. Culture media and cultivation conditions

The MM consists of 0.5 g/L (NH4)2SO4, 0.8 g/L Na2HPO4, 0.2 g/LKH2PO4, 0.1 g/L MgSO4, 0.02 g/L yeast extract and 10 mL trace ele-ment solution [23]. LB medium containing 5.0 g/L beef extract,10.0 g/L peptone, and 5.0 g/L NaCl was used as nutrient medium.The initial pH of all of the mediums was adjusted to 7.2 with 2 mol/LNaOH solution.

Strains were cultivated in 250 mL Erlenmeyer flasks on a rotaryshaker at 180 rpm and 30 ◦C. Phenolic compounds as carbon sourceswere filter-sterilized through membranes (pore size of 0.22 �m)before adding to MM.

2.3. Phenolic compounds biodegradation

The strain was cultivated in 50 mL LB medium with 2% (v/v)inoculation. After 24 h of incubation, cells were harvested at the lateexponential growth phase by centrifugation (10,000 × g) for 10 minat 4 ◦C, then washed twice with 50 mL MM and resuspended inMM as inocula at an appropriate concentration (OD600 ≈ 5.0). In allexperiments, 2 mL inocula were inoculated into 100 mL MM withvarying initial phenol, m-cresol and 4-CP concentrations, respec-tively.

2.4. Analytical methods

Cell density was monitored spectrophotometrically by mea-suring the optical density at 600 nm (OD600). Dry cell weight(DCW) was calculated from OD600 with linear correlation factor(1 OD600 = 437.48 DCW mg/L). The dry cell weight was determinedby filtering a known volume of cell suspension then drying thecells at 105 ◦C [13]. The phenolic compounds’ concentration wasdetermined by the methods as described [16,20,21] with somemodifications. The supernatant was prepared by centrifugationat 6000 × g for 10 min and then filtered through syringe fil-ters (0.22 �m). The residual substrate concentrations in preparedsupernatant were measured using high performance liquid chro-matography (HPLC, 1200, Agilent, USA) equipped with a ZorbaxEclipse XDB-C18 column (250 mm × 4.6 mm, 5 �m, Agilent, USA)and a UV-detector (G1313B, Agilent, USA) at 280 nm. The mobilephase of methanol/water (57/43, v/v) at a flow rate of 1.0 mL/minat 30 ◦C was adopted. The retention time for phenol, m-cresol and4-CP was 3.87 min, 5.46 min and 6.75 min, respectively.

2.5. Statistics

All experiments were repeated three times. Data shown in fig-ures of Section 4 were the mean values of the experiments and theerror bars indicated standard deviation.

3. Modeling

3.1. Model development

Activation of the benzene ring by monooxygenase and ring-fission by dioxygenase are two key steps in aromatic compoundsbiodegradation pathways [3]. Co-metabolism usually plays animportant role in the metabolism of aromatic compounds such asphenol, m-cresol and 4-CP ascribe to the two key steps catalyzedby the same enzyme system of a microorganism. Therefore, whenmixed substrates are used as carbon sources for cell growth, thecomplex interactions between them may usually be representedas competitive, noncompetitive and uncompetitive inhibition,

http://www.ncbi.nlm.nih.gov/

158 J. Liu et al. / Biochemical Engineering Journal 67 (2012) 156– 166

analogous to enzymatic kinetics [17]. Then cell growth on threephenolic compounds could be proposed based on enzymatic kinet-ics as the following reactions [8,12,17]:

X + a1S1k+1⇔k−1

X ′k+2→2X (1)

X ′ + a′1S1

k+3⇔k−3

X ′S1 (2)

X ′ + a′′1S2

k+4⇔k−4

X ′S2 (3)

X ′ + a′′′1 S3

k+5⇔k−5

X ′S3 (4)

X + a2S2k′+1⇔

k′−1

X ′′k′+2→2X (5)

X ′′ + a′2S2

k′+3⇔

k′−3

X ′′S2 (6)

X ′′ + a′′2S1

k′+4⇔

k′−4

X ′′S1 (7)

X ′′ + a′′′2 S3

k′+5⇔

k′−5

X ′′S3 (8)

X + a3S3k′′+1⇔

k′′−1

X ′′′k′′+2→2X (9)

X ′′′ + a′3S3

k′′+3⇔

k′′−3

X ′′′S3 (10)

X ′′′ + a′′3S1

k′′+4⇔

k′′−4

X ′′′S1 (11)

X ′′′ + a′′′3 S2

k′′+5⇔

k′′−5

X ′′′S2 (12)

X ′, X ′′ and X ′′′ are intermediate states of the microorganism, inwhich substrates are consumed without cell growth. The param-eters ai, a′

i, a′′

iand a′′′

i(i = 1, 2, 3) are stoichiometric coefficients,

which represents the mean mass of each substrate consumed byan organism to reach its intermediate state. The value of thesestoichiometric coefficients could be assumed to be 1 according toYoon et al. [8]. On the basis of pseudo-steady state assumptionfor the intermediates: X ′, X ′′, X ′′′, X ′S1, X ′S2, X ′S3, X ′′S1, X ′′S2, X ′′S3,X ′′′S1, X ′′′S2 and X ′′′S3, the following equations can be easily derived[12,17]:

�Xi= �mi

Si

Ksi + I−1i

S(13)

and

S = I1

(S1 + S2

1Ki1

)+ I2

(S2 + S2

2Ki2

)+ I3

(S3 + S2

3Ki3

)

+ f12S1S2 + f13S1S3 + f23S2S3 (14)

where �Xi, �mi

, Ksi, Kii are the specific cell growth rate (h−1), themaximum specific cell growth rate (h−1), the saturation constant(mg/L) and the self-inhibition constant (L/mg) on one substrate,respectively, while Ii (mg/L) and fij (L/mg) are the substrate inter-action coefficients. These parameters can be obtained from theoriginal reaction rate constants ki, k′

iand k′′

i. The global substrate

concentration S (mg/L), which means the integral effect of all sub-strates on the microorganism, is derived from individual impactand cross-interactions of all substrates on cell growth with differ-ent coefficients. Its impact on the degradation of different substratecould be characterized by the parameter Ii.

Consequently, the overall specified growth rate could bedescribed as follows:

�X =3∑

i=1

�Xi(15)

The Andrew model [24] (sometimes referred as Haldane model),which was widely used to evaluate the growth kinetics of inhibitorysubstrates, could be obtained from Eqs. (13)–(15) when the cellsgrow on single substrate:

�X = �mS

Ks + S + S2/Ki(16)

So parameters �mi, Ksi and Kii can be obtained separately from the

kinetics of individual cell growth on phenol, m-cresol and 4-CP,respectively. Theoretically, the parameters Ii and fij both dependon the ratios of the parameters Ksi, but they cannot strictly complywith the relationships of the original reaction rate constants dueto the complexity of the system [17]. They could be determinedas independent parameters by fitting the data of the mixed-substrate biodegradation, and the parameter I1 is often set to1 [12].

Based on the data of the exponential growth phase, the specificgrowth rate was calculated as follows:

�X = d ln(X)dt

(17)

where X is the cell concentration (mg/L).Confidence intervals (CI) of each kinetic parameter were calcu-

lated by the bootstrap method as described by Mishra et al. [25], byusing MATLAB R2008b. A fictional dataset was constructed by ran-dom sampling with replacement from the original dataset, whichwas used to estimate the new values of parameters. Random sam-pling process was repeated 1000 times to obtain a set of values ofeach estimated parameter. The confidence interval (95%) of eachparameter was calculated based on the upper and lower 95 per-centiles of the set of the parameter values.

To analyze the assimilation efficiency of substrates in differentexperiments, the observed overall yield coefficient YX/C (mg dry

cell formed/mg carbon consumed) was defined as milligrams of biomass(DCW) produced by the consumption of per milligram of carbon inthe substrates, and can be calculated as follows:

YX/C = Xmax − X0

�SC(18)

where Xmax and X0 are the maximum and initial biomass con-centration, �SC (mg/L) is the amount of carbon consumed by themicroorganism in the substrates when the biomass reached maxi-mum.

3.2. Sensitivity analysis

In order to find the most influential kinetic parameter on cellgrowth, sensitive analysis is carried out, as described by Shashi andKumar [26] with some modifications. Then the response variable�X is considered as a function of the fourteen parameters:

�X = f (�m1, Ks1, �m2, Ks2, �m3, Ks3, Ki1, Ki2, Ki3, I2, I3, f12, f13f23)

(19)

The first order Taylor’s expansion of Eq. (19) was obtained asfollows:

��X = ∂�X

∂�m1��m1 + ∂�X

∂Ks1�Ks1 + ∂�X

∂�m2��m2 + · · · + ∂�X

∂f23�f23

(20)


Then sensitivity of �X with respect to the parameter pi couldbe indicated by the sensitive function ∂�X/∂pi (the slope of �X

with respect to the parameter pi). During the analysis only oneparameter was changed at a time with the others fixed as constant.All parameters varied from −90% to +100% of the fitted values ofparameters (obtained in Section 3.1). Then a series of values of sen-sitive function ∂�X/∂pi is obtained. Moreover, normalization of theseries of values of ∂�X/∂pi was performed so that different slopescould be plotted on the same plot:

Normalized slope = (slope − slopes)SD

(21)

where slopes and SD indicates the mean value and standard devi-ation of the series of slope data, respectively.

4. Results and discussion

4.1. Characterization of the strain

Six strains were obtained and a further detection of biodegrada-tion capability of strains showed that the strain cbp1-3 displayedthe best performance on the biodegradation of phenol, m-cresoland 4-CP, especially of their mixture. Consequently, this strainwas selected for further investigations. The appearance for thecolony of cbp1-3 was a large, smooth, convex, light brown andround on the nutrient agar. Microscopic observation showed thestrain was a short-rod gram-negative strain with mobility. Align-ment of the partial 16S rRNA gene sequences of cbp1-3 (1440nucleotides, Accession no. JN426990) demonstrated that, this straincbp1-3 was highly homologous (99%) to the type strains Pseu-domonas putida GB-1 (Accession no. CP000712.1) and F1 (Accessionno. CP000926.1). Therefore, the strain cbp1-3 was identified as astrain of Pseudomonas sp. temporarily. As one of the most popularmicroorganisms, Pseudomonas species has shown great advantagesin biodegradation of phenolic compounds [6,14–17]. Pseudomonassp. cbp1-3 may not hold the highest degradation rate comparedwith some of the previously reported bacteria aiming at one specificphenol, but it was the first reported strain that could simultane-ously degrade phenol, m-cresol and 4-CP as carbon and energysources. This advantage of cbp1-3 provided us a very good opportu-nity to study the interactions of these three phenols and to developnew cell growth model based on complicated substrates.

4.2. Single-substrate biodegradation

The capabilities of cbp1-3 to degrade phenol, m-cresol and 4-CPindividually were investigated under different initial substrate con-centrations. As shown in Fig. 1(a), the phenol degradation and cellgrowth were examined with the initial concentration ranging from800 to 1600 mg/L. Results showed that the time for complete degra-dation of phenol extended from 28 h to 60 h with the augmentationof the initial phenol concentration. In addition, the final biomassalso increased gradually and the lag phase prolonged noticeablyfrom 0 h to 32 h. However, cbp1-3 failed to degrade 1600 mg/Lphenol within 60 h.

The cell growth and biodegradation of m-cresol and 4-CP bycbp1-3 were shown in Fig. 1(b) and (c), respectively. Similar tothe biodegradation of phenol, the time for the complete degrada-tion of 4-CP and m-cresol was prolonged with the increase of theinitial substrate concentration. However, different from the phe-nol biodegradation, there was no obvious final biomass increase inthe biodegradation of m-cresol and 4-CP. The maximum concentra-tion of 4-CP that the strain cbp1-3 could completely degrade wasmerely 150 mg/L, lower than that of m-cresol (800 mg/L) and phe-nol (1400 mg/L). These results indicated that 4-CP exhibited much

Fig. 1. Cell growth (open) and biodegradation (solid) of phenol (a), m-cresol (b) and4-CP (c) in single-substrate systems.

stronger inhibitory on the substrates degradation by cbp1-3 thanphenol and m-cresol.

4.3. Dual-substrate biodegradation

A series of dual-substrate biodegradation experiments contain-ing phenol, m-cresol and 4-CP under different initial concentrationswere performed. Fig. 2(a) showed the biodegradation of 600 mg/Lphenol and m-cresol varying from 100 to 500 mg/L. Althoughphenol and m-cresol almost began to degrade simultaneously,the complete biodegradation of phenol finished prior to that ofm-cresol. Moreover, higher initial concentration of m-cresol hadstronger inhibition on phenols biodegradation according to longertime for the lag phase and complete biodegradation of phenol.For instance, when the m-cresol concentration increased from 100to 500 mg/L, the time for complete degradation of 600 mg/L phe-nol extended from 16 h to 26 h. In addition, the final biomasswas generally enhanced by the increase of the initial substrateconcentration. In another situation, Fig. 2(b) represented the dual-substrate biodegradation of 400 mg/L m-cresol and phenol from200 to 600 mg/L. Just as the effect of m-cresol on phenol degra-dation, when the phenol concentration increased, the longer lagphase and time for complete degradation of m-cresol as well as


Fig. 2. Cell growth and substrate biodegradation in phenol-m-cresol dual-substrate systems. (a) 600 mg/L phenol with m-cresol of 100–500 mg/L and (b) 400 mg/L m-cresolwith phenol of 200–600 mg/L.

the higher final biomass were observed. For example, the timefor 400 mg/L m-cresol complete degradation increased from 22 hto 32 h when the phenol concentration was enhanced from 200to 600 mg/L. It can be concluded that mutual inhibition betweenphenol and m-cresol occurred in these biodegradation processes,and phenol was a preferential substrate for cbp1-3 compared withm-cresol.

Experiments were conducted by varying initial 4-CP concentra-tions from 60 to 150 mg/L with phenol of constant concentrationsof 600 mg/L (Fig. 3(a)). Although the concentration of 4-CP wasmuch lower than that of phenol, the degradation of 4-CP couldonly initialed after the almost fully depletion of phenol. Higherconcentrations of 4-CP could strongly inhibit phenol degradationand cell growth. In these experiments, the time for completebiodegradation of 600 mg/L phenol was prolonged from 20 h to40 h when 4-CP concentration enhanced from 60 to 150 mg/L.Fig. 3(b) showed the biodegradation of 100 mg/L 4-CP and phe-nol with the initial concentrations varying from 100 to 500 mg/L.Phenol under low concentration from 100 to 300 mg/L graduallyaccelerated the biodegradation of 4-CP. The time for 100 mg/L 4-CP complete degradation were 26 h and 31 h with 300 mg/L phenolor not, respectively. Although the lag phase of cbp1-3 growth on

100 and 300 mg/L phenol were almost vanished simultaneously,the biomass accumulation of the latter was faster than that of theformer. In this case, phenol was easily utilized as carbon and energysource by cbp1-3 to yield more biomass to promote the biodegrada-tion of 4-CP. But cell growth and 4-CP biodegradation was stronglyinhibited by high concentration of phenol (from 300 to 500 mg/L),where toxicity of phenol had stronger influence on cell growth thanits supplies of carbon and energy.

The biodegradation behaviors of mixture of m-cresol and 4-CPwere very similar to that of phenol and 4-CP as mentioned above.The effects of 4-CP varying from 60 to 150 mg/L on the biodegra-dation of 400 mg/L m-cresol were investigated (Fig. 4(a)). It wasobserved that m-cresol was utilized for cells growth at the begin-ning of biodegradation, then 4-CP was degraded by cbp1-3. Buthigher concentration of 4-CP exhibited stronger inhibition on cellgrowth, and longer time was required for complete removal of allsubstrates. In another case, Fig. 4(b) showed the biodegradationof 100 mg/L 4-CP and m-cresol from 100 to 500 mg/L. The pres-ence of m-cresol with low concentrations (from 100 to 300 mg/L)could accelerate the biodegradation rate of 100 mg/L 4-CP gradu-ally, whereas the opposite effect was observed with m-cresol over300 mg/L.


Fig. 3. Cell growth and substrate biodegradation in phenol-4-CP dual-substrate systems. (a) 600 mg/L phenol with 4-CP of 60–150 mg/L and (b) 100 mg/L 4-CP with phenolof 100–500 mg/L.

4.4. Ternary-substrate biodegradation

Little information was available for the biodegradation ofternary mixtures of phenolic compounds utilized as sole carbonand energy sources. Wang and Loh applied conventional carbonsources such as glucose to co-metabolize phenol and chlorophe-nol that could not be degraded alone [7,17]. Another study foundthat a mixture of 4-CP, 4-nitrophenol and phenol could be com-pletely co-metabolized by Arthrobacter chlorophenolicus A6 [27]. Inthis study, phenols biodegradation and cell growth on ternary sub-strates were further investigated. Typical profiles were shown inFig. 5. The data presented in Fig. 5(a) was consistent with the resultsof experiment T4 (Table 1). It was shown that phenol and m-cresolalmost began to degrade simultaneously after a 12 h lag phase.Compared with the corresponding single-substrate biodegrada-tion, time required for complete degradation of 4-CP shortened to24 h, while time for complete degradation of phenol and m-cresolwas prolonged to 18 h and 22 h, respectively. Additionally, whenthe concentration of phenol increased to 400 mg/L (correspondingto experiment T22), the complete degradation time was prolongedto 24 h and 40 h for complete transformation of m-cresol and 4-CP,

respectively (Fig. 5(b)). As shown in Fig. 5(c) and (d) (datacorresponding to experiment T6 and T7), the increase of the con-centration of m-cresol and 4-CP resulted in deceleration of the otherphenols degradation. In addition, low concentration of phenol andm-cresol promoted the 4-CP biodegradation, whereas high concen-tration of them exhibited the contrary effect. For example, time for

Table 1Summary of experiments in ternary-substrate systems.

Experiment no. Initial concentration (mg/L)

Phenol m-Cresol 4-CP

T1-T3 200 100, 250, 400 50T4-T6 200 100, 250, 400 100T7-T9 200 100, 250, 400 150T10-T12 300 100, 250, 400 50T13-T15 300 100, 250, 400 100T16-T18 300 100, 250, 400 150T19-T21 400 100, 250, 400 50T22-T24 400 100, 250, 400 100T25-T27 400 100, 250, 400 150

4-CP, 4-chlorophenol


Fig. 4. Cell growth and substrate biodegradation in m-cresol-4-CP dual-substrate systems. (a) 400 mg/L m-cresol with 4-CP of 60–150 mg/L and (b) 100 mg/L 4-CP withphenol of 100–500 mg/L.

100 mg/L 4-CP complete degradation with 200 mg/L phenol and100 mg/L m-cresol was 24 h, which was shorter than that of 32 hwithout phenol and m-cresol (Fig. 5(a)). However, a longer time of40 h was required for that with 400 mg/L phenol and 100 mg/L m-cresol as shown in Fig. 5(b). Overall, although the ternary-substratebiodegradation processes were complicated, the interactions of thephenols exhibited similar behaviors to that in the biodegradationprocesses of dual-substrate systems.

Additionally, it should be pointed out that the symbols ofFigs. 1–5 reflected the presupposed initial concentrations of thesethree phenols. However, in the actual experiments, the measuredconcentrations could be a little bit different from the prepared con-centrations.

4.5. Interaction analysis in phenols biodegradation

Generally, in a bacterium, phenolic compounds were usu-ally catalyzed by the same enzyme system. The key metaboliteswith same structure of catechol were generated and cleaved bymonooxygenase and dioxygenase in their biodegradation [3,4]. Thesubstrate specificity and enzymatic activity of the two key enzymeswere dependent on the structures of enzymes and substrates.

The chlorine substituent and methyl group on benzene ring hadelectron-attracting and steric hindrance effect to impede catalysisof the two oxygenases. Besides, stronger substrate inhibition of 4-CP and m-cresol decreased the activity of the two enzymes. Thus,4-CP and m-cresol were more difficult to be degraded than phenolas observed in this study. Other researchers also found the similarphenomenon: only 375 mg/L 4-CP could be completely degradedwithin 40 h by a good chlorophenols degrader, A. chlorophenoli-cus A6 [4,20], whereas the concentrations of phenol and cresolmineralized could reach to 3000 and 1500 mg/L, respectively[18,28].

Moreover, compared with 4-CP and m-cresol, phenol was lowertoxic to induce more easily the synthesis of oxygenases [29], andmore readily biodegradable with higher enzyme reactive activi-ties. Therefore, when the three coexisted phenolic compounds werecatalyzed by the same enzymes system, phenol was usually firstlyutilized as primary substrate to promote the degradation of cresoland 4-CP by co-metabolism [11,15–17]. And then m-cresol beganto degrade, which followed by the 4-CP biodegradation. However,further work for gaining insight into the interaction mechanisms isnecessary by investigating the global metabolism of cell growth onphenolic mixtures, which is being undergoing in our lab.


Fig. 5. Typical cell growth and substrate biodegradation in ternary-substrate systems.

4.6. Kinetics study of phenolic compounds biodegradation

4.6.1. Intrinsic kinetics parameters estimationBatch cultures were performed in the MM with different con-

centrations of phenol (0–1400 mg/L), m-cresol (0–800 mg/L) and4-CP (0–150 mg/L), respectively. On the basis of the single substrateexperiments, the parameters �mi

, Ksi and Kii for the biodegrada-tion of phenol, m-cresol and 4-CP as the sole carbon source weredetermined, respectively. The fitted values of parameters with con-fidence intervals (95%) (Table 2) were obtained by a nonlinearleast-square regression analysis using MATLAB R2008b. Larger sat-uration constant (Ksi) means lower substrate affinity, and smallervalue of self-inhibition constant (Ki) indicates stronger substrateinhibition [30]. The values of saturation constant and self-inhibitionconstant for the biodegradations of phenol, m-cresol and 4-CP werein ascending and descending orders, respectively. It was indicatedthat the strain cbp1-3 exhibited a substrate preference in the fol-lowing order: phenol, m-cresol and 4-CP, which is in agreementwith the conclusion of the substrate toxicity order as mentioned inSection 4.1.

The information derived from single- or dual-substratebiodegradations could be applied to more complex systems [31].Eqs. (13)–(15) were used to describe the dual- and ternary-substrate biodegradation. Then, coupled with the determinedparameters calculated based on single-substrate experiments asmentioned above, the parameters I2, I3, f12, f13 and f23 could beobtained on the basis of the data of dual- and ternary-substrateexperiments. All of these parameters with confidence interval (95%)were listed in Table 2. Finally, four nonlinear curve fits were per-formed for the determination of all the parameters, the smallresidual sum of squares (RSS, <5.65 × 10−3) and high correlationcoefficient (R2 > 0.979) of each nonlinear curve fit indicated thatthe model fitted the experimental data very well.

The newly developed models for cell growth on three substratesin this research have a similar part of Eq. (13) with models such asMonod model, the additive Monod model [8], the skip model [9]and other more complex models [12,17] deriving from cell growthprogresses reactions similar to enzymatic kinetics. The variable partbetween these models can be integrated to a parameter of globalsubstrate concentration (S), which reflects the complex interactionof substrates. Moreover, other appropriate and meaningful formsof S may be selected or obtained for the biodegradation in differentsystems.

4.6.2. Yield coefficients of cell growthThe yield coefficient of cell growth significantly depended

on the medium composition and environment conditions. Three

Fig. 6. Overall yield coefficient of cell growth on single, dual and ternary substrates.


Fig. 7. Plots of normalized ∂�X1/∂pi with normalized parameters pi over the range of −90% to +100% in four different experiments of T4 (a), T22 (b), T6 (c) and T7 (d).

patterns: the constant pattern [32], the linear decline pattern [33]and the allometric decline pattern [19] were summarized by Xiaoet al. [34] to describe different trends of yield coefficient correlatedwith substrate concentration. Pirt [35] proposed that yield coef-ficient had a direct relation with impact of substrate on specificgrowth rate and maintenance energy. Lower yield coefficient maybe obtained ascribe to lower specific growth rate and higher main-tenance energy which could be resulted in by stronger inhibition ofphenolic compounds with higher concentration. As shown in Fig. 6,the observed overall yield coefficient YX/C indicated the assimila-tion efficiency of substrates in single-, dual-, and ternary-substratesystems (mentioned in Figs. 1–5). Relatively small changes in theyield coefficient of cell growth on phenol (0.35–0.39) was observedowing to low growth rates at high substrate concentration, whilethe yield coefficient of cell growth on m-cresol (0.44–1.78) and 4-CP

(2.26–4.30) varied distinctly. The yield coefficient of cell growth onphenolic mixtures distributed above that on phenol and below thaton 4-CP, with a mean value of 0.88. Moreover, complex interactionsof phenols in their biodegradation usually had indistinct effects onthe metabolism of cells besides substrate inhibition. Thus, irregularfluctuation of yield coefficient of cell growth on phenolic mixtures(especially in the ternary-substrate systems listed in Table 1, butnot shown in figure) could be observed. However, in this work,the yield coefficient, taken as a whole, performed the allometricdecline pattern accompanying with the increase of substrate con-centration.

4.6.3. Sensitivity analysis of parametersSensitivity analysis was performed to identify sensitive parame-

ters in the cell growth model. Slopes ∂�X1/∂pi, ∂�X2/∂pi, ∂�X3/∂pi

Table 2Kinetic parameters of Pseudomonas sp. cbp1-3 in ternary-substrate systems.

Parameters Value RSS R2 CI (95%) Data sources

�m1 (h−1) 0.275 5.65 × 10−3 0.985 (0.253, 0.330)Phenol alonebiodegradation

Ks1 (mg/L) 6.90 (1.00, 20.00)Ki1 (mg/L) 530.7 (389.2, 655.1)�m2 (h−1) 0.541 2.63 × 10−4 0.997 (0.407, 0.696)

m-Cresol alonebiodegradation

Ks2 (mg/L) 12.58 (1.00, 25.79)Ki2 (mg/L) 73.9 (51.8, 109.2)�m3 (h−1) 0.416 6.34 × 10−4 0.979 (0.376, 0.456)

4-CP alonebiodegradation

Ks3 (mg/L) 26.50 (6.49, 38.27)Ki3 (mg/L) 14.3 (12.51, 16.55)I1 1 Set valueI2 0.315 4.29 × 10−4 0.989 (0.273, 0.358)

Dual-andternary-substratebiodegradation

I3 0.499 (0.434, 0.561)f12 (L/mg) 3.54 × 10−4 (1.78 × 10−4, 6.00 × 10−4)f13 (L/mg) 4.92 × 10−3 (4.11 × 10−3, 5.64 × 10−3)f23 (L/mg) 5.60 × 10−3 (4.32 × 10−3, 6.75 × 10−3)

4-CP, 4-chlorophenol; CI, confidence intervals; RSS, residual sum of squares.


Fig. 8. Effect of phenols concentration on ∂�X1/∂pi (black, squared), ∂�X2/∂pi (red, circle), ∂�X3/∂pi (blue, regular triangle) and ∂�X /∂pi (olive green, inverse triangle). A seriesof values of one slope, such as ∂�X1/∂�m1, was obtained to calculate the mean value with the corresponding parameter pi varying from −90% to +100% in one experiment,then four mean values of each parameter were obtained in different experiments of T4, T22, T6 and T7 (Table 1) corresponding to the four points from left to right, with errorbars indicating the variation range of the slopes. ∂�X /∂�mi was equal to ∂�Xi/∂�mi and ∂�X /∂Ksi was equal to ∂�Xi/∂Ksi (i = 1, 2 and 3). (For interpretation of the referencesto color in this figure legend, the reader is referred to the web version of this article.)

and ∂�X/∂pi with respect to the same parameter showed a similartrend ascribe to their similar mathematical structure (pi means aparameter). Thus the change trends of normalized slopes in fourtypical experiments (Table 1: T4, T22, T6 and T7) were shown bytaking ∂�X1/∂pi for example (Fig. 7). And the real values of slopes∂�/∂pi in the four experiments were compared in Fig. 8 (� was�X1, �X2, �X3 and �X , respectively).

Fig. 7 showed that the parameters could be divided into threegroups: the first group of �m1 and Ks1 was insensitive and had sta-ble impact on �X1; the second group of Kii (i = 1, 2 and 3) weresensitive when normalized parameter less than about −40%; thethird group of Ii (i = 2 and 3) and fi (i = 12, 13 and 23) had a simi-lar stable variation tendency. But Fig. 8 showed that the absolutevalues of ∂�/∂fi were much larger than ∂�/∂Ii or other slopes,which indicated that fi had the most important effects on cellgrowth. Moreover, great changes of ∂�/∂fi were observed in differ-ent experiments with different concentrations of phenols in Fig. 8,and it demonstrated that the interactions of the phenols had greatrelationship with the phenols concentration and played much sig-nificant role in cell growth. So the parameters fi were consideredas the most sensitive parameters.

5. Conclusions

Strain cbp1-3, newly isolated and identified as Pseudomonassp., could effectively degrade phenol, m-cresol, 4-CP and theirmixtures. Substrate interactions were investigated in mixed phe-nols biodegradation: Phenol and m-cresol had a mutual inhibitionto each other, both of which enhanced the biodegradation of 4-CP when their concentrations were below 300 mg/L, while 4-CPshowed strong inhibition on the other two phenols biodegradation.New kinetic models were developed to describe the cell growthbehavior on ternary substrates, which fitted the experimental datavery well. The self-inhibition constants Kii indicated a substratepreference order as follows: phenol, m-cresol and 4-CP. In sensi-tivity analysis, the most sensitive parameters of fi (i = 12, 13 and23) indicated that the interactions between the phenols play themost significant role in the cell growth. Finally, the observed over-all yield coefficient was calculated, which significantly depended on

the composition and concentration of the substrates and generallydecreased as the initial substrate concentration increased.

Understanding of the mechanisms of substrate interaction wasimportant as well as the new kinetic models proposed in thiswork, which help us better understand and utilize the complicatedphenolic compounds biodegradation processes, like progress opti-mization and scale-up. And a further insight into mechanisms ofsubstrate interaction from the cell physiology perspective is alsonecessary, which would be facilitated by global metabolism anal-ysis undergoing in our lab. It should also be pointed out that theunique ability of Pseudomonas sp. cbp1-3, to degrade these threetypical phenolic compounds, also makes this strain very promisingto be applied in practical bioremediation of complicated phenols ina wide range of concentrations.

Acknowledgements

The authors wish to acknowledge the financial support pro-vided by the National Natural Science Foundation of China (No.20906070), the Seed Foundation of Tianjin University and theProgram of Introducing Talents of Discipline to Universities (No.B06006).

References

[1] C.D. Silva, J. Gómez, R. Beristain-Cardoso, Simultaneous removal of 2-chlorophenol, phenol, p-cresol and p-hydroxybenzaldehyde under nitrifyingconditions: kinetic study, Bioresour. Technol. 102 (2011) 6464–6468.

[2] P. Christen, A. Vega, L. Casalot, G. Simon, R. Auria, Kinetics ofaerobic phenol biodegradation by the acidophilic and hyperther-mophilic archaeon Sulfolobus solfataricus 98/2, Biochem. Eng. J. (2012),http://dx.doi.org/10.1016/j.bej.2011.12.012.

[3] E. Dorn, H.J. Knackmuss, Chemical structure and biodegradability of halo-genated aromatic compounds. Substituent effects on 1,2-dioxygenation ofcatechol, Biochem. J. 174 (1978) 85–94.

[4] A.O. Olaniran, E.O. Igbinosa, Chlorophenols and other related derivatives ofenvironmental concern: properties, distribution and microbial degradationprocesses, Chemosphere 83 (2011) 1297–1306.

[5] R.K. Singh, S. Kumar, S. Kumar, A. Kumar, Biodegradation kinetic studies forthe removal of p-cresol from wastewater using Gliomastix indicus MTCC 3869,Biochem. Eng. J. 40 (2008) 293–303.

[6] Y.J. Wang, J. Song, W. Zhao, X.L. He, J. Chen, M. Xiao, In situ degradation ofphenol and promotion of plant growth in contaminated environments by asingle Pseudomonas aeruginosa strain, J. Hazard. Mater. 192 (2011) 354–360.

dx.doi.org/10.1016/j.bej.2011.12.012


[7] S.J. Wang, K.C. Loh, Biotransformation kinetics of Pseudomonas putida forcometabolism of phenol and 4-chlorophenol in the presence of sodium glu-tamate, Biodegradation 12 (2001) 189–199.

[8] H. Yoon, G. Klinzing, H.W. Blanch, Competition for mixed substrates by micro-bial populations, Biotechnol. Bioeng. 19 (1977) 1193–1210.

[9] K.F. Reardon, D.C. Mosteller, J.B. Rogers, N.M. DuTeau, K.H. Kim, Biodegrada-tion kinetics of aromatic hydrocarbon mixtures by pure and mixed bacterialcultures, Environ. Health Perspect. 110 (2002) 1005–1011.

[10] D.E.G. Trigueros, A.N. Módenes, A.D. Kroumov, F.R. Espinoza-Quinones, Mod-eling of biodegradation process of BTEX compounds: kinetic parametersestimation by using particle swarm global optimizer, Process Biochem. 45(2010) 1355–1361.

[11] Z. Alexieva, H. Yemendzhiev, P. Zlateva, Cresols utilization by Trametes versi-color and substrate interactions in the mixture with phenol, Biodegradation 21(2010) 625–635.

[12] Y. Jiang, J.P. Wen, J. Bai, D.Q. Wang, Z.D. Hu, Phenol biodegradation by theyeast Candida tropicalis in the presence of m-cresol, Biochem. Eng. J. 29 (2006)227–234.

[13] H. Yao, Y. Ren, X. Deng, C. Wei, Dual substrates biodegradation kinetics of m-cresol and pyridine by Lysinibacillus cresolivorans, J. Hazard. Mater. 186 (2011)1136–1140.

[14] T. Abuhamed, E. Bayraktar, T. Mehmetoglu, U. Mehmetoglu, Kinetics modelfor growth of Pseudomonas putida F1 during benzene, toluene and phenolbiodegradation, Process Biochem. 39 (2004) 983–988.

[15] K.C. Loh, T. Wu, Cometabolic transformation of 2-chlorophenol and 4-chlorophenol in the presence of phenol by Pseudomonas putida, Can. J. Chem.Eng. 84 (2006) 356–367.

[16] P. Saravanan, K. Pakshirajan, P. Saha, Biodegradation of phenol and m-cresol ina batch and fed batch operated internal loop airlift bioreactor by indigenousmixed microbial culture predominantly Pseudomonas sp., Bioresour. Technol.99 (2008) 8553–8558.

[17] S.J. Wang, K.C. Loh, Growth kinetics of Pseudomonas putida in cometabolism ofphenol and 4-chlorophenol in the presence of a conventional carbon source,Biotechnol. Bioeng. 68 (2000) 437–447.

[18] D.J. Lee, K.L. Ho, Y.Y. Chen, Degradation of cresols by phenol-acclimated aerobicgranules, Appl. Microbiol. Biotechnol. 89 (2010) 209–215.

[19] Y. Li, J. Li, C. Wang, P. Wang, Growth kinetics and phenol biodegradationof psychrotrophic Pseudomonas putida LY1, Bioresour. Technol. 101 (2010)6740–6744.

[20] N.K. Sahoo, K. Pakshirajan, P.K. Ghosh, A. Ghosh, Biodegradation of 4-chlorophenol by Arthrobacter chlorophenolicus A6: effect of culture conditionsand degradation kinetics, Biodegradation 22 (2011) 275–286.

[21] Y. Jiang, J.P. Wen, H.M. Li, S.L. Yang, Z.D. Hu, The biodegradation of phenol athigh initial concentration by the yeast candida tropicalis, Biochem. Eng. J. 24(2005) 243–247.

[22] C.S. Qiu, J.P. Wen, X.Q. Jia, Extreme-thermophilic biohydrogen productionfrom lignocellulosic bioethanol distillery wastewater with community anal-ysis of hydrogen-producing microflora, Int. J. Hydrogen Energy 36 (2011)8243–8251.

[23] M. Caldeira, S.C. Heald, M.F. Carvalho, I. Vasconcelos, A.T. Bull, P.M.L. Castro,4-Chlorophenol degradation by a bacterial consortium: development of a gran-ular activated carbon biofilm reactor, Appl. Microbiol. Biotechnol. 52 (1999)722–729.

[24] J.F. Andrew, A mathematical model for continuous culture of microor-ganisms utilizing inhibitory substrates, Biotechnol. Bioeng. 10 (1968)707–723.

[25] D.K. Mishra, K.D. Dolanl, L. Yang, Bootstrap confidence intervals for the kineticparameters of degradation of anthocyanins in grape pomace, J. Food ProcessEng. 34 (2011) 1220–1233.

[26] M.C. Shashi, S. Kumar, Sensitivity analysis of biodgradation of soil applied pes-ticides using a simulation model, Biochem. Eng. J. 19 (2004) 119–125.

[27] M. Unell, K. Nordin, C. Jernberg, J. Stenström, J.K. Jansson, Degradationof mixtures of phenolic compounds by Arthrobacter chlorophenolicus A6,Biodegradation 19 (2008) 495–505.

[28] H.Y. Zeng, H. Jiang, K. Xia, Y.J. Wang, Y. Huang, Characterization of phenol degra-dation by high-efficiency binary mixed culture, Environ. Sci. Pollut. Res. Int. 17(2010) 1035–1044.

[29] J.A. Field, R. Sierra-Alvarez, Microbial degradation of chlorinated phenols, Rev.Environ. Sci. Biotechnol. 7 (2008) 211–241.

[30] T. Essam, M.A. Amin, O. El Tayeb, B. Mattiasson, B. Guieysse, Kinetics andmetabolic versatility of highly tolerant phenol degrading Alcaligenes strainTW1, J. Hazard. Mater. 173 (2010) 783–788.

[31] J.B. Rogers, N.M. DuTeau, K.F. Reardon, Use of 16S-rRNA to investigate microbialpopulation dynamics during biodegaradation of toulene and phenol by a binaryculture, Biotechnol. Bioeng. 70 (2000) 436–445.

[32] O.J. Hao, M.H. Kim, E.A. Seagren, H. Kim, Kinetics of phenol and chloropohenolutilization by Acinetobacter species, Chemosphere 46 (2002) 797–807.

[33] S.J. Wang, K.C. Loh, Modeling the role of metabolic intermediates in kinetics ofphenol biodegradation, Enzyme Microb. Technol. 25 (1999) 177–184.

[34] Z.J. Xiao, F.F. Huo, Y.L. Huang, X.K. Zhu, J.R. Lu, A novel 2,3-xylenol-utilizing Pseu-domonas isolate capable of degrading multiple phenolic compounds, Bioresour.Technol. 104 (2012) 59–64.

[35] S.J. Pirt, The maintenance energy of bacteria in growing cultures, Proc. R. Soc.Lond. B. Biol. Sci. 163 (1965) 224–231.

Biodegradation of dual phenolic substrates in simulated wastewater byGliomastix indicus MTCC 3869

Shashi Kumar, Deepika Arya, Abhinav Malhotra, Surendra Kumar *, Brajesh Kumar

Department of Chemical Engineering, Indian Institute of Technology Roorkee, Roorkee 247667, Uttarakhand, India

Introduction

Environmental pollution due to phenolic compounds is a majorproblem, which is being faced by developing countries today.Phenol and its derivatives such as resorcinol, and cresols are widelyfound in the effluents of many industries which include oilrefineries, ceramic plants, steel plants, coal conversion processplants, and textile, phenolic resin, paper and pulp, pharmaceutical,fertilizers, pesticides, plastics, petrochemical, explosive produc-tion, rubber industries. The discharge range of phenolic com-pounds depends upon the type of industry. The discharge ranges ofconcentration of phenol, resorcinol and p-cresol from a fewindustries have been reported by Gonzalez-Munoz et al. [1],Gonzalez et al. [2], Kira et al. [3], Phutdhawong et al. [4], Kumaraand Paruchuri [5], Babich and Davis [6]. Once the phenoliccompounds present in untreated industrial effluents, are releasedinto the environment, they persist in the water for a week or moreand adversely affect life forms. They react with metal ions andother compounds present in the waste resulting into the formationof more toxic complex compounds [7]. Phenol and resorcinol arenot potential carcinogens. But United States environment protec-tion agency has classified p-cresol as pollutant of group C (possiblehuman carcinogens) and listed it as priority pollutant [8,9]. WorldHealth Organization (WHO) has set a limit level of 0.001 mg/L to

regulate the concentration of these phenolic compounds indrinking water [10]. The European Union listed phenol amongthe ‘‘substance undesirable in excessive amounts’’ and has decidedthat its amount in water (lakes, streams) should be limited to0.3 mg/L to protect human health from the possible harmful effectson exposure to phenol by drinking water and eating contaminatedwater plants [11].

Various treatment technologies have been studied for thetreatment of wastewater contaminated with phenols in order toreduce the toxicity of wastewater and thereby to reducethe environmental load of harmful phenolic compounds. In thebiodegradation technique a large number of reports on thebiodegradation of phenolic compounds by bacteria are availablein the literature [12–21]. But the application of yeast and fungalstrain has not been studied in detail. However, the filamentousfungi are being used in many biological processes like productionof organic acids, enzymes, hormones, antibiotics, steroids, andtreatment of environment problems [22,23]. In recent years ofresearch interest has been focused on the degradation of industrialeffluents by the fungi in order to solve the environmentalproblems. In the present research programme fungal strainGliomastix indicus MTCC 3869 has been used for biodegradationof phenol, resorcinol, and p-cresol in synthetic wastewaters.

During single substrate biodegradation of a toxic compound,the biomass growth inhibition occurs due to the enhancedsubstrate toxicity, beyond a certain initial concentration in theculture medium. This substrate inhibition effect results intothe slow biomass growth rate, lower biomass yield, and increasein the energy expenditure for maintenance of the cells. Usually the

Journal of Environmental Chemical Engineering 1 (2013) 865–874

A R T I C L E I N F O

Article history:

Received 3 June 2013

Received in revised form 24 July 2013

Accepted 28 July 2013

Keywords:

Biodegradation

Wastewater treatment

Interaction parameter

Maintenance energy

A B S T R A C T

The biodegradation of phenol, resorcinol, and p-cresol by Gliomastix indicus was carried out by

conducting batch experiments in single and the dual substrate systems: phenol–p-cresol, and phenol–

resorcinol, at various initial concentrations in simulated aqueous solution. The growth kinetic model was

selected and interaction parameters were estimated. Competitive inhibition type of substrate

interaction was involved in this dual substrate systems. The interaction parameter values were found

as Ia,1 = 0.044, Ia,2 = 1.17 for phenol–p-cresol, and Ia,1 = 1.09, Ia,2 = 0.052 for phenol–resorcinol system.

The variation in maintenance energy expenditure with specific growth rate was incorporated to model

specific degradation rate. A linear model for specific degradation rate was developed for single substrate

system, while a non linear model was developed for dual substrate system. A set of model equations was

proposed and solved to describe the biodegradation dynamics of substrates in the two dual substrate

systems. The simulation results were found to be consistent with the experimental data.

� 2013 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +91 9897077460; fax: +911332273560.

E-mail addresses: [email protected], [email protected]

(S. Kumar).

Contents lists available at ScienceDirect

Journal of Environmental Chemical Engineering

journa l homepage: www.e lsev ier .com/ locate / jece

2213-3437/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.jece.2013.07.027




http://www.sciencedirect.com/science/journal/aip/22133437


industrial effluents contain a mixture of pollutants in wastewatercreates problems for their biodegradation because differentoperating conditions such as pH and temperature may be requiredfor the biodegradation of different compounds. Besides, thebiodegradation of one substrate may be inhibited by the presenceof other substrate present in the wastewater. The interactionamong these multiple substrates is complex due to their toxicityand competition for microbial enzymes and cofactors [24].Therefore, it is necessary to study the way of interaction ofsubstrates with each other and its effect on their degradation. Indual substrate system the biodegradation rate remains low,usually due to substrate toxicity, competitive inhibition, and theformation of toxic intermediates by nonspecific enzymes[15,20,25]. Various types of substrate interaction patterns includ-ing competitive inhibition and non competitive inhibition havebeen observed in different dual substrate biodegradation system[26–29]. The pollutants like phenol resorcinol, p-cresol are toxic to

growing cells. They damage the cells inhibiting their metabolismand stop their growth when present in higher concentration.

In majority of biodegradation studies research workers havefrequently reported biomass growth kinetics and substratedegradation kinetics without taking into account the changes inmaintenance energy requirement of the culture. A significantlyhigher amount of maintenance energy is required during thebiodegradation of toxic substrates, in comparison to other cultureswhere energy providing substrate is nontoxic substance likeglucose, fructose, molasses etc. Experimentally it has beenobserved that the biomass growth yield and requirement ofmaintenance energy of microorganism vary with the differentconcentrations of toxic substrate in the medium [30–32]. Theenergy requirement for the cell maintenance depends upon themicroorganism and the toxic substrate under biodegradation.The operating condition such as temperature, concentration ofnutrients, and pH value in the medium also affect the maintenanceenergy requirement of microorganisms. The cells under biodegra-dation studies require a minimum, constant and continuousamount of maintenance energy to tolerate the toxicity of thesubstrate and for their survival at each growth phase. This amountof maintenance energy is consumed by the cells for maintenanceactivities while the rest of the maintenance energy is produced forthe growth of microorganism. Further, when substrate inhibitionto biomass growth takes place in the medium, the degree oftoxicity of substrate including the production of various inter-mediates and extracellular products, affect the biomass growthyield adversely and lead to the higher requirement of energy forthe maintenance of the cells [33,34]. Therefore, the biomassgrowth yield and the substrate degradation are not directlyproportional to each other. Substrate degradation takes place eventhough the biomass growth yield is low, because the consumedsubstrate is utilized for more energy generation to be utilized forhigher maintenance of microbial cells at enhanced concentrationof the toxic substrates like phenol, resorcinol, and p-cresol, causingthe inhibition to biomass growth and to their own degradation.Hence, the concept of energy expenditure for maintenance of cellsis needed to provide proper description of biodegradationdynamics. The knowledge of biodegradation dynamics is signifi-cant to design the biodegradation unit and to predict the changesin the concentration of a component in the wastewater during itsremoval by biodegradation technology.

The present study reports the quantification of maintenanceenergy during the degradation of a mixture of phenolic substrates,and an analysis of substrate interactions during biomass growth,and the degradation dynamics. To the best of our knowledge, thereis no other biodegradation study available in the literature whichquantifies the variation in the maintenance energy expenditurewhen substrate concentration in the mixture is in inhibition range,and biomass growth follows inhibition kinetics. The workpresented here, is a reasonable starting point for the developmentand validation of mathematical model to describe the specificdegradation rate, growth rate, maintenance energy expenditure,growth yield, and degradation dynamics for mixtures of twohomologous substrates. The results provide in depth knowledge ofthe degradation of organic pollutants for phenolic wastewatertreatment, required to design a biodegradation facility.

Materials and method

Microorganism and cultivation condition

The fungus G. indicus MTCC 3869 was procured from theInstitute of Microbial Technology (IMTECH), Chandigarh, India andmaintained on the potato dextrose agar (PDA) medium containing:potatoes (200 g/L), dextrose (20 g/L) and agar (15 g/L), at pH 6

Nomenclature

Abbreviations

PSSH pseudo-steady state hypothesis

SKIP sum kinetics with interaction parameters

Notations

f [–] substrate interaction coefficient

Ia,1 [–] inhibition to the degradation of substrate 1 in

the presence of substrate 2

Ia,2 [–] inhibition to the degradation of substrate 2 in

the presence of substrate 1

Ib,1 [L/mg] inhibition to the degradation of substrate 1

in the presence of both the substrates 1 and 2

Ib,2 [L/mg] inhibition to the degradation of substrate 2

in the presence of both the substrates 1 and 2

K [mg/L]2 substrate interaction coefficient for sub-

strate used in Eqs. (6)–(12)

Ki [mg/L] inhibition constant

KSi [mg/L] saturation constant of biomass growth for

substrate i

k2, k8 [h�1] reaction rate constants

mSi [h�1] maintenance energy coefficient for substrate i

qSi [h�1] specific degradation rate of substrate i

qSmax [h�1] maximum specific degradation rate

qS0 [h�1] initial specific degradation rate

qS0max [h�1] maximum initial specific degradation rate

[Si] [mg/L] concentration of substrate i

[S0] [mg/L] initial substrate concentration

t [h] time

[X] [mg/L] biomass concentration

[X0] [mg/L] initial biomass concentration

[XT] [mg/L] total biomass concentration

(Yx/s)oi [g/g] observed growth yield coefficient

(Yx/s)T [g/g] true growth yield coefficient

(Yx/s)Ti [g/g] maximum growth yield coefficient with

respect to substrate i

Greek letters

mg [h�1] specific growth rate

mgmax [h�1] maximum specific growth rate

mgi [h�1] specific growth rate for substrate i

S. Kumar et al. / Journal of Environmental Chemical Engineering 1 (2013) 865–874866

using 1N NaOH by serial transfer on the PDA medium after everytwo weeks duration and incubating at 28 8C. For biodegradationstudies, a mineral salt medium supplemented with phenol,resorcinol, and p-cresol as single substrate and phenol incombination with p-cresol and resorcinol as dual substrate alongwith the ingredients of K2HPO4 (1 g/L), FeSO4�7H2O (0.01 g/L),NH4NO3 (3 g/L), MgSO4�7H2O (0.5 g/L), KCl (0.5 g/L) was used [35].

Chemicals

All the chemicals used in the experimentation including phenol,resorcinol and p-cresol were of AR grade with more than 99%purity. These chemicals were procured from HiMedia LaboratoriesPvt. Ltd. Mumbai, Loba Chemie Pvt. Ltd. Mumbai, Ranbaxy FineChemicals Ltd., New Delhi, Reidel Chemicals, Hapur.

Inoculum development and experimentation

Acclimatization experiments were conducted for familiariza-tion of fungal strain to phenolic environment. In the presentresearch work, the three phenolic compounds phenol, resorcinoland p-cresol were degraded using fungal strain G. indicus MTCC3869 at high concentration of phenol up to 1000 mg/L, resorcinolup to 1300 mg/L, and p-cresol up to 700 mg/L. To acclimatize thefungus, 2% glucose was added for appropriate growth of the fungusin modified czapeck medium containing substrate (phenol/resorcinol/p-cresol). Cultures were acclimatized, by exposing themto the toxic substrate in a series of conical flasks (250 mL) withworking volume of 100 mL. Glucose concentration was decreasedgradually along with increasing concentration of the toxicsubstrate into the medium for a period of 2 months. The developedinoculum was used for all batch culture experiments during theexponential growth phase of the culture. pH 6 and temperature of28 8C in the incubator were maintained during the wholeexperimentation.

For each concentration of the substrates, fourteen flasks of250 mL capacity with 50 mL working volume were used and keptin BOD incubator-cum-orbital shaker at 28 8C and 150 rpm.Inoculation step was done in aseptic conditions of UV chamberand 5% V/V inoculum was taken. The single substrate biodegrada-tion experiments were conducted in the range of 70 to 1000 mg/Lfor phenol, 50 to 700 mg/L for p-cresol, and 90 to 1300 mg/L forresorcinol. The batch experiments of the interaction betweenmixed substrates were performed in the three combinations ofphenol with p-cresol, and phenol with resorcinol. The combina-tions of phenol and p-cresol were as 100 mg/L phenol with 300 mg/L p-cresol, 200 mg/L phenol with 200 mg/L p-cresol, and 300 mg/Lphenol with 100 mg/L p-cresol. Similarly, phenol and resorcinolwere used in the three combinations; 100 mg/L phenol with300 mg/L resorcinol, 200 mg/L phenol with 200 mg/L resorcinol,and 300 mg/L phenol with 100 mg/L resorcinol.

Determination of biomass and substrate concentrations in the sample

The experimental flasks were taken out at different timeintervals for sampling, to study the biomass growth and substratedegradation. The samples of 2 mL volume were taken from each ofthe experimental flasks and were subjected to centrifugation at8000 � g for 15 min at 25 8C, for the estimation of biomass andresidual substrate concentrations. Supernatant was separated outfor the analysis of substrate concentrations by high performanceliquid chromatography (HPLC, Waters system, USA) with aSymmetry1 C18, 5 mm (250 mm � 4.6 mm, Waters, Ireland)column. Elution was performed with 400/300 (v/v) methanol/water at a flow rate of 1.0 mL/min, and detection was realized witha photodiode array detector (Waters 2998) at 271, 274, 277 nm for

phenol, resorcinol, and p-cresol respectively while in dualsubstrate biodegradation system detection was realized at275 nm. The retention time for phenol, resorcinol, and p-cresolwere 3.35, 2.53, and 4.54 min respectively. For the analysis ofbiomass growth, oven dry method was used. The biomass wasfound in the form of a pellet at the bottom of the centrifuge tubeafter the centrifugation. The biomass pellet was washed with thedistilled water using centrifugation again. After washing, thebiomass pellet was taken out on filter paper in a petri plate and waskept in the oven at 80 8C for 24 h. After evaporation of water fromthe biomass pellet, dry biomass was weighed using a chemicalbalance for the analysis of biomass growth. Each batch experi-mental run was repeated three times under identical conditionsand the values were averaged to get true experimental values.

Kinetic modeling

During single or dual substrate batch biodegradation process,the growth rate of biomass is expressed as

d½X�dt¼ mg ½X� (1)

where mg is specific growth rate (h�1) and [X] is the biomassconcentration at any time t. In order to model the growth kineticsfor mixed substrates, various sequences of reactions analogous toenzymatic reactions have been proposed in the literature[14,40,41,43]. In the present situation, one possible sequence ofreactions based on the enzymatic reactions for the mixed substratesystem consisting of substrates S1 and S2, is given in Table 1. Inthese reactions X1 and X1 and X2 are different intermediate states ofthe microorganisms in which substrates S1 and S2 are consumedrespectively but no biomass growth has occurred. Other inter-mediates are X1S1, X1S2, X1S1S2, X2S2, X2S1, and X2S1S2 complexes.In most of the dual substrate systems during two substratereactions, it appears that a ternary intermediate complex X1S1S2

and X2S1S2 may be formed with both substrates [36]. These factsare incorporated in Eqs. (e), (f), (k), (l) of Table 1. Eqs. (b) and (h) are

Table 1Reaction schemes and concentration of active intermediates.

Reaction Equation Reaction Equations

X þ S1 !k1

k1b

X1 (a) X þ S2 !k7

k7b

X2 (d)

X1 þ S1 !k2

2X (b) X1 !k8

2X (e)

X þ S1 !k3

k3b

X1S2 (c) X2 þ S1 !k10

k9b

X2S2 (f)

X1 þ S2 !k4

k4b

X1S2 (g) X2 þ S1 !k10

k10b

X2S1 (j)

X1S1 þ S2 !k5

k5b

X1S1S2 (h) X2S2 þ S1 !k11

k11b

X2S2S1 (k)

X1S2 þ S1 !k6

k6b

X1S2S1 (i) X2S1 þ S2 !k12

k12b

X2S1S2 (l)

Concentration of

intermediates

Equation Concentration of

intermediates

Equation

X1½ � ¼X½ � S1½ �2

KS1

(m)X2½ � ¼

X½ � S2½ �KS2

(r)

X1S1½ � ¼ X½ � S1½ �2

KS1Ki1(n) X2S2½ � ¼ X½ � S2½ �2

KS2Ki2(s)

X1S2½ � ¼ X½ � S1½ � S2½ �KS1K11

(o) X2S1½ � ¼ X½ � S2½ � S1½ �KS2K12

(t)

X1S1S2½ � ¼ X½ � S1½ �2 S2½ �KS1Ki1K21

(p) X2S2S1½ � ¼ X½ � S2½ �2 S1½ �KS2Ki2K22

(u)

X1S2S1½ � ¼ X½ � S1½ �2 S2½ �KS1K11K31

(q) X2S1S2½ � ¼ X½ � S2½ �2 S1½ �KS2K12K32

(v)

1

KS1¼ k1

k1b þ k2;

1

KS2¼ k7

k7b þ k8;

1

Ki1¼ k3

k3b;

1

K11¼ k4

k4b;

1

K21¼ k5

k5b;

1

K31¼ k6

k6b;

1

Ki2¼ k9

k9b;

1

K12¼ k10

k10b;

1

K22¼ k11

k11b;

1

K32¼ k12

k12b

S. Kumar et al. / Journal of Environmental Chemical Engineering 1 (2013) 865–874 867

assumed to be irreversible first-order reactions, whereas remain-ing Eqs. (a), (c)–(g), (i)–(l) are assumed to be reversible, and are offirst order with respect to each of the reactants and products. InTable 1, Eqs. (a)–(c) represent substrate inhibition by S1 and Eqs.(d)–(f) represent substrate inhibition by substrate S2. Eqs. (g)–(i)and (j)–(l) represent the cross inhibition between substrates S1 andS2.

The active intermediates may follow two reaction pathways. Inone pathway, the active intermediate may be deactivated which isjust the reverse reaction of their formation (reversible reaction). Inthe alternative pathway, the active intermediate decomposesspontaneously to form stable products (irreversible reaction) [37].It is very difficult to measure the concentration of activeintermediates because they are highly reactive and very shortlived. Consequently, the evaluation of reaction rate laws in theirpresent form becomes quite difficult. Besides, in most of theinstances it is not possible to eliminate the concentration of activeintermediates in the differential form of the mass balance equationto obtain the solution. However, an approximate solution may beobtained, exploring the pseudo-steady state hypothesis (PSSH)method. In pseudo-steady state approximation, the rate offormation of active intermediates is assumed to be equal to itsrate of disappearance. As a result, the net rate of formation of activeintermediates is zero. Thus, the concentration of active inter-mediates can be expressed in terms of the concentrations ofbiomass and substrate. The approximation by PSSH is applicabledue to two conditions of intermediates: they have very short lifetime because of their high reactivity and are present in lowconcentrations. Further, it is known that the net rate of formationof any reaction species involved in many simultaneous reactions isthe sum of the rates of formation of that reaction species in eachreaction. On this basis, the net rate of formation of ith reactionspecies occurring in N different reactions can be generalized as,

ri ¼XN

j¼1

rij; j ¼ 1!N (1a)

In above series of reaction step Eqs. (a)–(l) of Table 1, eachreaction is elementary in which the reaction orders andstoichiometric coefficients are identical. Thus, on applyingEq. (1a) and PSSH on Eqs. (a)–(l), the net rate of formation ofthe active intermediates in Eqs. (a)–(l) are formulated. These netrate rates of formation of all intermediates are set to zero to getconcentrations of intermediates in terms of biomass and substrateconcentrations as represented by Eqs. [(m)–(v)] in Table 1.

The total biomass growth rate can be represented in terms oftotal biomass concentration [XT], as

d½XT �dt¼ d

dt

� ½X� þ ½X1� þ ½X2� þ ½X1S1� þ ½X1S2� þ ½X1S1S2� þ ½X1S2S1�þ½X2S1� þ ½X2S2� þ ½X2S1S2� þ ½X2S2S1�

� �(2)

On applying PSSH in Eq. (2) and combining it with Eqs. (1), (b),and (e) one gets

d½XT �dt¼ d½X�

dt¼ mg ½XT � ¼ k2½X1� þ k8½X2� (3)

Eq. (3) yields

mg ¼k2½X1� þ k8½X2�

½XT �(4)

The substitution of values from Eqs. (m)–(v) in Eq. (4), gives

mg ¼mgmax1½S1�

D1þ

mgmax2½S2�D2

(5)

where

k2 ¼ mgmax1 ; k8 ¼ mgmax2

D1 ¼ KS1 þ ½S1� þ½S1�2

Ki1þ ½S1�½S2�

K51þ ½S1�2½S2�

K41þ f ½S2� þ f

½S2�2

Ki2

þ f

K42½S2�2½S1� (6)

D2 ¼ KS2 þ ½S2� þ½S2�2

Ki2þ ½S1�½S2�

K52þ ½S2�2½S1�

K42þ ½S1�

fþ ½S1�2

fKi1

þ ½S1�2½S2�fK41

(7)

f ¼ KS1

KS2; (8)

1

K41¼ 1

Ki1K21þ 1

K11K31

� �; (9)

1

K42¼ 1

Ki2K22þ 1

K12K32

� �; (10)

1

K51¼ f

K12þ 1

K11; (11)

1

K52¼ 1

K12þ 1

fK11(12)

The values of mgmax1, KS1, Ki1 and mgmax2, KS2, and Ki2 can beobtained separately from the kinetics of individual biomass growthon phenol, resorcinol, and p-cresol as sole energy and carbonsource. The other parameters including ‘f’ can be determined asindependent parameter by fitting the experimental data.

Results and discussion

Growth kinetics

Single substrate system

The biodegradation ability of fungus G. indicus has beeninvestigated up to the initial concentrations of 1000, 700, and1300 mg/L for phenol, p-cresol, and resorcinol respectively.Biodegradation of initial phenol concentration of 700 mg/L takesplace in 98 h. The biodegradation of p-cresol and resorcinol withthe same initial concentration of 700 mg/L takes place in 130 and69 h respectively. Figs. 1 and 2 show the variation of specificgrowth rate and observed biomass growth yield with initialconcentrations of the three substrates. The specific growth rate andobserved biomass growth yield increase up to the initialconcentration of 70 mg/L of phenol, 50 mg/L of p-cresol, and90 mg/L for resorcinol. Beyond these inhibitory initial concentra-tions, the biomass growth yield and specific growth rate tend todecrease because of increased substrate toxicity at higher initialsubstrate concentrations. This decline trend shows that phenol, p-cresol, and resorcinol are inhibitory substrates.

Specific growth rate of resorcinol is much higher than thespecific growth rate of phenol and p-cresol, because phenol and p-cresol cause stronger substrate inhibition to the biomass growthin comparison to resorcinol. Therefore, resorcinol is utilized morethan the phenol and p-cresol by the fungal cells in singlesubstrate system. To study the biomass growth kinetics of thesingle substrates, four single substrate growth kinetic modelslisted in Table 2 have been selected [38]. The kinetic parametersof the models have been estimated by non linear least square


technique using MATLAB 7.2 based on Windows XP. Theestimated values of kinetic parameters are also given inTable 2 for phenol. The growth kinetic parameters for resorcinoland p-cresol have been estimated in previous study [38].Goodness of the fit of experimental data with the models understudy has been judged by measuring R2 and the percent standard

deviation Dmg%� �

between experimental and predicted values of

specific growth rate for each model [39]. It is clear that lower thevalue of percent standard deviation, the better is the fit of

experimental data. Yano R2 ¼ 0:974; Dmg%� �

¼ 1:73� �

and

Andrews and Noack R2 ¼ 0:993; Dmg%� �

¼ 1:03� �

models

were found best fit to the specific growth rate data of resorcinoland p-cresol respectively [39]. For phenol, the predictions ofAndrews and Noack model are in the best agreement with theexperimental data of specific growth rate with R2 = 0.98 and

percent standard deviation Dmg%� �

¼ 2:1 (Table 2). The equa-

tions of the best fit single substrate kinetic models for the threesubstrates, with estimated values of kinetic parameters can berestated as follows:

mg ¼0:462½S�

½S� þ 78:29þ ½S�244:49

For phenol (13)

mg ¼0:512½S�

½S� þ 91:87þ ½S�221:99

For p-cresol (14)

mg ¼0:185½S�

½S� þ 19:83þ ½S�2

376 1þ ½S�1790

� � For resorcinol (15)

Dual substrate system

A series of batch biodegradation experiments were conductedto study the biomass growth rate and interaction between the twosubstrates in the two dual substrate degradation systems. Eq. (5)indicates that the specific growth kinetic model for dual substratesystem can be expressed as

mg ¼ mg1 þmg2 (16)

mg1 and mg2 represent specific growth rates of the biomass onsubstrates S1 and S2 respectively. The mathematical expressions ofmg1 and mg2 indicate that mg1 and mg2 depend on both thesubstrates due to inhibition and interaction effects of one substrateon the other. Similar types of expressions have been proposed bymany authors [40–42]. Furthermore, in the study by Yan et al. [50]on phenol and m-cresol, the values of coefficients of S2

1S2, S22S1, S2

1

and S22 in Eqs. (6) and (7) are found to be negligibly small. Finally, an

alternative model analogous to Eq. (5) that takes into account theunspecified type of substrate inhibition has been formulated. Thismodel consists of the interaction parameters that are treated asunknown. According to this description, Juang and Tsai [15]expressed specific growth rate in dual substrate system as

mg ¼mgmax1½S1�

KS1 þ ½S1� þ ½S1 �2Ki1þ Ia;1½S2� þ Ib;1½S1�½S2�

þmgmax2½S2�

KS2 þ ½S2� þ ½S2 �2Ki2þ Ia;2½S1� þ Ib;2½S1�½S2�

(17)

where Ia,1 represents inhibition to the degradation of S1 in thepresence of S2. Ia,2 represents inhibition to the degradation of S2 inthe presence of S1. While Ib,1 shows inhibition to the degradation ofS1 due to the presence of both S1 and S2. Similarly Ib,2 shows theinhibition to the degradation of S2 due to the presence of both S1

and S2. These interaction parameters take into account allinteractions jointly as considered in Eq. (5). The values of kineticconstants mgmax, KS, Ki are the same as in the single substratebiodegradation system (1 and 2 on suffix refer to substrates S1 andS2 respectively). There is a concept of four substrate interactionpatterns for dual substrate system applied to various studies[40,44]. The conditions corresponding to these interaction patternsare applied to Eq. (17) to get final results to represent and discussthe growth kinetics. The conditions Ib,1 = 0, Ib,2 = 0, Ia,1 6¼ 0, Ia,2 6¼ 0represent the competitive cross inhibition between the twosubstrates (pattern 1). The conditions Ib,1 6¼ 0, Ib,2 6¼ 0, Ia,1 = 0,Ia,2 = 0 represent the uncompetitive cross inhibition (pattern 2).The conditions Ib,1 = 0, Ib,2= 0, but either Ia,1 6¼0 , Ia,2 = 0 or Ia,1 = 0,Ia,2 6¼ 0 indicate the competitive partial inhibition (pattern 3). Theconditions Ia,1 = 0, Ia,2 = 0, but either Ib,1 6¼ 0, Ib,2 =0 or Ib,1 = 0,Ib,2 6¼ 0 show the uncompetitive partial inhibition between the twosubstrates (pattern 4).

[(Fig._1)TD$FIG]

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0 200 400 600 800 1000 1200 1400

Spec

ific

grow

th r

ate

(h-1

)

Initial substrate concentration (mg/L)

Specific growth rate phenolSpecific growth rate p-cresolSpecific growth rate resorcinol

Fig. 1. Effect of initial substrate concentration on specific growth rate of phenol,

p-cresol and, resorcinol.

[(Fig._2)TD$FIG]

0.01

0.017

0.024

0.031

0.038

0.045

0.052

0.059

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0 200 400 600 800 1000 1200 1400

Obs

erve

d bi

omas

s gro

wth

yie

ld (g

/g)

Initial substrate concentration (mg/L)

Observed growth yield phenolObserved growth yield p-cresolObserved growth yield resorcinolMaintenance energy coefficient phenolMaintenance energy coefficient resorcinolMaintenance energy coefficient p-cresol

Mai

nten

ance

ene

rgy

coef

ficie

nt (h

-1)

Fig. 2. Effect of initial substrate concentration on observed biomass growth yield

and maintenance energy coefficient.


The results of single substrate degradation study reveal thatphenol, p-cresol, and resorcinol start to cause substrate inhibitionin the medium at the initial concentration of 70, 50, and 90 mg/Lrespectively [39]. Hence, the substrate interaction, in the dualsubstrate biodegradation systems: phenol–p-cresol, and phenol–resorcinol has been studied at the initial concentrations higherthan the inhibitory concentrations of these substrates. Thebiodegradation of Phenol and p-cresol has been studied in threecombinations of concentrations; 100 mg/L phenol–300 mg/L p-cresol, 200 mg/L phenol–200 mg/L p-cresol, and 300 mg/L phenol–100 mg/L p-cresol. Figs. 1–3 show the biodegradation trend ofphenol and p-cresol with biomass growth curve for the combina-tions of phenol and p-cresol. The values of interaction parametershave been estimated by Levenberg–Marquardt nonlinear regres-sion program in MATLAB 7.2. Eq. (17) describes four type ofsubstrate inhibition during dual substrate degradation. Theequation was solved applying the four conditions given for thefour patterns of substrate interaction in the dual substrate system.The conditions given for pattern 1 are found applicable to theexperimental data of phenol–p-cresol system. The estimatedvalues of interaction parameters Ia,1 and Ia,2 show the involvementof competitive cross inhibition in phenol–p-cresol system asthe substrate interaction. Values of interaction parameters(Ia,1 = 0.044, Ia,2 = 1.17) imply that the phenol inhibits p-cresoldegradation more than the p-cresol causes inhibition to phenoldegradation in the medium. Experimental study reveals that thephenol is consumed preferentially by fungus G. indicus in phenol–p-cresol system. The specific growth rate of fungal biomass in

phenol–p-cresol system can be expressed as

mg ¼0:462½S1�

78:29þ ½S1� þ ½S1 �244:29þ 0:044½S2�

þ 0:512½S2�91:87þ ½S2� þ ½S2 �2

21:99þ 1:17½S1�(18)

To study the effect of the presence of resorcinol on biodegra-dation, phenol, and resorcinol have been taken in combinations of100 mg/L phenol–300 mg/L resorcinol, 200 mg/L phenol–200 mg/Lresorcinol, and 300 mg/L phenol–100 mg/L resorcinol. The resultsof single substrate biodegradation study show that resorcinol is aninhibitory substrate like phenol. it causes substrate inhibitioneffect in the medium at an initial concentration of 90 mg/L. Hence,the biodegradation of resorcinol in presence of phenol has beenstudied at the concentrations higher than 90 mg/L.

The analysis of Eq. (17) for phenol–resorcinol system byLevenberg–Marquardt nonlinear regression program applied onthe four inhibition patterns, shows that competitive crossinhibition takes place during the biodegradation of phenol withresorcinol. The values of interaction parameters have beenobtained as Ia,1 = 1.09, Ia,2 = 0.052 for phenol–resorcinol degrada-tion system, which indicate that the resorcinol has strongerinhibition effect on phenol degradation in comparison to theinhibition caused by the phenol to resorcinol degradation. Duringthe experimental study it has been observed that the fungusconsumed resorcinol in preference to phenol. The specific growthrate of biomass in phenol–resorcinol system is expressed as

mg ¼0:462½S1�

78:29þ ½S1� þ ½S1 �244:29þ 1:09½S2�

þ 0:185½S2�19:83þ ½S2� þ ½S2 �2

376 1þ ½S2 �1790

� �þ 0:052½S1�

(19)

The expressions of Eqs. (18) and (19) are similar to SKIP modelor ‘Sum Kinetics with Interaction Parameters’ equation asproposed and discussed by Yoon et al. [43]. This study on phenol,resorcinol, and p-cresol in dual substrate system indicates theascending order of toxicity of these substrates for G. indicus asresorcinol > phenol > p-cresol. Phenol, resorcinol, and p-cresol arehomologous substrates; therefore, competitive cross inhibition islikely to take place [45]. The results of dual substrate degradationstudy are supported by the reported findings of Saravanan et al.[38] and Wang et al. [46], where the less toxic substrate causesstronger inhibition to the degradation of more toxic substrate inthe dual substrate degradation system. A few studies on differentdual substrate degradation systems with the estimated values oftheir interaction parameters have been summarized in Table 3along with the values of interaction parameters obtained for thetwo dual substrate systems in the present study. All the modelparameters are corresponding to SKIP model.

Table 2Estimated values of biomass growth kinetic model parameters for phenol.

Model Mathematical expression mgmax(h�1) KS (mg/L) Ki (mg/L) K (mg/L) R2 Dmg%� �

Equations

Andrews and Noack mg ¼mgmaxS

Sþ KS þ ðS2=KiÞ0.462 78.29 44.49 – 0.98 2.10 (35)

Haldane mg ¼mgmaxS

Sþ KS þ S2=Ki

� �þ SKS=Kið Þ

0.485 53.56 53.59 – 0.97 2.75 (36)

Yano mg ¼mgmaxS

Sþ KS þ S2

Ki1þ S

K

� 0.262 36.08 128.5 784.6 0.99 2.16 (37)

Webb mg ¼mgmaxS 1þ S

K

� Sþ KS þ S2

Ki

0.240 29.88 114.9 9510 0.95 6.42 (38)

[(Fig._3)TD$FIG]

0

50

100

150

200

250

300

350

0 10 20 30 40 50 60

Con

cent

ratio

n of

subs

trat

es, p

heno

l/p-c

reso

l (m

g/L

)

Degradation time (h)

Exp. phenol 100 mg/LExp p-cresol 300 mg/LExp. phenol 200 mg/LExp p-cresol 200 mg/LExp. Phenol 300 mg/LExp. p-cresol 100 mg/LModel

Fig. 3. Effect of degradation time on concentration of substrates, phenol/p-cresol.


Degradation kinetics

The specific degradation rates of substrates S1 qS1ð Þ and S2 qS2ð Þcan be represented by Eqs. (20) and (21) respectively as givenbelow

qS1 ¼mg1

YX=S

� T1

(20)

d½S1�dt¼ �qs1½XT � (20a)

qS2 ¼mg2

YX=S

� T2

(21)

d½S2�dt¼ �qs2½XT � (21a)

The corresponding substrate degradation rate equations forsubstrates S1 and S2 can be described by using Eqs. (20a) and (21a)respectively. Here YX=S

� T1

and YX=S

� T2

are the maximum growthyield coefficients with respect to substrates S1 and S2. Eqs. (20) and(21) hold true when the maintenance requirements are negligible.In the biodegradation, a part of substrate is used by the biomass toform new biomass cells and another part is used to performmetabolic activities irrespective of growth. These non-growthmetabolic activities are performed by the consumption of energy,termed as maintenance energy expenditure and for a substrate ’ i ’they are described by maintenance energy coefficient mSi. Pirt [44]initially defined maintenance coefficient as the minimum sub-strate consumption to maintain the cell activity. Thus, apart frombiomass growth, some measure of maintenance energy is neededto provide proper description of the substrate degradationdynamics in dual substrate system. Maintenance energy wasconsidered as a constant quantity specific to substrate–microor-ganism system in various dual substrate degradation kineticmodeling studies [14,43].

The maintenance energy coefficient msi can be incorporated indefining specific degradation rate qSi for the given substrate ’ i ’ indual substrate system as follows:

qSi ¼ mSi þmgi

YX=S

� Ti

(22)

In this approach, maintenance denotes extra substrate con-sumption not used for growth purposes. Neijssel and Tempest [47]and Hempfling and Mainzer [48] reported the measurements of themaintenance energy and found the variation in it. Pirt [44]described the maintenance as growth rate dependent parameter

and postulated a modification to this theory considering mainte-nance dependent on specific growth rate and by including aportion that decreases with the increasing specific growth rate.Accordingly, the Eq. (17) is rewritten as

qSi ¼mgi

YX=S

� Ti

þm1i þ ki 1�mgi

mgmaxi

!(23)

where mgi is specific growth rate, m1i denotes the constantmaintenance energy coefficient and the third term of the equationis the growth rate dependent maintenance energy coefficient forsubstrate ’ i ’ in the dual substrate system. ki is a constant.

The specific substrate degradation rate for substrate ’ i ’ can alsobe defined in terms of observed growth yield coefficient YX=S

� oi

asfollows:

qSi ¼mgi

YX=S

� oi

(24)

Here it is important to distinguish YX=S

� oi

from YX=S

� Ti

. YX=S

� Ti

considers consumption of substrate for biomass growth only,while YX=S

� oi

is the yield corrected for maintenance. It implies thatYX=S

� Ti

is supposed to be higher and less variable with substrateconcentration than YX=S

� oi:. Experimentally it has been found that

observed biomass growth yield for phenol, resorcinol, and p-cresolvaries with the initial substrate concentration in the singlesubstrate systems. At each initial substrate concentration, theobserved growth yield has been determined by linearizing biomassgrowth with substrate degradation. Fig. 2 shows observed growthyield coefficient profiles as a function of initial substrateconcentrations for phenol, resorcinol, and p-cresol. The maximumobserved growth yield value of 0.437 g/g has been estimated at theconcentration of 70 mg/L for phenol, 0.443 g/g at 90 mg/L forresorcinol, and 0.31 g/g at 50 mg/L for p-cresol. For the estimationof maintenance energy coefficient values at each initial concen-tration of phenol, resorcinol, and p-cresol the linear expression formSi used in Eq. (23) has been applied. The decreasing trend ofobserved growth yield coefficient YX=S

� oi

and increasing mainte-nance energy coefficient mSi beyond the inhibitory initial substrateconcentration results in the reduction of observed growth yield.This study concludes that the substrate inhibition reduces thespecific growth rate as well as biomass growth yield due to theincrease in the value of maintenance energy coefficient. Therefore,relating specific degradation rate with the growth rate in terms ofYX=S

� Ti

[Eqs. (20) and (21)] is incorrect [33]. On comparing Eq. (24)with Eq. (23), it is clear that YX=S

� oi

incorporates the maintenanceenergy expenditure. For phenol, resorcinol, and p-cresol Eq. (23)

Table 3The estimated values of interaction parameters by SKIP model for different dual substrate systems.

Dual substrate system Microorganism Interaction parameter values Type of substrate inhibition References

Ia,1 Ia,2 Ib,1 Ib,2

Phenol–m-cresol Alcaligenes faecalis 4.82 4.12 – – Competitive inhibition Bai et al. [14]

Phenol–sodium salicylate Pseudomonas putida 0.277 0.126 1.46 0.509 Competitive + Uncompetitive

inhibition

Juang and Tsai [15]

Phenol–m-cresol Mixed culture 3.9 9.9 – – Competitive inhibition Saravanan et al. [38]

Benzene –toluene 5.16 0.49 – –

Benzene –phenol Pseudomonas putida F1 1.08 0.27 – – Competitive inhibition Abuhamed et al. [42]

Toluene –phenol 1.03 0.14 – –

Phenol–m-cresol Candida albicans PDY-07 2.91 1.79 – – Competitive inhibition Wang et al. [46]

Phenol–resorcinol Trichosporon Cutanium R57 – – – Aleksieva et al. [51]

Phenol–p-cresol Trametes versicolor 4.72 7.46 – – Competitive inhibition Alexieva et al. [52]

Phenol–p-cresol Trichosporon cutaneum 1 – – Competitive inhibition Alexieva et al. [53]

Phenol–p-cresol Aspergillus awamori 8.6 0.3 – – Competitive inhibition Yemendzhiev et al. [54]

Phenol–p-cresol 0.044 1.17 Competitive inhibition This study

Phenol–resorcinol Gliomastix indicus MTCC 3869 1.09 0.052 – –


can be restated as

qS ¼mg

0:437þ 0:020þ 0:0212 1�

mg

0:129

� �For phenol (25)

qS ¼mg

0:437þ 0:0135þ 0:054 1�

mg

0:132

� �For resorcinol (26)

qS ¼mg

0:437þ 0:0229þ 0:011 1�

mg

0:102

� �For p-cresol (27)

where the values of m1i and ki have been calculated by plotting thevalues of specific degradation rate qSð Þ against specific growth ratemg

� �values for each substrate in single substrate system.

Variation of maintenance energy expenditure as a function ofinitial substrate concentration has been shown in the Fig. 2.

Pirt [44] defined maintenance energy coefficient and therebythe specific degradation rate as a linear function of specific growthrate as is clearly shown in Eq. (23). This type of linear relationshipbetween qSi and mgi is applicable to single substrate degradationsystem as shown above. In dual substrate system, the linearrelationship between qSi and mgi has been used in empiricalformulation [14,33,43]. In these formulations YX=S

� Ti

has beenderived empirically as the maximum yield after correcting forconstant maintenance energy expenditure.

The sensitivity analysis of maintenance done by Bodegom [49]indicates the importance of various non-growth components, andemphasizes that the overall maintenance depends nonlinearly onrelative growth rate, relative death rate, growth yield, andendogenous metabolism. The maintenance is a dynamic processand ideally maintenance description should incorporate thedynamics of each non-growth component. There is no constantrelation between these non-growth parameters. The simplecombinations of these parameters cannot be made due to partialoverlapping of these parameters. The conceptual analysis onvarious non-growth components by Bodegom [49] shows strongdependence of overall maintenance on growth rate. This overallmaintenance depends on the growth rate in a non linear way.Further, the analysis on growth yield in case of dual substratesystem indicates that it is difficult to find out the growth yield forsubstrate ’ i ’ present in the mixture because the biomass onlyrepresents the total growth, and its decomposition in two partscorresponding to two substrates is extremely difficult. Therefore,instead of Eq. (24), Eq. (22) is considered to describe specificdegradation rate for dual substrate system where YX=S

� Ti

may beassumed to be a constant parameter, not measured experimentallybut estimated empirically by fitting the experimental data toempirical mathematical model.

In view of the above discussion, to generate a non-linearfunction of mgi, it is convenient to express maintenance energycoefficient mSi for substrate ’ i ’ in the following second degreepolynomial form

mSi ¼ A1i þ A2imgi þ A3im2gi (28)

With the substitution of Eq. (28) in the model, Eq. (22) todescribe the specific degradation rate of substrate ’ i ’ in dualsubstrate system, can be transformed in the following Eq. (29)

qSi ¼mgi

YX=S

� Ti

þ A1i þ A2imgi þ A3im2gi (29)

where A1i, A2i and A3i are constants of polynomial expression. Onkeeping YX=S

� Ti

constant, Eq. (29) can be reduced to

qSi ¼ A1i þ A4imgi þ A3im2gi (30)

where, A4i ¼ 1YX=Sð ÞTi

þ A2i

� �

The experimental values of specific degradation rates qS1ð Þ andqS2ð Þ are calculated using Eqs. (20) and (21) for phenol, resorcinol,

and p-cresol in the two dual substrate systems; phenol–p-cresol,and phenol–resorcinol. To study the variation of specific degrada-tion rates of the substrates with specific growth rate of biomass ateach combination of phenol with p-cresol and resorcinol, theproposed model Eq. (30) has been applied. For each substrate indual substrate system, the constants involved in Eq. (30) areestimated by using curve fitting tool in MATLAB 7.2. Thus, thespecific degradation rates for each substrate can be expressed asfollows:

Phenol– p-cresol system

qS1 ¼ 0:038þ 1:071mg1 � 2:460m2g1 ðR2 ¼ 1Þ For phenol (31)

qS2 ¼ 0:046� 2:175mg2 þ 62:56m2g2 ðR2 ¼ 1Þ For p-cresol (32)

Phenol–resorcinol system

qS1 ¼ 3:369� 108:2mg1 þ 886:7m2g1 ðR2 ¼ 1Þ For phenol (33)

qS2 ¼ 2:520� 51:34mg2 þ 274:9m2g2 ðR2 ¼ 1Þ For resorcinol

(34)

The values of regression coefficient (R2 = 1) in above expres-sions conclude that simulated model predictions are wellconsistent with the experimental data. These results justify theidea of inclusion of maintenance energy expenditure varyingnonlinearly with the growth rate, to quantify the specificdegradation rate in dual substrate system. Thus, it is suggestedthat Eq. (30) may be very well adopted for the assessment ofspecific degradation rate for the substrate in dual substrate system.

Computed substrate degradation profiles in dual substrate system

In the present study, the substrate degradation profiles withtime have been computed for the two dual substrate degradationsystems under study; phenol–p-cresol and phenol–resorcinol. Forthis purpose a set of model equations mentioned in Table 4 alongwith boundary conditions has been used. In the exponentialgrowth phase of batch culture, since the substrate consumption inlag phase is negligibly small in comparison to high substrateconcentration in the mixture, the initial substrate concentration ofboth the substrates in dual substrate system can be used as initialboundary conditions required to solve the model equations. Themodel equations have been solved simultaneously using ordinarydifferential equation solver tool of MATLAB 7.2 for the both dualsubstrate systems. The simulated results are discussed below.

Phenol – p-cresol System

Fig. 3 shows the comparison of the model predictions and theexperimentally determined degradation data on phenol and p-cresol in dual substrate system at the total initial substrateconcentration of 400 mg/L. The model corroborates the experi-mental data of phenol and p-cresol degradation well for eachcombination of the two substrates. These trends conclude that themodel proposed for the specific degradation rate [Eqs. (31) and(32)] by incorporating maintenance energy is quite correct. In acombination of 100 mg/L phenol and 300 mg/L p-cresol theobserved biodegradation times of phenol and p-cresol are 33 hand 50 h respectively. In the single substrate degradation system, ittakes 10 h for the 100 mg/L phenol and 38 h for 300 mg/L p-cresol.Since observed rate of degradation of either substrate in dualsubstrate system is slower than the degradation rate of either


substrate alone in single substrate system, a competitive type ofcross inhibition between phenol and p-cresol is thus observed [27].Therefore, the presence of p-cresol in phenol–p-cresol systemdecreases the phenol degradation rate. On increasing the phenolconcentration in combination with p-cresol the biodegradationtime of both the substrates is further increased due to thecompetitive cross inhibition caused by both substrates on eachother. These observations support the findings of growth kineticstudies on phenol–p-cresol system (Section 3.1.2) carried out totest the four substrate interaction patterns (Eq. (18)).

Phenol–resorcinol system

Fig. 4 shows the experimental and model predictions ofdegradation profiles for phenol and resorcinol in phenol–resorcin-ol system, at the total initial substrate concentration of 400 mg/L.The simulated values of the model show good agreement with theexperimental data of phenol and resorcinol degradation for eachcombination of phenol and resorcinol. Fig. 4 shows that it takes lesstime for resorcinol degradation in comparison to phenol. Thebiodegradation time for 100 mg/L resorcinol in combination with300 mg/L phenol has been observed 36 h while it took 38 h for thedegradation of 300 mg/L phenol. Biodegradation of 100 and300 mg/L phenol occurs in 7 and 32 h respectively, in singlesubstrate degradation system. When resorcinol is present withphenol in the medium it causes inhibition to phenol degradation.Since the biodegradation time for both the substrates in dual

substrate system is larger than the biodegradation time for thesesubstrates in single substrate system, the competitive crossinhibition is expected to occur between phenol and resorcinol.This fact is supported by the growth kinetic studies on phenol andresorcinol described by Eq. (19) with condition Ib,1 = Ib,2 = 0. Thedegradation dynamic profiles clearly support to conclude that themodels represented by Eqs. (33) and (34) describe well the specificdegradation rates for phenol–resorcinol system.

Conclusion

G. indicus degraded phenol, p-cresol, and resorcinol in the dualsubstrate systems efficiently. Competitive inhibition type ofsubstrate interaction was found between phenol and p-cresol,phenol and resorcinol. Specific degradation rate model with thevariation of maintenance energy expenditure was proposed andthe model predictions were found consistent with the experimen-tal data. Further, biodegradation dynamics of these substrates wasmodeled. Simulated model predictions supported the findings ofgrowth kinetic studies well. The substrate degradation kineticmodels proposed in this study can be applied to the design oftreatment unit for the biodegradation of mixture of phenoliccompounds on commercial scale.

Acknowledgments

This work was financially supported by the University GrantCommission (UGC), Bahadur Shah Zafar Marg, New Delhi: 110002,India.

References

[1] M.J. Gonzalez-Munoz, S. Luque, J.R. Alvarez, J. Coca, Recovery of phenol fromaqueous solutions using hollow fibre contactors, Journal of Membrane Science213 (2003) 181–193.

[2] G. Gonzalez, G. Herrere, M.T. Garcia, M. Pena, Biodegradation of phenolic indus-trial waste water in a fluidized bed bioreactor with immobilized cells of Pseudo-monas putida, Bioresource Technology 80 (2001) 137–142.

[3] A.O. Kira, M.B. Hector, W.R. Campbell, Effect of temperature o the inhibitionkinetics of phenol biodegradation by Pseudomonas putida Q5, Biotechnology andBioengineering 70 (2000) 291–299.

[4] W. Phutdhawong, S. Chowwanapoonpohn, D. Buddhasukh, Electrocoagulationand subsequent recovery of phenolic compound, Analytical Sciences 16 (2000)1083–1084.

[5] P. Kumaran, Y.L. Paruchuri, Kinetics of phenol biotransformation, Water Resource31 (1997) 11–22.

[6] H. Babich, D.L. Davis, Phenol a review of environmental and health risks, Regula-tory Toxicology and Pharmacology 1 (1981) 90–109.

[7] D.J. Sufit, Some effect of exposing rainbow trout to phenol solution, Journal of FishBiology 13 (1978) 7–17.

[8] J.K. Fawell, S. Hunt, Environmental Toxicology Organic pollutants, Halsted Press,John Wiley and sons, New York, 1988p. 398.

[9] N.G. Buckman, J.O. Hill, R.J. Magee, M.J. Mc Cormick, Separation of substitutedphenols, including eleven priority pollutants using high performance liquidchromatography, Journal of Chromatography A 284 (1984) 441–446.

[10] World Health Organization (WHO), International Standards for Drinking Water,Geneva 40, 1963.

[11] J.D. Zuane, Handbook of Drinking Water Quality, second ed., Wiley Publishers,Hoboken, NJ, 1997.

[12] M. Bajaj, C. Gallert, J. Winter, Phenol degradation kinetics of an aerobic mixedculture, Biochemical Engineering Journal 46 (2009) 205–209.

[13] G.Y. Wang, Z.P. Wen, G.H. Yu, H.M. Li, Anaerobic biodegradation of phenol byCandida albicans PDY-07 in the presence of 4-chlorophenol, World Journal ofMicrobiology and Biotechnology 24 (2008) 2685–2691.

[14] J. Bai, J.P. Wen, H.M. Li, Y. Jiang, Kinetic modeling of growth and biodegradation ofphenol and m-cresol using Alcaligenes faecalis, Process Biochemistry 42 (2007)510–517.

[15] R.S. Juang, S.Y. Tsai, Growth kinetics of Pseudomonas putida in the biodegradationof single and mixed phenol and sodium salicylate, Biochemical EngineeringJournal 31 (2006) 133–140.

[16] K.M. Khleifat, Biodegradation of phenol by Ewingella Americana: effect of carbonstarvation and some growth condition, Process Biochemistry 41 (2006) 2010–2016.

[17] Kumar, S. Kumar, S. Kumar, Biodegradation kinetics of phenol and catecholusing Pseudomonas putida MTCC, Biochememical Engineering Journal 22 (2005)151–159.

[(Fig._4)TD$FIG]

0

50

100

150

200

250

300

350

0 5 10 15 20 25 30 35 40

Con

cent

ratio

n of

subs

trat

es, p

heno

l/res

orci

nol (

mg/

L)

Degradation time (h)

Exp. phenol 300 mg/LExp. resorcinol 100 mg/LExp phenol 200 mg/LExp. resorcinol 200 mg/LExp. Phenol 100 mg/LExp. Resorcinol 300 mg/LModel

Fig. 4. Effect of degradation time on concentration of substrate, phenol/resorcinol.

Table 4Model equations for substrate degradation dynamics.

Model equations Boundary conditions Equations

dXT

dt¼ mg1 þmg2

� �XT

At t = 0, XT = XTo,

S1 = S1o, S2 = S2o

(39)

dS1

dt¼ �qS1XT (40)

dS2

dt¼ �qS2XT (41)

qS1 ¼ A11 þ A41mg1 þ A31m2g1 (42)

qS2 ¼ A12 þ A42mg2 þ A32m2g2 (43)

mg1 ¼mgmax1S1

KS1 þ S1 þ S21=Ki1

� þ Ia;1S2 þ Ib;1S1S2

(44)

mg2 ¼mgmax2S2

KS2 þ S2 þ S22=Ki2

� þ Ia;2S1 þ Ib;2S1S2

(45)


http://refhub.elsevier.com/S2213-3437(13)00119-X/sbref0005











































[18] Y. Jiang, J.P. Wen, H.M. Li, S.L. Yang, Z.D. Hu, The biodegradation of phenol at highinitial concentration by the yeast Candida tropicalis, Biochemical EngineeringJournal 24 (2005) 243–247.

[19] V. Arutchelvan, V. Kanakasabai, S. Nagarajan, V. Muralikrishnan, Isolation andidentification of novel high strength phenol degrading bacterial strains fromphenol formaldehyde resin manufacturing industrial wastewater, Journal ofHazardous Material B 127 (2005) 238–243.

[20] Z. Alexievaa, M. Gerginova, P. Zlateva, N. Penava, Comparison of growth kineticsand phenol metabolizing enzymes of Trichosporon cutaneum R57 and mutantswith modified degradation abilities, Enzyme and Microbial Technology 34 (2004)242–247.

[21] G. Tchobanoglous, F.L. Burton, H.D. Stensel, Wastewater Engineering TreatmentDisposal Reuse, fourth ed., Metcalf & Eddy Inc., Tata McGraw-Hill PublishingCompany Limited, New Delhi, 2003.

[22] K. Luke, S.G. Burton, A novel application for Neurospora Crassa: progress frombatch culture to a membrane bioreactor for the bioremediation of phenols,Enzyme and Microbial Technology 29 (2001) 348–356.

[23] P. Gerin, M.N. Bellon-Fontaine, M. Asther, P.G. Rouxhet, Immobilization of fungalspores of adhesion, Biotechnology and Bioengineering 47 (1995) 677–687.

[24] Z. Hu, R.A. Ferraina, J.F. Ericson, A.A. Mackay, B.F. Smets, Biomass characteristics inthree sequencing batch reactors treating a wastewater containing syntheticorganic chemicals, Water Research 39 (2005) 710–720.

[25] H.K. Katayama, S. Tobita, K. Hirayama, Biodegradation of phenol and monochlor-ophenols by yeast Rhodotorula Glutinis, Water Science and Technology 30 (1994)59–66.

[26] S.J. Wang, K.C. Loh, Modelling the role of metabolic intermediates in kinetics ofphenol biodegradation, Enzyme Microbial Technology 25 (1999) 177–184.

[27] S. Kar, T. Swaminathan, A. Baradarajan, Biodegradation of phenol and cresolisomer mixtures by Arthrobacter, World Journal of Microbiology and Biotechnol-ogy 13 (1997) 659–663.

[28] D.H. Hutchinson, C.W. Robinson, Kinetics of simultaneous batch degradation of p-cresol and phenol by Pseudomonas putida, Applied Microbiology and Biotechnol-ogy 29 (1988) 599–604.

[29] J.S. Meyer, M.D. Marcus, H.L. Bergman, Inhibitory interaction of aromatic organicsduring microbial degradation, Environmental Toxicology and Chemistry 3 (1984)583–587.

[30] Y. Jiang, J. Wen, J. Bai, X. Jia, Z. Hu, Biodegradation of phenol at high initialconcentration by Alcaligenes faecalis, Journal of Hazardous Material 147 (2007)672–676.

[31] O.J. Hao, M.H. Kim, E.A. Seagren, H. Kim, Kinetics of phenol and chlorophenolutilization by Acinetobacter species, Chemosphere 46 (2002) 797–807.

[32] Banerjee, J.M. Modak, K. Bandopadhyay, D. Das, B.R. Maiti, Mathematicalmodel for evaluation of mass transfer limitations in phenol biodegradationby immobilized Pseudomonas putida, Journal of Biotecnology 87 (2001) 211–223.

[33] G. Minkevich, S.V. Andreyev, V.K. Eroshin, The effect of two inhibiting substrate ongrowth kinetics and cell maintenance of the yeast Candida valida, Process Bio-chemistry 36 (2000) 209–217.

[34] P.J. Allsop, Y. Chisti, M.M. Young, G.R. Sullivan, Dynamics of phenol degradation byPseudomonas putida, Biotechnology and Bioengineering 41 (1993) 572–580.

[35] S. Nagalaxmi, A. Subrahm, G. Ravikanth, Studies on the reduction of conventionalorganic pollutants from paper and pulp effluent using Gliomastix indicus, IndianJournal of Environmental Protection 23 (2003) 785–787.

[36] J.E. Bailey, D.F. Ollis, Biochemical Engineering Fundamentals, second ed.,McGraw-Hill, New York, 1986.

[37] H.S. Fogler, Elements of Chemical Reaction Engineering, third ed., Prentice-Hallof India Private Limited, New Delhi, 1999.

[38] P. Saravanan, K. Pakshirajan, P.K. Saha, Biodegradation of phenol and m-cresol in abatch and fed batch operated internal loop air lift bioreactor by indigenous mixedmicrobial culture predominantly Pseudomonas sp, Bioresource Technology 99(2008) 8553–8558.

[39] D. Arya, S. Kumar, S. Kumar, Biodegradation dynamics and cell maintenance forthe treatment of resorcinol and p-cresol by filamentous fungus Gliomastix indicus,Journal of Hazardous Material 198 (2011) 49–56.

[40] K.F. Reardon, D.C. Mosteller, J.D.B. Rogers, Biodegradation kinetics of benzene,toluene and phenol as single and mixed substrates for Pseudomonas putida F1,Biotechnology and Bioengineering 69 (2000) 385–400.

[41] S.J. Wang, K.C. Loh, Biotransformation kinetics of Pseudomonas putida for co-metabolism of phenol and 4-chlorophenol in the presence of sodium glutamate,Biodegradation 12 (2001) 189–199.

[42] T. Abuhamed, E. Bayrakter, T. Mehmetoglu, U. Mehmetoglu, Kinetic model forgrowth of Pseudomonas putida F1 during benzene toluene and phenol biodegra-dation, Process Biochemistry 39 (2004) 983–988.

[43] H. Yoon, G. Klinzing, H.W. Blanch, Competition for mixed substrates by microbialpopulations, Biotechnology and Bioengineering 19 (1977) 1193–1210.

[44] S.J. Pirt, Maintenance energy: a general modal for energy limited and energylimited growth, Archives of Microbiology 133 (1982) 300–302.

[45] J.V. Littilejohns, A.J. Daugulis, Kinetics and interaction of BTEX compounds duringdegradation by a bacterial consortium, Process Biochemistry 43 (2008) 1068–1076.

[46] G. Wang, J.P. Wen, H. Li, C. Qiu, Biodegradation of phenol and m-cresol by Candidaalbicans PDY-07 under anaerobic condition, Journal of Industrial Microbiologyand Biotechnology 36 (2009) 809–814.

[47] O.M. Neijssel, D.W. Tempest, Bioenergetic aspects of aerobic growth of Klebsiellaaerogenes NCTC 418 in Carbon-limited and carbon sufficient chemostat culture,Archives of Microbiology 107 (1976) 215–221.

[48] W.P. Hempfling, S.E. Mainzer, Effects of varying the carbon source limiting growthyield and maintenance characteristics of Escherichia coli in continuous culture,Journal of Bacteriology 123 (1975) 1076–1087.

[49] P.V. Bodegom, Microbial maintenance: a critical review on its quantification,Microbial Ecology 53 (2007) 513–523.

[50] J. Yan, W. Jianping, B. Jing, W. Daoquan, H. Zongding, Phenol biodegradation by theyeast Candida tropicalis in the presence of m-cresol, Biochemical EngineeringJournal 29 (2006) 227–234.

[51] Z. Aleksieva, D. Ivanova, T. Godjevargova, B. Atanasov, Degradation of somephenol derivatives by Trichosporon cutaneumR57, Process Biochemistry 37(2002) 1215–1219.

[52] Z. Alexieva, H. Yemendzhiev, P. Zlateva, Cresols utilization by Trametes versicolorand substrate interactions in the mixture with phenol, Biodegradation 21 (2010)625–635.

[53] Z. Alexieva, M. Gerginova, J. Manasiev, P. Zlateva, N. Shivarova, A. Krastanov,Phenol and cresol mixture degradation by the yeast Trichosporon cutaneum,Journal of Industrial Microbiology and Biotechnology 35 (2008) 1297–1301.

[54] H. Yemendzhiev, M. Gerginova, P. Zlateva, I. Stoilova, A. Krastanov, Z. Alexieva,Phenol and cresol mixture degradation by Aspergillus awamori strain: biochemicaland kinetic substrate interactions, Proceedings of ECOpole 2 (1) (2008) 153–159.












































































































Final Report

Documents

Transcript of Final Report