QSAR models for the removal of organic micropollutants in four different river
water matrices
Sairam Sudhakaran
a,
⇑
, James Calvin
a,b
, Gary L. Amy
a
a
King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
b
Texas A&M University, College Station, TX, USA
article info
Article history:
Received 22 September 2011
Received in revised form 3 December 2011
Accepted 5 December 2011
Available online 14 January 2012
Keywords:
PPCPs
QSAR
Advanced oxidation process (AOP)
Molecular descriptors
Quantum-chemical
Validation
abstract
Ozonation is an advanced water treatment process used to remove organic micropollutants (OMPs) such
as pharmaceuticals and personal care products (PPCPs). In this study, Quantitative Structure Activity
Relationship (QSAR) models, for ozonation and advanced oxidation process (AOP), were developed with
percent-removal of OMPs by ozonation as the criterion variable. The models focused on PPCPs and pes-
ticides elimination in bench-scale studies done within natural water matrices: Colorado River, Passaic
River, Ohio River and Suwannee synthetic water. The OMPs removal for the different water matrices var-
ied depending on the water quality conditions such as pH, DOC, alkalinity. The molecular descriptors
used to define the OMPs physico-chemical properties range from one-dimensional (atom counts) to
three-dimensional (quantum-chemical). Based on a statistical modeling approach using more than 40
molecular descriptors as predictors, descriptors influencing ozonation/AOP were chosen for inclusion
in the QSAR models. The modeling approach was based on multiple linear regression (MLR). Also, a global
model based on neural networks was created, compiling OMPs from all the four river water matrices. The
chemically relevant molecular descriptors involved in the QSAR models were: energy difference between
lowest unoccupied and highest occupied molecular orbital (E
LUMO
–E
HOMO
), electron-affinity (EA), number
of halogen atoms (#X), number of ring atoms (#ring atoms), weakly polar component of the solvent
accessible surface area (WPSA) and oxygen to carbon ratio (O/C). All the QSAR models resulted in a good-
ness-of-fit, R
2
, greater than 0.8. Internal and external validations were performed on the models.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Pharmaceuticals and personal care products (PPCPs) are gaining
importance as a class of organic micropollutants (OMPs) (Boyd
et al., 2003; Richardson et al., 2005; Carballa et al., 2007; Lapen
et al., 2008; Peng et al., 2008; Rahman et al., 2009; Kosma et al.,
2010; McClellan and Halden, 2010; Yoon et al., 2010). There are sev-
eral candidate water treatment processes (adsorption, membrane
separation and river bank filtration) to eliminate the micropollu-
tants. Adsorption processes are less efficient with polar compounds.
Reverse Osmosis (RO) is a very efficient water treatment process
but there are problems of brine solution and expense. River bank
filtration, a sustainable process governed by biodegradation, is less
efficient with respect to non-biodegradable compounds (persistent
organic micropollutants) such as primidone and atrazine. Ozona-
tion is a good option to remove OMPs since ozonation exhibits
selectivity towards certain organic compounds and easily trans-
forms them (Von Gunten, 2003). Advanced oxidation processes
(AOPs), exploiting hydroxyl radical (OH
) oxidation, are generally
considered less selective and hence may oxidize a wider range of
compounds.
The PPCPs are present in water at nano-grams/liter (ng L
1
)
levels. The analytical methods used to detect them are compli-
cated, expensive, and in certain cases, time-consuming. Predictive
models, Quantitative Structure Activity/Property Relationship
(QSAR/QSPR) models, are a rapid and cost-effective alternative to
experimental evaluation. The number of QSAR articles published
in the in water-related sciences is constantly increasing. QSAR
models are recognized by government regulatory bodies as a
method to screen toxic chemicals. Biowin, a software tool that pre-
dicts the biodegradability of toxic compounds in water, has its ba-
sis in QSAR models. With the increasing growth of reliable
software, it has become relatively easy to compute the important
properties related to micropollutants. QSAR models are also used
to study reaction mechanisms and degradation pathways of micro-
pollutants (Sabljic, 2001). QSAR models use relevant molecular
physico-chemical properties (molecular descriptors) to predict
important treatment responses (e.g., rate constants) (Kusic et al.,
2009) which can serve as indices for water treatment process
selection and performance assessment. Models have also been
0045-6535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2011.12.006
⇑
Corresponding author. Tel.: +966 565771055/2 808 2881.
E-mail address: sairam.sudhakaran@kaust.edu.sa (S. Sudhakaran).
Chemosphere 87 (2012) 144–150
Contents lists available at SciVerse ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere