Carbon 40 (2002) 2861–2869
E stimation of activated carbon adsorption efficiency for organic
vapours
I. A strategy for selecting test compounds
a, a a
*
¨¨¨
Ingrid E. Fangmark , Lars-Gunnar Hammarstrom , Marianne E. Stromqvist ,
bb b
Amanda L. Ness , Paul R. Norman , Neale M. Osmond
a
˚
Swedish Defence Research Agency
,
FOI
,
SE-
901 82
Umea
,
Sweden
b
Dstl
,
Chemical and Biological Systems
,
Porton Down
,
Salisbury
,
UK
Received 3 January 2002; accepted 22 June 2002
Abstract
Strategies for developing quantitative structure–affinity relationships (QSAfR) for the prediction of break-through
performance of 31 chlorinated hydrocarbons on activated carbon have been studied. Two different approaches for the
selection of a limited set of compounds for modelling were evaluated through the predictive power of the resulting QSAfR
models. When the model was based on a training-set selected without a rational strategy, the developed QSAfR model
showed poor predictive performance. Accordingly, such models have a limited capability to produce information concerning
the important adsorbate related parameters influencing adsorption. By using a strategy where multivariate data analytical
techniques are used in conjunction with statistical experimental design to select a balanced set of compounds for
break-through performance evaluation, it was possible to develop QSAfR models with high predictive capability.
2002 Elsevier Science Ltd. All rights reserved.
Keywords
:
A. Activated carbon; C. Adsorption; modeling; D. Adsorption properties
1 . Introduction savings in development of adsorbents for specific applica-
tions. For such preliminary evaluation, predictive methods
Activated carbon is used in filters for air purification that which need a limited experimental data input are advan-
range from large air pollution control devices to small tageous.
respiratory filters. It is one of the most versatile adsorbents Quantitative-structure-activity-relationships (QSAR)
known with high removal efficiency, low costs, reusability have been successfully applied in drug design and toxicol-
and possible product recovery. ogy to model physiological activity based on chemical
Key performance parameters for consideration in the structure and physical properties. These relationships,
design or selection of an activated carbon adsorbent bed which basically rely on receptor theory, may also be
are its adsorption capacity for different vapours and the applied for modelling carbon adsorption affinity. Quantita-
associated break-through time or effluent profile. In view tive structure–affinity relationships (QSAfR) could, thus,
of the time consuming nature of the experimental protocols serve as a tool to estimate unknown performance data of
used for the evaluation of filter efficiency and the vast carbon filters from the physico–chemical properties of the
number of chemicals produced world-wide, empirical adsorbate. Earlier a QSAfR model was reported by Nir-
models to predict carbon adsorption performance are malakhandan and Speece [1]. They evaluated two different
advantageous. Models that allow fast and simple screening QSAfR approaches; the molecular connectivity index and
of carbon filter performance would lead to time and cost the linear solvation energy relationship, as adsorbate
descriptors in multiple regression models. The best fit
between experimental data of 17 different compounds,
*
Corresponding author. Tel.: 146-90-106-642; fax: 146-90-
from two literature sources, and predicted values were
106-800.
¨
E-mail address
:
ingrid.fangmark@foi.se (I.E. Fangmark).
obtained when the molecular connectivity index was used
0008-6223/02/$ – see front matter 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6223(02)00222-1