|
|
Activated
Carbon Adsorption for Treatment of Agrochemicals in Water
Prof. Motoyuki Suzuki, United Nations University, Japan
ABSTRACT
For control of numerous chemicals found in drinking water sources,
activated carbon adsorption is expected to play an important role.
In order to make a rational design and operation of adsorption process
for removing low concentration organic compounds from water, aqueous
phase adsorption isotherms and adsorption rates of targeted chemicals
are necessary. A generic isotherm relation is presented here and compared
with the experimental results obtained from batch adsorption of eight
agrochemicals on activated carbon fiber. Adsorption rate is another
important parameter needed for adsorption processes. The correlation
of effective surface diffusion coefficients in activated carbon particles
proposed by the authors is introduced here.
INTRODUCTION
Numerous chemicals are used in agricultural activities for controls
of weeds, insects, worms and bacteria. Those chemicals used in open
systems eventually enter hydrospheres and may become annoying source
of contaminants of drinking water. For removing those trace organic
components, such as color, toxic, mutagenic or endocrine disrupting
compounds, from water, activated carbon adsorption is considered to
be one of the powerful methods. As a matter of fact, activated carbon
adsorption alone or together with ozone oxidation treatment plays
an important role in advanced treatment of drinking water in many
cases. For rational design and operation of adsorption processes,
fundamental adsorption properties of organics in concern on activated
carbons are necessary. Numbers of new chemicals, however, are flowing
into practical uses, which often makes a reliable operation of adsorption
processes difficult.
For organic chemicals used in anthropogenic spheres such as agricultural
activities, chemical structures and sometimes major molecular properties
are readily known. Hence, it is desirable if we have a general method
to predict adsorption characteristics of these organic chemicals on
activated carbon in aqueous phase.
The aim of this paper is to review progresses in fundamentals of adsorption
in general and to find a way to make a first approximate estimation
of adsorption characteristics of man-made chemicals of low concentration
often found in water on activated carbon. For evaluation of adsorption
equilibrium, adsorption potential expression is recommended and for
adsorption rate in granular activated carbon, empirical correlations
are intorduced.
FUNDAMENTALS OF ADSORPTION
Behaviors of molecules in adsorption processes are described in terms
of both adsorption equilibrium and adsorption rate. Adsorbable components
in water migrate in bulk phase and diffuse into the porous solid which
has large surface area in contact with the water phase. The extent
of accumulation of a specific molecule on a solid surface is influenced
by many factors involved in the interactions between the solid surface,
water and the species concerned. Adsorption processes for water treatment
can be designed by a chemical engineering principle on the basis of
adsorption equilibrium and rate.
Adsorption Equilibrium Relations in Aqueous Phase In general, the
more hydrophobic is the solute component, the more it is extracted
onto the solid surface. This is true especially on hydrophobic solids
such as carbon surfaces. Hydrophobicity of an organic molecule is
judged from the balance of lipophilic groups such as paraffinic and
aromatic groups and polar groups or hydrophilic groups such as hydroxyl
and carboxyl groups. Solubility of a molecule is one of the parameters
representing hydrophilicity. Also when a molecule is extracted from
the aqueous phase to the surface of an adsorbent, the surface area
of the molecule in contact with the water becomes smaller, which influences
interaction forces between the molecule and the surrounding water.
This effect is quantitized in part by the solvophobic interaction
approach which takes into account Van der Waals energy, static energy,
energy for formation of the cavity for a solute molecule, surface
tension for interface, etc. to explain affinity between a molecule
in water and the surface. This model can predict the tendency of adsorbability
among the homologues but is yet insufficient for quantitative explanation
of generic adsorption isotherms from aqueous solution.1)
Table 1 Typical isotherm equations for aqueous phase
For dependency of the equilibrium adsorbed amount on the concentration,
adsorption isotherm, many equations based on different concepts have
been proposed. The simplest equations are the Langmuir equation and
the Freundlich equation, each of which involves two parameters. Langmuir
equation is derived from adsorption and desorption kinetics on a uniform
surface.
In Equation (1) of Table 1,  represents
the saturated capacity of adsorption and b corresponds to the
strength of adsorption. /b
represents the Henry constant, K, of adsorption at the low coverage
(q=Kc).
The Freundlich equation is more empirical and is frequently used in
correlating aqueous phase experimental data. It is difficult to derive
Freundlich equation rigorously from a theoretical model, but assumption
of exponential shape energetic distribution on the surface will lead
after further simplification to the Freundlich equation. This equation
is not thermodynamically sound but well fits the experimental data
in a limited range. There are several equations proposed to combine
the Langmuir equation and the Freundlich equation, which attempt to
keep thermodynamic soundness and yet correlate the data well. The
Radke-Prausnitz equation2) and the Toth equation3) are the typical
of these equations. The Toth equation, which has three parameters
to be empirically determined, is shown in Table 1.
A more versatile concept seems to be given by the adsorption potential
model.4) Adsorption potential, A, is defined as an excess energy above
the heat of condensation (vaporization) corresponding to a concentration,
c, and is given as:
 (5)
represents the solubility of the component in concern. Then the model
assumes that adsorbed amounts of similar components at a given adsorption
potential are the same in volume. Functional form of dependence of
adsorbed amount on adsorption potential is considered to depend on
the micropore characteristics.
As a generalized expression, a Dubinin-Astakhov type equation5) is
given as shown in Table 1. E is the characteristic energy of
adsorption and n is the parameter given for each adsorption
system.6)
represents the saturation capacity of adsorption determined for each
adsorbent. The values of n and E depend on the affinity between
the solute and the surface, obviously determining the shape of each
isotherm (Figure 1).
The Freundlich isotherm, 
corresponds to the case of n = 1 in the DA equation. In this
case adsorption is considered to take place on an energetically heterogeneous
surface. Freundlich constants k and  are
expressed in terms of the parameters in the DA equation as:  (6)
When adsorption takes place in micropores, adsorbate molecules are
in the potential field determined by the surrounding walls. In this
case n = 2 may be applied. Similarly, on the adsorbent with
molecular sieving pores, n = 3 is used in the case of gas phase
adsorption.
Experimental data of adsorption isotherms obtained for agrochemicals
by the author are plotted against reduced concentration,  ,
in Figure 2. Solubilities of the adsorbates are taken from the literature
or estimated by BlogP method. Table 2 shows the molecular characteristic
parameters of the adsorbates calculated by BlogP method. Order of
adsorbability can be correlated well with the molecular density calculated
by dividing molecular volume by molecular weight. This appears to
suggest that the solvophobic effect is dominating the adsorption of
these chemicals.
Multicomponent adsorption isotherms have attracted the attention of
many scientists. For bicomponent isotherms of resembling organics,
ideal adsorbed solution (IAS) theory7) gives the most reliable prediction
from the single-component isotherm of each component. In the case
of water treatment, however, the combination of components of wide
variety makes prediction difficult.
TABLE 2 Molecular parameters of agrochemicals used for adsorption
equilibrium study
| Agrochemicals |
MW
|
Volume
|
Surface |
BlogW |
BlogP
|
MW/VOL |
| |
Da
|
ﾅ3 |
ﾅ2
|
|
|
Da/ﾅ3 |
| Benthiocarb |
257.5 |
229.69 |
288.56 |
-2.23 |
3.341 |
1.121 |
| Isoprothiolane |
290 |
252.34 |
318.85 |
-2.809 |
3.125 |
1.149 |
| Flutranil |
323 |
274.12 |
332.48 |
-2.85 |
2.689 |
1.178 |
|
Propyzamide |
256 |
214.31 |
265.02 |
-2.936 |
3.277 |
1.195 |
| Dichlorvos
(DDVP) |
190 |
153.55 |
202.93 |
178.23 |
-4.538 |
1.237 |
| Tolclofos-methyl
|
301 |
223.59 |
281.88 |
34.155 |
6.511 |
1.346 |
|
Chlorothalonil (TPN) |
266 |
171.76 |
208.48 |
-5.208 |
4.897 |
1.549 |
| Chloropyrifos
|
350.5 |
249.25 |
310.96 |
35.249 |
5.181 |
1.406 |
Adsorption Rate
Since molecular diffusivities in liquid phases are far smaller than
those in gaseous phases, mass transfer steps can become rate limiting
in most adsorption processes. In the case of water treatment by means
of packed beds of granular adsorbents such as activated carbons, intraparticle
diffusion is an important factor in determining uptake rates and the
overall behavior of the adsorption bed.
Intraparticle diffusivities in activated carbons are composed mainly
of two components. One is the transport of adsorbate molecules through
the liquid phase in the pores. This is easily defined if the porosity
and the pore tortuosity of the particle are given. Pore diffusion
is described by means of concentration gradient driving force. The
other transport mechanism is the migration of adsorbed molecules on
the surface of solid. This is called surface diffusion and the flux
is described in terms of surface diffusion coefficient defined by
using a gradient of the amount adsorbed as the driving force of diffusion.
These diffusion mechanisms take place in parallel; thus the total
flux of diffusion is expressed as the sum of these mechanisms.
 (7)
The pore diffusivity, Dp, is essentially in proportion to molecular
diffusivity which slightly changes with temperature and molecular
weight. For polar molecules such as dodecylbenzenesulfonate, pore
diffusion kinetics becomes dominant.8) In this case pore diffusivities
for different activated carbons fell close to each other.
 (8)
where e represents the porosity and k2 denotes the tortuosity of the
pore in the particle.
The effective surface diffusion coefficient, Ds, depends much on temperature
and volatility of the adsorbate (Figure 3).9) Thus surface diffusion
may become dominant in the case of low boiling point volatile molecules.
Change of surface diffusion coefficient with molecular weight was
determined for poly(oxyethylene)s and the effective surface diffusion
coefficient was found to be inversely proportional to the square root
of the molecular weight (Figure 4).10) Also, since surface diffusion
depends on the mobility of a molecule on the surface, surface diffusion
coefficient is expected to be a function of the amount adsorbed when
the adsorption surface is not uniform. Adsorption takes place from
strong sites to weaker sites and with increase in the amount adsorbed,
mobility of the molecules becomes higher.11)
CONCLUSION
Adsorption processes have more possibilities to contribute to the
removal of trace contaminants and recovery of reusable components.
Based on the progress in studies on the fundamentals of adsorption,
improvement of adsorbents and adsorption processes is becoming of
greater importance. There are also needs for more research in adsorption
phenomena involved in material accumulation and transport in natural
water environments. Generalized treatment of equilibrium relations
is desirable in order to describe aqueous phase partitioning for a
variety of combinations of solids and adsorbates.
REFERENCES
1. Belfort, G.: AIChE Journal, 30, 197 (1984).
2. Radke, C. J. and J. M. Prausnitz: Ind. Eng.Chem., Fundamentals,
11, 445 (1972).
3. Toth, J.: Acta Chim. Acad. Sci. Hungary, 69, 311 (1971).
4. Polnyi, M.: Verh. Deut. Chem., 57, 106 (1914).
5. Dubinin, M. M. and V. A. Astakhov: 2nd Int. Conf. On Molecular-Sieving
Zeolites (1970).
6. Kawazoe, K. and T. Kawai: Seisan Kenkyu, 22, 491 (1970) (in Japanese).
7. Myers, A. L. and J. M. Prausnitz: AIChE Journal, 11, 121 (1965).
8. Suzuki, M. and K. Kawazoe: J. Chem. Eng. Japan, 7, 346 (1974).
9. Suzuki, M. and K. Kawazoe: J. Chem. Eng. Japan, 8, 381 (1975).
10. Suzuki, M., T. Kawai and K. Kawazoe: J. Chem. Eng. Japan, 8, 203
(1975).
11. Suzuki, M. and T. Fujii: AIChE Journal, 28, 380 (1982).
|
|
