Environ. Sci. Technol. 1993, 27, 2079-2085
Oxidative Coupling of Phenols on Activated Carbon: Impact on Adsorption Equilibrium Radisav D. Vidic 943 Benedum Hall, University of Pittsburgh, Pittsburgh, Pennsylvania 1526 1
Makram T. Suldan'
University of Cincinnati, Cincinnati, Ohio 4522 1-007 1 Richard C. Brenner Risk Reduction Engineering Laboratory, U S . EPA, Cincinnati, Ohio 45268
Previously reported results by the authors revealed that the presence of molecular oxygen (oxic conditions) in the test environment can, in some instances, cause up to a 3-fold increase in the adsorptive capacity of granular activated carbon (GAC) for phenolic compounds. It was discovered that these compounds undergo oxidative coupling on the carbon surface under oxic conditions. The polymers formed as a result of these chemical reactions are very difficult to desorb from the surface of GAC. This led to significant irreversible adsorption in the presence of molecular oxygen. On the other hand, when the same compounds are adsorbed on the carbon surface under anoxic conditions, essentially all of the adsorbate can be recovered from the carbon surface by solvent extraction. The ionized species of phenolic compounds showed even higher susceptibility toward polymerization on the surface of GAC than the parent neutral molecules. GAC particle size did not influence the extent of polymerization. Oxygen uptake measurements revealed significant consumption of molecular oxygen during the adsorption of phenolic compounds. The amount of molecular oxygen consumed in these experiments was found to be linearly proportional to the amount of irreversibly adsorbed compound.
Introduction
The presence of low concentrations ofvarious refractory compounds can be a major obstacle to the use and reuse of water streams. Phenolic compounds can cause objectionable taste and odor problems in drinking water and can exert adverse effects on various biological treatment processes. Some phenolic compounds originate from natural sources, while others are manufactured. The urine of some animals and decay of vegetation release phenol to water bodies. Several industrial sources such as coal gasification sites, coke ovens, oil refineries, town gas sites, and petrochemical units generate large quantities of phenolic compounds that can, if improperly managed, cause long-term contamination of both surface water and groundwater bodies. Treatment methods for removing phenolic compounds that have been investigated in the past include both aerobic (1)and anaerobic biodegradation (2),chemical oxidation with ozone ( 3 , 4 ) ,adsorption by ion exchange resin (5,6), and activated carbon (7-14). One of the key parameters in designing the activated carbon adsorption process for the removal of organic ~~~
* Author to whom all correspondence should be addressed. 0013-936X/93/0927-2079$04.00/0
0 1993 American Chemical Society
compounds is the capacity of carbon for the retention of the compounds of interest. The experimental protocol widely used for obtaining adsorption equilibrium data is the bottle point technique. Due to the lack of a unified procedure for conducting this test, many different adsorption isotherms for the same adsorbent-adsorbate pair can be found in the literature (12). Recent studies by Vidic e t al. (15)and Vidic and Suidan (16) revealed that the adsorptive capacity of granular activated carbon (GAC) for several phenolic compounds is highly influenced by the presence of molecular oxygen (oxic conditions). The GAC adsorptive capacity for 2-methylphenol attainable under oxic conditions was as much as 200% above that obtained in the absence of molecular oxygen. A similar phenomenon was demonstrated for the adsorption of phenol, 2-chlorophenol, and 3-ethylphenol as well as natural organic matter (16). Furthermore, extraction of the carbon preloaded with 2-methylphenol revealed that almost 100%of the originally adsorbed compound can be recovered from the surface of GAC by solvent extraction if the adsorption was carried out in the absence of molecular oxygen. Conversely, only 10-30% of the adsorbed compound was recovered from GAC loaded under oxic conditions (16). This study was designed to further evaluate the role of molecular oxygen on the adsorption of phenolic compounds by activated carbon and to provide possible explanations for the observed phenomena. Theoretical Background
GC/MS analyses performed on the extracts from the GAC used in the oxic isotherm tests with 2-methylphenol revealed the presence of appreciable amounts of dimers, trimers, and even tetramers of 2-methylphenol. Therefore, a reasonable conclusion is that the increase in the GAC adsorptive capacity under oxic conditions is a result of adsorbate polymerization on the carbon surface. It is generally believed that the mechanism of phenolic compounds coupling involves radicals of these compounds (17, 18),although some researchers have proposed that nonradical oxidative coupling of phenols can also occur (19). The first step in the oxidative coupling of phenols is usually the formation of phenoxy radicals from the phenol molecule or phenolate ion. The radical formation is generally initiated by the removal of a hydrogen atom from each phenolic molecule to form phenoxy radicals. These radicals can then participate in direct coupling with other radicals, homolytic aromatic substitution with phenol molecules, or heterolytic coupling with phenolate ions to form dimers (19). Electron localization in the radicals Environ. Sci. Technol., VoI. 27, No. 10, 1993 2078
Table I. Adsorbates Used in the Study aqueous solubility at 25 "C (g/W
boiling point
("0'
compound phenol 2-methylphenol 3-methylphenol 4-methylphenol 2,4-dimethylphenol 2-ethylphenol 3-ethylphenol 4-ethylphenol 2-chlorophenol 3-chlorophenol 4-chlorophenol 2,4-dichlorophenol 2,4,6-trichlorophenol 2-hydrohybenzoicacid 3-hydrohybenzoicacid 4-hydrohybenzoicacid 2-nitrophenol 3-nitrophenol 4-nitrophenol aniline p-anisidine
182 191 202 202 211 207f 21g 218 204 214 220 210 246 211 NAh NA 215 194 279 184 246
pKab 9.89 10.20 10.01 10.17 10.58 10.Ze 10.07e 10.Oe 8.528 8.978 9.378 7.908 5.998 2.97; 13.40 4.06; 9.92 4.48; 9.32 7.17 8.28 7.15 NA NA
93 25 26 23 NA sparingly so1.d slightly s01.f slightly 501.1 28 26 27 4.5d 28.6d 2e 8e 2e
2.0 1.35 1.69 35 sparingly so1.d
*
a Ref 31. Ref 32. Ref 33. Ref 34. e Ref 35. f Ref 36. Ref 28. NA = not available.
determines the coupling position (ortho or para position to hydroxyl group). Coupling is predominantly achieved through carbon-carbon bonding and less frequently through carbon-oxygen bonding (17). Several studies have shown that molecular oxygen can act as an initiator in oxidative coupling reactions of phenols (20). Molecular oxygen can react directly with phenol (21):
+ 0,
PhOH
-
PhO'
+ HO,'
(1)
In addition, the phenolate ion can also react with oxygen:
+ H+ PhO' + 0,'
PhOH e PHO-
-
+ 0, 0; + H+
PhO-
PhOH + HO,'
-
-
PHO'
(2)
HO,'
The radicals formed through eqs 1 and 2 can react with another phenol molecule: PhO'
+ H,O,
(3) Hydrogen peroxide reacts with another phenol molecule according to the following equations: PhOH + H,O, PhOH
+ HO'
-
+ H,O + HO' PhO' + H,O
(4)
The above reactions were demonstrated to take place at elevated temperatures (180-210 OC) and pressures (35 atm) (21), indicating a high activation energy of radical formation while Hay et al. (22)reported oxidative coupling to take place at room temperature in the presence of copper(1)salt. The results of this study indicate that such reactions are also feasible at room temperatures with the GAC surface acting as a catalyst. 2080
Envlron. Scl. Technol., Vol. 27, No. IO, 1993
Materials and Methods All adsorption experiments were performed at pH 7.0 using autoclaved Milli-Q water (deionized water passed through a Millipore purification system, Millipore Corp., Bedford, MA) to prepare the adsorbate solutions. Water was buffered with 0.01 M phosphate buffer, and the pH was adjusted with a 10 M NaOH solution. The adsorbent used in this study was Filtrasorb 400 GAC (CalgonCarbon Corp., Pittsburgh, PA). Prior to use, carbon was thoroughly washed with Milli-Q water, dried at 105 OC, and stored in a desiccator until use. All experiments described in this study were performed using the same batch of GAC. Adsorbates used in this study are listed in Table I. All the adsorbates were reagent grade or better. Adsorbate concentration measurements were performed on a Hewlett Packard (HP) 8452 diode-array spectrophotometer (Hewlett-Packard Co., Palo Alto, CA) using both 1- and 5-cm quartz cells. The extracts from the GAC used in the adsorption isotherm tests were analyzed using an HP 5890A gas chromatograph equipped with a DB-1 30-m fused-silicacapillary column (J&WScientific,Folsom, CA) and a flame ionization detector (FID). Gas chromatography/mass spectroscopy (GC/MS) analyses were performed on an HP 5985A GUMS using the electron-impact positive-ion mode. The gas chromatograph was equipped with a DB-1 30-m fused silica capillary column (J&W Scientific, Folsom, CA). The oven temperature was programmed from 40 to 280 OC at 10 OC/min with a 5 min hold at 40 OC. Adsorption Equilibrium. A detailed description of the experimental protocol used to evaluate GAC adsorptive capacity in the presence and absence of molecular oxygen is provided elsewhere (16) and will not be repeated here. All isotherm tests were performed at room temperature, which varied in a very narrow range of 21 f 1 OC. GAC Extraction. Solvent extraction of GAC samples preloaded with organic compounds in the presence and absence of molecular oxygen was performed using methanol and methylene chloride according to the procedure described by Vidic and Suidan (16). Oxygen Uptake Measurements. The amount of molecular oxygen consumed during the adsorption of organic compounds was measured using a computerized respirometer (N-CON Comput-OX, Model WB512, NCON Systems Co., Inc., Larchmont, NY). 16 X 20 mesh size GAC was first wetted in air saturated buffered water with a DO concentration of 8.5 mg/L. This was done to minimize oxygen uptake due to GAC adsorption even though the virgin GAC adsorbed only 4.2 mg of Oz/g of GAC at equilibrium dissolved oxygen concentration of 8.5 mg/L. GAC was then transferred together with the buffered water to a 500-mL respirometer bottle where it was complemented to a total volume of 400 mL with airsaturated buffered water that contained known concentrations of the organic compound to be tested. The bottle was then connected to the oxygen supply system that maintains a constant partial pressure of oxygen in the bottle headspace. These experiments were carried out for a period of at least 2 weeks, which was the equilibration period for most of the adsorption isotherm tests. All oxygen uptake measurements were performed at 21 OC.
Results and Discussion Results of the oxic and anoxic adsorption isotherm experiments with 2-, 3-,and 4-methylphenol are presented
I
'7
lTreV rn A
2-Methylphenol 3-Methylphenol 4-Methylphenol
;
102
U
k
t
o 2-Methylphenol o
10'
3-Methylphenol 4-Methylphenol
.
1
10'
1
100
1 10'
108
io3
C , mg/L
Flgure 3. Reversible and irreversible adsorption of three methylphenols.
x
w
. 20
. -
. * *
2-Methylphenol 3-Methylphenol a 8 t 4-Methylphenol o - " " " ' " " " " " ' " ' ' ' ' o
0
$ 0
OXlC
.
in Figure 1. All the adsorption isotherms for these compounds were modeled using the Freundlich isotherm equation. It is important to note that the coefficient of correlation was in all six cases above 0.98, which is why this adsorption isotherm model was selected to approximate the experimental data. Extraction efficiency obtained for the GAC used in the oxic and anoxic adsorption isotherm tests with these compounds are shown in Figure 2. As is apparent in Figure 1, the position at which the methyl group is substituted on the parent phenol molecule had very little influence on the anoxic GAC adsorptive capacity for these substituents since there is no statistically significant difference between the three anoxic adsorption isotherms. Furthermore, Figure 2 shows that more than 90 % of each of the compounds adsorbed on the carbon in the absence of molecular oxygen was recovered by extraction with methanol and methylene chloride. I t is hypothesized that extraction efficiencywas less than 100% due to the fact that even the adsorption isotherm bottles prepared according to the anoxic procedure contained an average residual of 0.7 mg/L of dissolved oxygen, which could have induced some polymerization of these compounds and, consequently, irreversible adsorption. It is, therefore, reasonable to believe that physisorption is the primary adsorption mechanism under anoxic conditions. Another important conclusion from Figure 1is that the substitutional position of the methyl group had a significant impact on the adsorptive properties of activated carbon in the presence of molecular oxygen. The highest relative increase in the GAC adsorptive capacity due to polymerization on the carbon surface under oxic conditions
was observed for the adsorption of 2-methylphenol. Consequently, the extraction efficiency achieved for 2-methylphenol was the lowest among the three methylphenols (Figure 2). Figure 1alsoshows that the presence of molecular oxygen had the least impact on the GAC adsorptive capacity for 3-methylphenol, which was the compound that showed the highest recovery among the three methylphenols (Figure 2). Such behavior is not unexpected since the compounds that have functional groups substituted on the meta position are thermodynamically more stable. This greater stability is due to the strengthening of the bond holding the hydrogen atom to oxygen with meta-directing groups contrasted to the weakening of that bond with ortho- and para-directing groups (23). Latest measurements of the homolytic bond dissociation energies (BDE) for the 0-H bonds for phenols by Bordwell and Cheng (24)revealed that the BDE values for 2-, 3-, and 4-methylphenol were 88.2, 89.4, and 88.7 kcal, respectively. Consequently, the oxidative coupling (polymer formation) of 3-methylphenol occurs to a much lesser extent than the coupling of 2- or 4-methylphenol. Data on adsorbate recovery and oxic GAC adsorptive capacity were used to calculate the amount of reversibly and irreversibly adsorbed compound (Figure 3). It is important to note that only the amount of adsorbate measured in the extract from the GAC used in the isotherm experiments was denoted as reversibly adsorbed. The irreversible uptake included both the amount of adsorbate and reaction products still remaining on the carbon as well as some of the reaction products that were desorbed in methanol. This assumption is justified because the amounts of the reaction products detected in the extracts were very small and weight measurements of the extracted carbons agreed very well with calculated values. Figure 3 depicts the fact that the amount of irreversible uptake remained practically constant over 3 orders of magnitude of the equilibrium aqueous-phase concentrations for each of the three methylphenols. This suggests that the capacity of activated carbon to promote polymerization reactions and, consequently, the adsorptive capacity for the products of these reactions are constant. The behavior observed for the adsorption and extraction of three ethylphenols and three chlorophenolswas identical to that documented above in the case of three methylphenols. On the other hand, three nitrophenols showed markedly different behavior than the above mentioned organic compounds. Oxic and anoxic adsorption isotherms for 2-, 3-, and 4-nitrophenol are given in Figure 4. It is apparent from this figure that the substitutional position Environ. Sci. Technol., Vol. 27, No. 10, 1993 2081
1
2.5 -
ie
I
o e I
2-Methylphenol 2-Ethylphenol 2-Chlorophenol
-.---
---
20-
0 _.:'
0
\ m
1 10'
'
Ox''
n
anoxic
o anoxic
L-
,
,
,
, , , ,
,
,
,
,
4-Nitrophenol 3-Nitrophenol , , , ,
, ,;,:_..,
10'
100
2-Nitrophenol
102
,I
u 10
I
133 19
'
'
5
"
'
'
"
"
10
"
'
15
'
20534
~
~
,
20
25
Time, d a y s
Figure 5. Rate of oxygen uptake during the adsorption of 2-methylphenol.
of the NO2 functionality not only had very little influence on the exhibited GAC adsorptive capacity in the absence of molecular oxygen, but also the presence of molecular oxygen in the test environment had no effect on the GAG adsorptive capacity for these adsorbates. Close to 100% of the compound loaded on the GAG surface during both oxic and anoxic adsorption isotherm experiments was recovered by extraction with methanol and methylene chloride, indicating the absence of irreversible uptake under oxic conditions. Lim et al. ( 2 5 ) calculated that a high activation energy of 12.5Kcal/mol is required for the oxic coupling of phenol in an alkaline media and in the presence of a strong catalyst such as cuprous oxide. Therefore, the adsorption behavior of nitrophenols can be explained by the fact that activated carbon is a much weaker catalyst, that the adsorption experiments were conducted in a neutral medium, and that the addition of electron-withdrawing functional groups like NO2 increases the potential required to oxidize the organic compound (17 , 2 4 , 2 6 ) . Among the 35 phenolic compounds tested by Bordwell and Cheng ( 2 4 ) , nitrophenols had the highest bond dissociation energy for the 0-H bond. Oxygen uptake measurements conducted using an aerobic respirometer were always performed on triplicate samples to obtain statistically significant results. The oxygen consumption rate measured for the adsorption system with an initial 2-methylphenol concentration of 1957.7 mg/L and 2.0 g of 16 X 20 mesh size GAG at 21 O G is depicted in Figure 5. The total amount of oxygen consumed during the equilibration period ranged from 2082
Environ. Sci. Technoi., Vol. 27, No. 10, 1993
1
2
~-
i~~
4
Irreversible A d s o r p i l a n mmol
Figure6. Stoichiometry of polymerization reactions for 2-methylpheno1, 2-ethylphenol, and 2-chlorophenol.
82.8 to 88.0 mg. On the basis of the equilibrium liquidphase concentration measurements and the data presented in Figure 3, the corresponding amount of irreversibly adsorbed 2-methylphenol ranged from 532.2 to 535.8 mg. The results of this test together with several other tests involving different masses of GAG and different initial 2-methylphenol concentrations are summarized in Figure 6. The results of similar experiments performed with 2-ethylphenol and 2-chlorophenolare also shown in Figure 6. It is apparent in Figure 6 that the amount of oxygen consumed during the adsorption of phenolic compounds is linearly proportional to the amount of irreversibly adsorbed compound. The coefficient of proportionality was 0.50 (h0.017) for 2-methylphenol, 0.65 (h0.025) for 2-ethylphenol, and0.42 (4~0.008)for 2-chlorophenol(values in parentheses represent standard deviations obtained by the least-squares linear regression analysis). Equations 1-4 suggest that the proportionality coefficient should be 0.25. However, this value pertains only to the first step of oxidative coupling (radical formation) of phenol in aqueous solution. The system employed in this study was more complex and involved oxidative coupling of substituted phenols. Furthermore, some of the hydrogen peroxide formed through reactions 1-4 could be consumed by the reaction with activated carbon rather than phenol molecules. All the organic compounds tested for the effect of molecular oxygen on the adsorptive capacity of GAC belong to different classes of organic compounds and differ in their affinity toward oxidative coupling. In most cases, the relative affinity of different adsorbates toward polymerization on the surface of GAG can be related to their tendencies to undergo oxidative coupling as solutes. Reactivity of these compounds in oxidative coupling reactions can be characterized using the notion of "critical oxidation potential" (COP) as defined by Fieser (27).Fieser (27) tested several phenols and amines with oxidizing agents of known redox potentials to determine the threshold oxidation potential required to oxidize each compound in a water-alcohol mixture at 35 O G . COP was used in this study to determine the relative order of susceptibility of organic compoundstoward polymerization on the surface of GAG. Figure 7 shows the amount of irreversibly adsorbed compound as a function of COP. The COP values for most of the compounds presented in Figure 10 are as measured by Fieser (29, while several
m! "
. . -
i---.--L.-_iL.-i-p-__l.-
0
103
Figure 4. Adsorption isotherms for nitrophenols.
"
i
-
I
C , mg/L
0' 0
.''
3.5
e p-anistdinr M
3.0
A
-
: 2.5
1
c
:
ew
:
2.4-dmp 0
; : : : : : t: "p
e 2.0 . --
1
2
-2 m
1.5
e o-chlorophenol
'
o-chlorophen~l~
e
OpH=10
emhe
6-tcp
p-chlorophenol
:
~--CXSOI
.
e
2 4-dichlaraphenol
:
;
1 0 -
>
. phenol
os2-hydroxybenzoic
t 0 0 '
4 - h y d r o x y b e n z o ~ c acid "
'
"
1
'
1
'
"
-
acid
p-nltrophenol: '
S
'
1s
10'
2;Me~hy~p~en;l~ 2-Ethylphenol 2-Propylphenol
1 10'
100
102
103
C , mg/L
Flgure 8. Anoxic adsorption Isotherms for alkyl-substituted phenols.
COP values were calculated using the approximate values that Fieser established for substituents. Among the compounds investigated, the nitro group caused the highest increase in COP, which can explain the observed lack of polymerization of nitrophenols on the GAC surface. The other functionality that exhibits negative inductive effects (23),which significantly increase COP, is the carboxyl group. However, the magnitude of the increase is lower than that measured for the nitro group. Since 2- and 4-hydroxybenzoic acids polymerized on the carbon surface, it can be concluded that the threshold potential encountered in the adsorption system is somewhere between the COP of the nitro- and the carboxylsubstituted phenols. Another important observation of Fieser (27) is that the multiple substitution of the same functional group further emphasizes the change in COP in the same direction. Methyl functional group decreases COP, and consequently, the amount of irreversible adsorption for 2,4-dimethylphenol is above that observed for 2-methylphenol or 4-methylphenol (Figure 7). Similar behavior was demonstrated for the adsorption of 2-chlorophenol and 2,4-chlorophenol but in the opposite direction of the relationship observed for the methylphenols since the chlorine functionality belongs to the group of substitutes that increase COP. However, adsorption behavior of 2,4,6trichlorophenol does not follow the trend observed for 2-chlorophenol and 2,4-dichlorophenol. This can be attributed to the fact that the value of the dissociation constant, PKa, is 7.9 for 2,4-dichlorophenol and 5.99 for 2,4,6-trichlorophenol (28), indicating that most of 2,4dichlorophenol was present in the neutral form at pH 7.0 while the predominant species for 2,4,6-trichlorophenol was the phenolate ion. Denisov (20)and Shibaeva et al. (21)found that the reaction of phenolic anions with oxygen is much faster than the reaction of the corresponding phenols. Furthermore, Chin et al. (29)found the rate of aerobic coupling of phenol using cuprous chloride as a catalyst to be second order with respect to the solution pH when the pH values were below the pKa value. These studies suggest that phenoxy radicals are formed much more readily from phenolate ions than from the neutral species, which explains the observed behavior of polysubstituted chlorophenols. The comparison of the anoxic adsorption isotherms for 2-methylphenol, 2-ethylphenol, and 2-propylphenol, given in Figure 8, indicate that the increase in the molecular weight of the functional group substituted on the parent phenol molecule induced an increase in anoxic GAC
2-Methylphenol 2-Ethylphenol 2-Propylphenol
A
t 1
10' 100
0
;
I IO'
102
103
C, mg/L
Flgure 9. Recovery of alkylphenols from GAC surface.
adsorptive capacity. Such behavior is expected since the higher molecular weight compounds are more hydrophobic, which increases their affinity for the carbon surface. On the other hand, the relative increase in GAC adsorptive capacity resulting from the oxic conditions in the test environment, represented as the ratio of oxic to anoxic GAC adsorptive capacity for that compound, is the most pronounced for 2-methylphenol, followed by 2-ethylphenol and 2-propylphenol. Not only is the relative increase in GAC adsorptive capacity the most pronounced for the adsorption of 2-methylphenol, but the absolute value of the irreversible uptake is highest for 2-methylphenol (Figure 9). The average irreversible uptake for 2-methylphenol was 261.1 (f18.5) mg/g, while the corresponding values for 2-ethylphenol and 2-propylphenolwere 245.3 (f8.5) and 217.35 (*9.6) mg/g, respectively. This observed behavior is expected since the oxidative coupling reactions are sterically hindered in the case of the higher molecular weight compounds, and hence, polymerization of 2-methylphenol was the most pronounced among the three alkyl phenols. The effect of GAC particle size on the observed phenomenon was investigated by conducting oxic and anoxic adsorption isotherms using three particle sizes, namely 12 X 14,16 X 20, and 170 X 230 mesh size, having mean particle diameters of 1.54, 1.0, and 0.074 mm, respectively. The oxic and anoxic adsorption isotherms obtained for 2-methylphenol, shown in Figure 10, clearly demonstrate almost no effect of GAC particle size on adsorption equilibrium for this compound. Furthermore, GAC particle size had very little effect on irreversible adsorption as demonstrated by the extraction experiments Environ. Sci. Technol., Vol. 27, No. IO, 1993 208.5
lo3
pH 4.5
~
I
.
rev
irrev
\ M
g
102
12x14
GAC P a r t i c l e Size 16x20 :70x230 OXlC
anoxic
10'
100
10'
10'
io3
102
IO0
C , mg/L
TV' . '
'
'
"
J
15x20
I
10' 10'
10'
10E
13*
Figure 13. Effect of solution pH on 2-chlorophenol recovery from GAC surface.
" "
'12x1,
100
> - -
C mg/i
Figure 10. Effect of GAC particle size on GAC adsorptlve capaclty. lo3
i
t
1
L
I 03
102
C , mg/L
Figure 11. Effect of GAC particle size on adsorbate recovery.
13. It is apparent from these figures that the solution pH did not have a major effect on the exhibited anoxic and oxic GAC adsorptive capacities once the adsorbate was present in the solution primarily in its neutral form. Both oxic and anoxic GAC adsorptive capacities and the irreversible uptakes obtained at pH 4.5 and 7.0 were very similar. Conversely, the anoxic GAC adsorptive capacity for 2-chlorophenol at pH 10.0 was appreciably lower than the capacity exhibited at pH 4.5 or 7.0. Such behavior was previously explained by Rosene and Manes (30). The data in Figure 13 indicate that the irreversible uptake at pH 10.0 is higher than that observed at pH 4.5. or 7.0. Such observation is in agreement with the previously discussed increased reactivity of the ionized species compared to their neutral counterparts.
Conclusions
pH=lO.O 10' 100
10'
102
io3
c , mg/L
Figure 12. Effect of solution pH on GAC adsorptive capacity for 2-chlorophenoL
conducted on GAC used in the oxic isotherm tests with 1 2 X 14 and 16 X 20 mesh size GAC (Figure 11). No extraction experiments were conducted on 170 X 230 mesh size carbon due to the loss of some adsorbent during filtration for concentration measurements and the inability to adequately contain carbon in cellulose extraction thimbles used for Soxhlet extraction. The effects of pH on GAC adsorptive capacity and irreversible adsorption under oxic and anoxic conditions was evaluated for 2-chlorophenol, which has a dissociation constant, pKa, of 8.11. Three different values of the solution pH, 4.5, 7.0, and 10.0, were tested in this study. The oxic and anoxic GAC adsorptive capacities observed at these pH values are summarized in Figure 12, while the reversible and irreversible uptakes are presented in Figure 2084
Envlron. Scl. Technol., Vol. 27, No. 10, 1993
This study clearly delineates the important role of molecular oxygen on the adsorption of phenolic compounds by activated carbon. The presence of molecular oxygen induced polymerization of these compounds on the carbon surface, resulting in a significant increase in the adsorptive capacity of GAC compared to that attainable under anoxic conditions. The capacity of GAC to promote polymerization reactions and the adsorptive capacity for the reaction products remained constant over 3 orders of magnitude of equilibrium aqueous phase concentrations. Solvent extraction of GAC loaded with adsorbate in the presence of molecular oxygen yielded very low extraction efficiencies, while almost 100% of the adsorbate was recovered from the carbon surface when the adsorption phase was carried out under anoxic conditions. Consequently, the phenol number, which is one of the standard parameters that characterize activated carbon, is highly influenced by the experimental conditions used in that test. The amount of oxygen consumed during the adsorption of three different adsorbates was linearly proportional to the amount of irreversibly adsorbed compound. Critical oxidation potential was used to establish relative susceptibility toward polymerization and, consequently, the amount of irreversible adsorption for the compounds tested in this study. It was also established that GAC particle size affects neither the oxic and anoxic GAC adsorptive capacities nor the adsorbate recovery for the GAC used in these tests.
Phenolate ions were found to adsorb onto the GAC surface to a lesser extent than neutral molecules. Moreover, oxidative coupling and, consequently, irreversible adsorption were much more pronounced in the case of the ionic species. Acknowledgments
Funding for this work was provided by the U.S. Environmental protection Agency under COE-UC/RREL Cooperative Agreement CR-816700. The views expressed are those Of the authors and do not reflect the views of the agency. Mention of trade names endorsement products does not Or or recommendation for use.
Literature Cited Adams, C. E. Proceedings of the 29th Industrial Waste Conference, Purdue University: West Lafayette, IN, 1974, p 617. Suidan, M. T.; Strubbler, C. E.; Kao, S. W.; Pfeffer, J. T. J . Water Pollut. Control Fed. 1983,55, 1263. Gould, J.P.; Weber, W. J.,Jr. J. WaterPollut. ControlFed. 1976,48 ( l ) , 47. Chrostowski, P. C.; Dietrich, A. M.; Suffet, I. H. Water Res. 1983. 17 111).1627. Kim,'B. R.; Snoeyink, V. L.; Saunders, M. F. J . WaterPollut. Control Fed. 1976,48 (l),120. van Vliet, B. M.; Weber, W. J., Jr. J . Water Pollut. Control Fed. 1981,53 (11))1585. Weber, W. J., Jr.; Morris, J. C. J . Sanit. Eng. Diu., Am. Ser. Civ. Eng. 1964,185. Coughlin, R. W.; Ezra, F. S. Environ. Sci. Technol. 1968, 2,291. Snoeyink, V. L.; Weber, W. J., Jr.; Mark, H. B., Jr. Environ. Sci. Technol. 1969,3,918. Zogorski, J. S.; Faust, S. D.; Haas, J. H., Jr. J . Colloid Interface Sci. 1976,55, 329. Crittenden, J. C.; Weber, W. J., Jr. J . Environ. Eng. Diu., Am. SOC.Civ. Eng. 1978,185. Peel, R. G.; Benedek, A. Environ. Sci. Technol. 1980,14, 66. Seidel, A.; Tzscheutschler, E.; Radeke, K. H.; Gelbin, D. Chem. Eng. Sci. 1985,40,215. Magne, P.; Walker, P. L., Jr. Carbon 1986,24,101. Vidic, R.D.; Suidan, M. T.; Traegner, U. K.; Nakhla, G. F. Water Res. 1990,24 (lo), 1187.
(16) Vidic, R. D.; Suidan, M. T. Environ. Sci. Technol. 1991,25 (9), 1612. (17) Musso, H. In Oxidative Coupling of Phenols; Taylor, W. I., Battersby, A. R., Eds.; Marcel Dekker: New York, 1967. (18) Nonhebel, D. C.; Walton, J. C. Free-Radical Chemistry; Cambridge University Press: Cambridge, 1974. (19) McDonald, P. D.; Hamilton, G. A. In Oxidation in Organic Chemistry;Trahanovsky, W. S., Ed., Academic Press: New York, 1973. (20) Denisov, E. T. In ChemicalKinetics; Bamford, C. H., Tipper, C. F. H., Eds.; Elsevier Scientific Publishing Co.: New York, 1980; Vol. 16. (21) Shibaeva, L. V.; Metelitsa, D. I.; Denisov, E. T. Kinet. and Catal. 1969,10 (5)) 832. (22) Hay, A. S.; Blanchmd, H. S.; Endres, G. F.; Eustance, J. W. J . Am. Chem. SOC.1959,Bl (23))6335. (23) Gould, E. S. Mechanism and Structure in Organic Chemistry; Holt, Rinehart and Winston: New York, 1962. (24) Bordwell, F. G.; Cheng, J. P. J. Am. Chem. SOC. 1991,113, 1736. (25) Lim, P. K.; Cha, J. A.; Patel, C. P. Ind. Eng. Chem. Process Des. Dev. 1983,22,477. (26) Stone,T. J.;Waters, W. A. J . Am. Chem. SOC.1964,86,213. (27) Fieser, L. F. J . Am. Chem. SOC.1930,52,5204. (28) Drahanovsky, J.; Vacek, 2. Collect. Czech. Commun. 1971, 36,3431. (29) Chin, K. C.; Cha, J. A.; Lim, P. K. Ind. Eng. Chem. Process Des. Dev. 1985,24,339. (30) Rosene, M. R.; Manes, M. J . Phys. Chem. 1977,81,1651. (31) Catalog Handbook of Fine Chemicals; Aldrich Chemical Co.: Milwaukee, WI, 1991. (32) Handbook of Chemistry and Physics, 68th ed.; CRC Press: Boca Raton, FL, 1987. (33) Morrison, R. T.; Boyd, R. N. Organic Chemistry, 10th ed.; Allyn and Bacon: Boston, 1964. (34) Merck Index, 11th ed.; Merck & Co: Rahway, NJ, 1989. (35) Dean, J. A. Handbook of Organic Chemistry; McGrawHill: New York, 1987. (36) Handbook of Data on Organic Compounds, 2nd ed.; Waest, R. C., Grasselli, J. G., Eds., CRC Press: Boca Raton, FL, 1989; Vol. 5. Received for review November 23, 1992.Revised manuscript received June 14, 1993. Accepted June 25, 1993.' Abstract published in Advance ACS Abstracts, August 15,1993.
Environ. Sci. Technol., Vol. 27, No. 10, 1993
2085
