The surface characteristics of activated carbon as affected by ozone and alkaline treatment

(1)

The surface characteristics of activated carbon as aﬀected

by ozone and alkaline treatment

Hung-Lung Chiang

a

, C.P. Huang

b,*

, P.C. Chiang

c

a

Department of Environmental Engineering, Fooyin Institute of Technology, Kaohsiung Hsien, 831, Taiwan, ROC

b

Department of Civil and Environmental Engineering, University of Delaware, Newark, DE 19716, USA

c

Graduate Institute of Environmental Engineering, National Taiwan University, Taipei, Taiwan, ROC Received 17 April 2000; received in revised form 6 July 2001; accepted 13 August 2001

Abstract

The surface chemical characteristics of activated carbon treated by ozone and alkaline are determined in terms of surface functional groups and surface acidity. Surface functional groups are analyzed by the IRspectroscopic method and Boehm’s titration technique. The surface acidity of activated carbon is determined by electrophoretic mobility measurements. The oxygen concentration of activated carbon increases upon ozone and NaOH treatment. Surface functional groups increase mostly in the hydroxyl and carboxyl categories rather than the carbonyl category upon ozone and NaOH treatment. Ó 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Activated carbon; Ozone; Alkaline; Oxygen concentration; Surface functional groups

1. Introduction

Activated carbon can be produced from a great va-riety of carbonaceous materials including coconut shells, sawdust, wood char, coal, petroleum coke, bone char, molasses, peat and paper-mill waste (lignin) (Mattson and Mark, 1971; Hassler, 1974; Bansal et al., 1990). Depending on the extent of oxidation reaction, two types of activated carbon can be produced (Mattson and Mark, 1971; Hassler, 1974; Corapcioglu and Huang, 1987; Bansal et al., 1990): H-type and L-type. H-type activated carbons exhibit positive charge in water, ad-sorb strong acid and are hydrophobic. L-type activated carbons display negative charge in water, neutralize strong bases and are hydrophilic. Much has been re-ported on the oxidation of activated carbon. Steebberg (1944) characterized the activation and oxidation of

carbon at various temperatures and classiﬁed those carbons that were oxidized at low-temperature and ad-sorbed primarily hydroxide ion as L-carbons, and those activated at high temperatures and adsorbed strong acid as H-carbons. The L-carbon behavior is expected to intensify after long exposure to the atmosphere even at ambient temperatures. Cookson (1978) and Huang (1978) reported the adsorption of electrolytes and non-electrolytes and its eﬀect on the structure of the electrical double layer and the role of surface functional group on the nature of adsorption.

Bailey (1982) proposed two mechanistic extremes for the oxidation of carbon by ozone: a radical type, ozone-initiated autoxidation, and a concerted reaction, i.e., l, 3-dipolat insertion. Schubert and Pease (1956a) reported that in the high temperature range, molecular oxygen participates in the reaction as the temperature increases up to 270 °C. The reaction then becomes a slow combustion process that is indistinguishable from the reaction with oxygen alone. At low temperature, the mechanism of vapor-phase reaction is similar to that of liquid-phase ozonation, at least when only molecular

www.elsevier.com/locate/chemosphere

*

Corresponding author. Tel.: 831-8428; fax: +1-302-831-3640.

E-mail address: huang@mail.ce.udel.edu (C.P. Huang).

(2)

ozone is involved. At ordinary temperature, reactions in general occur via an ozone insertion mechanism in-volving the hydrotrioxideðRO3Þ intermediate, its

prod-uct, peroxidesðRO

2Þ and other radicals. If ozone is the

principal reactant, no chain reaction will take place. In the presence of metal catalysts, radical chain mecha-nisms are involved and become an important part of the oxidation reaction.

Reaction between activated carbon and alkaline is another chemical means for the modiﬁcation of surface property. Under alkaline environment, it is expected that OHwill react with the surface functional groups of the activated carbon. However, little is available in the literature on the eﬀect of alkaline and ozone treatment on the surface property changes of activated carbon. The objective of this study was to evaluate the surface property changes of an activated carbon upon treatment by ozone and alkaline. Infrared spectroscopy, elec-trophoretic measurements and alkalimetric titration were used to assess the changes of surface property of the activated carbon.

2. Experimental

2.1. Material

An activated carbon (8 30 mesh), made from coco-nut shells and provided by Kowa Cosmos Company, Japan was used in this study. Prior to chemical treat-ment, the surface property of this activated carbon was characterized. Detailed experimental procedures for the characterization of surface property were reported else-where (Chiang et al., 1999).This activated carbon has a BET surface area, average pore diameter, macropore volume (>500 AA), mesopore volume (between 20 and 500 AA), and macropore volume (<20 AA) of 795 50 m2_=g, _14:7_0:05_A_A, _0:011_{0:009 cm}3_=g, _0:041

0:012 cm3_{=g, 0:325}_{0:023 cm}3_{=g, respectively. All}

so-lutions were prepared from chemicals provided by Merck Chemicals Company, Germany. A pH meter (model 420 A) was used for pH measurements. Strong acid (0.1 N HCl) and baseð0:1 N NaHCO3, Na2CO3,

and NaOH) were used for the analysis of surface func-tional group. Unless otherwise mentioned, strong acid (0.l M HClO4) and strong base (0.2 M NaOH) were used

for all pH adjustments.

2.2. Chemical treatment of activated carbon

500 g of activated carbon was weighed and placed into each of the ﬁve 2-l polyethylene bottles. 1 l of NaOH solution at various concentrations was added to each of the ﬁve bottles containing the activated carbon. The bottles were then placed on a rotating vibrator and

mixed constantly for 24 h. The activated carbon was then separated from the solution and placed in an oven at 105°C and dried for 48 h. The NaOH-treated acti-vated carbon samples were divided into two parts. One part received further ozone treatment at an inﬂow con-centration of 40 mg/l and ﬂow rate of 2.5 l/min (or mass rate of 100 mg/min) for 30 min. The reaction tempera-ture was controlled at level below the combustion point of the activated carbon using a water tank.

Both activated carbon samples were washed with distilled water until the sodium ion concentration of the rinsed water reached that of the distilled water. The activated carbon was dried in an oven at 105°C for 48 h then transferred to a desiccator until use.

2.3. IR spectrum analysis

The activated carbon samples were ground in an agate motar to ﬁne powder. The activated carbon pow-der was mixed with KBr by a weight ratio of 700:1. A given amount of the mixture (200 mg) was used for the preparation of KBr pellets. The KBr pellets were stored in a desiccator until IRanalysis. The IRinstru-ment was a Bomem, model DA 3.002 FTIR. The IR spectrum was obtained over a frequency between 500 and 4000 cm1_.

2.4. Boehm titration

Procedures for the analysis of oxygen functional group follow those established by Boehm (Puri and Bansal, 1964; Boehm, 1966; Barton et al., 1973; Fabish and Schleifer, 1984; Magane and Dupong-Parlovskky, 1988; Arico et al., 1989). The activated carbon samples were ﬁrst dried in a vacuum oven (102₁₀3_{mm Hg,}

105 °C) for 24 h. 25 ml of an alkali solution (0.1 N NaHCO3, Na2CO3, or NaOH) were added to test tubes

containing a given amount of the activated carbon sample (5 g). The samples were constantly mixed over a vibrator (100 rpm) at 25°C for 24 h. A given amount of the supernatant (5 ml) was then drawn from the test tubes and back titrated with HC1 (0.1 N) solution. The concentrations of various functional groups were de-termined by the residual bases after back titration as described by Boehm (1966).

2.5. Electrophoretic mobility measurements

Zeta potential measurements of activated carbons were made with Zeta Meter System 3.0. The activated carbon (5 g) was placed in a porcelain mortar and ground to ﬁne powder. A given amount of the activated carbon powder (100 mg) was added to 1 l distilled water. After mixing for several minutes, the coarse particles

(3)

were allowed to settle, and the colloidal size particles were collected for electrophoretic mobility measure-ments. After electrolyte addition (102_{M NaCl) and pH}

adjustment, the activated carbon suspension was placed in the electrophoresis cell for zeta potential measure-ments.

2.6. Surface acidity determination

The surface acidity of the activated carbon was de-termined according to concept and procedures proposed by Huang and associates (Corapcioglu and Huang, 1987). Upon hydration, the surface acidity develops on the activated carbon

COH2¼ COH þ Hþ; Ka1int ð1Þ

COH¼ COþ Hþ; Ka2int ð2Þ

where COHþ₂, COH, and CO represent protonated, neutral and ionized surface hydroxyl groups, respec-tively. The intrinsic acidity constant, Kint

a1 and Ka2int are

deﬁned as follows:

K_a1int¼ fCOHgfHþ_g=fCOHþ

2g ð3aÞ

Kint a2 ¼ fCO

_gfHþ_g=fCOHg _ð3bÞ

where i stands for the surface concentration of the ith species. The total number of surface Bronsted acidity site, NB, is therefore the sum of all the three surface

hydroxo groups, i.e.,

NB¼ fCOHþ2g þ fCOHg þ fCO

_g _ð4Þ

The surface proton concentration is not directly mea-surable; rather it is calculated from the pH measure-ments in the bulk phase, i.e., {Hþ_g

b, and by the

Boltzmann equation

fHþg ¼ fHþgbexpðF wo=RTÞ ð5Þ

where w_o, F, R, T denote the surface potential, Faraday constant, gas law constant and absolute temperature, respectively. The surface potential is not a directly measurable quantity. In a dilute inert electrolyte solu-tion, the surface potential, w_o, is close to the potential at the outer Helmholtz plane, or the diffuse layer potential, w_d. Hþ and OH are considered the sole potential de-termining ions, as indicated by the change of elec-trophoretic mobility at different pH values. In the absence of specific adsorption such as in dilute inert electrolyte solution, the surface charge obtained from alkalimetric titration can be set equal to the diffuse layer charge, rd. The surface potential can therefore be

cal-culated by the Gouy–Chapman electrical double layer theory (Corapcioglu and Huang, 1987; Hunter, 1987)

Woffi Wd¼ ð2RT =zF Þ sinh1 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi p=2RT eI p h i rd; ð6Þ

where z; ; and I are absolute valence of (1:1) electrolyte, dielectric constants of water and ionic strength, respec-tively. In aqueous solution at 20 °C, Eq. (6) can be further simpliﬁed (Corapcioglu and Huang, 1987)

Wo¼ 0:05 sinh1

rd

11:7pﬃﬃI

: ð7Þ

The unit for wo;rd and I in Eq. (7) are volt, lC=cm2,

and M, respectively. As shown above, at pH < pH_zpc the surface is predominately positively charged, there-fore

NBﬃ COHþ2

þ COHf g ð8Þ

Similarly, at pH > pH_Zpc, the surface is negatively charged

NBﬃ COHf g þ COf g ð9Þ

The quantity for the term fCOHg and fCO} are de-termined from alkalimetric titration, i.e.,

fCOHþ

2g ¼ rþ=S ð10aÞ

and

fCO_{g ¼ r}

=S ð10bÞ

where rþ, r are positive surface charge and negative

surface charge, respectively. S is a charge conversion factor from C=cm2 _{to surface concentration (mol/g or}

mol/cm3_{). By substituting the above relationships into}

Eqs. (3a) and (3b) and with further mathematical ar-rangement, one has

1=fHþ_{g ¼ N} B=Ka1intð1=rþÞ ð1=Ka1intÞ at pH < pHzpc ð11aÞ and fHþg ¼ NB=Kinta2ð1=rÞ ð1=Ka2intÞ at pH < pHzpc: ð11bÞ A plot of 1=fHþ_{g versus 1=r}

þ for the positive surface

and of {Hþ} versus 1=r for the negative surface will

yield intercepts and slopes from which the intrinsic acidity constants, Kint

a1 and K int

a2, and total surface acidity

capacity, NB, can be calculated. This method has been

used previously by Huang and Stumm for their studies of hydrous oxide systems (Huang and Stumm, 1973; Huang, 1981; Corapcioglu and Huang, 1987).

(4)

3. Results and discussion

3.1. Surface functional groups – IR analysis

IRresults in Table 1 show clearly that the C–C band (normal paraﬃn alkanes), out of plane C–C band, and the aromatic CH band position are at 497–521, 661–680 and 777–802 cm1_{, respectively. There are two}

absorp-tion bands at 917–934 and 1382–1392 cm1_{, which can}

be attributed to the COOH band. The absorption po-sitions of the C-bands are 1187–1205 and 1272 cm1_,

individually. The absorption bands of C@C double bonds are 1560–1572 and 1633–1639 cm1_{. There is a}

weak band at 1723 cm1 _{in the sodium salt treated}

activated carbon, which may be a normal carbonyl group. Magane and Dupong-Parlovskky (1988) have proposed that NaOH reaction with unsaturated n-lac-tones will form both carboxylate anions and a normal carbonyl group in the form of an aldehyde. The spectra of activated carbon observed show a chemical shift which is due to the reaction of NaOH and O3 with

ac-tivated carbon.

3.2. Surface functional groups – Boehm’s titration carb-oxylic, lactone, and phenolic groups

The Boehm titration method allows the determina-tion of the surface funcdetermina-tional groups such as phenolic group (–OH), lactone group (C@O) and carboxylic group (–COOH). Fig. 1 shows the results of surface functional groups of various activated carbon samples by the Boehm method.

For virgin activated carbon (AC), the total oxygen containing function group is 0.196 meq/g with a break-down of 0.117 meq/g (60%), 0.046 meq/g (23%), and 0.034 meq/g (17%), individually, for phenolic, lactone and carboxylic groups. When treated with O3, the total

oxygen containing functional group increases to 0.240 meq/g, which is an increase of 22% compared to the untreated AC. Increase in oxygen containing functional groups, takes mostly in the phenolic and lactone cate-gories. The phenolic group increases from 0.117 to 0.144 meq/g and the lactone group increases from 0.034 to 0.052 meq/g. The concentration of the carboxylic group slightly increases from 0.034 to 0.044 meq/g upon O3

treatment. It is interesting to note that the composition among these three oxygen containing functional groups remains relatively unchanged compared to AC, that is, 60% versus 60%, 22% versus 23% and 18% versus 17% for the phonemic, lactone and carboxylic groups re-spectively, between AC(O3) and AC.

Treatment of AC with NaOH (1–5 N) increases the total concentration of the oxygen containing functional groups from 0.196 meq/g to between 0.275 and 0.309 meq/g dependent on the NaOH concentration. This is an increase of almost 40–88%. As far as the individual oxygen containing functional groups are concerned, the phenolic group increases from 0.117 to between 0.143 and 0.204 meq/g, the lactone group increases from 0.046 to between 0.073 and 0.096 meq/g and the carboxylic group increases from 0.034 to between 0.053 and 0.094 meq/g. Like the ozone-treated activated carbon, the major increase takes place in the phenolic group.

Tryk et al. (1984) proposed the following reaction between NaOH and activated carbon:

Table 1

IRspectra of activated carbon treated with ozone and NaOH (N@ 2) Vibration group Adsorption peaksðcm1_Þ

A B C D E F G H I J K L

C–C band (Normal paraﬃns alkanes)

509 520 520 508 515 515 521 508 505 497 507 515

Out of plane ring C–C band 680 673 661 667 661 680 667 675 667 663 – – Out of plane aromatic C–H band 790 790 777 802 784 796 777 – 802 – – – COOH 924 918 – 924 931 924 – 936 – 917 – – 1389 1389 1389 1395 1382 1395 – 1394 1389 1388 1389 1896 C–O–C 1199 1205 1187 1205 1205 1193 1193 1196 1199 1208 1193 1193 – – 1272 – – – – – – – – – C@C 1572 1567 1572 1572 1572 1572 1560 1587 1572 1562 1566 1572 1633 – – – – – – – 1639 – – – C@Oa _– _– _– _– _– _– _– ₁₇₂₃ _– ₁₇₂₃ _– _– CH3 – – – – – 2630 2373 2951 2366 – 2360 – –OH 3461 3472 3436 3436 3461 3448 3485 3435 3436 3453 3454 3442

Note. A: VAC, B: AC(O3), C: AC(1N NaOH), D: AC(2N NaOH), E: AC(3N NaOH), F: AC(4N NaOH), G: AC(5N NaOH), H:

ACð1N NaOH þ O3Þ, I: ACð2N NaOH þ O3Þ, J:ACð3N NaOH þ O3Þ, K: ACð4N NaOH þ O3Þ, L: ACð5N NaOH þ O3Þ. a_{Unsaturated b-lactones.}

(5)

Cþ 6OH_{¼ CO}2

3 þ 3H2Oþ 4e;

DG° ¼ 70:72 kcal=mole: ð12Þ It is further known that

4Hþ_{þ O}

2þ 4e¼ 2H2O;

DG° ¼ 113:38 kcal=mole: ð13Þ Combining Eqs. (12) and (13), one has

Cþ 2OHþ O2¼ CO23 þ H2O;

DG° ¼ 184:10 kcal=mole: ð14Þ The following shows the conceptual scheme of NaOH reaction with activated carbon:

Treatment of AC with combined NaOH and O3

in-creases the oxygen containing functional group even further. Results indicate that the total oxygen containing functional group increases from 0.197 (AC) to 0.275– 0.369 meq/g (NaOH–AC) to 0.328–0.371 meq/g (AC(O3)–NaOH). This is an increase by 67–85%

com-pared to the untreated AC. The concentration of

phen-olic group increases from 0.117 to 0.143–0.202 meq/g (NaOH–AC) to 0.165–0.184 meq/g ðACðO3Þ–NaOH).

The concentration of lactone group increases from 0.046 (AC) to 0.052–0.098 meq/g (NaOH–AC) to 0.086–0.115 meq/g (AC(O3)–NaOH). The concentration

of the carboxylic group increases from 0.034 (VAC) to 0.053–0.091 meq/g (NaOH–AC) to 0.063–0.094 meq/g (AC(O3)–NaOH).

3.3. Reaction pathways

Results show that major functional groups of the treated activated carbon include phenolic, lactone, carb-oxylic, carbon–hydrogen bond and carbon–carbon double bonds. Results of the Boehm titration and FTIR analysis clearly indicate that hydroxyl carbonyl, carb-oxylic, C–H and C@C bonds are the major surface functional groups. Based on the literature (Schubert and Pease, 1956b,c; Schubert and Pease, 1956c), a concep-tual reaction pathway can be proposed (Fig. 2).

The surface of activated carbon is thought to be as-semblies of hexagon structure (a) and the reaction site is at the carbon–hydrogen bond. The reaction of carbon– hydrogen bond and ozone follows two pathways, one is the formation of hydroxide compounds (b) and the other is the generation of hydrotrioxide intermindates (c). The hydroxide site may be further transformed to carbon–carbon double bond (d). The hydrotrioxide in-termindates will undergo three reaction pathways. First, hydrotrioxide losses a hydrogen peroxide ðH2O2Þ to

form the carbon–oxygen double bond (e). Second, the

Fig. 1. Intensity of surface oxygen containing functional group as aﬀected by O3and NaOH. Adsorbents: A¼ AC, B ¼ ACðO3Þ, 1N:

AC(1N NaOH), 2N: AC(2N NaOH), 3N: AC(3N NaOH), 4N: AC(4N NaOH), 5N: AC(5N NaOH), 1NO: ACð1N NaOH þ O3Þ,

(6)

hydrotrioxide loses an O2H to produce radicalðOÞ (f).

Third, the deprotonation of hydrotrioxide yields the peroxide radical ðOO_{Þ (g). The peroxide and oxygen}

radical compounds may be further transformed to carboxyl compounds and subsequently carbonyl com-pounds (h).

3.4. Surface acidity

The presence of surface functional groups such as hydroxyl, carboxyl, and carbonyl on activated carbon can bring about surface charge upon hydration. This is evident of the fact that a positive charge evolves when introducing an H-type activated carbon into water. Likewise, a negative charge is observed on the L-type activated carbon when it is placed in water (Huang, 1981).

Fig. 3 shows the zeta potential of various activated carbon samples as a function of pH. Generally, a ﬂesh H-carbon has a pH_zpc of 7. The pHzpc value shifts to

5.3 upon exposure to atmosphere over an extended pe-riod of time which is indicative of an L-type conversion. Exposing an H-carbon to the atmosphere tends to gradually oxidize the carbon and converting it to an L-carbon. Upon treatment with ozone and sodium

hydroxide, the pH_zpc shifts to pH < 2, which clearly indicates an oxidizing surface on the activated car-bon.

Table 2 summarizes the total surface acidity density obtained by alkalimetric titration and the Boehm me-thod. Results indicate that alkalimetric titration yields a total surface acidity larger than the Boehm titration. This can be attributed to the extra surface charge con-tributed by other functional groups such as sulfur and nitrogen which cannot be detected by the Boehm method.

4. Conclusions

Ozone and NaOH treatment of activated carbon re-sults in an increase of surface oxygen functional groups, especially in the phenolic and carboxylic categories. Because the increase in the formation energy of lactone is greater than that of other functional groups, the in-crease of lactone group intensity on activated carbon is not signiﬁcant by either ozone or NaOH treatment. Based on the results obtained from this study and re-ported literature information, a preliminary pathway for O3reaction with activated carbon can be proposed. The

reaction of carbon–hydrogen bond and ozone follows

(7)

two pathways, one is the formation of hydroxide com-pounds and the other is the generation of hydrotrioxide intermindates. The hydroxide site may be further

transformed to carbon–carbon double bond. The hy-drotrioxide intermindates will undergo three reaction pathways: First, the hydrotrioxide losses a hydrogen

Fig. 3. (a) Eﬀect of NaOH treatment on the zeta potential of activated carbon. (b) Eﬀect of O3treatment on the zeta potential of

(8)

peroxideðH2O2Þ to form carbon–oxygen double bond.

Second, the hydrotrioxide losses an O2H to form

radi-cal ðOÞ. Third, the hydrotrioxide deprotonates to form peroxide radicalðOO_{Þ. The peroxide and oxygen}

radical compounds may be further transformed to carboxylic compounds and subsequently lactone com-pounds. Alkalimetric titration gives a surface acidity larger than the Boehm titration due in part to the extent of neutralization reaction.

References

Arico, A.S., Antonucci, V., Minutoli, M., Giordano, N., 1989. The inﬂuence of functional groups on the surface acid–base characteristics of carbon blacks. Carbon 27, 337–347. Bailey, P.S., 1982. Ozonation in Organic Chemistry, vol. II.

Noleﬁnic Compounds. Academic Press, New York. Bansal, R.C., Donnet, J.B., Stoeckli, F., 1990. Active Carbon.

Marcel Dekker, New York.

Barton, S.S., Gillespie, D., Harrison, B.H., 1973. Surface studies of carbon: acidic oxides on Speron 6. Carbon 11, 649–654.

Boehm, H.P., 1966. Chemical identiﬁcation of surface groups. In: Eley, D.D., Pines, H., Weisz, P.B. (Eds.), Advances in Catalysis, vol. 16. Academic Press, New York, p. 179. Chiang, H.L., Huang, C.P., Chiang, P.C., Chang, E.E., 1999.

Eﬀect of metal additives on the physico-chemical charac-teristics of activated carbon exempliﬁed by benzene and acetic acid adsorption. Carbon 37, 1919–1928.

Cookson, J.T., 1978. Adsorption mechanism: the chemistry of organic adsorption on activated carbon. In: Cheremisinoﬀ, P.N., Ellerbursch, F. (Eds.), Carbon Adsorption Hand-book. Ann Arbor Science, New York, pp. 241–280.

Corapcioglu, M.O., Huang, C.P., 1987. The surface acidity and characterization of some commercial activated carbons. Carbon 25, 569–578.

Fabish, T., Schleifer, D.E., 1984. Surface chemistry and carbon black work function. Carbon 22, 19–38.

Hassler, J.W., 1974. Activated Carbon: Industrial Commercial Environmental. Chemical, New York.

Huang, C.P., 1978. Chemical interactions between inorganic and activated carbon. In: Cheremisinoﬀ, P.N., Ellerbursch, F. (Eds.), Carbon Adsorption Handbook. Ann Arbor Science, New York, pp. 281–330.

Huang, C.P., Stumm, W., 1973. Speciﬁc adsorption of ca-tions on hydrous r-Al2O3. J. Colloid Interface Sci. 43,

409.

Huang, C.P., 1981. In: Anderson, M.A, Rubin, A.J. (Eds.), The Surface Acidity of Hydrous Solid in Adsorption of Inor-ganics at Solid–Liquid Interfaces. Ann Arbor Science, New York.

Hunter, R.J., 1987. In: Zeta Potential in Colloid Science, Principles and Applications. Academic Press, San Diego, pp. 11–52.

Magane, P., Dupong-Parlovskky, N., 1988. Graphite–ozone surface complexes. Carbon 26, 249–255.

Mattson, J.S., Mark, H.B., 1971. Activated Carbon: Surface Chemistry and Adsorption from Solution. Marcel Dekker, New York.

Puri, B.R., Bansal, R.C., 1964. Studies in surface chemistry of carbon blacks – II: Surface acidity in relation to chemi-sorbed oxygen. Carbon 1, 457–464.

Schubert, C.C., Pease, R.N., 1956a. The oxidation of lower paraﬃn hydrocarbons. II. Observations on the role of ozone in the slow combustion of isobutane. J. Am. Chem. Soc. 78, 5553–5554.

Schubert, C.C., Pease, R.N., 1956b. The oxidation of low paraﬃn hydrocarbons room temperature reaction of meth-Table 2

Eﬀect of O3and NaOH treatment on the surface acidity of activated carbona

Adsorbentsb _{Ionic strength 5}₁₀2_Mc _pHd

zpc Zeta potential-NB(lC/cm2) Boehm’s method-NB(lC/cm2)

pKint a1 pKa2int A 4.09 6.25 5.30 3.01 2.42 B ()0.67) 4.67 2.00 10.54 2.72 C ()0.79) 4.79 2.00 6.85 3.21 D ()0.16) 4.16 2.00 6.60 4.23 E ()0.84) 4.84 2.00 9.54 3.92 F ()0.83) 4.83 2.00 8.84 4.40 G ()0.40) 4.40 2.00 7.29 3.88 H ()0.79) 4.79 2.00 13.00 3.88 I ()0.79) 4.79 2.00 8.78 3.97 J ()0.74) 4.74 2.00 8.97 4.26 K ()1.19) 5.19 2.00 7.92 4.28 L ()0.27) 4.27 2.00 9.28 4.24 a

The concentration of activated carbon is 100 mg-PAC/l-NaClO4(aq) and experiment is performed in closed system at 25°C. b

A: VAC, B: ACðO3Þ, C: AC(1N NaOH), D: AC(2N NaOH), E: AC(3N NaOH), F: AC(4N NaOH), G: AC(5N NaOH), H:

ACð1N NaOH þ O3Þ; I: ACð2N NaOH þ O3Þ, J: ACð3N NaOH þ O3Þ, K: ACð4N NaOH þ O3Þ, L: ACð5N NaOH þ O3Þ. c

NaClO4as inert electrolyte. d pHzpc¼ ðpK int a1þ pK int a2Þ=2.

(9)

ane, propane, n-butane and isobutane with ozonized oxy-gen. Am. Chem. Soc. 78, 2044–2047.

Schubert, C.C., Pease, R.N., 1956c. Reaction of paraﬃn hydrocarbons with ozonized oxygen: possible role of ozone in normal combustion. J. Chem. Phys. 24, 919– 923.

Steebberg, B., 1944. Adsorption and Exchange of Ions on Ac-tivated Charcoal. Almquist and Wiksells, Uppsala, Sweden. Tryk, D., Aldred, W., Yeager, E., 1984. In: Sarangapani, S., Akridge, J.R., Schumm, B. (Eds.), Proceedings of the Workshop on the Electrochemistry of Carbon. The Electrochemical Society, Pennington, NJ, p. 192.