Pertanika 13(1), 73-77 (1990)

Adsorption Potential of Activated Carbon in Some Acidicand Alkaline Media

M. BADRI, KARE A. CRO SE and CHUA SOOK YA

Department of ChemistryFaculty oj Science and Environmental Studies

Universiti Pertanian Malaysia43400 UPM Serdang, Selangor Darul Ehsan, Malaysia.

Key Words: Carbon activated, adsorption, adsorption potential, functional groups on carbon.

ABSTRAK

Satu elektrod telah diperbuat daripada karbonyang telah diaktijkan. Perubahan pada keupayaan permukaannya terdapatapabila ia direndamkan dalam larutan-larutan mengandungi HCI dan hidroksida dan karbonat bagi Na +, K+ danNH~. Perubahan-perubahan ini telah disabitkan sebagai kesan kepekatan Cl- dan OH- ke atas kuasa penjerapanpelbagai kumpulan berJungsi di atas permukaan karbon yang telah diaktijkan.

ABSTRACT

An electrode was constructed using activated carbon. Changes in its sUlJace potential were observed when it was dippedinto solutions containing HCI and hydroxide and carbonate ofNa +, K + andNH~. These changes have been atlributedto the effect ojconcentration ofCl- and 0 H- ions upon the adsorptive strength ofvariousJunctional groups on the surfaceoj activated carbon.

INTRODUCTIONAlmost every kind of activated carbon containscombined oxygen and hydrogen on its surfaceforming a variety of complexes (Puri & Bansel1964). Even though a complete list of thesecomplexes (or functional groups as they arecommonly called) is not yet available, some, suchas carboxyl, carbonyl, phenol, lactone andhydroxyl, are said to have been definitelyidentified (Boehm 1966). A summary of thereactions carried out for group identification hasalso been pu blished (Donnet 1968).

In addition, activated carbon samplesprepared by different methods show differentcharacteristics in the reaction with acids and bases(Epstein et al. 1971). These differences reflect thenature and the abundance of the functionalgroups on the surface of carbon. The acidicgroups such as carboxyl, for example, ionize torelease hydrogen ions into solution (Puri 1966)while the basic groups, such as oxides with a

pyrone-type structure, adsorb hydrogen ions Iromsolution (Boehm & Vo1l1970). In either case, thereaction of the functional groups with waterresults in the formation of immobile chargedspecies on the surface of carbon with theirconjugates staying alongside them in the solu tion.Such an arrangemen t ofcharges would give rise toan electrical double layer. The magnitude of thepotential of this double layer would depend on thedegree of ionization and adsorption as well as thesurface concentration of each type.

Measurement of this potential would bedifficult, if not impossible. However, its variationas a result of changes in the degree of ionizationand adsorption could be investigated. This can bedone by bringing opposite sides of a piece ofactivated carbon into contact with two differentsolutions. Under such conditions it is expectedthat an electrical double layer would bedeveloped at each side and that they would differin magnitude depending on the nature and

M. BADRI, KARE A. CROUSE AND CHUA SOOK YA

TABLE 1

Surface potential of carbon electrode

mY

o 1 234 5 6 7106 164 218 269 312 323 310 32377 101 137 245 287 325 307 333

102 124 138 180 276 314 330 31454 87 130 173 293 322 310 301

110 133 151 178 252 325 347 32964 98 140 234 292 330 324 312

117 146 188 235 270 303 313 305

50

.S'CQl

oa.

-"

100

350

at pC =

HCll\'aOHNa2COaKOHK 2COaNH40H(;\lH4)2COa

I n each case, it was found that the poten tialincreased with increasing pC reaching a maxi­mum at about pC~5. Between pC~5 and pC~ 7, the potential decreased slowly except forsolutions of HCI and TaOH. For these two, aminimum was observed at pC~6.

234-log concentration, pC

Fig. 1: Surface Potential of Carbon Electrode in HCl, NaOH and

Na2C0

3

SOlution

RESULTSThe results of the potential measurement on thetest solutions using the carbon electrode are givenin Table 1 and are plotted as a function of :logconcentration, pC, in Figs. 1,2 and 3. Fig.1 showsthe relations for HCl, NaOH and Na2C0

3; Fig.2

for Hel, KOH and K 2COa and Fig.3 for HCl,H 40H and (NH4)2COa. The values obtained for

HCl are shown in each figure for ease ofcomparison only.

MATERIALS AND METHODSAll reagents were of analytical grade unlessotherwise specified.

Large pieces of activated carbon wereprepared by a method described earlier (BritishPatent 1984). Samples thus prepared havevery low electrical conductivity (of the order of1 x 10 6S) and show very little increase inconductivity upon adsorption ofwater and simpleions (Badri et al. 1984). Fairly flat pieces wereground with fine sand paper into discs of about 6mm diameter and 2 mm thickness. Two discs werewashed, dried and used for the contruction of acarbon electrode as previously described (Badri &

Crouse 1989).An Ag/AgCl electrode (Shoemaker &

Garland 1962) was used as a reference. I twasinserted into a glass tube containing 1M HCl andhaving a sliver of Whatman Qualitative 3 filterpaper as a porous junction, The paper was soakedwith Araldite epoxy resin and dried before use sothat the leak of the acid through the junction intothe test solutions would be minimal (Ramli et at.

1989). The test solutions were HCI andhydroxides and carbonates ofNa +, K + and NH ~

of various concen trations. They were kept insmall, covered polyethylene bottles and wereequilibrated at 25.0 ± 0.2°C.

The potential developed when the carbonand the reference electrodes were dipped into atest solution was measured using a Hewlett­Packard 3476 A M ultimeter. The meter readingwas recorded when 'a constant value wasregistered. After such measurement, bothelectrodes were rinsed with distilled water andgently dried with .soft tissue paper before beingdipped into a solution of different concentrati­on. In addition, the carbon electrode was washedcontinuously for 24 hours with distilled waterdripping from a 10 litre container at a rate ofabout 3 drops per minute before being used againfor measurement with a test solution containing adifferent base.

concentration of the ionic species in the solutions.The nett potential developed can be measuredusing electrical circuitry similar to that used in pHmeasurements with a glass electrode. This paper isintended to illustrate such a measurement onactivated carbon prepared in our laboratory.

74 PERTANIKA VOL. 13 NO, I, 1990

ADSORPTION POTENTIAL OF ACTIVATED CARBON IN SOME ACIDIC AND ALKALINE MEDIA

phenomena of adsorption and desorption of H+when it comes into contact with water. Thesimplest among these is the dissociation of the·acidic groups (hydroxyl and carboxylic):

(1) ~OH + H 20 ~ ~O- + HP+

f--COOH + H 20 ~ ~COO- + HaO+(Hagiwara el at. 1971)

It has been suggested that H+ might also bereleased through hydrolysis of some other groupsas given below:

50 (2)

Fig. 2.

2 3 4 5 6 7-log concentration, pC

Surface Potential of Carbon Electrode in HCI, KOH and

K2C0

3

HO H 0 /.0I V '9~

CXO <:'>

o/I

CO) +

350 (Boehm el at. 1964)

oHCI

ONH 40H e(NH42co3

300

250

~

23456-log concentration, pC

7

(3)

or )C02 + HzO ~ )CO~- + 2H +, Puri 1970)

On the other hand, some of the basicfunctional groups can adsorb H + possiblythrough reactions such as (4) dan (5):

(5)

75

(Garten & \Vciss. 1957)

(Boehm & Voli. 1970)

Fig. 3: Surface Potential of Carbon Electrode in HCI, NHpH and

(NH;2C03

At high concentrations (pC=O to 2), thepotential found for the hydroxides were lowerthan those found for the carbonates. For the moredilute solutions, however, the opposite was foundtrue. At pC~5 the potentials approachedapproximately the same value.

DISCUSSIONSeveral mechanisms have been proposed toexplain the acidic and basic characteristics ofactivated carbon which are manifested by the

PERTA IKA VOL. 13 O. 1, 1990

\1. BAUR!. KARE.'J A. CRO SE ANI) CHUA SOOK VAl

or any other reaction involving electron-richgroups such as lactone, quinone, ester, ether, elc.,

that may be present on the surface of activatedcarbon (Hassler 1974 and ~rattson & ~rark

1971).\\'hen the solution contains an acid,

adsorption of H + would be more favorable thandissociation. Hence the double layer formedwould have positive charges residing on thesurface of the carbon (this we arbitrarily call theCi-type double la)er . This Ci-type was probablythe type formed on the internal surface of thecarbon electrode. The magnitude of the potentialde\'eloped. Ci" remained constant since theconcelllration of HCI was unchanged.

1\ similar double layer was developed on theexternal surface of the carbon electrode when itwas dipped into HCI test solu tions. I ts magnitude,Ci,., was in the opposite direction to Cii and changedwith changing concentrations. At the same time,ionization of the acidic groups might have alsotaken place leading to the developmen t ofanothertype ofdouble layer, ~ .. , whose field direction was,however, parallel to Ci i . Such a double layer ismost likely to develop at lower concentrations ofthe test olution. Thus at any instant, the nettpotential measured, P, could be expressed as:

p = Ct., - Ct., + [3,

I t is clear that as the concen tration of the test

solution was lowered, Ci.. became smaller and ~ ..became larger leading to an overall increase in pas obsen·ed. [The contribution of~.. to the nettpotential was small because of two reason.Firstly, the surface concentration of the acidicgroups is small (Puri & Banse11964, and Barton el

al. 1972) and secondly, the degree of dissociationof these acids did not reach 100% due to the factthat they are weak acids, and more importantstill, the hydrogen ions released were confined tothe vicinity of their conjugates.]

It has been found that carbon has a lowintrinsic ion-exchange property for simple ionssuch as Ci- (Tsuchida & YIuir 1986). The activesite for CI- adsorption is not known but positivecentres de\'eloped through equations (4) and (5)were the most probable ones. The effect of Cl­adsorption was to lower Ci.. bu t as the con­centration of HCI was lowered, and hencemore acidic groups dissociated, the residual

negative charge thus produced on the carbonsurface would, through resonance or induction,neutralize these positive centres. Under such acondition, carbon would begin to desorb Cl-.Desorption of Cl- increased Ci,. and this wasperh a ps the rea on for the decrease in p asobserved for HCI at pC between 5 and 6.

The increase in p with dilution for the othertest solutions can be similarly explained.H yd roxides and carbonates can neutralise theacidic groups on the surface of carbon (Boehm el

al. 1964). Such neutralization reactions wouldprod uce negative charges on the surface ofcarbonwhich then could be delocalized throughresonance or induction to the entire graphite ringsystem (Given & Hill 1969). Besides neutralizingthe positive centres as described above, thisprocess could also further enhance the abili ty ofthe electron-rich centres to adsorb cations.Potassium ions, for example, were thought to beadsorbed on the phenolic groups at pH> 11(Tsuchida & Muir 1986). 1\s the alkalis becamemore dilute, less neutralization occurred, leadingto less adsorption and hence p increased.

Hydroxide ions influence adsorption in amanner similar to chloride ions. That is, at ahigher concentration, the ad orption of OH­would contribute to the increase in the potentialby neutralizing the positive centres on carbonsurface. At lower concentration of the alkalis, thiseffect was less important and the ionization of theacidic groups through reactions (1) to (3) becamemore extensi\'e helping to desorb OH -. Thecombined effect would be to lower p as observedat pC ~ 5 and higher.

Assuming complete dissociation of thehyd roxides and carbonates used III theseexperiments, the concentration of the cation in acarbonate solution is expected to be twice that ofthe corresponding hydroxide. However, hydroxi­des are stronger bases than their correspondingcarbonates. Therefore, they were able toneutralize weaker acidic groups as well (Boehm el

al. 1964) leading to a greater degree ofdelocalization of charges which enhanced cationad orption. This was probably the reason forlower p for the hydroxides compared to thecarbonates on the high concentration side of Figs.1,2 and 3.

To a first approximation, it can be shownthat at low pC the concentration of OH - in the

76 PERTANIKA VOL. 13 0 I, 1990

ADSORPTION POTE TIAL OF ACTIVATED CARBON I SOME ACIDIC AND ALKALI E MEDIA

o. 2086864. Universiti

alkali metal hydroxide is greater than that in the

corresponding carbonate solutions. The reverse is

true for solutions with high pC. The change-over

point occurs at pC between 3 and 4. Thus as

discussed above, neutralization of the acidic

functional groups would be more extensive in

dilute carbonate solutions as compared to the cor­

responding hydroxides. Consequently, adsorp­

tion of cations was greater, leading to lower p for

the carbonates compared to the hydroxides in this

concentration region, Figs. 1 and 2. This wasprobably also true for the case ofammonia, Fig. 3.The slight \·ariation observed was probably due to

the multiple equilibria that can take place in

ammonium carbonate solutions.

CONCLUSION

In summary it can be said that the electron-rich

centres were the active sites for cation adsorption

on the carbon surface. The activity of these

centres could be enhanced either by the

development of negative charges on the acidic

functional groups or the adsorption of anions that

could neutralize positive charges developed as a

result of manifestation of basic character by

carbon. The development of negative centres

could either be due to ionization of the acidic

functional groups at low concentration of the

electrolytes or, more markedly, through the

ionization of these groups by bases.

ACKNOWLEDGEMENT

IRPA support (Project No. 2-06-03-009) IS ap­

preciated.

REFERENCESBADRI M. and K.A. CRO SF.. 1984. Conductance

fcasurcment of Water-sorbed Activated Carbon.Pertanika 7: 59-65.

BADRI M. and K.A. CROUSE. 1989. Changes in SurfacePotential ofActivated Carbon Due to Adsorption ofIons. Pertanika 12(1): 65-70.

BARTON S.S., G.L. BOULTON and B.H. HARRISON.1972. Surface Studies of Graphite: Acidic SurfaceOxides. Carbon 10: 395-400.

BOEH~I H.P. 1966. Chemical Identification of SurfaceGroups. Adv. Catal. 16: 179-274.

BOEH~I H.P., E. DIEHL, W. HECK and R. SAPPOK.

1964. Surface Oxide of Carbon. Angew. Chem.Internat. Edit. 3: 669-677.

BOEH~I H.P and M. VOLL. 1970. BasicheOberflachenoxide AufKohlenstoff - I. Adsorptionvon Sauren. Carbon 8: 227-240.

BRITISH PATE T. 1984.Pertanian Malaysia.

DON:-IET J.B. 1968. The Chemical Reactivity ofCarbons. Carbon 6: 161-176.

EpSTEl:\ B.D., E. DALLE-MoLLE and j.S. MATTSON.1971. Electrochemical Investigations of SurfaceFunctional Groups on Isotropic Pyrolytie Carbon.Carbon 9: 609-615.

GARTON V.A. and D.E. WEISS. 1957. A ewInterpretation of the Acidic and Basic Structure inCarbons II. The Chromene Carbonium Ion Couplein Carbon. Aust. ]. Chem. 10: 309-328.

GI\'E:-I P.H. and L.W. HILL. 1969. Analysis of SurfaceGroups on Carbon Blacks. Carbon 1: 649-658.

HAGI\I'ARA S., H. TAKAHASHI and K. TSUTSUMI.1971. Interaction between Active Hydrogen ofCarbon Surface and Water. Carbon 9: 693-694.

H.\SSLER K. W. 1974. Purification with ActivatedCarbon. Industrial, Commercial and Environmen­tal. p. 187 3rd Ed. Intertext Books.

MATTSON j.S. and H.B. MARK, Jr. 1971. ActivatedCarbon. Surface Chemistry and Adsorption fromSolution. p. 39. N.Y: Marcel Dekker.

PURl B.R. 1966. Chemi orbed Oxygen Evolved asCarbon Dioxide and its Influence on SurfaceReactivity of Carbons. Carbon 4: 391-400.

PURL B.P. and R.C. BANSAL. 1964. Studies in SurfaceChemistry of Carbon Blacks Part I. HighTemperature Evacuations. Carbon 1: 451-455.

RAMLI I., M. BADRl, W.M. DAUD, K.A. CROUSE andSHAHRIN HITAM. 1989. Karbon DiaktifkanSebagai Simpangan Berliang untuk ElektrodRujukan. Prosiding Simposium Kimia AnalisisKebangsaan fll. Jilid l.p. 306-318, UTM, JOHOR.

SHOEMAKER D.P. and C.W. GARLAND 1962.Experiments in Physical Chemistry. p. 445. ewYork: McGraw-Hill.

TSUCHIDA N. and D.M. MUIR. 1986. PotentiometricStudies on the Adsorption ofAu(C J2 and Ag(CNJ2onto Activated Carbon. Metall Trans B 17B: 523­528.

(Received 9 December, 1989).

PERTA IKA VOL. 13 NO.1, 1990 77