Journal of Colloid and Interface Science 224, 265–271 (2000)
doi:10.1006/jcis.2000.6721, available online at http://www.idealibrary.com on
Adsorption of Gemini and Conventional Cationic Surfactants onto
Montmorillonite and the Removal of Some Pollutants by the Clay
Fang Li and M. J. Rosen1
Surfactant Research Institute, Brooklyn College of the City University of New York, Brooklyn, New York 11210
Received August 2, 1999; accepted January 6, 2000
The adsorption of a series of gemini surfactants, [Cn H2n+1 N+
(CH3 )2 -CH2 CH2 ]2 · 2Br− , where n = 10, 12, 14, and 16, on clay (Na-
montmorillonite) from their aqueous solution in 0.01 M KBr and
the effect of this adsorption on the removal of 2-naphthol and 4-
chlorophenol have been studied. Compared to those of conventional
cationic surfactants with similar single hydrophilic and hydropho-
bic groups (Cn H2n+1 N+ (CH3 )3 · Br− , where n = 10, 12, 14, and 16),
the molar adsorptions of the gemini and conventional surfactants
are almost identical. This indicates that only one of the hydrophilic
groups in the gemini molecule is adsorbed onto the clay and that
the second hydrophilic is presumably oriented toward the aqueous
phase, in contrast to the adsorption of the conventional surfactants,
where the hydrophobic group is oriented toward the aqueous phase.
Stability studies on dispersions of clay treated with the two types
of surfactants confirm this. The slight increase in the moles of sur-
factant to values above the CEC of the clay with an increase in the
carbon number of the hydrophobic chain indicates that adsorption
through hydrophobic group interaction occurs in addition to the
major ion exchange. Adsorption studies of the pollutants onto the
clay treated by either the gemini or the conventional surfactants
show that the former are both more efficient and more effective at
removing the pollutants from the aqueous phase. °C 2000 Academic Press
Key Words: gemini surfactants; adsorption; montmorillonite; 2-
naphthol; 4-chlorophenol; pollutant removal; cationic surfactants;
clay.
INTRODUCTION
In recent years, cationic surfactants, especially quaternary am-
monium cationic surfactants, have been used to treat soil and
then remove pollutants from aqueous media (1–8). Since gem-
ini surfactants (surfactants containing two hydrophilic and two
hydrophobic groups in the molecule) have been shown to be
much more surface-active than conventional surfactants contain-
ing a single similar hydrophilic and hydrophobic group in the
molecule (9, 10), there was interest in determining whether the
former would be more efficient and/or effective than the latter
in removing pollutants from aqueous media. Although the ad-
sorption of gemini surfactants at the air/solution interface has
been much investigated (11), little attention has been paid to the
adsorption of gemini surfactants at the solid/solution interface.
1 To whom correspondence should be addressed.
265
So far as we know, only two papers (12, 13) involving the silica/
aqueous solution interface concern the adsorption of gemini sur-
factants at the solid/aqueous solution interface. In Ref. (13),
the gemini surfactants used by Zana et al. are alkanediyl-α, ω-
bis(dodecyldimethylammonium bromide), with the alkane-diyl
spacer group C2 H4 , C4 H8 , C6 H12 , and C10 H20 . They studied
mainly the influence of the spacer on the amount of surfactant
adsorbed and discussed the adsorption mechanism. They did not
study the effect of the gemini surfactants on the removal of any
pollutant. The only reported study of the effect of gemini sur-
factants on the removal of organic contaminants is that of Esumi
et al. (12), who used one gemini surfactant to treat silica and
studied the effect of this gemini surfactant on the removal of
2-naphthol. The only gemini surfactant they studied is ethane-
1,2-bis(dodecyldimethylammonium bromide). They found that
the molar amount of gemini surfactant adsorbed on silica was
lower than for the corresponding conventional surfactant, dode-
cyltrimethylammonium bromide, but that the ratio of the max-
imum amount of 2-naphthol adsorbed to the adsorbed amount
of surfactant on silica increases from conventional surfactant to
gemini surfactant.
The adsorption of surfactant onto silica is not strong, around
5 × 10−5 mol/g silica (12, 13). In this study, we chose Na-
montmorillonite, which has a much larger cation exchange
capacity (CEC = 7.64 × 10−2 eq/100 g). Adsorption of a series
of cationic gemini surfactants, [Cn H2n+1 N+ (CH3 )2 -CH2 CH2 ]2 ·
2Br− , where n = 10, 12, 14, and 16, onto the montorillonite is
investigated, and the effect of this adsorption on the removal of
two pollutants, 2-naphthol and 4-chlorophenol, is noted. These
two pollutants are widely used in surfactant-concerned envi-
ronmental studies and can be easily analyzed by UV spectro-
photometry (7, 8, 12). The adsorption and removal of the pol-
lutants are compared to those of conventional cationic
surfactants with similar single hydrophilic and hydrophobic
groups (Cn H2n+1 N+ (CH3 )3 · Br− , where n = 10, 12, 14, and 16).
EXPERIMENTAL PROCEDURES
Materials
Synthesis of the gemini surfactant. A 0.01 mol amount
of 1,4-dibromobutane (95%) and 0.022 mol of alkyl N,N-
dimethylamine were added to isopropanol, and the mixture was
0021-9797/00 $35.00
Copyright °C 2000 by Academic Press
All rights of reproduction in any form reserved.
266 LI AND ROSEN

heated to reflux for about 12 h. After the reaction, the solvent

was removed and the residue was recrystallized from tetrahy-
drofuran (THF) several times. The pure diquaternary ammonium
dibromide was finally obtained as a colorless powder by recrys-
tallization three times from ethanol/acetate mixed solvent. The
gemini surfactants are referred to as C10 gemini, C12 gemini,
C14 gemini, and C16 genimi, based on their hydrophobic chain
lengths. The structure of the gemini surfactants is
[R–N+ (CH3 )2 –CH2 –CH2 –CH2 –CH2 –N+ (CH3 )2 –R] · 2Br− ,
where R is the alkyl chain.
Nitrogen analysis data for two gemini surfactant are as fol-
lows: C12 gemini, N = 4.36% (calcd.), 4.32% (found); C16 gem-
ini, N = 3.71% (calcd.), 3.86% (found).
The conventional surfactants used in this study are decyl-
trimethylammonium bromide (C10 TMAB), dodecyltrimethyl-
ammonium bromide (C12 TMAB), tetradecyltrimethylam-
FIG. 1. Surface tension vs log of surfactant molar concentration for the
monium bromide (C14 TMAB), and hexadecyltrimethylam- gemini surfactants at 25◦ C in 0.01 M KBr: (m) C10 gemini; (r) C12 gemini;
monium bromide (C16 TMAB). (d) C14 gemini; (w) C16 gemini.
The clay [Na-montmorillonite (Wyoming)] was obtained
from Professor Stephen Aja, Department of Geology, Brook-
lyn College of CUNY. The surface area was 31.82 ± 0.22 m2 /g The amount of surfactant adsorbed onto the montmorillonite
[measured by BET(N2 )]. The cation exchange capacity (CEC) and the amount of either pollutant adsorbed onto the surfactant-
is 76.4 meq/100 g. treated montmorillonite is given by the following equation:
Surface tension measurements. Measurements were per-
(C 0 − Ceq ) × V
formed at 25 ± 0.1◦ C on a K-12 tensiometer, by use of the ns = , [1]
Wilhelmy plate technique, with a sandblasted platinum blade g
of ca. 4 cm perimeter. The instrument was calibrated against where n s is the number of moles of adsorbate (surfactant or
quartz-condensed double-distilled, previously deionized water pollutant) per gram of adsorbent (mol/g), C 0 is the initial con-
(the last distilled stage from alkaline KMnO4 through a 1-m high centration of adsorbate (mol/dm3 ), Ceq is the equilibrium con-
Vigreaux column). Values were taken until the surface tension centration of adsorbate (mol/dm3 ), V is the volume of solution
was constant for 0.5 h. Reproducibility of the surface tension (dm3 ), and g is the weight of adsorbent (clay) (g).
measurement is less than 0.2 mN/m.
Adsorption isotherm of surfactants on clay and the adsorp-
tion of the pollutants on surfactant-treated clay. A series of
aqueous surfactant solutions in 0.01 mol/dm3 KBr aqueous so-
lution was prepared in the absence and presence of either 2-
naphthol or 4-chlorophenol. The initial concentration of either
pollutant was 4 × 10−4 mol/dm3 . A 50-ml amount of the solu-
tion was added to 50 mg of clay in a capped centrifuge tube.
The solution was equilibrated by shaking for at least 24 h, and
then the solid was removed by centrifugation. Since the mont-
morillonite, when stirred with the aqueous phase containing no
surfactant, did not reduce the surface tension of the aqueous
phase, the surface tension of the supernatant solution in equi-
librium with the clay can be used to measure the concentration
of the surfactants in the solution. In addition, the concentra-
tion of the surfactant in the aqueous phase was measured by the
mixed indicator two-phase titration method (14, 15), using the
cationic or gemini solution as the titrant. The same results were
obtained by either method. The concentration of either pollutant FIG. 2. Surface tension vs log of surfactant molar concentration for the con-
in the aqueous phase was determined by UV spectrophotometry ventional surfactants at 25◦ C in 0.01 M KBr: (m) C10 TMAB; (r) C12 TMAB;
(Kontron Instruments, Unikon 9410). (d) C14 TMAB; (w) C16 TMAB.
 ADSORPTION OF SURFACTANTS ON MONTMORILLONITE 267

RESULTS AND DISCUSSION the form (16a):

µ ¶
Adsorption of the Gemini and Conventional Surfactants 1 × 10−3 ∂γ
0=− [2]
at the Air/Aqueous Solution and Montmontillonite/ 2.303n RT ∂ log C T
Aqueous Solution Interfaces
and
The surface tension (γ )–log surfactant molar concentration 1020 2
(C) curves at 25◦ C in 0.01 M KBr, used to determine the con- Amin = (in Å ), [3]
N 0max
centration of surfactants in the aqueous phase in equilibrium
with the adsorbed surfactant, are shown in Fig. 1 for gemini where γ is the surface tension in mN/m; 0 is the absorption
surfactants and in Fig. 2 for the conventional surfactants. amount in mol/m2 ; [∂γ /(∂ log C)]T is the slope in each case;
The concentration of the surfactant at the air/solution inter- T is absolute temperature, R = 8.314 J mol−1 K−1 ; and N is
face, 0, and the minimum area, Amin , occupied by the surfactant Avogadro’s number. The value of n (the number of species at
there can be determined from the Gibbs adsorption equation in the interface whose concentration at the interface changes with

FIG. 3. Adsorption isotherms of the conventional surfactants and the gemini surfactants on montmorillonite at 25◦ C in 0.01 M KBr. (a) C10 TMAB and
C10 gemini; (b) C12 TMAB and C12 gemini; (c) C14 TMAB and C14 gemini; (d) C16 TMAB and C16 gemini.
268 LI AND ROSEN

TABLE 1
Maximum Surfactant Adsorption onto Montmorillonite at 25◦ C from 0.01 M KBr and the Stability of the Resulting Dispersion

CMCa Amin
a Cs (max) Ceq (max)c Initial absorbance t1/2
Surfactant (mol/dm3 ) (A2 ) (mol/g Clay) (mol/dm3 ) (A0 ) (min) Abs/A0

C10 TMAB 6.22 × 10−2 70.6 7.43 × 10−4 1.82 × 10−2 0.17 9.96 /
C10 gemini 7.95 × 10−4 101.6 7.20 × 10−4 6.59 × 10−3 0.73 22.1 0.41
C12 TMAB 9.29 × 10−3 67.6 8.13 × 10−4 3.61 × 10−3 0.47 11.4 /
C12 gemini 1.64 × 10−4 89.8 7.54 × 10−4 1.43 × 10−3 1.56 >60 0.61
C14 TMAB 2.47 × 10−3 63.4 8.27 × 10−4 6.47 × 10−4 0.60 12.6 /
C14 gemini 7.82 × 10−6 21.7 8.12 × 10−4 1.90 × 10−4 1.84 >60 0.69
C16 TMAB 8.20 × 10−4 59.3 8.92 × 10−4 1.64 × 10−4 0.75 13.0 /
C16 geminib 2.25 × 10−6 9.3 / / / / /

a Literature values at 30◦ C in 1.25 × 10−2 M KBr [Trap, H. J. L., and Hermans, J. J., Koninki Ned Akad Weten. Proc. Ser B 58, 97 (1955)]: C TMAB, 5.9 × 10−2 ;
10
C12 TMAB, 1.08 × 10−2 ; C14 TMAB, 2.1 × 10−3 .
b Maximum amount adsorbed on clay cannot be reached due to poor solubility of surfactant in water.
c C (max) is the equilibrium surfactant concentration in solution phase at maximum amount adsorbed.
eq

change in C) is taken as 1 for the gemini surfactants because
the surfactant concentrations used to calculate Amin are much
less than one-tenth of the ionic strength (i.e., 0.01 M). For the
conventional surfactants, n is taken as 1 when the surfactant
concentrations used to calculate Amin are less than one-tenth of
0.01 M, otherwise, the value of n can be obtained by use of the
following equation (17):
Csurf
n=1+ , [4]
Csurf + I S
where IS is the ionic strength of the added electrolyte. Csurf is
the average surfactant concentration used for [∂γ /(∂ log C)]T .
The slopes of the plots in Fig. 1 indicate that the concentra-
tion of surfactant at the air/aqueous solution interface increases
as the number of carbons in the alkyl chain increases. Table 1
shows that the Amin values decrease with increase in the length
of the alkyl chain for the gemini surfactants, the same as for
the conventional surfactants studied here and for other conven-
tional surfactants (16b). The very large slopes (and the very
small Amin values) for the C14 and C16 gemini surfactants have
been observed before and may be due to multilayer formation
of the surfactants at the air/solution interface (18).
Adsorption isotherms of the conventional surfactants and
the geminis onto montmorillonite at 25◦ C from aqueous 0.01
M KBr solution are shown in Fig. 3a–d. Data are listed in
Table 1.

FIG. 4. Time dependence of the absorbance of the clay suspension after

treatment by the conventional and gemini surfactants. (1) C14 gemini; (2) FIG. 5. Highly simplified representation of the expected adsorption for
C14 TMAB; (3) C12 gemini; (4) C12 TMAB; (5) C10 gemini; (6) C10 TMAB; both cationic gemini surfactants and conventional surfactants on the clay surface.
(7) C16 TMAB. (A) Conventional surfactant; (B) gemini surfactant.
 ADSORPTION OF SURFACTANTS ON MONTMORILLONITE 269

FIG. 6. Adsorption of 2-naphthol and 4-chlorophenol onto surfactant-treated clay at 25◦ C in 0.01 M KBr. (a) C10 TMAB and C10 gemini; (b) C12 TMAB and
C12 gemini; (c) C14 TMAB and C14 gemini; (d) C16 TMAB and C16 gemini.

The typical S-shaped isotherm (16c) is apparent only in
the case of C10 TMAB. For the other surfactants, the amount
adsorbed on the clay increases sharply to a maximum with
increase in equilibrium surfactant concentration. For both the
gemini and the conventional surfactants, the maximum amount
of surfactant at the clay/aqueous solution interface, Cs (max),
increases with the length of hydrophobic chain. The maxi-
mum amount of surfactant absorbed onto the clay is larger
than the CEC of the clay in the case of C12 TAMB, C12 gemini,
C14 TMAB, C14 gemini, and C16 TMAB. This indicates that al-
though cation exchange plays a major key role in the adsorption,
hydrophobic bonding between chains also occurs and becomes
stronger as the hydrophobic chains become longer. The data on
the surfactant concentration in the solution phase Ceq (max) at
which this maximum adsorption of surfactant, Cs (max), occurs
are shown on Table 1. Noteworthy is the much lower concen-
tration of gemini, about 13 that of the conventional surfactant
with the same alkyl chain length, needed to achieve this max-
imum adsorption. This indicates that the gemini surfactant is
much more efficient than the conventional surfactant at achiev-
ing this maximum adsorption on the montmorillonite. The data
also show that, in spite of the fact that the gemini have two
quaternary nitrogen groups that can adsorb onto negative sites
onto the montmorillonite, the molar adsorptions of the gemini
and conventional surfactants are almost identical. This indicates
that only one of the hydrophilic groups in the gemini molecule is
adsorbed onto the clay and that the second hydrophilic group is
presumably oriented towards the aqueous phase. This mode of
adsorption would be in marked contact to the adsorption of the
conventional surfactants, which would also have their quaternary
270 LI AND ROSEN

nitrogens oriented toward the clay, but with their hydrophobic TABLE 2
groups oriented towards the aqueous phase. Pollutant Adsorption onto Surfactant-Treated Montmorillonite
The adsorption of the gemini surfactants onto the montmo-
Max. amount Pollutant/
rillonite would consequently make the surface of the latter hy- of pollutant surfactant
drophilic, while the adsorption of the conventional surfactants adsorbed ratio
would make the surface of the latter hydrophobic. That this is Pollutant Surfactant Cs (mol/g) (mol/g) (mol/mol)
indeed the case is shown by the absorbance in the visible region
of aqueous suspensions of the clay with the maximum amount 2-Naphthol C10 TMAB 7.43 × 10−4 1.48 × 10−4 0.20
C10 gemini 7.20 × 10−4 2.61 × 10−4 0.36
of surfactant adsorbed onto it (Fig. 4). After the clay suspension C12 TMAB 8.13 × 10−4 1.73 × 10−4 0.20
was shaken for 24 hrs, its absorbance was immediately mea- C12 gemini 7.54 × 10−4 2.95 × 10−4 0.38
sured in the visible at 580 nm. From the absorbance data, the C14 TMAB 8.27 × 10−4 2.09 × 10−4 0.25
clay suspension with the adsorbed gemini surfactant is much C14 gemini 8.12 × 10−4 3.14 × 10−4 0.38
more stable than that with the adsorbed conventional surfactant. C16 TMAB 8.92 × 10−4 2.40 × 10−4 0.27
C16 geminia / / /
In the case of the clay with adsorbed conventional surfactant, the
p-Chlorophenol C10 TMAB 7.43 × 10−4 4.33 × 10−5 0.058
absorbance rapidly decreases with time, indicating rapid floccu-
C10 gemini 7.20 × 10−4 1.68 × 10−4 0.24
lation of the clay as a result of its hydrophobic surface. In the C12 TMAB 8.13 × 10−4 7.01 × 10−5 0.084
case of the clay with adsorbed gemini surfactant, the solution C12 gemini 7.54 × 10−4 2.08 × 10−4 0.28
remains turbid for an extended period, due to dispersion of the C14 TMAB 8.27 × 10−4 1.10 × 10−4 0.13
clay in the aqueous phase as a result of the positive charge on C14 gemini 8.12 × 10−4 2.42 × 10−4 0.29
C16 TMAB 8.92 × 10−4 1.39 × 10−4 0.16
the surface of the particles and the hydrophilic surface character
C16 geminia / / /
imparted by the second hydrophilic group oriented away from
the clay surface. Figure 5 illustrates the expected gemini adsop- a Maximum amount adsorbed on clay can not be reached due to poor solubility
tion compared with that of conventional surfactant. The initial of surfactant in water.
absorbance, A0 , and t1/2 (the time when the absorbance reduces
to half of its initial value) of the gemini-treated clay suspen-
sion are both larger than those of its corresponding conventional factants with the carbon number of their hydrophobic chain, the
surfactant (Table 1). maximum adsorbed amount and ratio of the amount of pollutant
As a result of the increased adsorption with increase in the adsorbed to the mole of surfactant adsorbed also increase with
length of the hydrophobic group to values above the CEC, the the carbon number of the surfactant’s alkyl chain. The decrease
initial absorbance, t1/2 , and Abs/A0 (where Abs is the absorbance of the adsorption amount of either pollutant with increase in the
value at 60 min) increase from C10 TMAB to C16 TMAB and from equilibrium surfactant above its critical micelle concentration
C10 gemini to C14 gemini (Table 1). after the maximum adsorption amount has been reached is pre-
sumably due to solubilization of the pollutant in the micelles in
Adsorption of 2-Naphthol and 4-Chlorophenol the aqueous phase (12).
onto Montmonrillonite with Adsorbed Layers This increased adsorption of the pollutant onto the gemini,
of Gemini or Conventional Surfactant compared to that of the conventional surfactant, shows that the
Adsorption isotherms of 2-naphthol and 4-chlorophenol onto former is more effective than the latter in removing pollutants
montmorillonite with the maximum amount of the adsorbed from aqueous media. The observation that the maximum ad-
gemini or conventional surfactant on it are shown in Fig. 6a–d, sorption of the gemini surfactant onto the clay occurs at about 13
and the data are listed in Table 2. They show that the amount the concentration of the conventional surfactant in the solution
of 2-naphthol adsorbed per gram of clay at the point where the phase (Table 1) shows that the gemini is also more efficient at
surfactant is adsorbed at its maximum is 1.5 times for the gem- removing the pollutant from aqueous media.
ini surfactant, compared to that adsorbed onto the conventional
surfactant. More than twice as much 4-chlorophenol is adsorbed REFERENCES
onto the gemini-treated clay than onto the conventional-treated
1. Boyd, S. A., Lee, J.-F., and Mortland, M. M., Nature 333, 345
clay, at the point where the amount of adsorbed surfactant is max-
(1988).
imum. Table 2 also shows the moles of pollutant adsorbed per 2. Xu, S.-H., and Boyd, S. A., Environ, Sci. Technol. 29, 312 (1995).
mole of surfactant adsorbed, at this point of maximum adsorp- 3. Lee, J.-F., Crum, J. R., and Boyd, S. A., Environ. Sci. Technol. 23, 1365
tion of the surfactant. Again, the ratio of the amount of pollutant (1989).
adsorbed to the mole of surfactant adsorbed is 1.5–1.9 times on 4. Nye, J. V., Guerlin, W. F., and Boyd, S. A., Environ. Sci. Technol. 28, 944
(1994).
the gemini surfactant than on the conventional surfactant in the
5. Smith, J. A., Jaffe, P. R., and Chiou, C. T., Environ. Sci. Technol. 24, 1167
case of 2-naphthol and 2–3 times in the case of 4-chlorophenol. (1990).
For both the conventional and the gemini surfactants, because 6. Zhang, Z.-Z., Sparks, D. L., and Scrivner, N. C., Environ. Sci. Technol. 27,
of the slight increase of the maximum adsorbed amount of sur- 1625 (1993).
 ADSORPTION OF SURFACTANTS ON MONTMORILLONITE
7. Lumpp, E. K., Heitmann, H., and Schwuger, M. J., Colloids Surf. A 78, 93
(1993).
8. Schieder, D., Dobias, B., Klumpp, E., and Schwuger, M. J., Colloids Surf.
A 88, 103 (1994).
9. Rosen, M. J., Chem. Technol. 30 (1993).
10. Zana, R., Curr. Opin. Colloid Interface Sci. 1, 566 (1996).
11. Rosen, M. J., and Tracy, D. J., J. Surfactants Detergents 1, 547 (1998).
12. Esumi, K., Goino, M., and Koide, Y., J. Colloid Interface Sci. 183, 539
(1996).
271
13. Chorro, C., Chorro, M., Dolladille, O., and Zana, R., J. Colloid Interface
Sci. 199, 169 (1998).
14. Reid, V. W., Alston, T., and Heinerth, E., Tenside 4, 292 (1967).
15. Reid, V. W., Alston, T., and Heinerth, E., Tenside 5, 90 (1967).
16. Rosen, M. J., in “Surfactants and Interfacial Phenomena,” 2nd edition. John
Wiley, New York, 1989: (a) pp. 65–68; (b) pp. 70–80; (c) p. 48.
17. Matijevic, E., and Pethica, B. A., Trans. Faraday Soc. 54, 1382 (1958).
18. Rosen, M. J., Mathias, J. H., and Davenport, L., Langmuir 15, 7340
(1999).