Russian Journal of Applied Chemistry, Vol. 75, No. 6, 2002, pp. 894 899. Translated from Zhurnal Prikladnoi Khimii, Vol. 75, No. 6, 2002,
pp. 913 919.
Original Russian Text Copyright 2002 by Tikhomolova, Urakova.
SORPTION
AND ION-EXCHANGE PROCESSES
Interaction of Co(II), Ni(II), and Cu(II) with Quartz
Surface in Aqueous Solutions at Varied pH
K. P. Tikhomolova and I. N. Urakova
St. Petersburg State University, St. Petersburg, Russia
Received November 30, 2001
Abstract Sorption/desorption of Co(II), Ni(II), and Cu(II) on quartz was studied in aqueous solutions at
pH 3.0, 5.0, and 7.3 [for Co(II) and Ni(II)] and 3.0 and 5.0 [for Cu(II)].
Specific adsorption in disperse systems containing ployed the fraction with equivalent particle diameter
aqueous solutions of various metal salts has been a de of 40 70 m. The specific surface area Ssp of this
subject of a great number of experimental and theo- fraction was determined from adsorption of nitrogen
retical works. However, some points related to the
(BET) and Methylene Blue to be 0.20 0.21 m2 g 1.
adsorption mechanism are not understood yet.
Adsorption of the metals on dispersed quartz was
The classical theory of electrical double layer
monitored by the M(II) concentration in the solution
(Gouy s theory) formally treats the resultant action of
bulk, cM. Hereinafter quartz after adsorption is called
two forces, electrostatic (Coulombic) and diffusion,
modified. Also, we determined the amount of ad-
without taking into account the adsorbate (metal cat-
sorbed M(II) per unit area of the adsorbent surface
ion in our case) nature. In other words, this theory
(mol m 2). The M(II) concentration in the solu-
considers nonspecific adsorption. However, in many
tion was determined spectrophotometrically on a
cases the classical theory cannot reproduce experimen-
Specol-10 instrument: Co(II) with Nitroso R salt
tal results adequately. In these cases, the necessity
( = 520 nm), Ni(II) with PAN ( = 540 nm), and
arises to involve specific interactions into considera-
Cu(II) with Xylenol Orange ( = 580 nm). The was
tion, depending on the metal nature and speciation in
estimated from the change in the metal concentration
the solution bulk. Many questions of specific adsorp-
in the aqueous phase. The error of spectrophotometric
tion have been studied insufficiently. To solve these
determinations was within 0.5 3.0%.
questions, new approaches have been developed [1 19].
The working solutions were prepared from M(II)
A series of works have been based on the concept
nitrates. The characteristics of these solutions are as
of the determining role of complexation [7, 9 19],
follows: M(II) concentration 10 4 M; pH 3.0, 5.0, and
with the term complex is used in different senses. The
7.3 [only for Co(II) and Ni(II)]; and ionic strength
chemical (coordination) concept of specific adsorption
from aqueous solutions on oxides is also related to 10 3 M. The ionic strength and the desired pH were
this research [16]. adjusted by adding KNO3, KOH, or HNO3. In addi-
tion to the working solutions, we used reference
In this work, we studied adsorption of double-
solutions with the same pH values and ionic strengths,
charged cations of 3d metals [Co (3d7), Ni (3d8), and
but containing no M(II).
Cu (3d9)]. In the experiments we used quartz (SiO2)
from the Kyshtym deposit (99.8% -quartz). The The solution pH values were chosen was made in
sample preparation procedure was as follows. Quartz view of the following considerations. Firstly, the init-
was ground into a powder in a steel ball mill, washed ial necessary condition was that the adsorbate solution
first with dilute sulfuric acid to negative reaction for should be a true solution, i.e., a solution containing
iron and then with distilled water to neutral reaction, only species of molecular dimensions. The systems
sieved to obtain a desired fraction, dried at about with Co(II) and Ni(II) meet this requirement at all
110 C, and, finally, poured over with distilled water. the three pH values, and in the case of Cu(II), hydrox-
In the adsorption desorption experiments, we em- ide particles are formed at pH 7.3. Note that, accord-
1070-4272/02/7506-0894 $27.00 2002 MAIK Nauka/Interperiodica
INTERACTION OF Co(II), Ni(II), AND Cu(II) WITH QUARTZ SURFACE IN AQUEOUS SOLUTIONS 895
ing to the speciation diagrams given in [18] and con-
structed using potentiometric titration data,1 Co(II)
and Ni(II) occur as aqua ions at the indicated three
pH values, and Cu(II), practically totally as aqua ions
at pH 3.0 and 5.0. Secondly, in the case of Cu(II),
pH 5.0 is only slightly lower than pH 5.8 of dimeri-
zation with the formation of [Cu2(OH)2]2+, and pH 7.3
is lower than the pH of formation of Co(II) and Ni(II)
hydroxides (7.8 and 8.5, respectively). Finally, pH
strongly affects the amounts of negatively charged
Fig. 1. Specific adsorption vs. pH: (1) OH , (2) Co(II),
[ SiO] and neutral active sites [SiOH]0 on the quartz
(3) Ni(II), and (4) Cu(II).
surface. As stated previously [16, 17], these surface
groups can serve as ligands, i.e., they can participate
in the formation of surface complexes with an ad-
sorbed cation as acceptor; [ SiO] being a more re-
active ligand than [SiOH]0.
Taking into account the very high sensitivity of the
surface properties to even extremely slight variations
in the chemical composition of the surface, and to
make comparative analysis of results obtained for
various metals, all the experiments were carried out
with the same quartz sample.
To estimate the degree of dissociation of the quartz
surface sites at various pH values, we used the known Fig. 2. M(II) concentration in the liquid phase, cr (in
ads
percent relative to the metal concentration in the initial
method of continuous potentiometric titration [20].
working solution c0) as a function of the adsorption time
Before starting titration of the quartz dispersion,
tads. Adsorbed ion: (1) Co(II), (2) Ni(II), and (3) Cu(II).
the sample was dried a little on a paper filter, and
The subscripts denote pH of the initial working solution.
then a weighed portion was taken and equilibrated
with a solution at pH 3. The ionic strength of both
layer (stronger in the second case [21]). In all cases,
the solution and the dispersions was 10 3 M (KNO3).
we used in the desorption stage M(II)-free solutions
In the course of titration, pH varied from 3 to 10.
(supporting electrolyte solutions). Publications de-
The amount of OH ions consumed in the course of
voted to the potentialities of the electroosmotic tech-
titration is equal to the amount of the [ SiO] groups
nique for this purpose are extremely scarce [19, 21,
formed on the surface. The potentiometric titration
22]. The main concern of these works is mostly soil
data are presented in Fig. 1.
compaction or consolidation grouting [23].
Before starting the adsorption desorption exper-
In all the adsorption experiments, we used disper-
iment, a quartz sample was poured over with the
sions containing 2 g of a quartz powder and 20 ml
supporting electrolyte solution at a fixed pH. The so-
of fresh working solution. Weighing bottles were
lution was refreshed at regular intervals over a period
hermetically sealed. After a fixed time tads varied from
of 1 2 days. If no specific interaction occurs in the
1 to 7 days, the weighing bottle was opened, and
dispersion at a given pH, this time is enough for the
the solution was removed by decantation and analyzed
formation of an electrical double layer. Then the so-
for pH and M(II). The solid phase (modified quartz)
lution was decanted, the solid phase was washed with
was then used in the desorption experiment.
a fresh portion of the supporting electrolyte, and
the adsorption experiment started at once. In static desorption, modified quartz was placed
in a weighing bottle, poured over with 20 ml of sup-
Adsorption was studied by the static method, and
porting electrolyte at the same pH as in the adsorption
desorption, by the static and electroosmotic displace-
stage, and sealed. Then we used the same procedure
ment methods differing in the effect on the adsorption
as in the adsorption experiments.
1
Since the metal and supporting electrolyte concentrations in In electroosmotic desorption experiments, the
[18] differ from those in this study, to refine M(II) specia- electric field was applied to a diaphragm formed
tion, we performed potentiometric titration. from modified quartz (dc field strength 10 V cm 1).
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 6 2002
896 TIKHOMOLOVA, URAKOVA
3 days, becoming longer with increasing pH; and for
eq
Co(II) and Cu(II), tads =2 4 and 4 h at both pHvalues.
eq
The relationship between tads and the firmness of
metal fixation in the adsorption layer was studied in
the static and electroosmotic experiments with varied
tads. As a parameter characterizing the desorption ef-
ficiency we used cr written as
des
cr = (cdesVdes) 100/(cadsVads),
des
where cdes is the current M(II) concentration in the
course of desorption; Vads and Vdes are the volumes of
the liquid phase in the course of adsorption and de-
sorption, respectively; and cads is the M(II) concentra-
tion in the working solution after adsorption.
According to the static desorption data, tads strong-
ly influences the desorption efficiency. With increas-
ing tads, the firmness of M(II) fixation on the quartz
surface grows; the more so with increasing pH. Fig-
ure 3a shows the time dependences of cr , obtained
des
for Ni(II) in both the static and electroosmotic de-
Fig. 3. (a) Ni(II) concentration in the liquid phase at
sorption experiments (tads 1 h, 1 and 7 days; pH 5.0).
pH 5.0 in the course of the desorption experiment, cr
des
It is clearly seen from the electroosmotic desorption
(in percent relative to the concentration in the working so-
data that, with increasing tads, the same cr is at-
des
lution after the adsorption stage) and (b) the specific ad-
tained in a longer time.
sorption of Co(II), Ni(II), and Cu(II) (mol cm 2) at
pH 5.0 vs. the desorption time tdes . (a) (1 3) Static de-
Figure 3b shows the desorption curves obtained for
sorption and (4 6) electroosmotic displacement. tads: (1,
Co(II), Ni(II), and Cu(II) by the static method at
4) 7 days, (2, 5) 1 day, and (3, 6) 1 h. Adsorbed ion:
pH 5.0 (tads 1 h, 1 and 7 days). The results show that
(1) Co(II), (2) Ni(II), and (3) Cu(II); the same for Fig. 4.
the effect of tads on = (cdesVdes/Ssp), where
des ads
The subscripts denote the adsorption time in the adsorp-
[ = (c0 cads)Vads/mSsp and m is the adsorbent
tion stage, expressed in hours. ads
weight], is considerably lower for Co(II) than for
Ni(II). Note that this effect is quite clearly pronounced
The amount of desorbed metal was estimated from
in the case of dispersions consisting of a Cu(II) solu-
the M(II) concentration in the filtrate and the filtrate
tion at pH 5.0 and modified quartz (tads 1 h, 1 and
volume.
7 days).
Variation of the M(II) concentration in the aqueous
It has been demonstrated by several authors that,
phase, cr (in percent relative to the metal concentra-
ads
after an adsorbate enters the surface reaction zone,
tion in the initial working solution c0) in the course
surface processes can be extended in time relative to
of the adsorption experiment is shown in Fig. 2.
the time of the equilibrium attainment, determined
Adsorption of all the three ions grows with in- using the parameter cads. In particular, this refers to
creasing solution pH. The cr and adsorption rate are
the phenomenon of specific adsorption [21]. Our data
ads
different for different ions. At any pH, the adsorption
on static desorption provide an additional evidence in
rate decreases in the order Ni(II) > Cu(II) > Co(II).
favor of this view (Figs. 2, 3b).
For all the three ions, cr approaches saturation at
Figure 4 shows how the amount of adsorbed M(II)
ads
a sufficiently long adsorption time. The time of equi- varies with time in sorption and in static and electro-
eq
librium attainment, tads, is found to exceed 1 h and
osmotic desorption experiments (all the desorption
to be much longer than that in ion-exchange pro- experiments were performed with quartz modified
cesses, when counterions are fixed in the surface for 7 days).
layer only by the Coulombic forces.
First, the results show that, at any experimental pH,
The time of equilibrium attainment decreases in the limiting amount of adsorbed M(II) increases in
eq
the order Ni(II) > Cu(II) > Co(II). For Ni(II) tads = 1 the order Co(II) < Cu(II) < Ni(II) (Fig. 4a).
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 6 2002
INTERACTION OF Co(II), Ni(II), AND Cu(II) WITH QUARTZ SURFACE IN AQUEOUS SOLUTIONS 897
Let us now analyze the most significant features of
desorption (Figs. 4b, 4c). Under any experimental
conditions studied, the highest desorption efficiency is
observed for Co(II). In static desorption, the residual
amount of Co(II) adsorbed on the quartz surface was
estimated to be about 5, 20, and 45% at pH 3.0, 5.0,
and 7.3, respectively. Since the electroosmotic dis-
placement involves a more severe treatment, the fact
of rapid total desorption of the metal ions is quite
natural [19, 21, 22]. Therefore, our results show that
the bonds formed by Co(II) with the surface sites of
quartz are the weakest among the metals studied.
As for Ni(II), Fig. 4b shows that static desorption
of Ni(II) is rapid, but inefficient process [residual
amount of adsorbed Ni(II) is 80, 85, and 87% at
pH 3.0, 5.0, and 7.3, respectively]. Only electro-
osmotic displacement provides total desorption of
Ni(II), but it requires longer time than that for Co(II),
suggesting firmer fixation of Ni(II). Therefore, Ni(II)
forms fairly strong bonds with the active sites of
quartz at any pH studied.
The parameters of static desorption of Cu(II) at
pH 5.0 virtually coincide with those of Ni(II), and at
pH 3.0, they take an intermediate position between
Co(II) and Ni(II).
Similarly to the other metals investigated, Cu(II)
is totally desorbed by the electroosmotic method, re-
gardless of pH.
In addition to the potentiometric titration data,
Fig. 1 also shows the pH dependences of the amounts
of all the adsorbed cations. It is clear that the posi-
tive charge due to adsorption of M(II) at the quartz
solution interface ( e+ ) is twice the product NA.
M M
Recalculation of (mol m 2) given in Fig. 1 to
the corresponding amount of the positive charge in
Fig. 4. Amount of M(II) adsorbed on the unit surface area
EDL and its comparison with the amount of the neg-
of quartz, , vs. (a) adsorption time tads and (b, c) de-
ative charges localized on the surface groups [ SiO] sorption time tdes at tads = 168 h. Desorption method:
( e = NA) show that e > e+ in all (b) static and (c) electroosmotic. Subscripts denote pH of
SiO M
the initial working solution.
cases.
In this connection, let us estimate the significance
Evidently, the M(II) aqua cations are bound to
of the [ SiO] and [ SiOH]0 surface centers as centers
primarily [ SiO] groups on the quartz surface. At
determining the electric field of the surface layer and
the same time, the interaction of M(II) with [ SiOH]0
also as ligands of the surface complexes. Taking into
cannot be ruled out. Also, the forces retaining the
account the inequality given above, it may be sug-
adsorbate at the [ SiO] active site cannot be restricted
gested that the [ SiO] sites contribute significantly to
to the Coulombic interactions only. The following
the specific adsorption through an essentially Coulom- facts additionally count in favor of the specific ad-
bic interaction M(II) [ SiO] . The role of the Cou- sorption of M(II) on the active site of the quartz sur-
lombic forces, as the longest-range forces, is restricted face. In solutions with pH 3.0 and 5.0, in which the
to attraction of the M(II) ions to the zone of action of initial Ni(II) and Cu(II) concentrations were lower
short-range forces responsible for specific adsorption. by a factor of 10 than the total concentration of all
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 6 2002
898 TIKHOMOLOVA, URAKOVA
other cations, the electrical neutrality of EDL is Note also that, for d9 complexes of Cu(II), the most
provided by considerably higher amounts of the Ni(II) typical structures of the inner coordination sphere
and Cu(II) ions, compared to the amounts correspond- are square-planar and octahedral.
ing to their adsorption as indifferent ions. Indeed, it
follows from the recalculated data of Fig. 1 that in
CONCLUSIONS
the systems containing Ni(II) and, to a lower but also
remarkable extent, in Cu(II)-containing systems, the
(1) The adsorption efficiency of Co(II), Ni(II),
ratio |e+ |/ |e | is considerably above 0.1 0.2. More-
and Cu(II) on quartz increases in the order Co(II) <
M
over, in the case of Ni(II), this ratio only slightly dif-
Cu(II) < Ni(II), growing with pH. The time of attain-
fers from 0.85 at pH 3.0 and 0.8 at pH 5.0; the cor-
ment of the adsorption equilibrium varies in the same
responding values for Cu(II) are 0.5 and 0.6.
order, ranging from 2 h to 3 days.
At present, the coordination chemistry in aqueous
(2) Desorption of M(II) from quartz depends on
solution is extensively studied. There is a great body
a desorption procedure used. In the static desorption,
of data on the effect of the nature of metals and li-
the desorption efficiency of Co(II) is higher than that
gands on the strength and structure of coordination
of Ni(II) at all the experimental pH values and than
bonds M L. To our knowledge, such data on surface
that of Cu(II) at pH 5.0. At pH 3.0 the static desorp-
complexes are lacking in the literature.
tion of Cu(II) is lower than that of Co(II), but higher
than that of Ni(II). In electroosmotic replacement,
At the same time, the force field at the interface
all the cations are totally desorbed quite rapidly (in
is very asymmetrical. Its strength drops from a high
at least 2.5 h).
value to zero over molecular ranges [19]. Therefore,
we analyzed the correlation of the experimental data
(3) In both static and electroosmotic desorp-
obtained with coordination model of adsorption, by
tion modes, the desorption efficiency depends on the
analogy with consideration of the effect of metal
adsorption time. In static desorption, the metal cations
ions and ligands on the formation of the bulk and
adsorbed in 1 h are desorbed more efficiently that at
surface complexes. Of course, this is not the literal
an adsorption time of 1 and 7 days. In electroosmotic
meaning. We only point out the known general trends
displacement mode, the same equilibrium cr is at-
des
in the effect of the nature of M(II) and L on the prop-
tained in longer time with increasing tads.
erties of the surface complexes formed.
The observed relatively weak fixation of Co(II) on ACKNOWLEDGMENTS
the quartz surface can be attributed to the well-known
fact that the coordination numbers 4 and 6, corre- The study was supported financially by the Russian
sponding to the tetrahedral and octahedral surround- Foundation for Basic Research (project no. 00-15-
ings of Co(II), respectively, are the most typical of
97 357).
this cation. Complexes of Co(II) with monodentate
ligands, having an oxygen atom as a donor center, are REFERENCES
mostly labile [24 26].
1. Bibik, E.E., Zh. Prikl. Khim., 1999, vol. 72, no. 6,
The desorption results obtained by the static
pp. 916 920.
method suggest that, in the case of Ni(II), donor ac-
2. Yates, D.E., Levin, S., and Healy, T.W., J. Chem.
ceptor bonds must be relatively strong. For d8 com-
Soc., Faraday Trans., 1974, no. 10, pp. 1807 1812.
plexes of Ni(II), both coordination numbers 4 (tetra-
3. James, R.O., Stiglich, R.J., and Healy, T.W., J. Col-
hedral or square-planar structure) and 6 (octahedron)
loid Interface Sci., 1972, vol. 40, no. 1, pp. 42 45.
are typical. According to the crystal field theory, li-
4. Damaskin, B.B., Gorichev, I.G., and Batrakov, V.V.,
gand substitution in d8 complexes by the SN1 or
Elektrokhimiya, 1990, vol. 26, no. 4, pp. 400 406.
SN2 mechanism is slow [27]. Most likely, Ni(II) com-
5. Matievich, E., J. Colloid Interface Sci., 1973, vol. 43,
plexes formed on the -quartz surface have square-
no. 2, pp. 217 245.
planar or octahedral configuration.
6. Chernoberezhskii, Yu.M., Bykova, N.I., and Yan-
In the systems with Cu(II), like in those with klovich, A.I., in Termodinamika i struktura gidrokso-
Ni(II), the lack of noticeable desorption in the static kompleksov v rastvorakh (Thermodynamics and Struc-
mode suggests the formation of relatively strong do- ture of Hydroxo Complexes in Solutions), Leningrad:
Leningr. Gos. Univ., 1983, pp. 152 156.
nor acceptor bonds between Cu(II) as a central atom
and the oxygen atom of the surface ligand [ SiO] . 7. Gorichev, I.G., Batrakov, V.V., and Shaplygin, I.S.,
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 6 2002
INTERACTION OF CO(II), NI(II), AND CU(II) WITH QUARTZ SURFACE IN AQUEOUS SOLUTIONS 899
Izv. Ross. Akad. Nauk, Neorg. Mater., 1994, vol. 30, 18. Baes, C.F. and Mesmer, R.E., The Hydrolysis of Cat-
no. 10, pp. 1219 1234. ions, New York: Wiley, 1976.
8. Lyklema, J., Adsorption from Solution at the Solid/ 19. Tikhomolova, K.P., Elektroosmos (Electroosmosis),
Liquid Interface, Parfitt, G.D. and Rochester, C.H., Leningrad: Khimiya, 1989.
Eds., London: Academic, 1983. Translated under the
20. Ahmed, S.M., in Oxide and Oxide Films, New York:
title Adsprbtsiya iz rastvorov na poverkhnostyakh
Wiley, 1972, pp. 319 517.
tverdykh tel, Moscow: Mir, 1986, pp. 261 293.
21. Tikhomolova, K.P., Kaufman, Yu.V., and Urako-
9. Devis, I.A., James, R.D., and Leckie, J.O., J. Colloid
va, I.N., Zh. Prikl. Khim., 2001, vol. 74, no. 8,
Interface Sci., 1978, vol. 63, no. 3, pp. 480 499.
pp. 1258 1264.
10. Sparnay, M.J., The Electrical Double Layer, Oxford:
22. Tsukanova, V.M., Interaction of Aqua and Hydroxo
Pergamon, 1972.
Species of Pb(II), Cu(II), and Hg(II) with the SiO2
11. Lygin, V.I., Farikov, A.Dzh., and Tugushev, R.E.,
Surface in Context of the Chemical Concept of Spe-
Zh. Fiz. Khim., 1995, vol. 69, no. 2, pp. 315 317.
cific Adsorption, Cand. Sci. (Chem.) Dissertation,
St. Petersburg, 1996.
12. Huang, C.P. and Stumm, W., J. Colloid Interface Sci.,
1973, vol. 43, pp. 409 420.
23. Zhinkin, G.N., Elektroosmoticheskoe zakreplenie
gruntov v stroitel stve (Electroosmotic Grouting of
13. Stumm, W., Huang, C.P., and Jenkis, S.R., Croat.
Grounds in Construction), Leningrad: Stroiizdat, 1966.
Chem. Acta, 1970, vol. 43, no. 2, pp. 223 245.
24. Cotton, F.A. and Wilkinson, G., Basic Inorganic
14. Stumm, W., Huang, C.P., and Dalang, F., Croat.
Chemistry, New York: Wiley, 1976.
Chem. Acta, 1976, vol. 48, no. 4, pp. 491 504.
25. Kukushkin, Yu.N., Khimiya koordinatsionnykh soedi-
15. Shindler, P.W., Furst, B., Dick, R., and Wolf, P.U.,
nenii (Chemistry of Coordination Compounds), Mos-
J. Colloid Interface Sci., 1976, vol. 55, no. 3,
cow: Vysshaya Shkola, 1985.
pp. 469 475.
16. Aleksandrova, L.K. and Tikhomolova, K.P., Kol- 26. Spice, J.E., Chemical Binding and Structure, New
loidn. Zh., 1988, vol. 50, no. 1, pp. 100 107. York: MacMillan, 1964.
17. Tikhomolova, K.P. and Tsukanova, V.M., Zh. Prikl. 27. Cotton, F.A. and Wilkinson, G., Advanced Inorganic
Khim., 1997, vol. 70, no. 3, pp. 353 370. Chemistry, New York: Interscience, 1966.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 6 2002