Professional Documents
Culture Documents
The effects of the adsorption modes of several low molecular weight (LMW) organic anions (maleate,
oxalate, and citrate) on the colloidal stability of corundum-water suspensions have been examined using
electrokinetic and shear yield stress (τy) measurements over a broad range of pH conditions and LMW
organic anion concentrations. Consistent with previous studies, increasing concentrations of maleate,
oxalate, and citrate progressively shift the electrokinetic isoelectric point and pH of the maximum shear
yield stress (τy,max) to more acidic conditions. Due to its predominant electrostatic driving force for adsorption,
outer-spherically adsorbed maleate possesses a very limited ability to charge reverse the corundum-water
interface or bind to the negatively charged corundum surface. By contrast, inner-spherically adsorbed
oxalate and citrate can significantly charge reverse the corundum-water interface, with the extent of
charge reversal being related to the relative binding strengths of the oxalate and citrate anions. Adsorbed
maleate, oxalate, and citrate generate steric barriers to interparticle approach, leading to substantial
reductions in the magnitude of τy,max at low to intermediate concentrations of those LMW anions. At the
highest anion concentrations investigated, however, increases in τy,max are observed, and can be attributed
to the formation of bridging Al(III)-organic surface precipitates, as suggested by in situ attenuated total
reflectance Fourier transform infrared spectroscopic measurements of corundum-oxalate suspensions at
high oxalate concentrations. The extent of precipitate formation is greatest for the corundum-oxalate
system due to the strong dissolution-enhancing properties of the inner-spherically adsorbed oxalate anion
(i.e., its ability to generate enhanced concentrations of dissolved Al(III) which can then participate in
precipitate formation). The effects of the LMW organic anion adsorption modes on both the forms of the
measured τy versus pH data, and the ability to quantitatively compare τy and ζ potential data measured
at different corundum concentrations, are also discussed.
In addition to the above processes, the adsorption of as outer-spherically adsorbed anions are often held at
LMW anions is known to have a significant effect on positively charged surfaces by electrostatic means, the
properties affecting the colloidal stability of mineral extent of their adsorption is likely to have a strong
(oxyhydr)oxide suspensions. Adsorbed LMW anions com- dependence on the interfacial potential. As a result, anions
monly affect colloidal stability in two ways: first, by that tend to adsorb in an outer-sphere manner may have
modifying the effective interfacial charge,12-16 so alter- a limited ability to charge reverse mineral surfaces (due
ing the nature of the electrical double layer interac- to the electrostatic repulsion that will exist between like-
tion between the sorbate particles,17 and, second, by charged interfaces and anions), a factor that will affect
generating a short-range steric barrier to interparticle the nature of the electrical double layer interaction
approach,12,16,18,19 thus effectively truncating the short- between adsorbate particles. In contrast, inner-spherically
range strength of the attractive van der Waals force. adsorbed anions are unlikely to exhibit such behavior (at
Similar effects have been reported in the presence of sorbed least to the same extent) due to the presence of a significant
macromolecular organic matter20-26 and in combination chemical (nonelectrostatic) contribution to the adsorption
(where they are collectively referred to as electrosteric free energy. The dissolution-enhancing properties of
effects) can have a dramatic impact on the colloidal anions adsorbed in an inner-sphere, mononuclear, bi-
behavior exhibited by mineral particles in the environ- dentate manner may also impact the colloidal stability of
ment. For example, Lowry et al.27 and Slowey et al.28 have mineral suspensions in two manners: first, by markedly
demonstrated that the presence of common LMW organic increasing the ionic strength through the production of
anions such as malonate, oxalate, and citrate can sig- significant concentrations of dissolved metal cations, an
nificantly enhance the stability (and therefore the un- effect that decreases the effective interfacial potential and
desirable transport) of mercury- and arsenic-bearing so alters the strength of the electrical double layer
mineral nanoparticles. Adsorbed organic matter can also interaction between adsorbate particles,34-37 and, second,
act to physically stabilize particles in natural waters,29,30 by driving the dissolution of sufficiently high concentra-
while the stability-enhancing characteristics of a common tions of metal cations from mineral surfaces to allow the
LMW organic anion, citrate, have recently been investi- formation of metal-organic precipitates which, should
gated in the context of remediating aquifer sediments they bridge between particles, will generate an additional
containing heavy metal contaminants.31 Similarly, in an attractive interparticle force. A reduction in the suspension
industrial context, the enhancement of colloidal stability stability will then result. Such bridging precipitates are
that can be imparted by adsorbed LMW organic anions likely to be formed to a lesser extent (if at all) in the
has led to species such as citrate being investigated for presence of outer-spherically adsorbed anions due to the
use as dispersants in the aqueous-based formulation of dissolution-inhibiting effect these anions can exhibit under
advanced ceramic components.13,32,33 acidic solution conditions.11
While a significant number of studies have now ad- The purpose of the present study is to examine the effects
dressed the effects of LMW anion adsorption on colloidal of outer-sphere versus inner-sphere adsorption of LMW
stability, none to our knowledge have yet attempted to organic anions on the colloidal stability of a model mineral
distinguish between the relative impacts of inner-spheri- suspension. To this end, two common LMW organic anions,
cally versus outer-spherically adsorbed LMW anions. Such maleate and oxalate, and a model mineral substrate,
differences in adsorption mode are anticipated to be of corundum (R-Al2O3), were selected for study. Several
potential importance in a number of contexts. For example, recent studies have demonstrated that maleate adsorbs
in a predominantly outer-sphere manner to aluminum
(12) Leong, Y. K.; Scales, P. J.; Healy, T. W.; Boger, D. V.; Buscall, (oxyhydr)oxide particles, including corundum.11,38 In
R. J. Chem. Soc., Faraday Trans. 1993, 89, 2473. contrast, provided a sufficiently long equilibration period
(13) Hidber, P. C.; Graule, T. J.; Gauckler, L. J. J. Am. Ceram. Soc. is allowed,39 oxalate will adsorb to aluminum (oxyhydr)-
1996, 79, 1857.
(14) Hidber, P. C.; Graule, T. J.; Gauckler, L. J. J. Eur. Ceram. Soc.
oxides, including corundum, primarily in an inner-sphere,
1997, 17, 239. mononuclear, bidentate fashion.39-41 As a result, the use
(15) Yao, H. L.; Yeh, H. H. Langmuir 1996, 12, 2989. of maleate and oxalate, both of which are simple dicar-
(16) Johnson, S. B. The Relationship Between the Surface Chemistry boxylate LMW anions, allows the effects of outer-sphere
and the Shear Yield Stress of Mineral Suspensions. Ph.D. Thesis, The
University of Melbourne, Melbourne, Australia, 1998. and inner-sphere adsorption, respectively, to be probed.
(17) Hunter, R. J. Foundations of Colloid Science, 2nd ed.; Oxford Corundum was selected for study due to its well-
University Press: Oxford, U.K., 2001. understood electrokinetic,13,14,36,42 dissolution,11,41,43,44 and
(18) Leong, Y. K.; Boger, D. V.; Scales, P. J.; Healy, T. W.; Buscall,
R. J. Chem. Soc., Chem. Commun. 1993, 639. rheological16,37,42 properties.
(19) Biggs, S.; Mulvaney, P.; Zukoski, C. F.; Grieser, F. J. Am. Chem. In the present study, the properties affecting the
Soc. 1994, 116, 9150. colloidal stability of model corundum suspensions have
(20) Davis, J. A. Geochim. Cosmochim. Acta 1982, 46, 2381.
(21) Tipping, E.; Higgins, D. C. Colloids Surf. 1982, 5, 85. (34) Wiese, G. R.; Healy, T. W. J. Colloid Interface Sci. 1975, 51, 427.
(22) O’Melia, C. R. Colloids Surf. 1989, 39, 255. (35) Larson, I.; Drummond, C. J.; Chan, D. Y. C.; Grieser, F. J. Am.
(23) Tiller, C. L.; O’Melia, C. R. Colloids Surf., A 1993, 73, 89. Chem. Soc. 1993, 115, 11885.
(24) Amirbahman, A.; Olson, T. M. Colloids Surf., A 1995, 95, 249. (36) Johnson, S. B.; Scales, P. J.; Healy, T. W. Langmuir 1999, 15,
(25) Killmann, E.; Adolph, H. Colloid Polym. Sci. 1995, 273, 1071. 2836.
(26) Mosley, L. M.; Hunter, K. A.; Ducker, W. A. Environ. Sci. Technol. (37) Johnson, S. B.; Franks, G. V.; Scales, P. J.; Healy, T. W. Langmuir
2003, 37, 3303. 1999, 15, 2844.
(27) Lowry, G. V.; Shaw, S.; Kim, C. S.; Rytuba, J. J.; Brown, G. E., (38) Rosenqvist, J.; Axe, K.; Sjoberg, S.; Persson, P. Colloids Surf.,
Jr. Environ. Sci. Technol., in press. A 2003, 220, 91.
(28) Slowey, A. J.; Johnson, S. B.; Rytuba, J. J.; Brown, G. E., Jr. (39) Axe, K.; Persson, P. Geochim. Cosmochim. Acta 2001, 65, 4481.
Environ. Sci. Technol., in press. (40) Yoon, T. H.; Johnson, S. B.; Musgrave, C. B.; Brown, G. E., Jr.
(29) Ali, W.; O’Melia, C. R.; Edzwald, J. K. Water Sci. Technol. 1985, Geochim. Cosmochim. Acta 2004, 68, 4505.
17, 701. (41) Johnson, S. B.; Yoon, T. H.; Slowey, A. J.; Brown, G. E., Jr.
(30) Weilenmann, U.; O’Melia, C. R.; Stumm, W. Limnol. Oceanogr. Langmuir 2004, 20, 11480.
1989, 34, 1. (42) Johnson, S. B.; Russell, A. S.; Scales, P. J. Colloids Surf., A
(31) Johnson, C. R.; Hellerich, L. A.; Nikolaidis, N. P.; Gschwend, P. 1998, 141, 119.
M. Ground Water 2001, 39, 895. (43) Carroll-Webb, S. A.; Walther, J. V. Geochim. Cosmochim. Acta
(32) Sacks, M. D.; Tseng, T. Y. J. Am. Ceram. Soc. 1983, 66, 242. 1988, 52, 2609.
(33) Luther, E. P.; Yanez, J. A.; Franks, G. V.; Lange, F. F.; Pearson, (44) Samson, S. D.; Stillings, L. L.; Eggleston, C. M. Geochim.
D. S. J. Am. Ceram. Soc. 1995, 78, 1495. Cosmochim. Acta 2000, 64, 3471.
6358 Langmuir, Vol. 21, No. 14, 2005 Johnson et al.
been examined over a broad range of maleate and oxalate In this study, electroacoustic measurements were generally
concentrations and solution pH conditions. The electrical performed at φ ) 0.02, although for comparison, several systems
properties of the corundum-water interface have been containing substantially higher solids volume fractions (φ ) 0.25)
probed using electrokinetic (electroacoustic) means over were also examined. A temperature of 25 οC and a constant
stirring speed of 500 rpm were used in all experiments.
a range of different corundum volume fractions, φ (φ ) Calculations of particle-averaged ζ potentials from the net ESA
0.02-0.25). The shear yield stresses of the corresponding responses were performed using the AcoustoSizer software (DOS
concentrated (φ ) 0.25) suspensions were then measured version 1.26).
as an indicator of the net interparticle force at each pH 2.3. Shear Yield Stress Measurements. Shear yield stress
condition examined. Analogous results obtained for an (τy) measurements were performed on concentrated particulate
additional LMW organic ligand, citrate (which strongly suspensions using the vane technique of Keentok, Nguyen, and
adsorbs to corundum in an inner-sphere manner13), are Boger.49-51 Experimentally, the shear yield stress apparatus
also presented for the purpose of comparison. In selected consists of a small four-bladed vane attached to the spring-driving
cases, in situ attenuated total reflectance Fourier trans- motor of a Haake RV-3 viscometer. The vane is immersed in a
form infrared (ATR-FTIR) measurements have addition- concentrated colloidal suspension of interest, with the spring
then being wound up at a low and constant rate. As a result, the
ally been undertaken to probe for the existence of metal- spring torque is increased without perceptible movement of the
organic precipitates. The results serve to demonstrate vane, with the maximum observed torque corresponding to the
some novel differences in colloidal stability promoted by moment of yielding. The maximum torque can then be related
outer-spherically versus inner-spherically adsorbed LMW to τy using the shear yield stress theory of Nguyen and Boger.51
organic ligands. In this study, an MK-500 driving head and a stainless steel
vane with dimensions Hv ) 30.17 mm and Dv ) 25.02 mm and
2. Experimental Section a blade thickness of 0.55 mm were used for all τy measurements.
Shear yield stress experiments were performed at the ambient
2.1. Materials. A well-characterized, high-purity (>99.99%)
temperature (23-24 °C), and using a rate of spring rotation of
AKP-30 corundum with a BET surface area of 7 m2/g, a mean
0.2 rpm (a condition at which the measured shear yield stress
particle diameter of 0.30 µm, and a density of 3.97 g/cm3 was
is expected to be independent of the speed of rotation50,52).
obtained from Sumitomo Chemical Co., Japan. It was used as
received. Maleic acid, oxalic acid, citric acid, crystalline aluminum 2.4. Attenuated Total Reflectance Fourier Transform
oxalate, KNO3, and KCl were obtained from a variety of suppliers, Infrared Measurements. ATR-FTIR measurements of oxalate
and were of reagent grade or higher. Milli-Q grade water and maleate adsorbed to corundum were performed in situ on
(resistivity 18.2 MΩ cm at 25 °C) was used in all experiments. wet corundum-oxalate and corundum-maleate pastes using a
All suspensions were prepared by briefly ultrasonicating the Nicolet NEXUS 470 FTIR spectrometer equipped with a deu-
aqueous corundum-organic mixtures using a Branson model terated triglycine sulfate (DTGS) detector, a horizontal attenu-
450 digital sonifier equipped with a 1.3 cm horn. The sonifier ated total reflectance attachment, and a germanium crystal. Data
was operated at a frequency of 20 kHz, and the power output was collection and spectral calculations were performed using OMNIC
maintained at 50% of the limiting power (350 W). The maximum software (version 6.0a, Nicolet Instrument Corp.). A total of 500
suspension volume ultrasonicated at any one time was ap- scans (with a spectral resolution of 4 cm-1) were performed and
proximately 90 cm3. Following ultrasonication, all suspensions averaged for each sample.
were allowed to stand for several hours prior to measurement. Wet pastes for ATR-FTIR analysis were obtained by centri-
For electroacoustic and shear yield stress measurements, a fuging aqueous corundum-oxalate or corundum-maleate sus-
background electrolyte of 0.01 M KNO3 was utilized, and the pH pensions (see section 2.1) at a centrifugal force of ∼10000g for
was adjusted using small volumes of 1-5 M HNO3 and KOH. 20 min in a Beckman Avanti 30 benchtop centrifuge. The
The concentrations of organic anions utilized were selected to supernatant liquids were then decanted and passed through 20
allow comparison with previously published measurements of nm syringe filters (Whatman Anotop 25 Plus) to remove any
maleate, oxalate, and citrate adsorption on corundum.11,13,41 residual corundum particles. The concentrated wet pastes were
All electroacoustic and shear yield stress experiments were applied directly to the germanium crystal. To hinder evaporation
undertaken with the pH progressing from acidic to more basic during ATR-FTIR measurements, a small volume of the corre-
conditions. For ATR-FTIR measurements, a background elec- sponding filtered supernatant was placed on top of each paste
trolyte of 0.01 M HCl was instead utilized to avoid spectral and the sample-holding region was sealed with a lid.
complications introduced by the FTIR response of the nitrate To obtain the ATR-FTIR signals from surface-associated
anion. In the latter cases, the suspension pH was instead adjusted oxalate and maleate species alone, it was first necessary to
using 1-5 M HCl and KOH solutions. eliminate the strong infrared absorbance of water. This was
2.2. Electroacoustic Measurements. All electroacoustic achieved by measuring the ATR-FTIR spectrum of each filtered
measurements were performed using an AcoustoSizer instrument supernatant, and subtracting this from the ATR-FTIR spectrum
(Colloidal Dynamics, Warwick, Australia). The AcoustoSizer measured for the corresponding corundum-oxalate or corundum-
measurement cell consists of a sample-holding vessel of volume maleate wet paste. In addition, the hydrated corundum surface
∼400 cm3, a variable-speed stirrer, built-in pH and conductivity itself was also found to possess a significant ATR-FTIR response.
probes, a dual titrator syringe module for acid and base additions, For each wet paste, the latter unwanted spectral contribution
and a heat exchange thermocouple for maintenance of a preset was removed by measuring the spectrum of a corundum-KCl
suspension temperature. The AcoustoSizer utilizes two parallel paste at an equivalent pH condition, and then subtracting this
plate electrodes embedded in opposite walls of the measurement from the ATR-FTIR spectrum measured for each corundum-
cell to generate pulsed alternating electric fields in the megahertz oxalate or corundum-maleate paste, as described elsewhere.11,41
frequency domain. Upon application of an ac voltage pulse,
charged particles are induced to oscillate with a velocity 3. Results and Discussion
dependent upon their charge and size, and the frequency of the
applied voltage. Such particle oscillations produce pressure
3.1. Electrokinetic Studies of LMW Organic An-
changes at the suspension boundaries, a phenomenon referred ions Adsorbed on Corundum. The ζ potential proper-
to as the electrokinetic sonic amplitude (ESA) effect. The ties of corundum are shown as a function of maleate
AcoustoSizer utilizes a combination of glass rods and piezoelectric
transducers to quantify the ESA signal.45 The ESA signal can (47) O’Brien, R. W. J. Fluid Mech. 1988, 190, 71.
then be related to the particle-averaged complex dynamic (48) O’Brien, R. W.; Cannon, D. W.; Rowlands, W. N. J. Colloid
mobility, 〈µD〉, and in turn the electrokinetic ζ potential, by the Interface Sci. 1995, 173, 406.
(49) Keentok, M. Rheol. Acta 1982, 21, 325.
theory of O’Brien and co-workers.46-48 (50) Nguyen, Q. D.; Boger, D. V. J. Rheol. 1983, 27, 321.
(51) Nguyen, Q. D.; Boger, D. V. J. Rheol. 1985, 29, 335.
(45) Hunter, R. J. Colloids Surf., A 1998, 141, 37. (52) Liddell, P. V.; Boger, D. V. J. Non-Newtonian Fluid Mech. 1996,
(46) O’Brien, R. W. J. Fluid Mech. 1990, 212, 81. 63, 235.
Organic Matter Adsorption at Mineral/Water Interfaces Langmuir, Vol. 21, No. 14, 2005 6359
Figure 1. Variation of the ζ-potential of corundum as a function Figure 2. Variation of the ζ-potential of corundum as a function
of maleate concentration (Γ) and pH. The corundum volume of oxalate concentration (Γ) and pH. The corundum volume
fraction was 0.02, and the background electrolyte concentration fraction was 0.02, and the background electrolyte concentration
was 0.01 M KNO3 in all cases. Key: b, Γ ) 0 µmol/m2; O, Γ ) was 0.01 M KNO3 in all cases. Key: b, Γ ) 0 µmol/m2; O, Γ )
0.36 µmol/m2; 2, Γ ) 0.71 µmol/m2; 4, Γ ) 1.4 µmol/m2; [, Γ 0.36 µmol/m2; 2, Γ ) 0.71 µmol/m2; 4, Γ ) 1.4 µmol/m2; [, Γ
) 3.6 µmol/m2; ], Γ ) 7.1 µmol/m2; 1, Γ ) 14 µmol/m2. ) 3.6 µmol/m2; ], Γ ) 7.1 µmol/m2.
concentration and solution pH in Figure 1. The background case for corundum in the presence of maleate, Figure 2
electrolyte concentration is 0.01 M KNO3 in all cases. In demonstrates that the position of the iep of corundum
the absence of maleate, Figure 1 shows that the isoelectric decreases markedly as a function of increasing oxalate
point (iep) of the bare corundum surface is 9.4, in good concentration. Except at very short reaction times, oxalate
agreement with values in the vicinity of 9 previously adsorbs to aluminum (oxyhydr)oxide minerals including
reported for corundum powders dispersed in aqueous corundum41 in a predominantly inner-sphere manner to
media.13,14,36,42,53,54 Figure 1 also shows that the position form bidentate, mononuclear surface complexes with
of the iep systematically decreases as a function of surface Al(III) cations.39,40 As a result, in addition to an
increasing maleate concentration, reaching a value of 5.8 electrostatic attraction between the oxalate anion and the
at the highest maleate concentration examined (Γ ) 14 positively charged corundum surface, a significant chemi-
µmol/m2, where Γ is defined as the total (i.e., adsorbed + cal component of the adsorption energy will exist as a
free) concentration of maleate normalized by the corundum result of the direct surface Al(III)-oxalate bonds that are
surface area). In addition, at pH < 10, the measured ζ formed. As a consequence of this chemical component,
potential data at each pH condition are observed to Figure 2 shows that oxalate is able to charge reverse the
progressively decrease as the concentration of maleate corundum surface to a significantly greater extent than
rises. Such results are driven by the significant surface maleate, with ζ potentials as low as -34 mV being observed
association of maleate (primarily in its fully deprotonated at pH 7.1 and the highest oxalate concentration examined
form, Mal2-) that has previously been observed by Johnson (Γ ) 7.1 µmol/m2). Interestingly, however, Figure 2 also
et al.11 on the positively charged corundum surface. shows that the ζ potential versus pH behavior observed
Maleate adsorbs at aluminum (oxyhydr)oxide surfaces, at pH g 11 is relatively insensitive to oxalate concentra-
including corundum, predominantly as an outer-sphere tion. The latter finding is in accordance with the recent
complex.11,38 As a result, the adsorption of maleate is adsorption study of Johnson et al.,41 which demonstrated
expected to be largely governed by its electrostatic that the extent of oxalate adsorption on corundum is very
interactions with the corundum surface. This expectation low above the iep of the bare corundum surface (i.e., the
is borne out in Figure 1, where at pH conditions above the surface in the absence of oxalate). The latter finding may
iep of bare corundum (i.e., pH > 9.4), the measured ζ indicate that the unfavorable electrostatic repulsion that
potential is relatively insensitive to maleate concentration. exists between oxalate anions and the corundum surface
Such results are consistent with the very low extent of under such high pH conditions is sufficient to overcome
maleate adsorption measured by Johnson et al.11 at pH the chemical component of adsorption. Alternatively, it
> 9.5, and indicate that anionic maleate is unable to adsorb may be a function of a different outer-sphere binding mode
to the negatively charged corundum surface to a significant dominating adsorbed oxalate speciation in the vicinity of
extent. Similar electrokinetic results have previously been the iep of the bare surface.
reported for the interaction of maleate with δ-Al2O3.15 The electrokinetic behavior exhibited in Figure 2 can
Figure 1 also demonstrates that, at pH < 9.4, maleate is be contrasted with that obtained in the presence of a more
able to charge reverse the corundum surface to only a strongly binding LMW ligand, citrate, as shown in Figure
small extent, even at the highest maleate concentrations 3. In agreement with past electrokinetic studies of aqueous
examined. The latter result can again be attributed to the corundum-citrate systems,13,14,33 Figure 3 shows that both
low affinity of maleate anions for like-charged mineral- the corundum iep and the corundum ζ potentials measured
water interfaces. Similar results have previously been at each pH condition are significantly reduced as the citrate
reported for the interaction of sulfate (which adsorbs to concentration is raised. Past in situ studies have dem-
(oxyhydr)oxide surfaces in an outer-sphere manner at all onstrated that citrate adsorbs to (oxyhydr)oxide minerals,
but quite acidic conditions55) with corundum.16 including corundum,13 in a predominantly inner-sphere
Figure 2 shows the ζ potential properties of corundum manner involving one or two of the citrate carboxyl
as a function of oxalate concentration and pH. As was the groups.56 The citrate R-hydroxyl group also participates
in the formation of inner-spherically adsorbed citrate
(53) Franks, G. V.; Meagher, L. Colloids Surf., A 2003, 214, 99.
(54) Kosmulski, M. J. Colloid Interface Sci. 2004, 275, 214.
(55) Wijnja, H.; Schulthess, C. P. J. Colloid Interface Sci. 2000, 229, (56) Lackovic, K.; Angove, M. J.; Wells, J. D.; Johnson, B. B. J. Colloid
286. Interface Sci. 2003, 257, 31.
6360 Langmuir, Vol. 21, No. 14, 2005 Johnson et al.
Figure 3. Variation of the ζ-potential of corundum as a function Figure 4. Variation of the ζ potential of corundum versus pH
of citrate concentration (Γ) and pH. The corundum volume at a maleate concentration, Γ, of 3.6 µmol/m2 and a background
fraction was 0.02, and the background electrolyte concentration electrolyte concentration of 0.01 M KNO3. This concentration
was 0.01 M KNO3 in all cases. Key: b, Γ ) 0 µmol/m2; O, Γ ) of maleate corresponds to the τy versus pH curve at which τy,max
0.071 µmol/m2; 2, Γ ) 0.14 µmol/m2; 4, Γ ) 0.36 µmol/m2; [, is at its minimum value. Key: b, φ ) 0.02; O, φ ) 0.25.
Γ ) 0.71 µmol/m2; ], Γ ) 1.4 µmol/m2; 1, Γ ) 3.6 µmol/m2.
Figure 7. Variation of the shear yield stress (τy) of corundum Figure 8. Variation of the shear yield stress (τy) of corundum
as a function of maleate concentration (Γ) and pH. The corundum as a function of oxalate concentration (Γ) and pH. The corundum
volume fraction was 0.25, and the background electrolyte volume fraction was 0.25, and the background electrolyte
concentration was 0.01 M KNO3 in all cases. Key: b, Γ ) 0 concentration was 0.01 M KNO3 in all cases. Key: b, Γ ) 0
µmol/m2; O, Γ ) 0.36 µmol/m2; 2, Γ ) 0.71 µmol/m2; 4, Γ ) 1.4 µmol/m2; O, Γ ) 0.36 µmol/m2; 2, Γ ) 0.71 µmol/m2; 4, Γ ) 1.4
µmol/m2; [, Γ ) 3.6 µmol/m2; ], Γ ) 7.1 µmol/m2; 1, Γ ) 14 µmol/m2; [, Γ ) 3.6 µmol/m2; ], Γ ) 7.1 µmol/m2.
µmol/m2.
(58) Lyklema, J. Fundamentals of Interface and Colloid Science; (59) Scales, P. J.; Johnson, S. B.; Healy, T. W.; Kapur, P. C. AIChE
Academic Press: London, 1991; Vol. 1. J. 1998, 44, 538.
6362 Langmuir, Vol. 21, No. 14, 2005 Johnson et al.
while malate and lactate data are from ref 16. Values of Γ were
selected to allow direct comparison of electrokinetic and rheological
data obtained under conditions where τy,max is at its minimum
Figure 10. Comparison of the changes in the electrokinetic value.16
iep (obtained at φ ) 0.02) and changes in the pH of τy,max
(measured at φ ) 0.25) measured for aqueous corundum- minimum value at Γ ) 1.4 µmol/m2 for both oxalate and
organic acid suspensions as a function of various LMW organic citrate, and Γ ) 3.6-7.1 µmol/m2 for maleate. Similar
anion concentrations. Data for maleate, oxalate, and citrate observations have previously been reported by Leong et
are from Figures 1-3 and 4-6, while data for malate and lactate
are from ref 16. Key: b, maleate; O, oxalate; 2, citrate; 4, malate;
al.,12,18 and are attributable to the finite sizes of the strongly
[, lactate. adsorbed LMW organic anions on (oxyhydr)oxide surfaces.
As a result, in this case, the adsorbed maleate, oxalate,
concentrations in Figures 1-3. A quantitative comparison and citrate anions generate short-range steric barriers to
between the pH positions of τy,max and the iep data obtained interparticle approach, so truncating the strength of the
from Figures 1-3 is shown in Figure 10, and demonstrates strongly attractive van der Waals attraction, and leading
that while the pH of τy,max and iep data are in close to an increase in the physical stability of the suspended
correlation for inner-spherically adsorbed oxalate and corundum particles (i.e., a decrease in τy,max). The minimum
citrate, substantial deviations are apparent for maleate. values of τy,max observed in Figures 7-9 decrease in the
The latter findings are presumably due to the different order oxalate > maleate > citrate, implying that the size
corundum concentrations utilized in the electrokinetic (φ of the steric barrier imposed in the vicinity of the iep follows
) 0.02) and τy (φ ) 0.25) measurements, which evidently the reverse order. Such findings are consistent with the
has little effect on adsorbates with a strong chemical relative sizes of (and so the size of the steric barriers
component of adsorption, but significantly affects results presented by) the maleate, oxalate, and citrate anions
obtained for LMW organic adsorbates instead adsorbed (i.e., citrate > maleate > oxalate).16 Further details
in an outer-sphere manner. Similar findings are obtained regarding the effects of adsorbed organic anions on the
for corundum particles suspended in aqueous solution in magnitude of τy,max are given in the studies of Leong et
the presence of other LMW organic anions. For example, al.12,18
Figure 10 shows that similar to the corundum-oxalate Interestingly, Figures 7-9 show that, at the highest
and corundum-citrate data, the pH of τy,max and iep data maleate, oxalate, and citrate concentrations examined,
obtained in the presence of malate are closely correlated. increases in τy,max are observed compared with its mag-
This finding is understandable given that malate strongly nitude at intermediate concentrations of those LMW
adsorbs to corundum to an extent comparable with that anions. Several possible explanations exist for these
of citrate,14 and via an adsorption mode that (on the basis findings. First, a rise in the magnitude of τy,max at the
of ATR-FTIR observations of its complexation to metal maleate, oxalate, and citrate concentrations examined
cations in solution57) is expected to be predominantly inner- could be consistent with a lower extent of adsorption of
sphere in nature. By contrast, the pH of τy,max data obtained the LMW anions at low pH, leading to a decrease in the
in the presence of lactate exhibit significant deviations size of the effective steric barriers presented by the
from the corresponding iep data (Figure 10). The latter maleate, oxalate, and citrate species. Such an explanation
finding is not surprising given that lactate adsorbs is particularly applicable to inner-spherically adsorbed
comparatively weakly to mineral (oxyhydr)oxides such as oxalate, which acts to significantly increase the dissolution
corundum14 in a combination of weak inner-sphere and kinetics of aluminum (oxyhydr)oxide materials3,40 includ-
outer-sphere adsorption modes.60,61 By contrast, a com- ing corundum,41 leading to the generation of substantial
parison of the pH of τy,max and iep data instead obtained dissolved Al(III) concentrations in solution. As oxalate
at φ ) 0.25 demonstrates close correlations for all organic forms a range of stable complexes with Al(III) in solution,62
anions examined in Figure 10 (see Table 1). In combina- dissolved Al(III) can then strongly compete with the
tion, the data of Figure 10 indicate that a quantitative corundum surface for oxalate, leading to a decreased extent
comparison of electrokinetic data (typically obtained at of oxalate adsorption under low pH conditions, as has
low φ) and rheological data (obtained at higher φ) obtained been recently demonstrated by Johnson et al.41 An Al-
for aqueous mineral-LMW organic anion systems only (III)-induced reduction in the extent of oxalate adsorption
appears reasonable in the case of strongly inner-spheri- compared with that obtained at intermediate oxalate
cally bound LMW organic anions. concentrations would, however, be expected to lead to a
Figures 7-9 further demonstrate that the magnitude corresponding upward shift in the iep and pH position of
of τy,max initially decreases as a function of increasing τy,max at the highest oxalate concentration examined (Γ )
maleate, oxalate, and citrate concentrations, reaching a 7.1 µmol/m2), an expectation that is clearly inconsistent
with the data of Figures 2 and 8 (where shifts to lower pH
(60) Filius, J. D.; Hiemstra, T.; Van Riemsdijk, W. H. J. Colloid are instead observed). Similarly, substantial downward
Interface Sci. 1997, 195, 368.
(61) Ha, J.; Johnson, S. B.; Brown, G. E., Jr., Manuscript in (62) Sjoberg, S.; Ohman, L. O. J. Chem. Soc., Dalton Trans. 1985,
preparation. 2665.
Organic Matter Adsorption at Mineral/Water Interfaces Langmuir, Vol. 21, No. 14, 2005 6363