Accepted Manuscript Regular article The Role of Metal Ion-Ligand Interactions during Divalent Metal Ion Adsorption Daniel S. Eldridge, Russell J. Crawford, Ian H. Harding PII: DOI: Reference:

S0021-9797(15)00433-6 http://dx.doi.org/10.1016/j.jcis.2015.04.056 YJCIS 20436

To appear in:

Journal of Colloid and Interface Science

Received Date: Accepted Date:

23 March 2015 26 April 2015

Please cite this article as: D.S. Eldridge, R.J. Crawford, I.H. Harding, The Role of Metal Ion-Ligand Interactions during Divalent Metal Ion Adsorption, Journal of Colloid and Interface Science (2015), doi: http://dx.doi.org/ 10.1016/j.jcis.2015.04.056

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Role of Metal Ion-Ligand Interactions during Divalent Metal Ion Adsorption Daniel S. Eldridge*, Russell J. Crawford and Ian H. Harding Faculty of Science, Engineering and Technology, Swinburne University of Technology, PO BOX 218, Hawthorn, Victoria, 3122, Australia. ABSTRACT: A suite of seven different divalent metal ions (Ca(II), Cd(II), Cu(II), Mg(II), Ni(II), Pb(II), Zn(II)) was adsorbed from solution onto two Fe2O3 samples, quartz SiO2 and three different amphoteric polystyrene latices (containing amine and carboxyl functional groups). For the metal oxides, a high correlation was observed between the pH at which 50% of the metal was removed from solution (pH50) and the first hydrolysis constant for the metal ion (pK1). For the polystyrene latices, a much higher correlation was observed between the pH50 and pKc (equilibrium constant describing metal-carboxyl affinity) as opposed to pK1. These observations provide evidence of a strong relationship that exists between a metal’s affinity for a particular ligand in solution and for that metal ion’s affinity for the same ligand present as part of an adsorbing surface. The isoelectric point of the amphoteric latex surface can be increased by decreasing the carboxyl content of the latex surface. For all 7 metal ions, this resulted in a substantial decrease, for any given pH, in adsorption. We suggest that this may be partly due to the decreased carboxyl content, but is dominantly attributable to the presence of less favorable electrostatic conditions. This, in turn, demonstrates that electrostatics play a controlling role in

1

metal ion adsorption onto amphoteric latex surfaces and, in addition to the nature of the metal ion, also controls the pH at which adsorption takes place. KEY WORDS: Adsorption; divalent metal; polystyrene latex; carboxyl; SiO2; Fe2O3; hydrolysis INTRODUCTION: Over the past few decades, the toxicological effects of aqueous heavy metals on humans and animals have been well recognized.1-4 Low concentrations of metal ions can be found naturally in the environment, however discharge from industrial activities such as electroplating, battery manufacture and mining can result in an increase in the concentration of aqueous metal ions in water streams.

Consequently, metal ion adsorption has been investigated as a means for

removing waste metals from solution. These studies have been performed using a wide range of adsorbents and metals, and under greatly varied conditions. They have not only led to improvements in the ability to treat contaminated water, but also have provided a better understanding of metal speciation and mobility in the environment where adsorption processes may well control the final aqueous concentration of metals. 5-7 A number of previous studies have adsorbed metal ions onto well characterized surfaces such as metal oxides in order to provide a greater understanding of the adsorption process, 8-10 while others have selected substrates that are less well defined, such as biological materials, with a view to using these as cheap, environmentally friendly adsorbents for removing toxic metals from waste-water.11-14 This previous research has demonstrated that there are a large number of parameters that profoundly influence the extent of metal adsorption, including temperature,15-18 time,19,20 the nature of the adsorbent,19-23 sorbent:sorbate ratio,22,24,25 ionic strength25,26 and pH.16,19,20

2

James and Healy27 proposed that metal ions must first hydrolyze before they are able to undergo adsorption onto a substrate. This would then suggest that a metal species’ affinity to hydrolyze would strongly influence its ability to undergo adsorption. Many subsequent studies have observed a link between metal hydrolysis and metal adsorption.28-31 An alternative explanation is that metal ion adsorption occurs primarily through a surface complexation mechanism, whereby the metal species forms a chemical bond with the functional groups present on the adsorbent surface.15,32,33 The link is then not a causal correlation, but rather suggests commonality in the thermodynamics that controls hydroxide interactions in solutions and at surfaces. Several collations of the order (with respect to pH) at which adsorption takes place for different metals ions can be found in the literature (see, for example, Pagnanelli 11). These generally show how wide-spread the finding is that metal ion surface affinity (i.e. pH of adsorption onset) and metal ion solution affinity (i.e. pK1) is strongly correlated. An extended and updated list of such studies is given here in Table 1. The adsorbents used in these studies include metal oxides, synthetic polymers, natural materials, activated carbon, biological waste products and bacteria. Table 1. Literature metal adsorption studies and the reported metal adsorption order as a function of pH. Substrate

Metal ion adsorption efficiency

Reference

Activated Carbon

Pb > Cu > Zn > Ni

34

Activated Carbon

Pb > Cu > Cd

35

Al2O3

Cu > Pb > Zn > Co

36

Carbon nanotubes

Pb > Ni > Zn > Cd

37

Coal (low grade)

Pb > Cr > Cd > Zn ≈ Ni > Ca

38

3

Fe2O3

Pb > Cu > Zn > Co

31

Goethite

Zn > Ni > Cd

9

Goethite

Cu > Pb > Zn > Co > Ni > Cd > Mn

28

Goethite

Cu > Pb > Zn > Co > Cd

39

Goethite

Cu > Pb > Zn

10

Hematite

Pb > Cu > Zn > Co > Ni > Mn

39

Hydrous Cr2O3

Cr > Zn > Ni

40

Hydrous Fe2O3

Pb > Cu > Zn > Cd

41

Kaolinite

Pb > Cu > Zn > Co > Mg

42

MnO2

Pb > Cu > Zn > Cd

43

P. chrysogenum

Pb > Cu > Zn > Cd > Ni > Co

44

PolyEGDMA

Pb > Cd > Hg

45

Polyhydroxyethylmethacrylate

Pb > Zn > Cu > Cd

46

Pyrolusite

Pb > Zn > Cd > Mg

47

R.arrhizus

Pb > Zn > Ni

48

S. cinnamoneum

Pb > Zn = Cu > Cd > Ni > Co

44

S. natans

Pb > Cu > Zn > Cd

11

SiO2

Fe > Cr > Co > Ni > Ca

8

SiO2

Fe > Cr > Co > Ca

27

SiO2

Pb > Cu > Zn > Ni > Cd > Co > Mg

49

The majority of these studies have been qualitative in their analysis, i.e. simple reporting of the order in which metal ions adsorb. Others28,29,31 have quantitatively explored the relationship between metal ion hydrolysis (pK1) and metal ion adsorption onto metal oxides (pH50) and found a strong correlation exists between these two parameters.

4

By contrast, other authors50-52 have shown a greater correlation existing between the pH50 and metal-carboxyl ligand stability constants (pKc) for metal ion adsorption onto biological substrates. Interestingly, these authors then used that information to suggest the presence of a high proportion of carboxyl surface functionality. Natural biological adsorbents can contain a suite of different surface functional groups, and hence definitive proof of a correlation with pKc has proven elusive The aim of the current investigation is to explore the relationship between metal ion adsorption and metal-ligand interactions in solution, using well characterized (model) surfaces having either oxide or carboxyl surface groups.

MATERIALS AND METHODS: Iron oxide (Brunauer Emmett Teller (BET) Specific Surface Area (SSA) = 3.0 m2/g, isoelectric point (IEP) = 5.9) was obtained from BDH Chemicals. Natural hematite, supplied by Eaglehawk Geological Consulting Pty. Ltd., Broken Hill, Australia, was ground in a ball mill for 8 hours before use (BET SSA = 12.3 m2/g, IEP = 6.5). Quartz silica (BET SSA = 2.2 m2/g, IEP ≈ 2) was obtained from Fluka (Sigma-Aldrich). The method used to prepare the amphoteric polystyrene latex was adapted from that previously reported by Harding.53 Briefly, 100 g styrene, 10 g methacrylic acid and 10 g diethylaminoethyl methacrylate were mixed in a baffled flask. The pH was adjusted to 1.5 using analytical reagent grade (AR) HNO3 and the total volume was adjusted to 1 L. AR nitrogen gas was diffused through the solution to expel any dissolved O2 before the addition of 0.5 g (NH4)2S4O8. The temperature was increased to 70°C and the mixture was stirred for 24 h, after which time the suspension was allowed to cool and then filtered through glass wool to remove any coagulum.

5

Three different latex samples were prepared using this method by varying the mass of methacrylic acid from 5 to 20 g in order to vary the ratio between the carboxyl and amine surface functional groups and consequently the IEP of the adsorbent. Each latex sample was purified by repeated dispersion into water followed by centrifugation until the conductance of the supernatant remained constant at a conductivity less than 50 µS. The three final products, designated Latex-L, Latex-M and Latex-H (indicating low, medium and high IEP) had IEPs of 4.3, 5.4 and 5.9, and BET SSAs of 13.6, 21.8 and 10.6 m 2/g, respectively. EDX analysis confirmed the increasing nitrogen:carbon ratio – and consequently the increased amine content – as the amount of diethylaminoethyl methacrylate was increased from Latex-L to Latex-H. The specific surface area of each adsorbent was determined using the Brunauer Emmett Teller (BET) method with a Micromeritics ASAP 2000 surface area analyzer. The surface area of the solid was determined in triplicate using low pressure nitrogen gas adsorption data, where P/P o < 0.1. Measurements were shown to be reproducible to within ±0.1 m2/g. Zeta potential analyses were carried out using a Brookhaven Instruments BIC90 particle analyzer, using phase analysis light scattering technology. Zeta potential as a function of pH was then calculated according to the method reported by Kosmulski.54 The isoelectric points of the oxides and latices are consistent with literature values.53,55 Metal ion adsorption studies were conducted individually for each metal by adding 50 mL of 0.001 M KNO3 to 430 mL of MilliQ H2O in a jacketed reaction vessel. A solid mass of adsorbent was added to provide a total surface area of 50 m2/L. The pH was then adjusted to approximately 3 unless adsorption was known not to occur until substantially higher pH levels. After de-gassing and equilibration, a 20 mL aliquot of 0.005 M metal nitrate solution was added to the 480 mL suspension. The suspension was then further equilibrated for 15 minutes before a 5 mL sample was taken from the reaction vessel and filtered through a 0.22 µm cellulose acetate

6

filter. The filtrate was acidified and analyzed using atomic absorption spectroscopy to determine the concentration of metal ion remaining in solution. The pH of the reaction vessel was then increased using ~0.1 M KOH, the solution allowed to equilibrate for 15 minutes at the new pH, and sampling repeated. Precipitation edges (no colloid present) were determined under the same experimental conditions.

The experimental precipitation experiments were combined with hydrolysis

equilibrium constants from Baes and Mesmer56 to calculate the solubility product (Ksp) of each metal hydroxide.

RESULTS AND DISCUSSION: The removal of the 7 aqueous metal ions, as a function of pH, is given for the oxide substrates in Figure 1.

7

100 Ca Cd Cu Mg Ni Pb Zn

% Metal Removed

80

BDH Fe2O3 60

(a)

40

20

0

3

4

5

6

100

8

9

10

11

8

9

10

11

8

9

10

11

Ca Cd Cu Mg Ni Pb Zn

80

% Metal Removed

7

pH

Natural Fe2O3 60

(b) 40

20

0

3

4

5

6

100 Ca Cd Cu Mg Ni Pb Zn

80

% Metal Removed

7

pH

SiO2

60

(c) 40

20

0

3

4

5

6

7

pH

8

Figure 1. Removal of 0.0002 M metal onto 50 m2/L of adsorbent as a function of pH at 25oC in 10-3 M KNO3. Figure a) BDH Fe2O3 b) Natural Fe2O3 c) Quartz SiO2 The order of onset of metal ion adsorption is commonly expressed in terms of the pH at which 50% of the aqueous metal ion is removed from solution (pH50).28,29,31 For the three adsorbents used in the adsorption studies above, the pH50 follows the trend, Cu(II) ≥ Pb(II) > Zn(II) > Ni(II) ≥ Cd(II) > Mg(II) > Ca(II). This trend closely follows many of the data sets reported in the literature, as earlier summarized in Table 1. It is also largely independent of the choice of pH50 since in most cases the curves have a similar shape and gradient. This order of onset of adsorption closely matches the order of the onset of hydrolysis, expressed either in terms of the first hydrolysis constant, pK1 or the solubility product, pKsp, as detailed in Table 2. The relationship between hydrolysis and adsorption was first articulated by James29 and later by (amongst others) Fischer et al.28 and Tamura.31

9

Table 2. pK1, pKsp and pKc data for metal ions used in this study.56,57 Reaction

pK1

pKsp

Reaction

pKc

Ca2+ + H2O  Ca(OH)+ + H+

12.8

5.3

Ca2+ + CH3COO-  CH3COO-Ca+

-0.77

Cd2+ + H2O  Cd(OH)+ + H+

10.1

13.2

Cd2+ + CH3COOCH3COO-Cd+

-1.70

Cu2+ + H2O  Cu(OH)+ + H+

7.6

18.3

Cu2+ + CH3COO-  CH3COO-Cu+

-2.24

Mg + H2O  Mg(OH)+ + H+

11.4

18.7

Mg + CH3COO  CH3COO-Mg+

-0.82

Ni2+ + H2O  Ni(OH)+ + H+

9.9

10.2

Ni2+ + CH3COO-  CH3COO-Ni+

-1.13

Pb2+ + H2O  + + Pb(OH) + H

7.7

14.5

Pb2+ + CH3COO-  + CH3COO-Pb

-2.45

Zn2+ + H2O  Zn(OH)+ + H+

9.0

15.7

Zn2+ + CH3COO-  CH3COO-Zn+

-1.59

2+

2+

-

The correlation (pH50 vs. pK1) is shown graphically using natural Fe2O3 as an example, in Figure 2. As earlier discussed, pH50 is the pH at which 50% of adsorption is reached and is used as an indicator of the order in which, on the pH scale, adsorption is taking place.

10

13

R2 = 0.9921 12 11

pH50

10 9

Fe2O3 Natural

8

Cu Pb Zn Ni Cd Mg Ca

7 6 5

7

8

9

10

11

12

13

pK1

Figure 2. pH50 as a function of pK1 for adsorption onto natural Fe2O3. For the calcium data point, it is acknowledged that there is some uncertainty in the pH 50 value, due to the fact that it occurs above a pH of 11 and therefore at a higher ionic strength than the other the adsorption experiments. Irrespective of this, the experiment was continued above pH 11 in order to provide an estimate of the pH50 value for calcium. This was the case for all 3 metal oxide substrates. A strong correlation exists between the order of metal ion adsorption (expressed in terms of the pH50) onto the Fe2O3 surface and the pK1 for the metal first hydrolysis reaction. A similarly high correlation is observed between pH50 and the solubility product (pKsp) of the metal oxide. The pattern is similar for the BDH Fe2O3 and SiO2 samples. Table 3 shows the correlation coefficient when plotting pH50 vs. pK1 and pH50 vs. pKsp for all 3 metal oxide adsorbents.

11

Table 3. Correlation coefficients for pH50 vs. pK1 and pH50 vs. pKsp metal adsorption onto metal oxides. Adsorbent

pK1

pKsp

BDH Fe2O3

0.9794

0.9885

Natural Fe2O3

0.9921

0.9928

SiO2

0.9884

0.9701

A strong correlation is evident for both the pH50-pK1 and pH50-pKsp data sets.

The high

correlation between both pK1 and pKsp with the metal adsorption is expected, as pK1 and pKsp are themselves strongly correlated. Because both parameters correlate well with metal adsorption, the metal oxide adsorption data is insufficient to evaluate which of the two parameters may be mechanistically more important, and therefore also cannot distinguish between surface complexation and surface precipitation. James and Healy27 first argued that a metal ion needs to hydrolyse prior to, or at the same time, as adsorption. Their argument was that the metal ion must reduce its charge from M2+ to M(OH)+ in order to mitigate the strong solvation barrier which prevents a charged ion from approaching closely to a surface. If the surface charge is positive, this will also mitigate any electrostatic repulsion the highly charged ion may experience, but will lower any electrostatic attraction if the surface is negatively charged. Other adsorption models, such as the triple-layer models pioneered by (for example) Davis and Leckie,32 proposed that adsorption takes place as a result of interaction between the metal species and the functional groups present on the adsorbent surface, via surface complexation. Like the 27

James Healy model, this model predicts that the adsorption of metal ions onto an oxide surface

12

should closely follow the onset of metal ion hydrolysis, however the mechanism for this is not explicitly required, and it could be a simple correlation of the affinity of the metal ion for solution hydroxide and the affinity of the metal ion for surface hydroxide. The carboxyl and amine surface functionality of amphoteric polystyrene latex surfaces provide an interesting substrate to further test this correlation, since the surfaces of these materials do not have hydroxyl functionality. The surface functionality of amphoteric polystyrene latex can be described as follows: Surface-COOH  Surface-COO- + H+ +

+

Surface-NH3  Surface-NH2 + H

These polystyrene latices exhibit several useful properties. The inclusion of both the positively charged amine surface site and the negatively charged carboxyl surface site on the latices means that they have an IEP that can be modified by minor alterations to the synthesis procedure, allowing for the same adsorbent to be tested while varying the IEP, which in turn, allows for investigation of the role that electrostatic interactions play in the adsorption process.

The

surfaces of the latices have been previously well characterized53,58,59 and as the surface is free of hydroxyl functionality, metal ions cannot adsorb onto the surface through the mechanism of surface hydrolysis. The amine groups present on the surface could be either positively charged or neutral and are therefore not expected to be electrostatically favourable for the adsorption of positively charged metal species. The carboxyl surface sites possess a pH dependent charge that can be either neutral or negatively charged. It is be expected (but not assumed) that adsorption would occur via the carboxyl functionality. The three latex samples used in this study differed only in the proportion of carboxyl compared to amino surface sites, with the IEP of the latex varying accordingly (see Table 2 for details). Thus the IEP, and subsequently the charge (or zeta

13

potential) at any given pH, is different for the three different substrates; but the chemical nature of the surface does not otherwise change (other than in terms of surface site density). The removal of aqueous metal ions as a function of pH is given for the latex substrates used in this study in Figure 3. To our knowledge, this is the first such reported data of a metal ion sequence onto amphoteric latex.

14

100 Ca Cd Cu Mg Ni Pb Zn

% Metal Removed

80

Latex-L IEP = 4.3

60

40

20

(a) 0

3

4

5

6

100

8

9

10

11

Ca Cd Cu Mg Ni Pb Zn

80

% Metal Removed

7

pH

Latex-M IEP = 5.4

60

40

20

(b) 0

3

4

5

6

100

8

9

10

11

Ca Cd Cu Mg Ni Pb Zn

80

% Metal Removed

7

pH

Latex-H IEP = 5.9

60

40

20

(c) 0

3

4

5

6

7

8

9

10

11

pH

15

Figure 3. Removal of 0.0002 M metal onto 50 m2/L of adsorbent as a function of pH at 25oC in 10-3 M KNO3. Figure a) Latex-L b) Latex-M c) Latex-H Comparing the adsorption of metal ions onto each latex, the role of electrostatic attraction in metal ion adsorption can be clearly seen. As the IEP of the latex increases (in the order of LatexL > M > H) the onset of adsorption occurs at a higher pH level, but the order in which the metals adsorb onto the substrate remains the same. This effect is illustrated in Figure 4 where the pH50 is shown for each metal in ascending order, for each of the three latices. 11 Latex-L

10

Latex-M Latex-H

9

pH50

8

7

6

5

4

Pb

Cu

Zn

Cd

Ni

Mg

Ca

Metal

Figure 4. pH50 of all 7 metals adsorbed onto 50 m2/L of Latex-L, Latex-M and Latex-H as a function of pH at 25oC in 10-3 M KNO3. The trend observed in Figure 4 illustrates that irrespective of the metal species being adsorbed, the pH at which adsorption occurs is relative to the IEP of the adsorbing surface. This demonstrates that electrostatic interactions play a significant role in the adsorption behaviour of the metal ions onto the latex surface, and is controlling factor in determining the pH at which adsorption starts to occur. It is now of interest to consider the role of solution hydrolysis on the adsorption of metal ions 27

onto latex. If, as originally suggested by James and Healy, solution hydrolysis is required

16

before adsorption takes place, then a correlation should exist, even for latex surfaces, between pH50 and pK1. The pH50 values obtained for the adsorption of the metals onto Latex-L is shown as a function of pK1 in Figure 5. 8

R2 = 0.8190

pH50

7

6

Latex-L Cu Pb Zn Ni Cd Mg Ca

5

4

7

8

9

10

11

12

13

pK1

Figure 5. pH50 as a function of pK1 for adsorption onto Latex-L While there is still a correlation between pH50 and pK1, it is a substantially weaker correlation than that observed for the metal oxides. The correlation coefficients for the pH50-pK1 and pH50pKsp data sets reflecting divalent metal adsorption onto the 3 polystyrene latices are shown in Table 4. Table 4. Correlation coefficients for pH50 vs. pK1 and pH50 vs. pKc, metal adsorption onto polystyrene latices. Adsorbent

pK1

pKsp

pKc

Latex-L

0.8190

0.7320

0.9880

Latex-M

0.8999

0.8261

0.9852

Latex-H

0.9314

0.8492

0.9311

17

For all 3 latex adsorbents, compared to adsorption onto the metal oxides, the correlation between pH50 and pK1 is weaker, and the correlation between pH50 and pKsp is poor. For both Latex-L and Latex-M (not shown), the pattern of divergence from good correlation was the same (i.e. the same metal ions were above the line, and the same below) suggesting that the pH50 value may be closely correlated to the same parameter, but not the pK1. A correlation to the surface functional group (presumably carboxyl) responsible for adsorption is the obvious candidate. The equilibrium constants for the interaction of a metal ion with a carboxyl ligand in solution (using acetate as an example), denoted pKc, are given in Table 2.59 The correlation coefficients between pH50-pK1 and pH50-pKc are presented in Table 4. For the adsorption of divalent metal ions onto both Latex-L and Latex-M samples, the pH50 vs. pKc data correlates very well, and a much greater correlation is observed than that obtained for pH50 vs. pK1. The same, however, cannot be said for the Latex-H adsorbent and this substrate therefore requires further consideration. In the case of Latex-H, adsorption occurs at pH levels close to the precipitation edge of the metal, and this may invalidate discussion which assumes the mechanism of metal ion removal is adsorption rather than precipitation. An alternative approach is to investigate the amount of metal adsorption that occurs at pH levels substantially lower than that at which precipitation occurs. Table 5 shows the correlation coefficient for the adsorption onto the Latex-H adsorbent, as a function of the degree of metal ion removal from solution (e.g. pH30 is the point at which 30% of metal ion is removed).

18

Table 5. Variation of the correlation coefficient for adsorption vs. pKx onto Latex-H where x represents a hydroxyl (pK1) or carboxyl (pKc) surface. Adsorption point

R2 (pK1)

R2 (pKc)

pH20

0.8378

0.9929

pH30

0.8830

0.9840

pH40

0.9206

0.9548

pH50

0.9314

0.9311

pH60

0.9432

0.8934

pH70

0.9496

0.8775

The data presented in Table 5 highlights that as the point of comparison on the adsorption isotherm is taken further away from the precipitation edge (towards lower pH levels), the correlation with pKc increases and the correlation with pK1 decreases. At the lower pH levels where precipitation no longer interferes with adsorption, a clear correlation between adsorption (pH20) and metal-ligand interaction (pKc) is observed. We hypothesize that removal from solution of most of the metal ions in the presence of LatexH is dominated by true adsorption onto the surface at low percent removal, but is dominated by hydrolysis of metal ions in solution followed by precipitation at the higher removals (which correspond to higher pH values). The results for Latex H are, therefore, fully consistent with those for Latex-L and Latex-M. The apparent good correlation between pH50 and pK1 for the Latex-H sample thus illustrates the danger of assuming a mechanism is proven because a good correlation is obtained. The better correlation between pH50 and pKc, equally, does not prove that the primary mechanism for metal

19

adsorption is via surface complexation, but it does show that the data is more consistent with surface complexation than with surface hydrolysis or the adsorption of a hydrolysed species. The causal relationship is much more likely to between adsorption and metal-ligand interactions, rather than between adsorption and metal-hydroxide interactions. A similar comparison of literature data for any correlation between pH 50 and both pK1 and pKc requires data where a large range of divalent metals of constant molarity have been adsorbed onto surfaces containing predominantly carboxyl surface functionality and preferably with well characterised (model) surfaces. Two studies identified that meet these criteria involve a low grade coal sample believed to have a poly-functional surface with a large number of carboxyl surface sites38 and a poly-functional activated carbon shown to contain carboxyl surface functionality.60 2

For the coal adsorbent, correlating pH50 with pKc as opposed to pK1 increases R from 0.7644 to 0.9185. Applying the same treatment to the activated carbon, R2 increases from 0.8357 to 0.9889. This provides further evidence that the metal ion adsorption order is dictated by reactions between the metal species and the surface functional groups, rather than preliminary hydrolysis reactions in solution. The converse is also true. Finding a correlation between the pattern of different metals adsorbing onto a surface, and the pKc of the surface sites, provides evidence for the existence and importance of those surface sites50-52.

CONCLUSIONS:

The adsorption behaviour of seven divalent metal ions onto six different substrates was investigated under constant, controlled conditions.

A strong correlation was observed for the

20

oxide surfaces between the pH at which 50% of the metal had been removed from solution (pH50) and the first hydrolysis constant pK1, which is consistent with data obtained from similar studies. A strong correlation between pH 50 and pKc for adsorption onto controlled carboxyl surfaces was observed. This result shows the importance of the interaction between the metal species and the surface functional groups. It also provides indirect evidence that the relationship between pH50 and pK1 for adsorption onto oxide surfaces is due to reactions between the metal species and OH groups on the surface, as opposed to free metal ion hydrolysis followed by non-specific physical adsorption. Electrostatic attraction was determined to be a factor that equally affected the adsorption of all metals onto the polystyrene latex surfaces. A shift in IEP of the adsorbing surfaces caused a subsequent shift of the adsorption isotherms, but did not influence the strong correlation between pH50 and pKc.

The latex with the highest IEP resulted in conditions where experimental

interference was observed from metal ion hydroxide precipitation. For amphoteric polystyrene latices, then, the conclusion is that the pH at which adsorption starts to take place is strongly controlled by both the specific metal ion adsorbing, and the electrostatics of the system. This combination of observations further suggests that a plot of pH50 versus pKx (where x represents a ligand in solution) can yield information about the functional groups present on the surface. A high correlation between pH50 and pKx would be indicative of ligand x being present at the adsorbing surface.

AUTHOR INFORMATION Corresponding Author *Daniel Eldridge can be contacted at [email protected]

21

Funding Sources The research was conducted using funding generously provided in the form of an Australian Postgraduate Award. REFERENCES 1.

Ekstrand, J.; Nielsen, J.B.; Havarinasab, S.; Zalups, R.K.; Söderkvist, P.; Hultman, P.

'Mercury toxicokinetics-dependency on strain and gender'. Toxicol. Appl. Pharmacol. 2010, 243, 283-291. 2.

Kisku, G.C.; Pandey, P.; Singh Negi, M.P.; Misra, V. Uptake and accumulation of potentially toxic metals (Zn, Cu and Pb) in soils and plants of Durgapur industrial belt. J. Environ. Biol. 2011, 32, 831-838.

3.

Habeebu S.S.; Liu Y.; Park J.D.; Klaassen C.D. Strain differences in the toxicity of

cadmium to trigeminal ganglia in mice. Toxicol. Appl. Pharmacol. 2001, 177, 200-207. 4.

Sharma, V. Sharma, A.; Kansal, L. The effect of oral administration of Allium sativum

extracts on lead nitrate induced toxicity in male mice. Food Chem. Toxicol. 2010, 48, 928-936. 5.

Jenne, E.A.; Luoma, S.N. Forms of Trace Elements in Soils, Sediments and Associated

Waters: An Overview of Their Determination and Biological Availability. In Biological implications of metals in the environment; Wildung, R. E., Drucker, H., Eds Washington, 1977, 110-143. 6.

Randall, S.R.; Sherman, D.M.; Ragnarsdottir, K.V.; Collins, C.R. The mechanism of

cadmium surface complexation on iron oxyhydroxide minerals. Geochim. Cosmochim. Acta 1999, 63, 2971-2987.

22

7.

Schipper, A.M.; Wijnhoven, S.; Leuven, R.S.E.W.; Ragas, A.M.J.; Hendriks, J. Spatial

distribution and internal metal concentrations of terrestrial arthropods in a moderately contaminated lowland floodplain along the Rhine River. Environ. Pollut. 2008, 151, 17-26. 8.

Agashe, K.B.; Regalbuto, J.R. A revised physical theory for adsorption of metal

complexes at oxide surfaces. J. Colloid Interface Sci. 1997, 185, 174-189. 9.

Barrow, N.J.; Brümmer, G.W.; Strauss, R. Effects of surface heterogeneity on ion

adsorption by metal oxides and by soils. Langmuir 1993, 9, 2606-2611. 10.

Rodda, D.P.; Johnson, B.B.; Wells, J.D. The Effect of Temperature and pH on the

Adsorption of Copper(II) Lead(II) and Zinc(II) onto Goethite. J. Colloid Interface Sci. 1993, 161, 57-62. 11.

Pagnanelli, F.; Esposito, A.; Toro, L.; Vegliò, F. Metal speciation and pH effect on Pb,

Cu, Zn and Cd biosorption onto Sphaerotilus natans: Langmuir-type empirical model. Water Res. 2003, 37, 627-633. 12.

Kaya, K.; Pehlivan, E.; Schmidt, C.; Bahadir, M. Use of modified wheat bran for the

removal of chromium(VI) from aqueous solutions. Food Chem. 2014, 158, 112-117. 13.

Azouaou, N.; Sadaoui, Z.; Mokaddem, H. Removal of lead from aqueous solution onto

untreated coffee grounds: A fixed-bed column study. Chem. Eng. Trans. 2014, 38, 151-156. 14.

Kausar, A.; Bhatti, H. N.; MacKinnon, G. Equilibrium, kinetic and thermodynamic

studies on the removal of U(VI) by low cost agricultural waste. Colloids Surf., B 2013, 111, 124133.

23

15.

Angove, M.J.; Wells, J.D.; Johnson, B.B. The Influence of Temperature on the

Adsorption of Cadmium(II) and Cobalt(II) on Goethite. J. Colloid Interface Sci. 1999, 211, 281290. 16.

Funes, A.; de Vicente, J.; Cruz-Pizarro, L.; de Vicente, I. The influence of pH on

manganese removal by magnetic microparticles in solution. Water Res. 2014, 53, 110-122. 17.

Malamis, S.; Katsou, E. A review on zinc and nickel adsorption on natural and modified

zeolite, bentonite and vermiculite: Examination of process parameters, kinetics and isotherms. J. Hazard. Mater. 2013, 252-253, 428-461. 18.

Rodda, D.P.; Johnson, B.B.; Wells, J.D. Modeling the effect of temperature on adsorption

of lead(II) and zinc(II) onto goethite at constant pH. J. Colloid Interface Sci. 1996, 184, 365-377. 19.

Nguyen, N.V.; Jeong, J.; Jha, M.K.; Lee, J.; Osseo-Asare, K. Comparative studies on the

adsorption of Au(III) from waste rinse water of semiconductor industry using various resins. Hydrometallurgy 2010, 105, 161-167. 20.

Harding, I.H.; Healy, T.W. Adsorption of aqueous cadmium(II) on amphoteric latex

colloids. I. General kinetics and thermodynamics. J. Colloid Interface Sci. 1985, 107, 362-370. 21.

Bradl, H.B. Adsorption of heavy metal ions on soils and soils constituents. J. Colloid

Interface Sci. 2004, 277, 1-18. 22.

Criscenti, L.J.; Sverjensky, D.A. A single-site model for divalent transition and heavy

metal adsorption over a range of metal concentrations. J. Colloid Interface Sci. 2002, 253, 329352.

24

23.

Velazquez-Jimenez, L. H.; Pavlick, A.; Rangel-Mendez, J. R. Chemical characterization

of raw and treated agave bagasse and its potential as adsorbent of metal cations from water. Ind. Crops Prod. 2013, 43, 200-206. 24.

Benjamin, M.M.; Leckie, J.O. Adsorption of Metals on Oxide Interfaces: Effects of the

Concentrations of Adsorbate and Competing Metals. Contam. Sediments 1980, 2, 305-322. 25.

Janusz, W.; Patkowski, J.; Chibowski, S. Competitive adsorption of Ca2+ and Zn(II) ions

at monodispersed SiO2/electrolyte solution interface. J. Colloid Interface Sci. 2003, 266, 259268. 26.

Chibowski, S.; Janusz, W. Specific adsorption of Zn(II) and Cd(II) ions at the -

Fe2O3/electrolyte interface structure of electrical double layer. Appl. Surf. Sci. 2002, 196, 343355. 27.

James, R.O.; Healy, T.W. Adsorption of hydrolyzable metal ions at the oxide-water

interface. III. A thermodynamic model of adsorption. J. Colloid Interface Sci. 1972, 40, 65-81. 28.

Fischer, L.; Brümmer, G.W.; Barrow, N.J. Observations and modelling of the reactions of

10 metals with goethite: Adsorption and diffusion processes. Eur. J. Soil Sci. 2007, 58, 13041315. 29.

James, R.O. The adsorption of hydrolysable metal ions at the oxide-water interface. Ph.D.

Thesis, University of Melbourne, 1971. 30.

Lowson, R.T.; Evans, J.V. Adsorption and Desorption of Heavy Metals on -Alumina.

Aust. J. Chem. 1984, 37, 2165-2178.

25

31.

Tamura, H. Theorization on ion-exchange equilibria: Activity of species in 2-D phases. J.

Colloid Interface Sci. 2004, 279, 1-22. 32.

Davis, J.A.; Leckie, J.O. Surface ionization and complexation at the oxide/water interface

II. Surface properties of amorphous iron oxyhydroxide and adsorption of metal ions. J. Colloid Interface Sci. 1978, 67, 90-107. 33.

Therdthianwong, S.; Tongsima, A.; Therdthianwong, A. Adsorption of Palladium on

MgO/Al2O3 and CuO/Al2O3. ScienceAsia 2004, 30, 399-410. 34.

Corapcioglu, M.O.; Huang, C.P. The adsorption of heavy metals onto hydrous activated

carbon. Water Res. 1987, 21, 1031-1044. 35.

Zayat, M. E.; Smith, E. Modeling of heavy metals removal from aqueous solution using

activated carbon produced from cotton stalk. Water Sci. Technol. 2013, 67, 1612-1619. 36.

Yasungada, T. and Ikeda, T. Adsorption - Desorption Kinetics at the Metal-Oxide –

Solution Interface Studied by Relaxation Methods. American Chemical Society, Washington D.C., 1986. 37.

Kumar, R.; Khan, M. A.; Haq, N. Application of Carbon Nanotubes in Heavy Metal

Remediation. Crit. Rev. Environ. Sci. Technol. 2014, 44, 1000-1035. 38.

Burns, C.A.; Cass, P.J.; Harding, I.H.; Crawford, R.J. Adsorption of aqueous heavy

metals onto carbonaceous substrates. Colloids Surf., A 1999, 155, 63-68. 39.

McKenzie, R. The Adsorption of Lead and Other Heavy Metals on Oxides of Manganese

and Iron. Aust. J. Soil Res. 1980, 18, 61-73.

26

40.

Crawford, R.J.; Harding, I.H.; Mainwaring, D.E. Adsorption and coprecipitation of single

heavy metal ions onto the hydrated oxides of iron and chromium. Langmuir 1993, 9, 3050-3056. 41.

Benjamin, M.M.; Leckie, J.O. Multiple-site adsorption of Cd, Cu, Zn, and Pb on

amorphous iron oxyhydroxide. J. Colloid Interface Sci. 1981, 79, 209-221. 42.

Ikhsan, J.; Johnson, B.B.; Wells, J.D. A comparative study of the adsorption of transition

metals on kaolinite. J. Colloid Interface Sci. 1999, 217, 403-410. 43.

Della Puppa, L.; Komárek, M.; Bordas, F.; Bollinger, J.; Joussein, E.; Adsorption of

copper, cadmium, lead and zinc onto a synthetic manganese oxide. J. Colloid Interface Sci. 2013, 399, 99-106. 44.

Puranik, P.R.; Paknikar, K.M. Influence of co-cations on biosorption of lead and zinc - A

comparative evaluation in binary and multimetal systems. Bioresour. Technol. 1999, 70, 269276. 45.

Kesenci, K.; Say, R.; Denizli, A. Removal of heavy metal ions from water by using

poly(ethyleneglycol dimethacrylate-co-acrylamide) beads. Eur. Polym. J. 2002, 38, 1443-1448. 46.

Arpa, Ç.; Alim, C.; Bektaş, S.; Genç, Ö.; Denizli, A. Adsorption of heavy metal ions on

polyhydroxyethylmethacrylate microbeads carrying Cibacron Blue F3GA. Colloids Surf., A 2001, 176, 225-232. 47.

Ajmal, M.; Rao, R.A.K.; Siddiqui, B.A. Adsorption studies and the removal of dissolved

metals using pyrolusite as adsorbent. Environ. Monit. Assess. 1995, 38, 25-35. 48.

Fourest, E.; Roux, J.C. Heavy metal biosorption by fungal mycelial by-products:

Mechanisms and influence of pH. Appl. Microbiol. Biotechnol. 1992, 37, 399-403.

27

49.

Kinniburgh, D.G.; Jackson, M.L.; Syers, J.K. Adsorption of Alkaline Earth Transition,

and Heavy Metal Cations by Hydrous Oxide Gels of Iron and Aluminium. Soil Sci. Soc. Am. J. 1976, 40, 796-800. 50.

Burnett, P.G.G.; Handley, K.; Peak, D.; Daughney, C.J. Divalent metal adsorption by the

thermophile Anoxybacillus flavithermus in single and multi-metal systems.

Chem.

Geol.

2007, 244, 493-506. 51.

Fein, J.B.; Daughney, C.J.; Yee, N.; Davis, T.A. A chemical equilibrium model for metal

adsorption onto bacterial surfaces. Geochim. Cosmochim. Acta 1997, 61, 3319-3328. 52.

Fein, J.B.; Martin, A.M.; Wightman, P.G. Metal adsorption onto bacterial surfaces:

Development of a predictive approach. Geochim. Cosmochim. Acta 2001, 65, 4267-4273. 53.

Harding, I.H. Amphoteric polystyrene latex colloids: polymerization pathway and the

control of particle size and potential. Colloid Polym. Sci. 1985, 263, 58-66. 54.

Kosmulski, M. Zeta for Windows. Department of Electrochemistry, Lublin University of

Technology, 2002. 55.

Kosmulski, M. The pH-dependent surface charging and points of zero charge: V. Update.

J. Colloid Interface Sci. 2011, 353, 1-15. 56.

Baes, C.F. and Mesmer, R.E. The Hydrolysis of Cations. Wiley, New York, 1976.

57.

Sillen, L.G. and Martell, A.E. Stability Constants. The Chemical Society, London, 1964.

58.

Harding, I.H.; Healy, T.W. Electrical double layer properties of amphoteric polymer latex

colloids. J. Colloid Interface Sci. 1985, 107, 382-397.

28

59.

Homola, A.; James, R.O. Preparation and characterization of amphoteric polystyrene

latices. J. Colloid Interface Sci. 1977, 59, 123-134. 60.

Puziy, A.M.; Poddubnaya, O.I.; Zaitsev, V.N.; Konoplitska, O.P. Modeling of heavy

metal ion binding by phosphoric acid activated carbon. Appl. Surf. Sci. 2004, 221, 421-429.

29

Graphical Abstract

30