REMOVAL OF HEAVY METALS FROM SOIL COMPONENTS AND

SOILS BY NATURAL CHELATING AGENTS

Part II. Soil Extraction by Sugar Acids

K. FISCHER

1

∗

and H.-P. BIPP

2

1

University of Trier, FB VI – Geography/Geosciences, Department of Analytical and Ecological

Chemistry, Trier, Germany;

2

Infineon Technologies AG, Munich, Germany

(

∗

author for correspondence, e-mail: [e-mail], fax: [phone])

(Received 14 March 2000; accepted 28 September 2001)

Abstract. Aqueous solutions of the natural chelating agents D-gluconic acid and D-glucaric acid
(D[+]-saccharic acid) were tested for their ability to remove heavy metals (Cd, Cr, Cu, Ni, Pb, Zn)
from a soil polluted by long-term application of sewage sludge. Batch equilibrium experiments were
performed under variation of fundamental process parameters, i.e. pH value, sugar acid concentra-
tion, batch solution volume, solid:liquid ratio and number of treatment cycles. The extractability of
heavy metals was low under near-neutral and slightly basic pH conditions. It increased drastically
between pH 12.0 and 13.0. Pb and Cu were preferentially extracted metals. Compared with the
extraction efficiency of pH adequate pure sodium hydroxide solutions, the sugar acids enhanced the
solubilisation of Pb and Cr especially. The metal depletion from soil was the highest when applying
20 or 50 g L

−1

solutions of the chelating agents. Under strongly basic conditions solid:liquid ratios

of 1:10 or 1:20 were proofed to be advantageous. Except Ni, multi-step extraction improved the
metal removal strongly. This effect was the greatest for Cr extraction. Under optimised conditions the
following metal extraction degrees were achieved with strongly alkaline D-gluconic acid solutions:
Ni 43%, Cr 60%, Cd 63%, Zn 70%, Pb 80%, and Cu 84%.

Keywords: D-glucaric acid, D-gluconic acid, heavy metals, metal extraction from polluted soil,
natural chelating agents, sugar acids

1. Introduction

Sugar acids, such as D-gluconic acid and 2-keto-D-gluconic acid, are natural oxid-
ation products of D-glucose and D-glucose containing polysaccharides. They are
also intermediate compounds in tartrate and ascorbic acid biosynthesis of various
plants (Sponholz, 1990). Various micro-organisms, e.g. Aspergillus niger, Gluc-
onobacter (sub)oxidans, Pseudomonas spp. and acetic acid bacteria generate acidic
oxidation products of D-glucose (Rowatt and Katznelson, 1961; Sokatch, 1969;
Gosselé et al., 1980). Due to the exudation of carbohydrates by plant roots and the
corresponding high microbial activity in the rhizosphere soil, 2-keto-D-gluconic
acid is commonly found as a rhizosphere product of wheat plants especially (Duff
and Webley, 1959; Webley and Duff, 1965; Stevenson and Ardakani, 1973; Moghimi
et al., 1978a). Sugar acids, in particular 2-keto-D-gluconic acid, are able to solu-
bilise phosphate minerals and phosphate rocks by chelation of the earth alkaline

Water, Air, and Soil Pollution 138: 271–288, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.

272

K. FISCHER AND H.-P. BIPP

mineral cations and by surface-coordinated ligand exchange reactions, thus in-
creasing the bioavailability of phosphate for plants (Duff et al., 1963; Moghimi
et al., 1978b). Furthermore 2-keto-D-gluconic acid plays a role in the reductive
dissolution of metal oxides in soils (Schwertmann et al., 1986; Eckhardt, 1985;
Jauregui and Reisenauer, 1982).

D-gluconic acid is non-toxic and easily biodegradable (Leroy, 1976). Combined

with its ability to chelate earth alkaline elements and heavy metals, these properties
led to diverse applications in medicine and industry (Sawyer, 1964; Hustede et al.,
1989). Amongst other applications, the use as sequestering agent for heavy metals
in alkaline cleansing solutions and as scale inhibitor is worth mentioning.

In a few studies conducted by Peters and coworkers to select chelating agents

for use in ‘on site’ or ‘off site’ soil washing procedures, D-gluconic acid is con-
sidered as a possible extractant for metal polluted soils (Peters, 1999; Peters et
al., 1994, 1995; Brewster et al., 1995). Applying the polyhydroxy carboxylic acid
under acidic and neutral pH conditions at concentrations of 10 mmol L

−1

(1.96 g

L

−1

) and 0.1 mol L

−1

(19.6 g L

−1

), the authors assessed this compound to be inef-

fective in mobilising heavy metals from contaminated soils. Since a deprotonation
of hydroxylic groups is necessary for the formation of very stable coordination
compounds by sugar acids, which requires strongly alkaline pH values (Pecsok
and Juvet, 1955, 1956; Coccioli and Vicedomini, 1978; Blomqvist and Still, 1985)
this conclusion is not surprising.

Actually the investigation of the metal extraction properties of sugar acids re-

ceives additional interest from a new approach for the remediation of polluted
soils, which is based on the utilisation of biomass residues (Fischer et al., 1998).
The oxidising conversion of carbohydrate containing residues from agriculture and
food industry by nitric acid, e.g. molasses, potato peel sludge, spent hops, and vine
yeast, results in hydrolysates, which contain the sugar acids glucaric, gluconic,
threonic and glyceric acid as main or minor components together with various other
aliphatic carboxylic acids (Bipp, 1996; Bipp et al., 1997). It could be shown that
these hydrolysates are effective in mobilising e.g. Cu and Pb from polluted soils
under alkaline conditions (Fischer et al., 1998).

Therefore it was the main objective of the current work to investigate the leach-

ing characteristics and the extraction selectivities of sugar acids which affect the
application properties of biomass hydrolysates. This included the examination of
optimised extraction conditions for sugar acids, providing highest leaching effi-
ciencies.

The present study is an integrated part of a greater project to assess the metal

leaching properties of natural chelating agents. Further details concerning the back-
ground and the experimental set-up of the investigation are provided in part I of the
article series (Fischer, 2002).

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

273

2. Materials and Methods

The soil, a humus, sandy silty loam, classified as Rendzic Leptosol (FAO) with
cambic B horizon, was collected at 0–30 cm depth from a former agricultural site
north of Munich, Germany. The metal burden and the high organic matter content
were caused by decades of sewage sludge amendment. The physical and chem-
ical properties of the air-dried and thoroughly mixed soil sample were determined
according to several DIN (German Institute of Standardisation) norms and further
soil analytical standard procedures (see part I of the article; Fischer et al., 1998;
Schlichting et al., 1995). The total heavy metal content was analysed by inductively
coupled plasma–atomic emission spectroscopy (ICP-AES) in aqua regia digests
(10 replicates) according to DIN (1983).

Based on these methods, the following soil characteristics were determined: tex-

ture: 27% sand (63 µm–2 mm), 45% silt (2–63 µm), 28% clay (<2 µm), 121 g kg

−1

organic matter, 157 g kg

−1

carbonate content, pH 6.5 (0.01 M CaCl

2

), 280 mmol

c

kg

−1

CEC, heavy metal content (means and standard deviations of 10 digests)

(mg kg

−1

): Cd: 44

±2, Cr: 224±9, Cu: 287±20, Ni: 72±3, Pb: 1242±54, Zn:

1688

±106.

A seven-step sequential extraction procedure, detailed elsewhere (Zeien and

Brümmer, 1989; Fischer et al., 1998), was used to divide the metals into mobile,
easily available, Mn oxide bound, organically bound, poorly crystallised Fe oxide
bound, well crystallised Fe oxide bound, and residual fraction.

The sugar acids D-gluconic acid (sodium salt, Merck, Darmstadt, Germany) and

D-glucaric acid (potassium salt, Sigma, St. Louis) were chosen as metal extracting
agents. A compilation of thermodynamic equilibrium constants for the formation
of 1:1 complexes between D-gluconic acid and the interesting metal ions at neutral
and strongly alkaline pH is given in Table I. Some ranges of stability constants of
D-glucaric acid metal chelates were determined with different methods at various
pH values by Velasco et al. (1976): Cu(II), pH 3.65, lgK 13.36–14.30; Cu(II), pH
5.56–9.5, lgK 6.13–7.37; Ni(II), pH 9.5, lgK 8.64–9.27; Cd(II), pH not specified,
lgK 8.85–9.07.

To maintain predefined pH values during metal extraction, the following buffer

solutions were used:

– 0.05 mol L

−1

MES [2-(N-morpholino)ethanesulfonic acid monohydrate], buf-

fering range pH 4.5–5.5;

– 0.1 mol L

−1

HEPES [N-(2-hydroxyethyl)piperazine-N

-1-ethanesulfonic acid],

buffering range pH 6.5–7.5;

– sodium carbonate/bicarbonate buffer, pH 10.0.

Other pH values were adjusted by adding of sodium hydroxide or hydrochloric acid
to the batch solutions.

274

K. FISCHER AND H.-P. BIPP

TABLE I

Stability constants (lg K

1

:1

) of various 1:1 metal gluconate

complexes at weak acidic/neutral and at strongly alkaline pH

Metal ion

Weak acidic/neutral pH

Strongly alkaline pH

lg K

1

:1

lg K

1

:1

Cd(II)

1.15

a

15.6

d

Cu(II)

2.15

a

18.3

e

Ni(II)

1.82

b

n.d.

Pb(II)

2.13

a

16.7

f

Zn(II)

1.70

c

n.d.

n.d. = No data.

a

Vicedomini, 1983.

b

Sillen and Martell, 1971.

c

Martell and Smith, 1974, 1977.

d

Complex type [M

2

L

2

]

2

−

; Blomqvist and Still, 1985.

e

Pecsok and Juvet, 1955.

f

Coccioli and Vicedomini, 1978.

Batch-shaker tests were conducted to extract the metal polluted soil in duplicate.

Therefore centrifuge tubes were filled with 1.0 g of soil and with 5 to 40 mL of
the batch solution. The concentration of the sugar acids varied between 5.0 and
100.0 g L

−1

. The pH values of the leaching tests spanned from 4.5 to 13.5. Single

and multiple step extraction experiments were performed, each step lasting 24 hr.
The soil samples were kept in suspension by means of a reciprocating shaker. In
measuring the total aqueous metal concentrations, aliquots were withdrawn from
the reaction mixtures and passed through 0.2 µm cellulose nitrate filters (Sartorius,
Göttingen, Germany). After acidification of the filtrates by adding of small amounts
of concentrated nitric acid, metal analysis was carried out using ICP-AES. All test
series were accompanied by pH aligned control experiments omitting the sugar
acid component.

3. Results and Discussion

3.1. M

ETAL BONDING IN SOIL

As the sequential leaching procedure reveals (Table II), the greatest proportions
of the investigated metals (except Cr) are associated with fraction 4 ‘(organically
bound)’ which is operationally defined by extraction with 0.025 mol L

−1

(NH

4

)

2

H

2

EDTA at pH 4.6. This bonding type is predominant in the case of Cu and Pb
especially. More than 80% of soil Cr are bound onto Fe oxides. This bonding mode
accounts for significant amounts of Ni also. Higher proportions of Cd (30.5%), as

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

275

TABLE II

Soil metal speciation

Metal fraction

(unit)

Cd

Cr

Cu

Ni

Pb

Zn

1. Mobile

(mg kg

−1

)

1.0

0.1

b

4.1

b

0.6

a

0.6

b

12.8

(%)

2.2

<0.1

1.4

0.8

<0.1

0.7

2. Easily available

(mg kg

−1

)

13.6

0.3

b

7.1

4.2

a

28.1

178.0

(%)

30.5

0.1

2.5

5.3

2.2

9.4

3. Bound onto

(mg kg

−1

)

9.5

0.3

a

6.1

3.3

b

89.3

260.0

Mn-oxides

(%)

21.3

0.1

2.1

4.1

7.0

13.7

4. Organically bound

(mg kg

−1

)

15.7

13.7

193.0

24.2

976.0

901.0

(%)

35.2

6.0

67.0

30.4

76.9

47.4

5. Bound onto poorly

(mg kg

−1

)

0.4

a

108.0

55.9

15.4

21.4

201.0

crystallized Fe-oxides

(%)

0.9

47.2

19.4

19.3

1.7

10.6

6. Bound onto

(mg kg

−1

)

2.3

77.5

2.4

a

19.4

66.0

265.0

crystallized Fe-oxides

(%)

5.2

33.8

0.8

24.4

5.2

13.9

7. Residually bound

(mg kg

−1

)

2.1

a

29.2

19.9

12.5

87.1

84.5

(%)

4.7

12.8

6.9

15.7

6.9

4.4

Sum of fractions 1–7

(mg kg

−1

)

44.6

229.0

288.0

79.6

1269.0

1902.0

Percent of aqua regia digest (%)

100.7

102.2

101.4

109.9

102.2

112.7

Main bonding fractions are marked in bold. Means of three replicates are given. The variation
coefficients are less than 10% typically. Higher variations are indicated:

a

10–25%;

b

>25%.

a commonly mobile element, were brought into solution with NH

4

OAc (fraction 2)

and attributed as ‘easily available’. Bonding onto Mn oxides is relevant for Cd and,
with lesser importance, for Zn. The residual fraction plays a minor role in the metal
bonding spectrum. About 16% of soil bound Ni and 13% of Cr are apportioned to
this fraction. Portions between 4.5 and 7.0% are typical for the other elements.

It can be concluded, that the metal distribution reflects the contamination path

to a high degree. Obviously the initial metal-organic matter association remained
stable during more than 20 yr of soil forming and of the proceeding conversion of
the organic substrate. As one considers metals bound in fractions 1–4 as readily or
moderately available under normal soil chemical conditions, roughly 70–90% of
Cd, Cu, Pb and Zn come within this category, indicating a serious environmental
burden.

4. Metal Extraction

The time course of the solubilisation of heavy metals by a 20 g L

−1

aqueous

solution of D-gluconic acid at pH 7.0 is summarised in Table III. The solute con-
centrations of Ni, Cd, Cr, and Zn increased more or less continuously during the

276

K. FISCHER AND H.-P. BIPP

TABLE III

Time dependent metal leaching from contaminated
soil by D-gluconic acid at pH 7.0

Time

Cd

Cr

Cu

Ni

Zn

(hr)

Depletion degrees (%)

2

3.0

0.5

4.2

9.0

5.6

4

4.0

1.0

4.6

10.3

6.0

8

4.4

1.3

5.4

11.1

6.6

16

4.5

1.9

5.6

11.6

7.0

24

5.3

2.4

6.0

12.7

7.1

48

6.6

3.2

4.5

14.6

7.6

72

9.1

3.9

4.8

17.6

11.8

Concentration of aqueous D-gluconic acid solution:
20.0 g L

−1

. Solid:liquid-ratio 1:10, Pb extraction

<1.0%.

Figure 1. pH dependent metal extraction by D-glucaric acid. Treatment time 24 hr, acid concentra-
tion: 20.0 g L

−1

.

whole time span of 72 hr, whereas the solute concentration of Cu reached its max-
imum after 24 hr. Pb (not listed) remained totally immobile. Except Ni, only very
small metal portions were mobilised under chosen conditions.

A series of batch tests with 20 g L

−1

solutions of D-glucaric (saccharic) and D-

gluconic acids were conducted in the range from neutral to highly alkaline pH. The
relative metal amounts extracted over 24 hr are comprised in Figure 1 (D-glucaric
acid) and 2 (D-gluconic acid), both indicating a strong pH influence on the metal

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

277

Figure 2. pH dependent metal extraction by D-gluconic acid. Treatment time 24 hr, acid concentra-
tion: 20.0 g L

−1

.

extractability. The amounts of added sodium hydroxide necessary to maintain a
certain pH value, are mentioned in the figure legend to allow comparisons with the
extraction performance of differently concentrated solutions of the pure base.

The pH-related extraction profiles of the chelating agents express a low ex-

traction efficiency at neutral and moderately alkaline pH combined with a sharp
increase of the extraction power between pH 12 and pH 13. Applying D-glucaric
acid, this property holds for Cd, Cr and Pb especially. Using D-gluconic acid, the
solubilisation of Zn is increased by a factor of 45, rising the pH from 12.0 to 12.9.
In both cases, the liberation of Ni gradually increases with increasing pH.

Except for Cd, highest leaching degrees were achieved at highest pH values

(NaOH content of 30 g L

−1

). Under these experimental conditions, the follow-

ing metal leaching selectivities were determined (in parentheses: metal extraction
degrees):

D-gluconic acid:

Pb (66%) > Cu (62%) > Cd (48%) > Zn (45%)

≈ Ni (43%) > Cr (39%)

D-glucaric acid:

Cu (66%)

≈ Pb (64%) > Cr (43%) > Ni (35%) > Zn (22%) ≈ Cd (21%) .

These results elucidate that the differences between the extraction efficiency of the
D-gluconic acid solution and of the D-glucaric acid solution are very small in the
case of Cu, Pb and Cr. As Figures 1 and 2 illustrate, D-gluconic acid solution was
twice as effective as the aldaric acid solution at mobilising Cd and Zn. Since D-
glucaric acid is expected to form coordination compounds being more stabile than

278

K. FISCHER AND H.-P. BIPP

TABLE IV

Metal extraction by diluted sodium hydroxide solutions

Extraction pH

Cd

Cr

Cu

Ni

Pb

Zn

Extraction degrees (%)

8

1.9

<0.1

1.5

3.4

<0.1

<0.1

10

2.2

1.0

12.8

6.7

0.7

0.7

12

2.0

0.2

2.0

a

7.6

0.2

0.1

13.0 (5.0 g L

−1

NaOH)

12.5

3.2

43.1

10.3

2.4

2.4

13.3 (10.0 g L

−1

NaOH)

15.0

2.7

46.1

20.8

3.8

4.1

13.5 (20.0 g L

−1

NaOH)

12.0

2.4

51.0

23.8

7.6

12.8

13.7 (30.0 g L

−1

NaOH)

10.8

2.8

53.1

23.8

11.3

21.2

a

Measurement uncertain.

those of D-gluconic acid, which is proofed for some tetravalent metal ions (Sawyer,
1964), the formation of polymeric coordination compounds and of other metal salts
having low solubilities (Velasco et al., 1976) might be one of the reasons for the
lower extraction effectiveness of this sugar acid. Calcium ions, which are brought
into solution by the dissolution of soil carbonates, are limiting the solubility of
D-glucaric acid additionally.

At least three factors may contribute to the heavy metal mobilisation under

strongly alkaline conditions:

a) formation of soluble coordination compounds of the organic chelating agents;

b) formation of soluble metal hydroxide complexes; and

c) mobilisation of the organically bound metal fraction in consequence of the

solubilisation of humic and fulvic acids at NaOH concentrations of 5.0 g L

−1

and higher.

The extraction of humic and fulvic acids by 5.0 g L

−1

solutions of NaOH is a

standard method in soil analysis (Kononowa, 1966). Considering the fact that high
proportions of Pb, Cu, and Zn were operationally defined as organically bound
(Table II), the latter effect may be of special importance.

To investigate the mobilization of heavy metals under alkaline conditions in the

absence of chelating agents, leaching tests with diluted solutions of pure NaOH
were performed.

As it can be deduced from Table IV, the extractability of most of the metals

increased with increasing NaOH concentration. Only small metal portions were
leached with NaOH concentrations below 5.0 g L

−1

. The elevation of the pH value

from 12.0 to 13.0 favoured the solubilisation of Cu and Cd especially. An even
higher hydroxide concentration fostered the extraction of Zn, Pb, and Ni above all.

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

279

TABLE V

Calculated total metal activities in the liquid phase (conditions: equilibrium with solid hydrox-
ides, formation of soluble hydroxide complexes) and measured metal concentrations in NaOH
soil extracts

Unit

Metal ions

Cd(II)

Cr(III)

Cu(II)

Ni(II)

Pb(II)

Zn(II)

pH 12

Calculated activities

(mg L

−1

)

0.03

2.10

0.62

0.004

0.009

10.79

Measured conc.

(mg L

−1

)

0.09

0.04

0.58

0.550

0.260

0.19

pH 13

Calculated activities

(mg L

−1

)

0.06

18.20

1.63

0.040

0.610

163.80

Measured conc.

(mg L

−1

)

0.55

0.73

12.40

0.740

2.950

4.08

The activities were calculated using the PHREEQC2.3 program (Parkhurst, 1995). Data base
was MINTEQA2 (Allison et al., 1990). The data refer to a temperature of 25

◦

C.

The experimental design does not allow to distinguish between metal mobil-

isation via hydroxide complex formation and via extraction of the organic binding
matrix clearly. To offer a certain orientation for the discussion of these factors,
the total activities of the metals in the liquid phase at equilibrium with their solid
hydroxide compounds were calculated for pH 12 and 13 on the basis of the pH-
dependent distribution of their solved hydroxide species (Table V). The speciation
was modelled using the PHREEQC2.3 software (Parkhurst, 1995) in combination
with the MINTEQA2 database (Allison et al., 1990). It must be pointed out that
this speciation does not claim to draw a realistic picture of the chemical composi-
tion of the soil extracts, since all other solved ions and compounds were ignored.
The calculated data were compared with the measured metal concentrations of the
sodium hydroxide extracts at pH 12 and 13 (Table IV).

As Table V clarifies only the found concentrations of Ni and Pb exceeded

drastically the metal activities calculated for pH 12. At least for Ni the calcu-
lated activity of 6.9

× 10

−8

mol L

−1

seems to be at the lower limit of the com-

monly accepted value range. For instance Jenne (1968) calculated a Ni activity of
6.5

× 10

−7

mol L

−1

.

More interesting is the situation at pH 13 referring to conditions often applied

for soil organic matter (SOM) extraction. Remarkable is the high concentration of
Cu, which surpasses the calculated activity by a factor of about 7.5. Having in mind
that about two third of Cu are organically bound this might be an indication for Cu
mobilisation by SOM extraction. Supposed that all of the solved Cu originated
from the organically bound fraction, about 64% of this fraction were extracted.

280

K. FISCHER AND H.-P. BIPP

TABLE VI

Metal leaching factors (ratios between metal amounts mo-
bilized by D-gluconic acid and by pH-adequate sodium
hydroxide solutions)

NaOH

Cd

Cr

Cu

Ni

Pb

Zn

(g L

−1

)

Leaching factor

10.0

2.9

10.2

1.1

1.7

14.0

8.7

20.0

4.1

14.3

1.1

1.5

7.8

3.3

30.0

4.5

13.9

1.2

1.8

5.7

2.1

On the other hand the other metals, which are characterised by high portions

associated with SOM, i.e. Pb and Zn (Table II), remained to a high extent fixed
on soil particles. This is especially remarkable in the case of Zn whose measured
concentration is far below its calculated activity. Estimated in the same way as for
Cu, only 3.1% (Pb) and 5.1% (Zn) of the organically bound fraction of these metals
were mobilised.

To explain this phenomenon, at least two hypothesis can be formulated. First,

the binding affinity of Cu for specific organic substrates differed greatly from Pb
and Zn and the organic components associated with Cu were much more soluble in
NaOH solutions than the Pb and Zn binding components. Second, the extraction of
fulvic and humic acids initiated a redistribution of Zn and Pb between the different
soil phases. Such processes are repeatedly reported to occur as side reactions during
sequential leaching procedures performed to investigate heavy metal speciation in
soils and sediments (Beckett, 1989).

To illustrate the relative importance of sugar acid complex formation for the

solubilisation of the different metal ions, leaching factors were calculated, which
express the ratios between metal concentrations in D-gluconate extracts and in the
pH-adequate NaOH extracts (Table VI). These calculations support the assumption
that the solubilization of Cr and Pb under strongly alkaline conditions is in the main
attributable to organic chelation. In contrast to this, the adding of D-gluconic acid
to NaOH solutions does not considerably enhance the extraction of Cu. At pH 13
the sugar acid solution was much more effective at mobilising Zn than the pure
NaOH solution. This difference is partially levelled out by a further increase of the
hydroxide ion concentration. Based on the means of the calculated leaching factors
summarised in Table VI, the following sequence can be arranged, which elucidates
the relative contribution of D-gluconate complex formation to the overall leaching
of heavy metals at strongly alkaline pH:

Cu < Ni < Cd < Zn < Pb < Cr .

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

281

Figure 3. Metal extraction by differently concentrated solutions of D-gluconic acid under addition of
10 g L

−1

NaOH.

Besides the already mentioned effects of SOM extraction and hydroxide complex
formation other factors have to be taken into account to explain the strong increase
of the leaching efficiencies of the sugar acid solutions provoked by a rise of the
pH from 12.0 to 13.0 (Figures 1 and 2). For this the reason seems to be the de-
protonation of alcoholic hydroxy groups of the sugar acids at strongly basic pH,
leading to altered complex structures such as chelate rings and to drastic increases
of the thermodynamic stability of the formed coordination compounds (Table I)
(Sawyer, 1964). According to Coccioli and Vicedomini (1978) and to Velasco et
al. (1976), D-gluconic acid and D-glucaric acid can act as trisdentate ligands in the
complex formation with Pb(II)- and Cu(II)-ions in alkaline media. D-glucaric acid
is supposed to function as tetradentate ligand in coordinating trivalent ions such as
Fe(III).

A further reason for the low leaching efficiencies of the sugar acids at neutral

and slightly basic pH is the low solubility of some of their metal salts in this
pH range. Pecsok and Juvet (1956) described the formation of sparingly soluble
lead salts in the pH range of 6–10. Joyce and Pickering (1965) investigated the
formation of an insoluble Ni-gluconate complex in the pH range of 7–9.

The influence of the D-gluconic acid concentration on the metal release was

tested with batch solutions containing 10 and 30 g L

−1

of NaOH. The concen-

tration of the sugar acid varied between 5.0 and 100.0 g L

−1

. As a consequence,

the resulting pH values spanned from approximately 12.4 to 13.0 after adding of
10.0 g L

−1

of NaOH (Figure 3) and from approximately 12.8 to 13.5 after adding

of 30.0 g L

−1

of NaOH (Figure 4). Due to the hydroxide ion consumption by the

sugar acid the pH values and the gluconic acid concentrations of the solutions are
inversely correlated.

As illustrated in Figure 3, a significant influence of the sugar acid concentration

on metal release at pH values

≤13.0 is ascertained for the elements Cd, Cr and

282

K. FISCHER AND H.-P. BIPP

Figure 4. Metal extraction by differently concentrated solutions of D-gluconic acid under addition of
30 g L

−1

NaOH.

Zn. The highest extraction yields were achieved with the 50.0 g L

−1

solution for

these metals. These yields are about twice as high as those of the 5.0 g L

−1

solution

(lowest extraction efficiency). The extractability of Pb was nearly uninfluenced by
the applied sugar acid concentration. Generally the 100.0 g L

−1

solution was less

effective at metal mobilising than the 50.0 g L

−1

solution. The difference between

the extraction efficiency of the 50.0 g L

−1

and of the 100.0 g L

−1

solution of D-

gluconic acid was greatest for Cd, presumably indicating a limited solubility of
the Cd-gluconate species formed. Two findings support this assumption. First, a
rise of the pH did not led to an increase of the maximal Cd release and second,
the moderately concentrated D-gluconic acid solution was most effective at Cd
mobilisation under strongly basic conditions (Figure 4). Highest amounts of Cu
were liberated by the 5.0 g L

−1

solution of the sugar acid. Since SOM extraction

plays an important role in the solubilisation of Cu it can be assumed that this result
reflects differences in the pH values of the extracting solutions. The solution with
the lowest sugar acid concentration has the lowest buffer capacity and the highest
pH value after adding of a base.

Increasing the amount of added NaOH the effects of sugar acid concentration

on the metal release remain more or less the same (Figure 4). Highest portions of
Pb, Zn, Cr, and Cu were mobilised by the 50.0 g L

−1

solution of D-gluconic acid.

In the case of Ni and Cd, the 20.0 g L

−1

solution was slightly more effective. The

pH increase favored the removal of Cr, Cu, and Pb especially. Between one fourth
and one fifth of the total amounts of these metals were additionally removed under
optimum conditions (Figures 3 and 4). The maximum removal degrees were 84%
(Cu), 80% (Pb), 70% (Zn), and 60% (Cr) (Figure 4).

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

283

Figure 5. Pb extraction by D-gluconic acid under variation of the solid:liquid-ratio and of the pH.

Usually the extraction yield is affected by the solid: liquid-ratio also. Applying a

20.0 g L

−1

solution of D-gluconic acid, the solid:liquid ratio was stepwise extended

from 1 g:5 mL to 1 g:40 mL. The experiments covered the whole relevant pH range.

By way of example the findings for Pb are presented in Figure 5. Independent

from the solid:liquid ratio the extraction of Pb was very low at pH values below
12. Nevertheless a certain correlation between the extraction volume and the metal
amounts extracted is already discernible at pH 10.0. The difference between the
Pb amounts extracted by the 5 mL and by the 40 mL solution of D-gluconic acid
is highest at pH 12.9. The overall highest amounts were leached by the 40 mL
solution at pH 13.4. With respect to the stoichiometric ratio between the applied
chelating agent and the amount of recovered metal, the use of a solid:liquid ratio
of 1:10 or 1:20 is advantageous.

To determine metal portions, which are additionally removable by repeated ex-

traction, a four-step batch procedure was carried out (Figure 6). For this, a 20.0 g
L

−1

solution of D-gluconic acid, containing 5 g L

−1

of NaOH (initial pH ap-

proximately 12.9) was used. Each extraction step lasted 24 hr. Simultaneously the
same procedure was conducted with the sodium hydroxide solution omitting the
chelating component.

Except Ni, the metal liberation was significantly improved by repeated extrac-

tion with D-gluconic acid. The release of Cr and Cu was enhanced especially. The
extraction degree of Cr mounts up to 54.6% and surpasses the result of the single
extraction by a factor of about 2.3. Interesting is the high constancy of the Cr
portions (about 9% of the total content) solubilised during steps 2–4. A similar
situation is found for Cu and, on a very low level, for Ni. In contrast to this the
removable portion of Cd seems to be exhausted after the second soil treatment.
Compared with the above described single step extraction with a solution volume
of 40 mL, the fourfold extraction with 10 mL of liquid is always more efficient.

284

K. FISCHER AND H.-P. BIPP

Figure 6. 4-fold extraction of the contaminated soil by 20.0 g L

−1

solutions of D-gluconic acid.

Initial pH

≈ 12.9. Solid:liquid-ratio 1:10. Treatment time per cycle 24 hr.

TABLE VII

Four-step soil extraction: total removal of metals

Extracting solution

Cd

Cr

Cu

Ni

Pb

Zn

(Removal degrees (%))

20.0 g L

−1

D-gluconic acid

63.1

54.6

79.0

37.7

73.5

57.2

and 5.0 g L

−1

NaOH

5.0 g L

−1

NaOH

25.6

7.3

72.9

16.2

7.7

14.1

Solid:liquid-ratio 1:10.

The juxtaposition of the metal portions mobilised during the four-step soil ex-

traction by the alkaline sugar acid solution and by a 5.0 g L

−1

solution of NaOH

(Table VII) confirms the conclusions formerly drawn on the basis of the single step
extraction tests. The specific ratios between the metal amounts extracted by the
chelating agent and by the pure NaOH solution were as follows: Cu 1.1, Ni 2.4, Cd
2.5, Zn 4.1, Cr 7.5, and Pb 9.5. This sequence completely agrees with the order of
the corresponding metal leaching factors calculated for the single-step extraction
at a NaOH concentration of 10.0 g L

−1

(Table VI). Despite of the high portion

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

285

of Pb initially bound onto soil organic matter, more than 90% of its total amount
remained in the solid phase after fourfold NaOH extraction at pH 13.

5. Summary and Conclusions

The ability of the polyhydroxy carboxylic acids D-gluconic acid and D-glucaric
acid to leach heavy metals from polluted soil is strongly influenced by the pH
value. The leaching capability is very low at near-neutral and slightly basic pH. It
drastically increases between pH 12.0 and 13.0.

At least three processes contribute to this effect: a) the deprotonation of al-

coholic groups of the sugar acids resulting in the formation of very stable metal
coordination compounds, b) the formation of soluble metal hydroxide complexes,
and c) the extraction of soil humic and fulvic acids by NaOH, leading to a cosol-
ubilisation of organically bound metal ions. The metal leaching factors calculated,
the computation of metal activities at equilibrium with their hydroxide species and
correlations between organically bound portions and amounts of metals extracted
by pure NaOH solutions give rise to the assumption that Pb and Cr are primarily
extracted by organic chelation whereas Cu is mainly mobilised by the alkaline
extraction of soil organic matter.

Apart from pH the sugar acid concentration affects the metal removal. In the

pH range from 12.5 to 13.5, a 50.0 g L

−1

solution of D-gluconic acid was most

effective at metal mobilising. At pH values above 13.0, highest extraction degrees
of Ni and Cd were achieved applying a 20.0 g L

−1

solution of the chelating agent.

Under strongly alkaline conditions a solid:liquid ratio of 1:10 or 1:20 was found
to provide a workable compromise between an optimum metal recovery and an
economic application of D-gluconic acid.

Except Ni, multi-step extraction improved the metal liberation considerably.

After 4 extraction steps, the totally solubilised amount of Cr was more than twice
as high as that of the first step alone.

Under optimised conditions the following metal removal degrees were achieved

with strongly alkaline D-gluconic acid solutions: Ni: 43%, Cr: 60%, Cd: 63%, Zn:
70%, Pb: 80%, and Cu: 84%.

A possible utilisation of these chelating agents for soil remediation requires pH

values of 12.5 or higher, thus ruling out in situ techniques. Furthermore it must be
pointed out that adverse effects on soil occur at this high pH such as destruction of
humic substances, reduction of the stability of carbonate minerals, loss of cations
through exchange with sodium ions, damage of soil (micro)organisms, etc. These
drawbacks seriously restrict the practical applicability of the tested sugar acids in
soil washing processes. On the other hand, if soil washing under alkaline conditions
is considered principally, the adding of sugar acids can strongly elevate the extrac-
tion efficiency for most of the heavy metals except Cu. Alternatively a lowering
of the pH value to 12.9 or 12.8 seems to be possible without loss of extraction

286

K. FISCHER AND H.-P. BIPP

performance, compared with the effectiveness of higher concentrated solutions of
sodium hydroxide.

Acknowledgements

The authors are indebted to Dr. R. Bierl, University of Trier, who modelled the
species distribution. This work was integrated in the Bavarian Research Associ-
ation for Waste Research and Reuse of Residues (BayFORREST) and was funded
by the Bavarian Stateministeries ‘für Landesentwicklung und Umweltfragen’ and
‘für Unterricht, Kultur, Wissenschaft und Kunst’. The research was carried out
at the GSF – National Research Centre for Environment and Health, Institute of
Ecological Chemistry (director: Prof. Dr. A. Kettrup), Munich-Neuherberg.

References

Allison, J. D., Brown, D. S. and Novo-Gradac, K. J.: 1990, MINTEQA2/PRODEFA2, a Geochemical

Assessment Model for Environmental Systems, Version 3.00 User’s Manual. USEPA, Athens,
GA.

Beckett, P. H. T.: 1989, ‘The Use of Extractants in Studies on Trace Metals in Soils, Sewage Sludges,

and Sludge-treated Soils’, Adv. Soil Sci. 9, 143–176.

Bipp, H.-P.: 1996, ‘Einsatz von Zuckersäuren und Hydroxycarbonsäuren zur Dekontamination

schwermetallbelasteter Böden sowie ihre Freisetzung aus kohlenhydrathaltigen Reststoffen’,
Ph.D. Thesis, Technical University of Munich, 208 pp.

Bipp, H.-P., Fischer, K., Bieniek, D. and Kettrup, A.: 1997, ‘Application of Ion Exclusion Chro-

matography (IEC) for the Determination of Sugar and Carboxylic Acids in Hydrolysates from
Carbohydrate Containing Residues’, Fresenius J. Anal. Chem. 357, 321–325.

Blomqvist, K. and Still, E. R.: 1985, ‘Solution Studies of Systems with Polynuclear Complex

Formation: Copper(II)- and Cadmium(II)-D-Gluconate-Systems’, Anal. Chem. 57, 749–752.

Brewster, M. D., Peters, R. W., Miller, G. A., Ti, W., Patton, T. L. and Martino, L. E.: 1995, ‘Phys-

ical/Chemical Treatment of Metals-Contaminated Soils’, in R. S. Ramachandra (ed.), Proc. 2nd.
Int. Symp. Waste Process. Recycling Metal Ind. II, Canadian Inst. Mining, pp. 539–565.

Coccioli, F. and Vicedomini, M.: 1978, ‘On the Dissociation of Gluconate Ions and their Complex

Formation with Lead(II) in Alkaline Solution’, J. Inorg. Nucl. Chem. 40, 2106–2110.

DIN (Deutsches Institut für Normung): 1983, DIN 38414-S7, Beuth-Verlag, Berlin.
Duff, R. B. and Webley, D. M.: 1959, ‘α-Ketogluconic Acid as a Natural Chelator Produced by Soil

Bacteria’, Chem. and Ind. (London), 1376–1377.

Duff, R. B., Webley, D. M. and Scott, R. O.: 1963, ‘Solubilization of Minerals and Related Materials

by 2-Ketogluconic Acid-Producing Bacteria’, Soil Sci. 95, 105–114.

Eckhardt, F. E. W.: 1985, ‘Solubilization, Transport and Deposition of Mineral Cations by

Microorganisms-Efficient Rockweathering Agents’, in J. J. Drever (ed.), The Chemistry of
Weathering, NATO ASI Series C, Vol. 149, pp. 161–173.

Fischer, K.: 2002, ‘Removal of heavy metals from soil components and soil by natural chelating

agenets. Part I: Displacement from clay minerals and peat by L-Cysteine and L-Penicllamine’,
Water, Air and Soil Pollut. 137, 267–286.

Fischer, K., Bipp, H.-P., Riemschneider, P., Leidmann, P., Bieniek, D. and Kettrup, A.: 1998, ‘Utiliz-

ation of Biomass Residues for the Remediation of Metal-Polluted Soils’, Environ. Sci. Technol.
32, 2154–2161.

REMOVAL OF HEAVY METALS BY NATURAL CHELATING AGENTS. PART II

287

Gosselé, F., Swings, J. and Deley, F.: 1980, ‘A Rapid, Simple and Simultaneous Detection of 2-Keto-,

5-Keto and 2,5-Diketogluconic Acids by Thin-layer Chromatography in Culture Media of Acetic
Acid Bacteria’, Zbl. Bakt. Hyg., Abt. Orig. C1, 178–181.

Hustede, H., Haberstroh, H.-J. and Schinzig, E.: 1989, ‘Gluconic Acid’, in B. Elvers, S. Hawkins,

J. F. Ravenscraft, M. Rounsaville and G. Schulz (eds), Ullmann’s Encyclopedia of Industrial
Chemistry, 5th ed., Vol. 12A, VCH-Verlag, Weinheim, Germany, pp. 449–456.

Jauregui, M. A. and Reisenauer, H. M.: 1982, ‘Dissolution of Oxides of Manganese and Iron by Root

Exudates’, Soil Sci. Soc. Am. J. 46, 314–317.

Jenne, E. A.: 1968, ‘Controls on Mn, Fe, Co, Ni, Cu, and Zn Concentrations in Soils and Water: The

Significant Role of Hydrous Mn and Fe Oxides’, in Trace Inorganics in Water, Adv. in Chem.
Ser. 73 (pp. 337–387), Am. Chem. Soc. Publ.

Joyce, L. G. and Pickering, W. F.: 1965, ‘An Investigation of the Nickel Gluconate System’, Aust. J.

Chem. 18, 783–794.

Kononowa, M. M.: 1966, Soil Organic Matter, 2nd ed., Pergamon Press, Oxford.
Leroy, P.: 1976, ‘Le Gluconate de Soude et les Problèmes de L’Environment’, Surfaces 105, 55–56.
Martell, A. E. and Smith, R. M.: 1974, Critical Stability Constants, Vol. 1, Plenum Press, New York.
Martell, A. E. and Smith, R. M.: 1977, Critical Stability Constants, Vol. 3, Plenum Press, New York.
Moghimi, A., Tate, M. E. and Oades, J. M.: 1978a, ‘Characterization of Rhizosphere Products

Especially 2-Ketogluconic Acid’, Soil Biol. Biochem. 10, 283–287.

Moghimi, A., Lewis, D. G. and Oades, J. M.: 1978b, ‘Release of Phosphate from Calcium Phosphates

by Rhizosphere Products’, Soil Biol. Biochem. 10, 289–292.

Parkhurst, D. L.: 1995, User’s Guide to PHREEQC, a Computer Program for Speciation, Reaction-

path, Advective-transport, and Inverse Geochemical Calculations. U.S. Geological Survey Water
– Resources Investigations Rep. 95–4227.

Pecsok, R. and Juvet, R. S.: 1955, ‘The Gluconate Complexes. I. Copper Gluconate Complexes in

Strongly Basic Media’, J. Am. Chem. Soc. 77, 202–206.

Pecsok, R. and Juvet, R. S.: 1956, ‘Gluconate Complexes. II. Lead Gluconate Complexes in Strongly

Basic Media’, J. Am. Chem. Soc. 78, 3967.

Peters, R. W.: 1999, ‘Chelant Extraction of Heavy Metals from Contaminated Soils’, J. Haz. Mat.

66, 151–210.

Peters, R. W., Miller, G. and Brewster, M. D.: 1994, ‘Desorption of Arsenic from Contaminated Soils

Using Chelant Extraction: Batch Feasibility Studies’, Conf. Proc. Emerging Technol. Haz. Waste
Managem. VI, Am. Chem. Soc. Atlanta, September 19–21, pp. 429–432.

Peters, R. W., Miller, G., Brewster, M. D. and Redwine, J. C.: 1995, ‘Columnar Studies for Remedi-

ation of Arsenic Contaminated Soils by Chelant Extraction’, Conf. Proc. Emerging Technol. Haz.
Waste Managem. VII, Am. Chem. Soc. Atlanta, September 17–20, pp. 1133–1135.

Rowatt, J. W. and Katznelson, H.: 1961, ‘A Study of the Bacteria on the Root Surface and in the

Rhizosphere of Crop Plants’, J. Applied. Bacteriol. 24, 164.

Sawyer, T. D.: 1964, ‘Metal-Gluconate Complexes’, Chem. Rev. 64, 633–643.
Schlichting, E., Blume, H.-P. and Stahr, K.: 1995, Bodenkundliches Praktikum, 2nd ed., Blackwell,

Berlin.

Schwertmann, U., Kodama, H. and Fischer, W. R.: 1986, ‘Mutual Interactions Between Organics

and Iron Oxides’, in P. M. Huang and M. Schnitzer (eds), Interactions of Soil Minerals with
Natural Organics and Microbes. Soil Sci. Soc. Am. Spec. Publ. No. 17, Madison, WI., U.S.A.,
pp. 223–250.

Sillen, L. G. and Martell, A. E.: 1971, ‘Stability Constants of Metal-Ion Complexes’, Supplement

No. 1, Spec. Publ. No. 25, Chem. Soc. London, Alden Press, Oxford.

Sokatch, J. R.: 1969, Bacterial Physiology and Metabolism, Academic Press, London.
Sponholz, W. R.: 1990, ‘Analysis of Galacturonic Acid and other Oxosugar Acids in Plant Material’,

GIT Fachz. Lab. 2/90, 107–116.

288

K. FISCHER AND H.-P. BIPP

Stevenson, F. J. and Ardakani, M. S.: 1973, ‘Organic Matter Reactions Involving Micronutrients in

Soils’, in J. J. Mortvedt, W. L. Lindsay and P. M. Giordano, (eds), Micronutrients in Agriculture,
Soil Sci. Soc. Am., Madison, WI., U.S.A., pp. 79–114.

Velasco, J. G., Ortega, J. and Sancho, J.: 1976, ‘On the Composition and Stability of Some

D-(+)Saccharic Acid Complexes’, J. Inorg. Nucl. Chem. 38, 889–895.

Vicedomini, M.: 1983, ‘Potentiometric Investigation of the Copper(II)-Gluconate System in Acid

Solution’, J. Coord. Chem. 12, 307–312.

Webley, D. M. and Duff, R. B.: 1965, ‘The Incidence, in Soils and Other Habitats of Microorganisms

Producing 2-Ketogluconic Acid’, Plant Soil 22, 307–313.

Zeien, H. and Brümmer, G. W.: 1989, ‘Chemische Extraktion zur Bestimmung von Schwermetall-

bindungsformen in Böden’, Mitteilgn. Dtsch. Bodenk. Gesell. 59, 505–510.