Short

Communication

Hang, Qin, Shen

957

Yiping Hang
Yongchao Qin
Jing Shen

College of Chemistry and
Molecular Science, Wuhan
University, Wuhan City, Hubei
Province 430072, P.R. China

Separation and microcolumn preconcentration
of traces of rare earth elements on nanoscale TiO

2

and their determination in geological samples by
ICP-AES

A simple, rapid microcolumn preconcentration and separation technique is described
for the determination of the trace rare earth elements Sm, Tm, Ho, and Nd in geologi-
cal samples by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-
AES). The technique was based on adsorption of analytes on Nanometer-Size Tita-
nium Dioxide (NSTD) packed in a microcolumn and separation from the matrix at
pH = 7.0. Quantitative adsorption of more than 94% of analytes was obtained in this
work. The dynamic adsorption capacities of NSTD were found to be 14.59, 12.86,
14.38, and 15.58 mg g

– 1

for Tm, Sm, Ho, and Nd, respectively. The analytes were

recovered quantitatively from NSTD with 2.0 mol L

– 1

HCl. The analytical procedure

was optimized in terms of sample acidity, elution, flow rate of sampling, and sample
volume. At a flow rate of 1.0 mL min

– 1

, the detection limits (3r) of the technique for

Tm, Sm, Ho, and Nd with an enrichment factor of 50 were 0.06, 0.18, 0.08, and
0.1 ng mL

– 1

, and the RSD were 1.8%, 4.7%, 2.0%, and 1.6%, respectively

.

Key Words: Adsorption; Nanometer-size titanium dioxide; Microcolumn; Preconcentration; Rare
earth elements

;

Received: April 26, 2002; revised: August 29, 2002; accepted: December 16, 2002

DOI 10.1002/jssc.200301270

1 Introduction

Rare earth elements (REEs) have been widely used in
agriculture, in industry, and in functional materials [1].
They may harm public health through their accumulation
in the food chain. REEs exist in geological materials at
trace and ultra-trace levels, and their direct determination
is beyond the scope of ICP-AES. Therefore, the determi-
nation of trace or ultra-trace rare earth elements in geolo-
gical samples requires preconcentration/separation and
monitoring methods.

Numerous research papers have been published world-
wide on preconcentration and separation techniques [2 –
6]. Compared with other techniques, the microcolumn
preconcentration and separation technique possesses
some unique advantages, such as simple operation; low
cost; high enrichment factor; high sensitivity; strong
resistance to jamming of matrix; and easy combination
with various modern analytical techniques [atomic
absorption spectrometry (AAS) and inductive coupled
plasma atomic emission spectrometry/mass spectrome-
try (ICP-AES/MS)]. Many substances have been used as
sorbents, such as chelex100 [7], alumina [8], zirconium

oxide [9], active carbon [10], cellulose [11], chelating
resins [12], microorganism [13], etc., in the technique.
NSTD is a new solid material that has gained importance
in recent years due to its excellent properties [14]. The
atoms on its surface are unsaturated, possess a high
chemical activity, and can bind with other atoms, and
can, therefore, adsorb metal ions with a high adsorption
capacity [15, 16].

Liang et al. [18] have investigated and optimized the ana-
lytical conditions, such as acidity, elution, sampling flow
rate, and sample volume for the REEs Y, Yb, Eu, La, Dy
on NSTD packed in a microcolumn coupled with ICP-
AES. In this work, the conditions were investigated and
optimized to preconcentrate and separate the REEs Tm,
Sm, Ho, and Nd by this technique [17], and then success-
fully applied to the determination of trace REEs in geologi-
cal samples.

2 Experimental

2.1 Apparatus

A 2-kW power rated, 27 l 3-MHz ICP generator, a PGS-2
plane grating spectrometer with 1 300 grooves mm

– 1

(Zeiss, Germany), and a conventional silica plasma torch
were used. The operating conditions and the wavelength
of the emission lines are summarized in Table 1.

Correspondence: Yongchao Qin, College of Chemistry and Mo-
lecular Science, Wuhan University, Wuhan City, Hubei Province
430072, P.R. China. Phone: [phone].
Fax: [phone]. E-mail: [e-mail].

J. Sep. Sci. 2003, 26, 957–960

i

2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1615-9306/2003/0907–0957$17.50+.50/0

958

Hang, Qin, Shen

J. Sep. Sci. 2003, 26, 957–960

The pH values were controlled with a Mettler Toledo 320-
S pH meter [Mettler Toledo Instruments (Shanghai) Co.
Ltd.] supplied with a combined electrode.

A HL-2 peristaltic pump (Shanghai Qingpu Instrument
Factory, China) was used in the preconcentration/separa-
tion process.

2.2 Column preparation

A polytetrafluroethylene (PTFE) microcolumn (20 mm
6

3.0 mm ID) was prepared in-house as follows: The col-

umn was filled with 50 mg of NSTD and plugged at both
ends with a small amount of glass-wool. Before use,
1.0 mol L

– 1

HCl was passed through the column followed

by twice-distilled water to clean and condition it to a neutral
state.

2.3 Standard solutions and reagents

Stock solutions of Sm, Tm, Ho, and Nd with concentra-
tions of 1 mg mL

– 1

were prepared by dissolving the corre-

sponding oxides (specpure) in dilute HCl, followed by dilu-
tion to a certain volume with water. Standard solutions
were prepared by serially diluting the stock solution. All
other chemicals used in the experiment were of specpure
grade. Twice-distilled water was used throughout the
experiment. NSTD (d a 30 nm), provided by the laboratory
of Inorganic Chemistry, Institute and Molecular Science of
Chemistry, Wuhan, University.

2.4 Experimental procedure

The pH value of the sample solutions was adjusted to 7.0
with 1.0 mol L

– 1

HCl and NH

3

N

H

2

O. Sample solution was

passed through the prepared column at a flow rate of
1.0 mL min

– 1

by a peristaltic pump, followed by twice-dis-

tilled water, after which the analytes were eluted with
2.0 mL of 2.0 mol L

– 1

HCl solution. The analytes in the

eluate were determined by ICP-AES.

3 Results and discussion

3.1 Effect of acidity

pH was one of critical factors affecting quantitative
adsorption of trace elements, because the pH of solution
decided the distribution of active sites on the NSTD sur-
face, and the ability of surface 1OH groups to bind
cations [18]. The following scheme illustrates the binding
process:

Thus the pH value plays an important role in the adsorp-
tion of different ions on the surface. The pH of sample
solution was adjusted within a range of 1 to 9 with
1.0 mol L

– 1

HCl and NH

3

N

H

2

O to evaluate the effect of

pH on the adsorption ratio (R %). The experimental result
show that the analytes were poorly adsorbed at pH a 6,
while quantitative recovery of A 92% was found between
pH 6.5 and 9 for the studied samples. pH 7.0 was selected
as a compromise.

3.2 Effect of flow rate of sample solution

As the quantity of elements on adsorbent has much to do
with the flow rate of the sample solution, the rate effect
was examined by passing 10 mL of sample solution
through the microcolumn with a peristaltic pump while
keeping other conditions constant. The flow rates were
adjusted in the range from 0.5 to 3.0 mL min

– 1

. We found

that recoveries of A 94% of the studied samples were
obtained at flow rates a 1.0 mL min

– 1

. Thus, a flow rate of

1.0 mL min

– 1

was employed in this work.

3.3 Effect of eluent volume, concentration, and

flow rate

The adsorption of cations at pH a 3 was shown to be negli-
gible. Therefore, HCl was used as eluent in this experi-
ment. In order to evaluate the effect of eluent concentra-
tion, various concentrations of HCl were tried for desorp-
tion of analytes from the microcolumn. The results are
given in Table 2. As seen, 2.0 mol L

– 1

HCl was sufficient

for complete desorption.

The effect of eluent volume on the recovery of analytes
was studied by maintaining the HCl concentration at
2.0 mol L

– 1

. It was found that quantitative recoveries of

more than 94%, (enrichment factor 50) could be attained
with 2.0 mL of 2.0 mol L

– 1

HCl. Therefore, a sample

Table 1. ICP-AES operating conditions and wavelengths of
emission lines examined.

Parameters

Value

Incident power [kW]

1.1

Carrier gas (Ar) flow rate [L min

– 1

]

0.8

Auxiliary gas (Ar) flow rate [L min

– 1

]

0.4

Coolant gas (Ar) flow rate [L min

– 1

]

16

Observation height [mm]

10

Entrance slit width [lm]

20

Integration time [s]

45

Wavelength [nm]

Sm 341.8

Tm 353.6

Ho 341.6

Nd 378.4

J. Sep. Sci. 2003, 26, 957–960

Separation and preconcentration of rare earth elements

959

volume of 100 mL and an eluent volume of 2.0 mL were
used in this work.

Various flow rates of eluent were studied and the optimal
one was found to be 1.0 mL min

– 1

which was used

throughout the determination.

3.4 Effect of amount of NSTD

The effect of the quantity of NSTD between 20 and
100 mg on the adsorption ratio was investigated by main-
taining other conditions constant. The result is given in
Table 3. The optimum quantity was found to be 50 mg.

3.5 Adsorption capacity

The adsorption capacity was an important factor for the
preconcentration/separation technique. Under dynamic
conditions, 100-mL volumes of pH 7 sample solutions of a
series of concentrations were passed one by one through
a microcolumn containing 50 mg NSTD. The concentra-
tions of the solutions before and after preconcentration/
separation were determined by ICP-AES. The capacity is
presented in Table 4, which indicates that NSTD had a
high capacity when used in this technique.

3.6 Effect of coexisting ions

The effect of common coexisting ions on the adsorption of
REEs in NSTD was studied. Solutions of 1.0 lg mL

– 1

of

Sm, Tm, Ho, and Nd containing the added interfering ions

were treated according to Section 2.4. The tolerance of
the coexisting ions, defined as the amount that lowered
the recoveries of the elements studied to less than 94%,
are listed as follows: Na

+

, K

+

, 20 000 mg L

– 1

; Ti

4+

,

50 000 mg L

– 1

; Al

3+

, Zn

2+

, 2000 mg L

– 1

; Ca

2+

, Mg

2+

,

10 000 mg L

– 1

; Fe

3+

, 500 mg L

– 1

; NO

3 –

, SO

4

2 –

, Cl

–

, 5 000

mg L

– 1

.

The result showed the presence of major cations and
anions had no obvious influence on the determination of
REEs under the selected conditions.

3.7 Detection limits and RSD

According to the IUPAC definition, the detection limits
(3 r) of the technique with an enrichment factor 50 were
calculated and listed as follows: 0.06 ng mL

– 1

Tm,

0.18 ng mL

– 1

Sm, 0.08 ng mL

– 1

Ho, and 0.1 ng mL

– 1

Nd.

The RSD (n = 6) of the technique for Tm, Sm, Ho, and Nd
were 1.8%, 4.7%, 2.0%, and 1.6%, respectively.

4 Analysis of real samples

1 000 g of standard reference sample (GSR-1) was
weighed, and transferred to a PTFE beaker. Then certain
volumes of HF, concentrated HNO

3

, and HCl were added.

The solution was heated almost to dryness. The residue
was dissolved in water and determined. The results, given
in Table 5, agreed with the certified values.

After being acidified to about pH 1, a water sample col-
lected from Jianghan oilfield, Hubei, China, was filtered
through a 0.45 lm membrane filter (Tianjin Jinteng Instru-
ment Factory, Tianjin, China). The pH was adjusted to 7.0
with 0.1 mol L

– 1

HCl and NH

3

N

H

2

O. The results, listed in

Table 6, indicate that the technique is applicable for the
determination of trace and ultra-trace REEs.

Table 2. Effect of eluent concentration on sorption of REEs
onto NSTD.

Eluent
[HCl, mol L

–1

)

Sorption [%]

Sm

Tm

Ho

Nd

0.5

86.2

89.4

90.3

87.5

1.0

93.5

96.6

95.8

94.2

2.0

95.0

97.3

96.4

95.8

3.0

95.2

97.7

96.8

95.5

Note: Values are average of three determinations.

Table 3. Percent sorption of Sm, Tm, Ho, and Nd as a func-
tion of amount of NSTD.

NSTD in mg

Sorption [%]

Sm

Tm

Ho

Nd

10

56.6

58.3

62.0

49.8

30

89.4

92.8

87.9

88.5

50

92.7

95.2

94.3

93.8

70

93.5

96.8

96.3

95.7

100

94.0

97.0

96.4

96.2

Note: Values are average of three determinations.

Table 4. Adsorption capacity (mg g

– 1

) of NSTD under

dynamic conditions.

Extractant

Tm

3+

Sm

3+

Ho

3+

Nd

3+

Capacity

14.59

12.86

14.38

15.58

Table 5. Analysis of standard reference material (GSR-1,
l

g g

– 1

).

Element

Found

Reference value

Tm

1.02 l 0.2

1.06 l 0.1

Sm

9.78 l 0.4

9.70 l 0.2

Ho

2.08 l 0.2

2.05 l 0.1

Nd

45 l 5

47 l 4

960

Hang, Qin, Shen

J. Sep. Sci. 2003, 26, 957–960

5 Conclusions

NSTD is an efficient sorbent for the microcolumn precon-
centration/separation technique of trace REEs. The ana-
lytes of interest were enriched 50 times and obtained in
high recoveries in the technique. The capacity of NSTD is
sufficiently high to preconcentrate and separate Sm, Tm,
Ho, and Nd from real samples. The technique is thus sim-
ple, sensitive, and reproducible.

References

[1] E. Diatloff, F.W. Smith, C.J. Asher, J. Plant. Nutr. 1995, 18,

1991.

[2] A.K. Singh, Talanta 1996, 43, 1843.

[3] M.I. Rucandis, Fresenius’ J. Anal. Chem. 1997, 357, 661.

[4] P. Roychowdhury, N.K. Roy, D.K. Das, A.K. Das, Talanta

1989, 36, 1183.

[5] H.G. Linge, Twenty-Second Annual Hydrometallurgical

Meeting of the Metallurgical Society of C.I.M., Canada,
1992.

[6] R.R. Barefoot, J.C. Vanloon, Talanta 1999, 49, 1.

[7] T. Yabutani, F. Mouri, A. Itoh et al., Anal. Sci. 2001, 17, 399.

[8] M. Hiraide, M.H. Sorouraddin, H. Kawaguchi, Anal. Sci.

1994, 10, 125.

[9] Z. Fang, Flow Injection Separation and Preconcentration.

Wiley-VCH, Weinheim 1993.

[10] G. Chakrapani, P.L. Mahanta, D.S.R. Murty, B. Gomathy,

Talanta 2001, 53, 1139 – 1147.

[11] V.L. Dressler, D. Pozebon, A.J. Curtius, Anal. Chim. Acta

2001, 438, 235 – 244.

[12] Q. Suai, Y.C. Qin, B. Hu, H.C. Xiong, Z.C. Jiang, Anal. Sci.

2000, 16, 957.

[13] Y. Madrid, C. Camara, Trends Anal. Chem. 1997, 16, 36.

[14] A. Henglein, Chem. Rev. 1989, 89, 1861.

[15] E. Vassileva, N. Furuta, Fresenius’ J. Anal. Chem. 2001,

370, 52 – 59.

[16] P. Liang, Y.C. Qin, B. Hu, T.Y. Peng, Z.C. Jinag, Anal. Chim.

Acta 2001, 440, 207 – 213.

[17] P. Liang, B. Hu, Z.C. Jinag, Y.C. Qin, T.Y. Peng, J. Anal. At.

Spectrom. 2001, 16, 863 – 866.

[18] C. Merterra, J. Chem. Soc. Faraday Trans. 1988, 84, 1617.

Table 6. Application to water sample (lg g

– 1

).

Element

Found

Certified

a)

Sm

1.92 l 0.14

2.0

Tm

2.44 l 0.20

2.5

Ho

0.56 l 0.10

0.6

Nd

0.19 l 0.008

0.22

a)

Other method of determination.