Geochemical characterization pathways “production site - water - sediment - soil - food

MINERALOGICAL INVESTIGATIONS OF SOILS IN SELECTED SETTLEMENTS OF THE NAM §ÞnH REGION, VIÖT NAM

JOERN KASBOHM¹, Lª THÞ LµI², MARIA-THERESIA SCHAFMEISTER¹
TRÇN TRäNG HUÖ², KARL-HEINZ HENNING¹

1 Institute of Geological Sciences, University of Greifswald / Germany

^{2 Institute of Geological Sciences, NCNST, Hà Nội}

Abstract: A traditional metal working trade is the economical basics for the Nam §Þnh province. The recent development of industrial production in this province is connected with serious environmental problems. Additionally to the geochemical characterization also soil samples of V©n Chµng and §ång Cãi settlements (Nam §Þnh province) have been investigated by X-ray diffraction and transmission electron microscopy.

These soil samples have following qualitative composition of its mineral matter: non clay minerals: quartz, feldspar, different Fe-hydroxides and hematite,traces of apatite, rutile, gibbsite; clay minerals: dioctahedral vermiculite (rarely also illite), dioctahedral vermiculite-smectite mixed layers, illite-smectite-mixed layers, kaolinite, Fe-rich chlorite, rarely beidellite, very rarely trioctahedral vermiculite, and saponite.
The estimated hydraulic conductivity shows values, which are typically for silty/ clayey sediments. Quartz and feldspar are the dominating minerals in all samples. In the fraction < 2mm expandable minerals, e.g. smectite or smectite-rich mixed layer minerals, have been detected only in traces. By this background the mineral matter offers only a low absorption capacity to limit additionally a further migration of waste water through the sediments.

A cutting of the direct pathway for polluted water from the production site in the surrounding soil is to consider for mutual environmental programs. The high sediment thickness in the Red River delta can not avoid that the recent handling of the waste water from production site will contaminate in near future also the main aquifers for the drinking water management in the Nam §Þnh region.
I. INTRODUCTION

Nam §Þnh is an administrative province about 100 km SE of Hµ Néi and lies in the river delta of the Red River. A traditional metal working trade is the economical basement of this region. Family manufactures and other small companies are the typical producers there. Mostly the people of this district are living in small settlements with less than 3.000 inhabitants of each. Additionally to the metal working production the inhabitants of such settlements cultivate rice and vegetables for their own consumption mostly. The region is flat and nearly free of any trees (for more details – see [8, 9]).

Table 1. Sample list for the investigated soils and sediments

Samples	Settlement	Description	Degree of pollution
LA2 LA3 LA4	V©n Chàng	Outside of the settlement; field (cultivation of rice & vegetable)	None exception: LA3a (0-25 cm): Cd 61 mg/kg
LA6		Iron-working production site and living house	medium: Zn, Pb, Cu, (partially Cd)
LA7		Iron-working production site and living house (only mechanical processes)	none
LA8, LA9 LA13		Soil in bank zone of Vân Chàng stream	low: Zn (Cu)
LA14		Garden of metal working production site/living house (with electrochemical processes)	high: Zn, Pb, Cu, (partially Cd)
LA21a		Sediment of a pond	high: Zn, Cd, Cr, Cu
LA16, LA17 LA18	§ång Cãi	Soil in bank zone of stream	low: Zn, Cu
LA19		Soil in bank zone of a pond	low: Zn, Cu
LA22a		Sediment of a pond	medium: Zn, Cd, Cr
LA23a		Sediment of a pond	high: Zn, Cd, Cr, Cu

The administrative authorities draw for the province Nam §Þnh an average life expectancy with less than 55 years accompanied with a high degree of cancerous, neurotic and pulmontic diseases.

Mineralogical investigations of contaminated soil are a part of analytical program to characterize the degree of pollution and its migration caused by productions sites in the settlements. Cations exchange capacity, which could be specifically for each mineral group, and grain size distribution have a large influence to sorption properties and hydraulic conductivity of soil.

The geochemical characterization of soils are summarized in [7].

II. METHODOLOGY

1. Sampling

In the two settlements, §ång Cãi and V©n Chµng (Nam §Þnh province), soil have been sampled by 1 m deep drills. The drilled material was divided in four samples for each drill representing areas from 0 – 25 cm, 26 – 50 cm, 51 – 75 cm, and 76 – 100 cm. Samples were taken from soils closed to and far from productions sites/living houses, from production sites only with mechanical processes and from production sites, which have included also chemical procedures for treatment of their products. Also soils were sampled in coastal zone of local streams and ponds. Sediments from few ponds were involved additionally (see table 1, [7]).

2. Analytical methods
X-Ray Diffraction (XRD)

For randomly oriented powder samples a Siemens D5000-equipment with Cu-beam and secondary curved graphite-monochromator (40 kV, 30 mA; variable slits: 20 mm, soller: 0.5/25, step & time: 0.02° 2 Θ for 3 s) was used. Oriented mounts were analysed in a Präzitronic Freiberg HZG 4-diffractometer controlled by Seifert-C3000 unit (30 kV, 30 mA; Co-beam, Fe-filter, fixed slits: 1.09 mm/ 6.0 mm, soller: 0.5/25, detector slit: 0.35 mm; step & time: 0.03 0.02° 2 Θ for 2 s) with air dried, ethylene-glycol solvated and 550°C heated (4 h) specimen.

In order to recognize any asymmetries of individual interferences the WinFit-software [6] was used for deconvolution procedures. The Fourier-transformation implemented in this software, was used for the determination of "coherent scatter domains" (CSD) [2, 3].

Transmission electron microscopy (TEM)

The suspension of the fraction < 2 mm was prepared after 10 min ultrasonic treatment on carbon-coated Cu-grids by air drying.

A Jeol-microscope with a JEM-1210 the individual particles were characterized by means of morphology, electron diffraction and element distribution (voltage: 120 kV, LaB6-cathode). That´s why also an OXFORD-LINK EDX-system was connected with the TEM-equipment.

Morphology of particles was described according to Henning & Stoerr (1986) [4]. The electron diffraction allows an evaluation of the stack order. Ring-like structures of the electron spots indicate for clay minerals a turbostratical order of the layers. Zoller (1993) [12] has demonstrated the possibility to distinguish between a 2M- or 1M-order for 2:1 sheet silicates, based upon the comparison of the intensities of (020)- and (110)-interferences in a convergent beam-system.

2M-polytype: | 110|/| 020| > 1

1M-polytype: | 110|/| 020| < 1

In order to evaluate the mineral formula of the individual particles the photons of each analysed particle were detected for 20 sec. The excited XRF-area of each particle is less than 100 nm in a diameter. The EDX-measurements were carried out standardless and by using of k-factors following the procedures of Cliff & Lorimer [1].

The procedure of Koster (1997) [5] for the conversion the EDX-analyse to a mineral formula was adopted. The measured data-set have been evaluated in relation to the total charge, e.g. 22 for dioctahedral sheet silicates.

A sophisticated system described by Srodon et al. (1992) [11] was used to distinguish by the mineral formula between montmorillonite and illite-smectite mixed layer structures (IS-ML) described a strict correlation of ´fixed´ cations with the expandability of IS-ML. They found such correlation also with the amount of Al in the tetrahedral layer. Furthermore they have pronounced their opinion that in illite-smectite mixed layer minerals (IS-ML) montmorillonite and illite are characterized by a typical total charge - 0.4 for montmorillonite and 0.89 for illite. This situation allows a verification of the estimated expandability.

(d(20) = grain size that is 20% finer by weight = effective grain size in mm; empirical estimation by graphical interpolation)

Figure 1. Overview – X-ray diffractograms of randomly oriented samples
(bulk sample and fractions > 63 mm, 2-63 mm, < 2 mm)

sample: LA002 = soil of a rice field, outside of the V©n Chàng settlement, no contamination

Figure 2. Overview – X-ray diffractograms of oriented samples (fraction < 2 mm)

sample: LA019 = soil in a bank zone of a pond in the §ång Cãi settlement, no contamination

Legend: ML80…90 – illite-smectite mixed layer minerals with 80…90 % of illitic layers; dioct. vermiculite – dioctahedral vermiculite; ML Sme/diVerm – dioctahedral vermiculite-smectite mixed layer phases; dioct. Fe-vermiculite – dioctahedral vermiculite with more than 0.3 Fe in the octahedral layer

III. RESULTS AND DISCUSSION

1. Grain size distribution

For selected samples the grain size separation documents mostly a silty character. The samples show values mainly lower than 20 % for the amount of the clay size fraction (see table 2). The d(20) parameter have been estimated graphically.

2. Mineral matter

Figure 1 shows the XRD patterns of randomly oriented samples containing quartz, feldspar and illitic minerals located outside of the V©n Chàng settlement in its future business area. In the fractions < 2 mm all samples are dominated by illite or I-S and chlorite with a variable kaolinite content. The powder XRD pattern of bulk samples in he profile from 0 – 25 cm down to 75 –100 cm offer no remarkable differences in their qualitative mineral composition. Also the samples from high contaminated localities show the same dominance of quartz and feldspar.

The dominating d₀₆₀ spacing at 1.504 Å confirms the mainly dioctahedral character of detected clay minerals. The identified chlorite is only a minor admixture in the mineral matter.

Especially soils samples, which were collected in bank areas of ponds and local streams, contain also minerals with a swelling behavior (Figure 2). Their swelling ability by ethylene-glycol saturation is limited to 16.3 – 16.6 Å. This behavior is typically for illite-smectite mixed layer minerals with more than 60 % of illitic layers. In contaminated soil samples such swelling minerals are not to identify.

The low intensities of the 14.3 Å and 4.72 Å interferences of chlorite in comparison to the 7.17 Å and 3.53 Å peaks indicate an Fe-rich composition.

The clay mineral matter of the high contaminated sample La014 was investigated also by transmission electron microscopy (TEM). The analysis of individual particles in the TEM allows also a calculation of mineral formulas for the detected phases (Table 3).

Two main groups are to distinguish in table 3. Mica phases with a sufficient K-content for the illitic components are identified as illite or illite-smectite mixed layer minerals. Mica structures with K-deficit in comparison to illite are considered as dioctahedral vermiculitic phases. Such vermiculitic particles are dominating in the full profile of the sample LA14. K-content looks like to be decreased with the depth of the profile. This decreasing K-amount is the only remarkable trend comparing the calculated mineral formulas. The analysed dioctahedral vermiculite-smectite mixed layer phases show a low value as average for Si in the tetrahedral layers (< 3.4). This value indicates that these particles have only a low number of smectitic layers in their structure.

Furthermore Fe-rich chlorite, kaolinite and traces of trioctahedral vermiculite as well as saponite have been detected. The upper horizon of the sample (0 – 25 cm) contains also beidellite.

Especially in horizons deeper than 50 cm also Zn and Ni have been detected by TEM-EDX. A TEM-mapping of chlorite demonstrates a preferred adsorption of Zn at the edges of such crystals. Otherwise Pb are Ni are distributed statistically on the surface of this mapped chlorite.

Table 4. Empirical Approximation of hydraulic conductivity using data
from the grain size distribution of selected soil samples

(years (cumm.) = cumulative time in years to pass the 1 m deep profile)

Finally soil samples from V©n Chàng and §ång Cãi settlements have following qualitative composition of its mineral matter in this order:

non clay minerals:

quartz, feldspar,

different Fe-hydroxides and hematite

traces of apatite, rutile, gibbsite

clay minerals:

dioctahedral vermiculite (rarely illite)

dioctahedral vermiculite-smectite

mixed layers,

illite-smectite-mixed layers,

Fe-rich chlorite,

kaolinite,

rarely beidellite trioctahedral vermi-

culite, and saponite

IV. CONCLUSIONS

The listed general mineral matter of the soil sample from the settlements is typically for Red River sediments. The short distance and the large height gradient from the sediment source in the mountains to the river mouth, the Red River delta, produce mainly illitic and chloritic minerals in the delta sediments. Such processes are discussed as arenization [10]. Leaching processes (spreading of layers and K-loss) alterate detrital illite particles to dioctahedral vermiculite structures.

Beidellite and poor ordered kaolinite a typical intermediate results of soil forming processes.

The main function of these mineralogical investigations is to support an estimation of the expecting migration rate through the sediments of V©n Chàng and §ång Cãi settlements. The Bialas-method allows an approximation to the paramater "hydraulic conductivity" using the data from the grain size distribution (K_f~ 0.0036 * d₂₀^2.3). Applying the data from table 2 the estimated values for the hydraulic conductivity describe mainly an interval from 10^-8 – 10^-10 m/s (Table 4). This interval is usually for a grain size distribution between silt and clay. Additionally table 4 contains also a column "years (cumm.)". These values shall visualize in which time waste water from production sites will have passed the sampled 1 m distance for each soil sample. In background of this approximation it looks like, that in the high contaminated soil of sample LA14 waste water will need only about three years to pass this 1 m deep profile.

Minerals with a high surface, e.g. smectites, offer well absorption properties to fix cationic pollutants from the water. Quartz, feldspar and illitic/vermiculitic phases represent mainly the mineral composition of the investigated samples. These minerals show generally only a low absorption capacity. Also in the clayey grain size fraction montmorillonite or other smectite-rich minerals were not detected in a remarkable amount. From this viewpoint also the mineral composition can not offer additional factors reducing in a significant scale the migration of waste water through these sediments.

A cutting of the direct pathway for polluted water from the production site in the surrounding soil is to consider for mutual environmental programs. The high sediment thickness can not avoid that the recent handling of the waste water from production site will contaminate in near future also the main aquifers for the drinking water management in the Nam §Þnh region.

ACKNOWLEDGEMENTS

The authors thank the Federal Ministry for Education and Research (BMBF/ Germany) and the Ministry of Science and Technology (MOST/ViÖt Nam) for the financial and logistical support (VNM 00/004). Special thanks we would like to send to all authorities and colleagues from the district management of Nam §Þnh province.

The laboratory analyses were carried out at the Greifswald University, Germany by Monika Schaefer and Manfred Zander. They are greatly appreciated.

REFERENCES

1. Cliff G., Lorimer G. W., 1975. The quantitative analysis of thin specimens. Journal for Microscopy 103: 203-207

2. Drits V. A., Eberl D. D., Środoń J., 1998. XRD measurement of mean thickness, thickness ditribution and strain for illite and illite/smectite crystallites by the Bertaut-Warren-Averbach technique. Clays and Clay Minerals 46: 461-475

3. Eberl D. D., Drits V., Środoń J., 1998. Deducing growth mechanisms for minerals from the shapes of crystal size distributions. American Journal of Science 298: 499 - 533

4. Henning K.-H., Störr M., 1986. Electron micrographs (TEM, SEM) of clays and clay minerals. Akademie-Verlag Berlin [Schriftenreihe für geologische Wissenschaften, Bd. 25]: 352 pp..

5. Köster H. M., 1977. Die Berechnung kristallchemischer Strukturformeln von 2:1-Schichtsilikaten unter Berücksichtigung der gemessenen Zwischenschichtladungen und Kationenaustauschkapazitäten, sowie die Darstellung der Ladungsverteilung in der Struktur mittels Dreieckskoordinaten. In: Clay Miner. 12: 45-54

6. Krumm S., 1994. WINFIT 1.0 - A Computer Program for X-ray Diffraction Line Profile Analysis: Acta Universitatis Carolinae Geologica, 38, XIIIth Conference on Clay Mineralogy and Petrology, Praha, 253-261

7. Lª ThÞ Lµi, Kasbohm J., TrÇn Träng HuÖ, §µo Huy Quý, Schafmeister M.-T., 2003. Geochemical characterization pathways "production site - water - sediment - soil - food – residents" as basis for an in-situ treatment system in the craft-settlements of Nam §Þnh province, Journal of GEOLOGY, Series B, No.21/2003

8. §µo Huy Quý, 2003. Environmental issues in socio-economic development strategy of Nam §Þnh, Journal of GEOLOGY, Series B, No.21/2003

9. Steingrube W., Bïi Quang Thµnh, 2003. Basis for a Sustainable Development in Rural Areas in Northern ViÖt Nam, Journal of GEOLOGY, Series B, No.21/2003

10. Środoń J., 1999. Use of clay minerals in reconstructing geological processes: recent advances and some perspectives. Clay Minerals 34: 27-37

11. Środoń J., Elsass F., Mchardy W. J., Morgan D. J., 1992. Chemistry of illite-smectite inferred from TEM measurements of fundamental particles.- Clay Minerals 27(2): 137-158

12. Zöller M. H., 1993. Charakterisierung von Illitkristallen durch konvergente Elektronenbeugung. In: Berichte der Deutschen Ton- und Tonmineralgruppe e.V.: 211-220 [Beiträge zur Jahrestagung der DTTG 1992]