ABSTRACT We examined the mechanism of alkali soil reclamation by gypsum from the change in hydraulic conductivity and the solute transport in soil. Solute transport mechanism was investigated by using a soil column equipped with tensiometers and four-electrode sensors (FES). Moreover, the pH and EC were measured, and quantitative analysis of the cations (Na+, Ca2+, Mg2+ and K+) and the anion Cl- in leachate was carried out. As a result, the initial Na leaching was confirmed by FES measurement, EC of leachate and quantitative analysis of leachate. Correlation coefficients between all the parameters (pH, EC, cations (Na+, Ca2+, Mg2+ and K+) and Cl-) were confirmed, to different levels of significance (P value < 0.01). Lastly, the water content balance and the evapotranspiration by Penman-Monteith method in the paddy field predicted the soil reclamation time, taking into consideration the meteorological data and the column experimental results. Consequently, it could be predicted that all salts in case of 0.5 wt% application and almost salts (97%) in case of 1.0 wt% application are leached by the end of October.
Journal of Water and Environment Technology, Vol 7, No 2, 2009 Salts Transport in Alkali Soil Reclamation by Gypsum and Prediction of Na Leaching in Field in China Yuji SAKAI*, Shohei SHITARA**and Masayoshi SADAKATA* * Department of Environmental Chemical Engineering, Kogakuin University, Tokyo 192-0015, Japan **Department of Chemical System Engineering, The University of Tokyo, Tokyo 113-8656, Japan ABSTRACT We examined the mechanism of alkali soil reclamation by gypsum from the change in hydraulic conductivity and the solute transport in soil Solute transport mechanism was investigated by using a soil column equipped with tensiometers and four-electrode sensors (FES) Moreover, the pH and EC were measured, and quantitative analysis of the cations (Na+, Ca2+, Mg2+ and K+) and the anion Cl- in leachate was carried out As a result, the initial Na leaching was confirmed by FES measurement, EC of leachate and quantitative analysis of leachate Correlation coefficients between all the parameters (pH, EC, cations (Na+, Ca2+, Mg2+ and K+) and Cl-) were confirmed, to different levels of significance (P value < 0.01) Lastly, the water content balance and the evapotranspiration by Penman-Monteith method in the paddy field predicted the soil reclamation time, taking into consideration the meteorological data and the column experimental results Consequently, it could be predicted that all salts in case of 0.5 wt% application and almost salts (97%) in case of 1.0 wt% application are leached by the end of October Keywords: alkali soil, China, gypsum, Na leaching, prediction, solute transport INTRODUCTION About 23% of the total cultivated area (1.5×107 km2) (Massoud, 1981) of the world is saline soil, and saline and sodic soils cover about 10% of all potentially arable lands (Szabolcs, 1989) Moreover, about 25% of all irrigated land has been affected by salts (Suarez and Rhoades, 1991) In countries with arid and semiarid regions, the decrease in agricultural production due to excessive salts is a very serious problem This is because insoluble salts that have accumulated in soil obstruct the growth of vegetation due to the osmotic pressure they exert Moreover, in saline soil, accumulation of monovalent cations (mainly Na+) causes deterioration of the soil physics, due to the dispersion and swelling of clay, and this makes it difficult for plants to grow These processes adversely affect the water penetration, the infiltration speed, and the moisture transport property of the soil, all of which are involved in soil moisture maintenance (Shainberg and Levy, 1992) Therefore, leaching with large amounts of water is used to improve this type of soil However, a natric layer of low permeability is generated by Na adsorption to the surface of the soil particles in the salts accumulation process As a result, there is a possibility of soil erosion, because this not only causes difficulties in the removal of salts by leaching due to the decrease in water permeability, but it also increases the outflow water on the soil surface Therefore, the use of soil amendments such as Ca compounds is the generally used method that is effective Moreover, it is known that the water penetration of clay strongly depends on the concentration and the composition of the soil solution, and, in general, decreases when the soil solution has a low Address correspondence to Yuji Sakai, Department of Environmental Chemical Engineering, Kogakuin University, Email: sakai@cc.kogakuin.ac.jp Received February 6th, 2009, Accepted February 25th, 2009 - 121 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 concentration and a high ESP (Quirk and Schofield, 1955) In recent years, the air, water and soil in China have been polluted by economic and industrial development Carbon dioxide (CO2) emissions from fossil fuel combustion are estimated to lie between 2,050 and 2,445 million tons of CO2 per year CO2 emissions from industrial processes are estimated to lie between 81 and 104 million tons per year (World Bank, 2004) Moreover, sulfur dioxide (SO2) emissions increase as coal consumption increases, and these emissions not only cause acid rain, but are also harmful to humans and vegetation However, desulfurization technologies such as flue gas desulfurization (FGD) of coal-fired power plants have not been put into widespread use because of economical and technological problems On the other hand, in northeastern China, the amount of salt-affected soil is increasing due to increases in the evaporation rate and excessive cultivation In order to solve these air pollution and soil desertification problems simultaneously, Sakai et al (2002, 2004, 2008) have reclaimed alkali soil with gypsum from FGD processes in China Concretely, desulfurization gypsums from wet, simple-wet, and semi-dry FGD process and bio-briquette ash have been reclaimed in Shenyang of China since 1996 Moreover, these soil reclamation methods have performed in Tianjin, Inner Mongolia (Wang et al., 2008), Ningxia Province of China (Wang et al., 2009) They showed that gypsums from FGD processes and bio-briquette ash are effective at reclaiming alkali soil Moreover, we quantified the metal (B, Cr, Mn, Ni, Cu, As, Cd, Pb) contents in soil, FGDG (FGD gypsum), and corn grains for the safety confirmation Consequently, there was no effect of gypsum application on the metal content in the corn grains Almost all of the metal contents were lower than the standard values set by FAO/WHO for human intake (Sakai et al., 2004) A simulation model based on a numerical solution to equations for moisture and solute transport in soil has been developed by the USSL (US Salinity Laboratory, 1954) These models can be applied to a wide variety of fields such as soil and moisture control, but forecasts by these models depend strongly on the input parameters The soil permeability properties that characterize water retention and infiltration, the transport rates of chemicals, and the transport and chemical parameters that influence the distribution of the solid and liquid phase are very important for these models (Sakai, 2002) It is very difficult to apply these models to actual test fields, as it is necessary to measure all these values to perform calculations with these models Therefore, in order to evaluate this environmental reclamation technology, it is necessary to develop effective methods for predicting salts transport in fields in China The objective of this study was to examine the change in hydraulic conductivity that occurs as a result of adding the gypsum, and to examine the transport of solutes in alkali soil reclamation The inner hydraulic conductivity and the transport of salts such as cations (Na+, Ca2+, Mg2+ and K+) and Cl- in a column were investigated by using a tensiometer and a four-electrode sensor, and the amounts of these salts in soil and leachate were measured by ICP-MS Moreover, by taking the water balance for evapotranspiration in test fields in China into consideration, salt leaching in fields could be predicted using the results of the soil column experiment MATERIALS AND METHODS Soil sampling and soil chemical analysis Soil samples were collected at Kangping (42゜ 70' N, 123゜ 50' E), which is located about - 122 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 130 km north of Shenyang in northeastern China Alkali soil reclamation with by-products from the flue gas desulfurization process has been performed there since 1996 (Sakai et al., 2002, 2004; Sakai, 2008) Soil samples from the surface to a depth of 20 cm were air-dried and passed through a mm sieve The pH, EC, and solution cations were measured by using 1:5 water extracts A pH meter (HM-14P, TOA Electronics Ltd.) was used for the pH measurement and an electrical conductivity meter (B-173, Horiba Ltd.) was used for EC measurement The cations and anions in the soil solution (1:5 water extraction) were measured by Inductively Coupled Plasma Mass Spectroscopy (ICP-MS) (HP4500, Yokogawa Co., Ltd.) and ion chromatography (LC-6A, Shimadzu Co., Ltd.), respectively The cation exchange capacity (CEC) was determined using 1N NaOAc at pH 8.2 In this research, pure gypsum was added to the alkali soil The amount of CaCO3 was measured by using 1.0 g of soil sample placed in a closed flask One milliliter of 2N H2SO4 was added to react with the CaCO3, and the pressure of the evolved CO2 in the headspace was measured immediately using gas chromatography (GC-8A, Shimadzu Co., Ltd.) The exchangeable sodium percentage (ESP) was calculated from the sodium adsorption ratio (SAR) of the soil solution (US Salinity Laboratory Staff, 1954) The soil chemical properties are given in Table Table 1- Chemical properties of Chinese soil used in the experiments Property Value pH [H2O] 10.5 EC (Electrical conductivity) [dS/m] 0.79 CEC (Cation exchange capacity) [(+) cmol/kg] 8.4 CaCO3 [g/kg] 22.5 ESP (Exchangeable sodium percentage) [%] 25 Column experiment measurements Solute transport measurement Soil under mm in size was mixed with gypsum at application rates of 0.5 wt% and 1.0 wt% Alkali soil and gypsum were mixed uniformly and were packed in an acrylic ring with an outer diameter of 60 mm, inner diameter of 50 mm, and height of 20 mm at a bulk density of 1.53 g/cm3 The acrylic rings were filled up to a height of 20 cm Degassed water was poured onto the top of the column at a rate of 1.0×10-3 to 2.0× 10-3 cm3/s using a peristaltic pump (SJ-1211L, Atto Corp.) The lower parts of the column, the porous cup, and the experimental water had been degassed in desiccators for about one hour Tensiometers and four-electrode sensors (Inoue at al., 2000) were inserted into the column at depths of 5, 9, 13 and 17 cm, and at and 17 cm (where soil surface z = cm), respectively The tensiometers were equipped with a transducer (HTVN-001, HI-NET Co., Ltd.) in order to measure the pressure (cmH2O) of the soil in the experimental column Transducers (TD (z=5 cm), TD (z=9 cm), TD (z=13 cm), and TD (z=17 cm)) connected to four tensiometers and two four-electrode sensors (FES; FES (z=5 cm) and FES (z=17 cm)) were connected to a data logger (CR10X, Campbell Scientific, Inc.) with a PC (Fig.1) The measurements could be controlled in the program by the application software (PC208W, Campbell Scientific, Inc.) Moreover, in order to examine the soil solution that seeped from the lower part of the experimental column, we collected the infiltration solution at constant time intervals by using a fraction collector (CHF122SB, Advantec Toyo Kaisha, Ltd.) The water volume was - 123 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 measured at each time interval The ion concentrations (Na+, Ca2+, Mg2+, K+ and Cl-) in the effluent solution were quantified by using ICP-MS The soil pH and soil EC of each solution were also measured These experiments were performed in triplicate at 25 °C Electrical conductivity measurements Direct measurement of the salinity of soil can be performed by using buried porous electrical conductivity sensors, four-electrode probe systems, electromagnetic induction sensors, or time domain reflectometry systems (Rhoades and Oster, 1986; Rhoades, 1992) In particular, a four-electrode probe is suitable for observing the salt concentration in a column with a water flow because the response is good The four-electrode sensor developed for this study was described in detail by Inoue and Shiozawa (Inoue and Shiozawa, 1994) In our research, a four-electrode sensor was used as an indicator of the total salt concentration in the soil column, to observe the runoff rate of salts The four-electrode sensor consisted of four stainless steel rods inserted in parallel into an acrylic column with an inner diameter of 50 mm, outer diameter of 60 mm, and a height of 20 mm (Fig 2) This structure exposes 18 mm of the point of the current supply electrode and 12 mm of the point of the voltage measurement electrode while insulating the other parts, and measures the electric conductivity of the soil in the center of the column The inner and outer diameters of the stainless steel rods were mm and 16 mm, respectively, and the ratio (I/V2) of the current of an outer electrode (I) to the potential difference (V2) of an inner electrode was measured This ratio is inversely proportional to the specific resistance of the medium, and proportional to the electric conductivity Because hardly any current flows to a voltage measurement electrode inside the column, the influence of local contact resistance in the current electrode surroundings, etc can be avoided However, it is necessary to have a high frequency exchange current in the current supply electrode to avoid polarization of the electrode Current (I) that flows in the current supply electrode is obtained from the relation I= V1/Rf by putting a resistance (Rf) of known value in the circuit, and measuring the potential difference V at both ends Therefore, the ratio (V1/V2) of the voltages V1 and microtube pump data logger (CR10X) z=0cm four-electrode sensor porous cup + transducer column ring water tank 5cm V1 Rf porous cup relay (z=5,17cm) (z=5,9,13,17cm) 1kHz 20cm V2 transducer voltage measurement electrode 50.00 mm data logger fraction collector electric current supply electrode 60.00 mm PC Fig 1- Schematic illustration of the experimental column setup Fig - Four-electrode sensor and connection with data logger - 124 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 V2 (V1: voltage applied to reference resistance, V2: voltage applied to measurement electrode for voltage) is proportional to the electric conductivity of the medium This measurement was made by a data logger CR10X and a relay controller RESULTS AND DISCUSSION Change in hydraulic conductivity resulting from the addition of gypsum to the soil In soil with salt accumulation, such as alkali soil, excessive Na carbonate or bicarbonate cause an increase in the pH of the calcareous soil This degrades the physical properties such as the breakdown of aggregated soil particles, and affects the growth of plants due to a decrease in the permeability (Suarez et al., 1984; Gupta and Abrol, 1990) Gypsum amendment to sodic soils can increase the permeability by means of both EC and cation exchange effects (Loveday, 1976) In alkali soil reclamation, it is very important for the Na soil particles to form a soil aggregate, and this can be achieved by adding Ca soil amendments such as gypsum This can result in improved soil permeability In our previous research, corn production was confirmed in plots in China with 0.5 wt% and 1.0 wt% added FGDG (Sakai et al., 2002) The hydraulic conductivity was dramatically increased with a decrease in the dry density (1.2, 1.4, and 1.6 g/cm3) and an increase in the gypsum application rate (0, 0.5, and 1.0 wt%) (Sakai, 2003) Therefore, it has been understood that gypsum addition and dry density have a large influence on the water penetration In the column transport experiment, the change in hydraulic conductivity was investigated in columns with both 0.5 wt% and 1.0 wt% applications, by using four transducers (TD 1, TD 2, TD and TD 4) Each hydraulic conductivity between porous cups was determined as K1-2, K2-3, K3-4 and the total hydraulic conductivity was determined as K1-4 The changes in K1-2, K2-3, and K3-4 were high during the initial stage, from to about 1.5 PV (pore volume, PV=172.3 cm3), and decreased after that initial stage All hydraulic conductivities at the 1.0 wt% application rate were higher than those at the 0.5 wt% application rate From Fig 3, it could be confirmed that the hydraulic conductivity (K1-4) between TD and TD increased rapidly and then decreased, from about 0.25 PV for the 0.5 wt% application and from about 0.76 PV for the 1.0 wt% application It took constant values from about 1.3 PV in both cases Moreover, the hydraulic conductivity at the 1.0 wt% application rate was higher than that at the 0.5 wt% one 4.0 0.5wt% 3.5 1.0wt% K1-4 [×10-4 cm/s] 3.0 2.5 2.0 1.5 1.0 0.5 0.0 PV [-] Fig - Change in hydraulic conductivity (K1-4) in the column (0.5 wt%, 1.0 wt%) - 125 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 Change in salts transport in soil column In alkali soil improvement, it is very important to understand the cation transport mechanism, in order to remove sodium from the surfaces of soil particles amended with soil amendments such as gypsum Therefore, the movement of salts obtained by pouring water over alkali soil with added gypsum was evaluated Fig shows the results from the four-electrode sensors at depths of cm and 17 cm at application rates of 0.5 wt% and 1.0 wt% In this research, because the sensors could be used to indicate salt movement in the column, calibration between the solution concentration and the volume of water content (Inoue et al., 2000) could not be performed The ECv value in the vertical line was calculated by gradually introducing CaCl2 solution (0.00230, 0.0105, 0.0172, 0.0552, 0.100 mol/dm3) after the end of the flow experiment From Fig 4, the value of ECv at the upper sensor that was cm below the soil surface decreased slightly and the lower sensor at a depth of 17 cm indicated early leaching of salts in the case of the 0.5 wt% application The results of the lower sensor showed that most of the salts in the soil could be leached in PV and after that the ECv only decreased slightly These results confirm that most of the soluble salts in the upper parts were leached past the lower sensor at about PV (about 350 cm3 of effluent) in the case of the 0.5 wt% application In the case of the 1.0 wt% application, the sensor value of the upper part (5 cm depth) was higher than that for the 0.5 wt% application Moreover, the behavior of both lower sensors looks like the EC behavior in Fig It could be confirmed that the difference of behavior between lower sensor and EC of leachate is due to time difference (Figs 4, 5) The changes in the concentrations of all the different ions can be discussed At the beginning of the leaching process for both the 0.5 wt% and the 1.0 wt% applications, the Na concentration in the leachate increased and then decreased drastically (Fig 6) This indicates Na+ leaching due to a Na+-Ca2+ exchange reaction resulting from the gypsum application Moreover, a difference between the 0.5 wt% and 1.0 wt% application rates could be confirmed The necessary water volumes for Na+ leaching from the 0.5 wt% and 1.0 wt% applications were about 1.2 PV (200 cm3) and about 2.3 PV (400 cm3), respectively (Fig 6) The Ca2+ behavior in both cases was similar (Fig 7) Naturally, the total amount of Ca2+ was larger for the 1.0 wt% application than the 0.5 wt% one In both cases, the Ca2+ concentration increased gradually from the beginning of leaching, decreased after a peak, and then increased and decreased again This behavior indicates that the initial leaching is like Na+ leaching, and the surplus Ca2+ is drained slowly because the Ca2+ ion is divalent Finally, almost depletion could be confirmed at about 3.0 PV (520 cm3) and 8.1 PV (1400 cm3) of total leachate for 0.5 wt% and 1.0 wt% applications, respectively The complete leaching time of Ca was longer than that of Na The Mg2+ behavior was similar in the 0.5 wt% and 1.0 wt% cases (Fig 8) The Mg concentration increased gradually from the start of leaching and decreased gradually after a peak Almost all the Mg2+ had leached from the 0.5 wt% and 1.0 wt% applications by about 2.3 PV (400 cm3) and about 8.1 PV (1400 cm3), respectively In both cases, K+ behavior was similar to that for Mg2+ (Fig 9) Sodic soil reclamation requires the removal of most exchangeable Na+ and its replacement by Ca2+ or Mg2+ and K+ in the soil of the root zone (Aydemir and Najjar, 2005) The Cl- ion is a non-adsorptive solute in the presence of sulfate ions, which have higher adsorption regardless of the soil type Therefore, the Cl- behavior was similar to the water movement in the column The water volume for Cl- leaching was about 0.58 PV (100 - 126 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 cm3) in the 0.5 wt% application and about 0.75 PV (130 cm3) in the 1.0 wt% application (Fig 10) The pH of leachate increased gradually at first, and then took a constant value from about 2.3 PV (400 cm3) (Fig 11) This is indicative of the behavior of the ions in the leachate The pH value in the 1.0 wt% application was lower than that in the 0.5 wt% application because of the lower pH value of gypsum Finally, the change in leachate volume over time was determined It was found that the leachate volume for the 1.0 wt% case was larger than that for the 0.5 wt% one (Fig 12) This is in agreement with the hydraulic conductivity result (Fig 3) The initial increase and decrease in the FES shows leaching of mainly Na+ (Figs 4, 6) The constant FES value observed after the initial decrease (Fig 4) was confirmed to be due to Ca2+, Mg2+ and K+ (Figs 7, 8, 9) This shows that the Na+-Ca2+ exchange reaction on the soil colloid surface occurred by water infiltration This indicates the stronger adsorption power of two valence ions over single valence ions (Bresler et al., 1982) 20 0.5wt% (z=5cm) 0.5wt% (z=17cm) 1.0wt% (z=5cm) 1.0wt% (z=17cm) 18 16 0.5wt% 1.0wt% EC [dS/m] ECv [dS/m] 14 10 12 10 6 2 0 PV [-] Fig - Change in ECv in upper (z=5 cm) and lower (z=17 cm) parts of the column (0.5 wt%, 1.0 wt%) PV [-] Fig - Change in EC of leachate (0.5 wt%, 1.0 wt%) 200 16 180 0.5wt% 160 1.0wt% 0.5wt% 14 1.0wt% 12 Ca conc [mmol/dm3] 140 Na conc [mmol/dm3] 120 100 80 60 40 10 20 0 PV [-] Fig - Change in Na+ concentration of leachate (0.5 wt%, 1.0 wt%) PV [-] Fig.7- Change in Ca2+ concentration of leachate (0.5 wt%, 1.0 wt%) - 127 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 25 0.3 0.5wt% 15 K conc [mmol/dm3] Mg conc [mmol/dm3] 0.5wt% 0.25 1.0wt% 20 10 1.0wt% 0.2 0.15 0.1 0.05 0 PV [-] Fig - Change in Mg2+ concentration of leachate (0.5 wt%, 1.0 wt%) PV [-] Fig.9- Change in K+ concentration of leachate (0.5 wt%, 1.0 wt%) 30 0.5wt% 1.0wt% 25 Cl conc [mmol/dm3] Cl conc [mmol/dm3] 30 20 15 10 25 20 15 10 0 PV [-] 0 PV [-] Fig 10 - Change in Cl- concentration of leachate (0.5 wt%, 1.0 wt%) 8.5 350 0.5wt% 300 1.0wt% 250 pH [-] time [h] 7.5 200 150 100 0.5wt% 50 1.0wt% PV [-] Fig 11 - Change in pH of leachate (0.5 wt%, 1.0 wt%) PV [-] Fig 12 - Relationship between PV and leaching time (0.5 wt%, 1.0 wt%) - 128 - Journal of Water and Environment Technology, Vol 7, No 2, 2009 1 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 C/C0 [-] C/C0 [-] The differences in the leaching rates of all the ions were obtained and they were leached in the order Cl->Na+>Mg2+>Ca2+>K+ in case of 0.5 wt% and 1.0 wt% application (Figs 13, 14) Moreover, all ion concentrations in soil before and after the infiltration experiment were measured The recoveries were very high (Na+: 98%, Mg2+: 94%, Ca2+: 83%, K+: 83%, Cl-: 96%) and showed the contribution of the cation exchange reaction 0.5 0.4 0.5 0.4 0.3 0.3 0.2 0.2 Na 0.1 Ca Mg K Cl Na 0.1 Ca Mg K Cl 0 PV [-] PV [-] 10 Fig 13 - Breakthrough curves of all ions Fig 14 - Breakthrough curves of all ions (0.5 wt%) (1.0 wt%) The correlations between pH, EC, cation (Na+, Ca2+, Mg2+, and K+) and Clconcentrations in leachate were investigated for both application rates Regression analysis was performed to investigate correlations using Excel Statistics 2006 (Social Survey Research Information Co., Ltd.) and the correlation coefficient and level of significance were calculated Correlation coefficients were found between all parameters, at different levels of significance (P value< 0.01) (Table 2) The pH of leachate had a negative correlation with all other data (EC, Na+, Ca2+, Mg2+, K+ and Cl-) In particular, the correlation coefficient between Na+ and EC was high (0.939**) The pH, EC, and ESP of soil reclaimed after the column experiments for 0.5 wt% and 1.0 wt% application rates were 9.1, 0.11 dS/m, and 2.5% and 9.1, 0.13 dS/m, and 2.9%, respectively This indicates that the experimental soil was reclaimed by the gypsum Table - Correlation coefficients between pH, EC, cation (Na+, Ca2+, Mg2+, and K+) and Cl- ion concentrations of leachate pH EC Na+ K+ Ca2+ Mg2+ Cl- pH EC Na+ K+ Ca2+ Mg2+ Cl1.000 -0.557** 1.000 -0.425** 0.939** 1.000 -0.730** 0.661** 0.659** 1.000 -0.648** 0.637** 0.267** 0.747** 1.000 -0.621** 0.691** 0.809** 0.975** 0.735** 1.000 -0.631** 0.621** 0.581** 0.441** 0.253** 0.558** 1.000 **: P