6. LABORATORY METHODS FOR ANALYSING GYPSIFEROUS SOILS

6.1 Introduction
6.2 Gypsum Content of Soils
6.3 The Particle-Size Distribution of Gypsiferous Soils
6.4 Cation Exchange Capacity and Exchangeable Cations

6.1 Introduction

The use of fertilizers has increased substantially as a result of the recent development of intensive agriculture in arid and semi-arid regions. During the energy crisis in the mid-1970s, fertilizer prices almost doubled, increasing the need to use them efficiently and economically. More field research is thus needed to evaluate application of commercially available fertilizer to produce the most economic yield. More laboratory research is also needed to translate the values found in the chemical tests into the fertilizer requirements of a specific crop on a specific soil.

The presence of gypsum and calcium carbonates interferes with some laboratory measurements including cation exchange capacity and particle-size distribution. Soil analysis for land evaluation and mapping for land utilization, for reclamation and for determining fertilizer status is becoming indispensable for economic development. Generally, most developing countries and also some developed countries are using routine methods. Such standardized methods and criteria cannot be used for all kinds of soils. Soils should be considered as systems, each system differing from another.

The laboratory methods used for the analyses of gypsiferous soils are basically the same as those used for non-gypsiferous soils. The total elements, the available nutrients and the organic matter content of soils are determined according to common procedures of soil chemical analysis. The following is a list of the various methods developed for gypsiferous soils.

6.2 Gypsum Content of Soils

6.2.1 Gypsum determination
6.2.2 The BaSO₄ method (Richards 1954)
6.2.3 The amended BaSO₄ method
6.2.4 Gypsum determination (Sayegh et al. 1978)
6.2.5 Gypsum determination (Lagerwerff et al. 1965)
6.2.6 Semiquantative gypsum determination by X-ray diffraction
6.2.7 Quantitative determination of gypsum content by X-ray diffraction (Friedel 1978)
6.2.8 Gypsum determination by extraction with boiling water (Kovalenko 1972)
6.2.9 Gypsum determination from loss of crystal-water content (Nelson et al. 1978)

Gypsum is present in soils in the form of calcium sulphate dihydrate (CaSO₄. 2H₂O). A transitional form of calcium sulphate hemihydrate (CaSO₄. 0.5H₂O) is sometimes found at the soil surface in extremely dry climates (Doner and Lynn 1977).

Crystals of gypsum dissolve partly in water to reach a saturation concentration of about 30 mEq dm^-3. The dissolution rate of gypsum in water increases as the particles become finer. The soil sample therefore should be finely ground before the analysis for gypsum content. Various methods have been developed for the determination of gypsum content in soils.

6.2.1 Gypsum determination

The following procedure is given by Richards (1954):

i. dissolve soil gypsum in a 1:5 soil - water ratio by shaking

ii. precipitate gypsum from the soil extract by adding acetone

iii. dissolve gypsum precipitate by adding water

iv. measure the electrical conductivity of the solution

v. determine the gypsum content by comparison with standard curve that relates gypsum content to electrical conductivity

6.2.2 The BaSO₄ method (Richards 1954)

Richards (1954) also gives the following reliable but slow method:

i. extract the soil sample with enough water to dissolve all the gypsum (Adams 1971)
ii. precipitate sulphate from solution by addition of BaCl₂ solution
iii. filter, dry and weigh the BaSO₄ precipitate
iv. make correction for the non-gypsum sulphate.

6.2.3 The amended BaSO₄ method

Coutinet (1965) gives the following BaSO₄ method:

i. boil 5 g of soil in 100 cm³ 5 percent ammonium carbonate solution for 15 minutes

ii. filter and acidify the extract using HCl, heat to boiling, then add sufficient 10 percent hot barium chloride solution to precipitate sulphate in the extract

iii. filter the precipitate, wash to eliminate Cl, and heat at 900°C

6.2.4 Gypsum determination (Sayegh et al. 1978)

1. Place 1 g of air-dried, sieved soil (270 mesh sieve) in a flask and add 500 cm³ of distilled water

2. Stopper the flask and shake for 20 minutes

3. Filter the extract through Whatman No. 41 filter paper

4. Precipitate the gypsum in an aliquot of the filtered extract using acetone

5. Redissolve the gypsum precipitate using distilled water

6. Measure the electrical conductivity and determine the concentration from a standard curve.

6.2.5 Gypsum determination (Lagerwerff et al. 1965)

Prepare a water saturated paste; calculate percent water/air-dry soil (PW_SE). Determine Ca and SO₄ concentrations in saturation extract (Ca_SE and SO_{4 SE}, mmol/l).

Prepare a dilute water extract by adding enough distilled water to dissolve all gypsum then filter (gypsum concentration should be 5 mmol/l or less). Calculate percent water/air-dry soil (PW_DE); determine sulphate concentration (SO_{4 DE}, mmol/l).

Recalculate concentrations in the two extracts on soil-weight basis:

(Ca)_SS = (Ca)_SE. PW_SE/100 (mmol/kg air-dry soil)

(SO₄)_SS = (SO₄)_SE. PW_SE/100

(SO₄)_DS = (SO₄)_DE. PW_DE/100

The total gypsum content of the soil therefore is:

(SO₄)_DS in cases where (SO₄)_SS is less than, or equal to (Ca)_SS; or

(SO₄)_DS - (SO₄)_SS + (Ca)_SS in cases where (SO₄)_SS exceeds (Ca)_SS (because of the presence of SO₄ from other sources).

6.2.6 Semiquantative gypsum determination by X-ray diffraction

Gypsum in soil can be determined by X-ray diffraction with potassium chloride as an internal standard. Ratios of counts on diffraction peaks of maximum intensity for gypsum and KCl are used for the semiquantitative determination of gypsum in soil.

6.2.7 Quantitative determination of gypsum content by X-ray diffraction (Friedel 1978)

The (020) X-ray reflection of gypsum at 7.56 A is nearly specific and may be used for detection and quantitative measurement of the mineral in soil samples. By X-ray powder diffractometry, gypsum can be detected when as little as 0.5 percent by weight. This limit may be extended down to 0.005 percent by concentration in the density fraction 2.1-2.5 g cm^-3. For quantitative measurements references may be prepared from sample parts by dehydration of the original gypsum and adding known quantities of this mineral. Thus it is possible to compare samples of nearly identical grain size distribution and mineral contents. This is important in the quantitative measurement by powder diffractometry of a mineral with low hardness and good cleavage where mechanical stress during homogenisation lowers the intensities of X-ray reflections. Under comparable conditions for sample and reference, gypsum contents can be determined in the range of 0.5 to 20 percent by weight with an accuracy of ±3% (relative).

6.2.8 Gypsum determination by extraction with boiling water (Kovalenko 1972)

This method is as follows:

i. 0.2-0.5 g finely ground air-dried soil is put in boiled water at a ratio of 1:1000
ii. the mixture is slightly heated to 40°C and left standing (periodically mixed) for 10-12 hrs
iii. then the mixture is heated to 80°C, and calcium is determined titrimetrically
iv. the gypsum content is then calculated

6.2.9 Gypsum determination from loss of crystal-water content (Nelson et al. 1978)

This method is sufficiently accurate for taxonomic uses. It is based on loss of crystal water of gypsum upon heating to 105°C.

6.3 The Particle-Size Distribution of Gypsiferous Soils

6.3.1 Gypsum removal by ammonium oxalate (Coutinet 1965)
6.3.2 Gypsum removal by hydrochloric acid (Loveday 1974)
6.3.3 Pretreatment of soil with BaCl₂ to coat gypsum with BaSO₄ (Hesse 1974)
6.3.4 Pretreatment of soil with BaCl₂ solution followed by ethanol (Matar and Douleimy 1978)
6.3.5 Pretreatment with BaCl₂ solution (amended method, Vieillefon 1979)

The presence of gypsum limits dispersion of soil colloids. In the traditional procedures for particle-size analysis of gypsiferous soils, gypsum present in small amounts is leached by using distilled water (Baldwin et al. 1938, Barzanji 1973). Solutions of strong electrolytes can be used to increase the solubility of gypsum and shorten the leaching time (Abrukova and Isayev 1983, Agboola and Corey 1973).

Rivers et al. (1982) suggest a new method to reduce the time necessary to leach gypsum from soils by half. The technique involves heating the soil samples at 105°C overnight to convert gypsum into more-soluble bassanite. Heating the samples does not change the determined clay mineralogy or the cation exchange capacity of the soils, but gives slightly higher values for clay content.

The presence of gypsum in soils affects their physical properties greatly and affects the results of particle-size analysis. Fine-textured soils usually have a higher gypsum content than otherwise similar but coarser soils. Comparison of field and laboratory determinations of the texture of gypsiferous soils can be misleading because gypsum occurs as crystals in the various size fractions. Consequently field textures are normally coarser than laboratory determinations. Laboratory assessment of the texture is misleading if determined by standard methods.

During particle-size analysis of soils containing gypsum the soil suspension usually flocculates partially or totally after a period of time, because of the calcium ions released from the gypsum.

Vieillefon (1977, 1978, 1979) in comprehensive studies on the improvement of analytical methods for gypsiferous soils, deals especially with the effects of gypsum on particle-size distribution. He concludes:

1. Partial flocculation of the fine soil particles results in their underestimation.

2. Gypsum has a density of 2.317 Mg m^-3, so methods using Stokes' Law which assumes an average particle density of 2.65 Mg m^-3 overestimate the clay and silt fractions of soils if not corrected.

3. In the pipette method, during drying of each particle-size fraction, gypsum loses its water of crystallization and consequently, the fractions of soils containing most gypsum are underestimated unless correction is made.

The methods described below offer advances in the determination of the particle-size distribution of gypsiferous soils.

6.3.1 Gypsum removal by ammonium oxalate (Coutinet 1965)

i. boil 40 g of soil with H₂O₂ to destroy the organic matter

ii. if the gypsum content is less than 25 percent add 5 g of ammonium oxalate to the treated soil with sufficient water (300 cm³), and boil the mixture for one hour, and discard the solution

iii. repeat this washing technique until no more calcium ions are detected in the soil solution

iv. add 20 cm³ of sodium-hexametaphosphate solution (40 g dm^-3) and boil the mixture for 3 hours

v. when the gypsum content is more than 25 percent, use the same techniques as above, but remove the gypsum by using sodium chloride solution (131.6 g dm^-3)

6.3.2 Gypsum removal by hydrochloric acid (Loveday 1974)

This procedure is used for soils containing visible gypsum in which the electrical conductivity EC_1:5 is >3 mS/cm.

i. to 25 g of air-dry soil, add 25 cm³ of 2M HCl to destroy the CaCO₃ and H₂O₂, to destroy organic matter, then add 500 cm³ of water

ii. stir gently with an electric stirrer for one hour, allow to clear and decant. Shaking and decanting is repeated until no further crystals of gypsum appear on the sides of the beaker

iii. 5 cm³ of 1M NaOH and 10 cm³ of 10 percent sodium tripolyphosphate are added, and the soil is dispersed for 20 minutes using a motor dispersing unit at low speed

iv. decant into the cylinder and proceed to determine the particle-size fractions

6.3.3 Pretreatment of soil with BaCl₂ to coat gypsum with BaSO₄ (Hesse 1974)

i. shake 10 g of soil gently with 40 cm³ of solution containing 50 g dm^-3 barium chloride and 20 cm³ of 45 cm³ dm^-3 tri-ethanolamine for one hour

ii. decant and wash with water until no barium ions can be detected in the leachate by using potassium chromate

iii. add 15 cm³ of a solution containing 40 g dm^-3 sodium hexametaphosphate, 10 g dm^-3 anhydrous sodium carbonate and 10 percent sodium hydroxide to the treated soil and disperse for 30 minutes

iv. add distilled water to dilute the soil suspension and determine the particle-size distribution

6.3.4 Pretreatment of soil with BaCl₂ solution followed by ethanol (Matar and Douleimy 1978)

i. leach 10 g soil with 0.1M BaCl₂ solution

ii. leach the excess Ba ions with pure ethanol

iii. stir the suspension for 12 minutes after adding 10 cm³ of calgon (40 g din^-3) and distilled water

iv. determine the various fractions

6.3.5 Pretreatment with BaCl₂ solution (amended method, Vieillefon 1979)

i. shake 10 g soil with 10 cm³ BaCl₂ (50 g dm^-3) and 20 cm³ triethanolamine in a centrifuge tube

ii. shake for 1 hour and then centrifuge the suspension

iii. discard the supernatant liquid

iv. add 40 cm³ distilled water, centrifuge, and discard the supernatant liquid

v. add 15 cm³ of dispersing solution (4 percent sodium hexametaphosphate + 1 percent Na₂CO₃ at pH 8.2) and shake

vi. determine the various fractions

6.4 Cation Exchange Capacity and Exchangeable Cations

6.4.1 Sodium oxalate method (Sayegh et al. 1978)
6.4.2 Saturation of soil with Ba ions
6.4.3 Cation exchange capacity of calcareous and gypsiferous soils
6.4.4 Determination of exchangeable cations and CEC

The presence of gypsum interferes with the measurement of cation exchange capacity, which is a useful parameter for assessing the mineralogy and fertility of soils. Conventional methods underestimate cation exchange capacity in gypsiferous soils because of incomplete saturation by the index cation. Hydrolysis of adsorbed Na and fixation of NH₄ also contribute to this underestimation. Most procedures attempt to reduce the solubility of gypsum in CEC determination. As gypsum is not very soluble in alcohol, a displacement solution of 0.5M NaCl in 60 percent alcohol saturated with gypsum is used to determine adsorbed Ca and Mg in gypsiferous soils. Sodium oxalate has been shown to be effective in precipitating calcium, and has been used in combination with sodium acetate and sodium carbonate. Barium, as barium acetate and barium chloride, has been used with varying degrees of success for CEC determination in calcareous and gypsiferous soils. Polemio and Rhoades (1977) proposed a procedure involving extraction with 0.5M Mg(NO₃)₂ (pH 7.0) whereby both Na and Cl were determined in the extract from which CEC was calculated. They found that values obtained by this method were correlated with air-dry water content, which reflects CEC. Many of these procedures take time and detailed mathematics, which may rule them out for routine work.

6.4.1 Sodium oxalate method (Sayegh et al. 1978)

Sayegh et al. (1978) developed an effective procedure for accurate CEC determination in soils containing gypsum. They give the following method:

i. weigh 4 g soil and transfer to a 50 cm³ centrifuge tube

ii. pretreat the sample with a saturated solution of sodium oxalate and shake for two days

iii. after two days, centrifuge the suspension and decant the supernatant liquid

iv. repeat the saturation procedure three times, each time with a 33 cm³ solution

v. wash the samples with isopropyl alcohol to remove the excess sodium, until the conductivity of the supernatant liquid is less than 0.04 dS m^-1

vi. replace the adsorbed sodium by using three 33 cm³ increments of ammonium acetate (1M) filtering each washing into a 100 cm³ flask

vii. determine the sodium concentration

6.4.2 Saturation of soil with Ba ions

This method described by Garman and Hesse (1975) exchanges the exchangeable cations with Ba²⁺ at pH 8.1, followed by exchanging with magnesium from a standard (0.05M) MgSO₄ solution. The CEC is calculated from the difference of Mg remaining in the standard solution.

Polemio and Rhoades (1977) show that appreciable Ca is found in the replacing solution, thus competing with the Mg from the standard solution on the soil colloidal complex and hence giving lower CEC results. It seems, therefore, that BaCl₂ does not satisfactorily prevent the dissolution of gypsum.

6.4.3 Cation exchange capacity of calcareous and gypsiferous soils

Polemio and Rhoades (1977) suggest the following method:

i. if the soil is high in salts (EC_extract >2 dS m^-1, first wash the soil with one 33 cm³ increment of water. Excessive washing should be avoided to prevent loss of particles during decantation

ii. saturate about 4 g of soil by four successive 33 cm³ increments of a 60 percent ethanol solution of 0.4M NaOAc - 0.1M NaCl at pH 8.2

iii. equilibrate by shaking the mixture for 5 minutes and decant

iv. extract with three 33 cm³ increments of 1M, pH 7 solution of Mg(NO₃)₂
v. total sodium (Na_t), and chloride (Cl_t) are then determined in this solution. Chloride is determined so that the soluble sodium (Na_sol) from the excess saturating solution carried over from the saturation step to the extraction step may be deducted from the total sodium to obtain exchangeable sodium (Na_exch) as follows: CEC = (Na_t-Na_sol) = Na_t - [Cl_t(Na/Cl)_{saturating solution}]

where (Na/Cl)_{saturating solution} is the ratio of Na to Cl in the saturating solution

6.4.4 Determination of exchangeable cations and CEC

For these methods described by Tucker (1974) the following solutions are required:

Solution A	Solution of 100 cm3 of ethylene glycol and 900 cm3 of pure ethanol.
Solution B	54 g of ammonium chloride are dissolved in 320 cm3 water and 670 cm3 of ethanol are added. pH is brought to 8.5 by adding 25 percent ammonia solution.
Solution C	50 cm3 of solution B are diluted with 310 cm3 water and 640 cm3 of ethanol and the pH is adjusted to 8.5 by adding sufficient 25 percent ammonia solution.
Solution D	150 g potassium nitrate are dissolved in 800 cm3 water. 60 g Ca(NO3)2. 4H2O are dissolved in this solution and made up to 1 dm3 with water.

To determine exchangeable cations:

i. remove soluble salts by shaking 2.0 g of air-dry soil with 25 cm³ of solution A for half an hour, centrifuge and decant the solution

ii. extract the exchangeable cations by shaking the residue with 20 cm³ of solution B for half an hour and decant the solution into a 100 cm³ flask after centrifuging

iii. repeat three times with 20 cm³ of reagent B and once with 15 cm³ solution C. Extracts are made up to 100 cm³ by adding 4M HCl

iv. split the sample into two:

2. Determine cation exchange capacity as follows: Replace the NH₄⁺ by shaking the other half of the split sample with 25 cm³ of Solution D. Determine NH₄ by titration and calculate as mEq/100 g of soil sample. Determine chloride by chloride titration and calculate similarly. Multiply the chloride content by the ratio NH₄Cl and subtract from NH₄ to arrive at the CEC.

A limitation to this method is that many arid soils contain micaceous minerals that fix NH₄ and K. Thus, saturation and extraction with K and NH₄ index cations may result in anomalously low CEC values (Bower et al. 1952).