Quality of liming materials used in aquaculture in Thailand

Aquaculture International 12: 161 168, 2004. Quality of liming materials used in aquaculture in Thailand TAWORN THUNJAI 1, CLAUDE E. BOYD 1, * and MALI BOONYARATPALIN 2 1 Department of Fisheries and Allied Aquacultures, Auburn University, Auburn AL 36849, USA; 2 Department of Fisheries, Kasetsart University Campus, Bangkok 10900, Thailand; *Author for correspondence (e-mail: ceboyd@acesag.auburn.edu; phone: +1-334-844-4078; fax: +1-334-844-5933) Received 24 April 2003; accepted in revised form 21 October 2003 Abstract. Samples of 45 brands of liming materials were obtained in Thailand and analyzed for chemical and physical properties. Eight of 10 products sold as ground calcium carbonate (calcitic agricultural limestone) were properly identified by vendors and of high quality, that is, neutralizing value and fineness rating above 85%. Seven of 15 products sold as ground dolomite (dolomitic agricultural limestone) were properly identified, seven were ordinary pulverized limestone instead of dolomite, and one was lime. The seven dolomitic agricultural limestone samples were of high quality, that is, fineness ratings above 85% and neutralizing values above 95%. Only two of eight misidentified samples were of high quality. Only one of four products sold as marl had neutralizing value and efficiency rating above 85%, but all were properly identified. Five products sold as crushed seashell had been burned and should have been identified as lime. However, neutralizing values (72 103%) were lower than those of good quality lime. All 13 samples sold as lime were properly identified, and eight were of good quality, that is, neutralizing value above 120% and fineness rating above 85%. The cost of liming materials ranged from US$ 0.01 to 0.02 kg 1 for marl and from US$ 0.10 to 0.14 kg 1 for lime. There was no relationship between product quality and cost. Fish and shrimp farmers in Thailand should insist that manufacturers and vendors of liming materials provide data on product composition. Key words: Agricultural limestone, Lime, Pond liming Introduction Liming is practiced widely in aquaculture for neutralizing acidity in pond bottom soils and water (Boyd and Tucker 1998). Common liming materials are agricultural limestone, burned lime, and hydrated lime. Agricultural limestone is made by finely pulverizing limestone, marble, chalk, marl, or seashells (Jones 1979). Raw materials consist primarily of the following: calcium carbonate or calcite limestone; calcium and magnesium carbonates in roughly 1:1 proportions, or dolomitic limestone; and calcium and magnesium carbonates in some other proportion, or ordinary limestone. Burned lime is made by burning the source material at high temperature in a kiln to drive off carbon dioxide and convert carbonates to oxides (Wingate 1985). If calcitic limestone is the raw material, burned lime will consist primarily of calcium oxide. Hydrated lime is prepared by treating burned lime with water to convert oxides to hydroxides. Raw materials are widely available, and liming materials often are produced locally by small-scale manufacturers. Analyses of liming materials seldom are

provided by manufacturers and vendors. Furthermore, the outward appearance of different kinds and qualities of liming materials may be similar. Shrimp and fish farmers often have little or no information on the composition and quality of liming materials to be applied to ponds. Thailand has a large aquaculture industry, and fish and shrimp producers commonly apply liming materials to ponds. There is a large market for liming materials, and many brands and several grades of the three basic materials are available. Thus, this study was conducted to determine the composition and quality of liming materials used in Thailand. Materials and methods Samples of 45 different brands and grades of liming material were obtained from aquaculture supply stores and shrimp and fish farms in Thailand. These samples were stored in plastic containers and transported to Auburn University for chemical analyses. Neutralizing value (Jones 1979) was determined by treating 0.500 g samples in a 500 ml Erlenmeyer flask with exactly 25 ml of 1.00 N hydrochloric acid and gently heating the mixture over a Meker burner. After the reaction appeared complete, 100 ml of distilled water were added and the solution was boiled for 2 min and held in a boiling water bath for 15 min to assure complete reaction. Excess hydrochloric acid was back-titrated to the phenolphthalein endpoint with 1.00 N sodium hydroxide. The neutralizing value was estimated as: NV ¼ ðv an a V b N b Þ 50 100 S where NV is the neutralizing value (% equivalent CaCO 3 ), V a is the volume HCl (ml), N a is the normality HCl (meq ml 1 ), V b is the volume NaOH (ml), N b is the normality NaOH (meq ml 1 ), 50 is the milliequivalent weight of CaCO 3 (mg meq 1 ), and S is the sample weight (mg). The acidic digestion was repeated on another 0.500 g sample to dissolve calcium and magnesium for analysis. The resulting solution was diluted to exactly 100 ml in a volumetric flask. A 5 ml aliquot was diluted to 100 ml with distilled water, 1 N sodium hydroxide was added dropwise to raise ph above 12 and precipitate magnesium as its hydroxide, and calcium was complexed by titration with 0.010 M ethylenediaminetetraacetic acid (EDTA) to the murexide endpoint. A second, 5 ml aliquot of the solution was diluted to 100 ml and neutralized to ph 7 by dropwise addition of dilute sodium hydroxide solution. The ph was adjusted to 10 with an ammonium hydroxide ammonium chloride buffer, and calcium and magnesium were complexed by titration with 0.010 M EDTA to the eriochrome black-t endpoint. Results of the two titrations were used to determine percentages of calcium and magnesium in liming materials as follows: Calciumð%Þ ¼ ðv 1ÞðMÞð40:08Þð20Þ 100 S

where V 1 is the volume of EDTA used to titrate calcium (ml), M is the molarity of EDTA (mm ml 1 ), 40.08 is the molecular weight of calcium (mg mm 1 ), 20 is the dilution factor for digestate, and S is the sample weight (mg). Magnesiumð%Þ ¼ ðv 2 V 1 ÞðMÞð24:31Þð20Þ 100 S where V 2 is the volume of EDTA used to titrate calcium plus magnesium (ml), and 24.31 is the molecular weight of magnesium (mg mm 1 ). Non-equilibrium ph tests were made in slurries of 10 g of liming material and 50 ml of distilled water (Boyd and Masuda 1994). The slurry was stirred vigorously with a magnetic stirrer for 1 min, a ph electrode was inserted into the slurry, and ph was recorded after 15 s while still stirring. Samples of liming material (100 g) were passed through nested sieves with 1.70, 0.85, and 0.25 mm openings. The separates retained on each screen and the one passing the bottom screen (0.25 mm openings) were weighed and percentages of particles in each of the four size classes (separates) were calculated. Boyd and Hollerman (1982) assigned fineness factors to separates based on their solubility as follows: Separate (mm) Fineness factor >1.70 0.036 1.69 0.86 0.127 0.85 0.25 0.522 <0.24 1.000 The percentage of each separate was multiplied by the corresponding fineness factor, and the results were summed for each sample to provide an overall fineness rating. Results and discussion Samples of liming materials were separated in groups according to product designation by vendors as follows: ground calcium carbonate, ground dolomite, marl, ground seashells, and burned lime. The results of the analyses of the samples are listed in Table 1. The ph of the non-equilibrium slurries of products designated as calcium carbonate were between 9.4 and 9.9 (Table 1). The equilibrium ph between calcium carbonate, distilled water, and atmospheric carbon dioxide is 8.3 (Boyd 2000). However, because carbon dioxide will be removed from solution by reaction with carbonate, the equilibrium concentration of carbon dioxide will not be reached quickly. The initial ph may be as great as 9.5 10.0, but ph above 10 should not occur in distilled water slurries of calcitic, dolomitic, or ordinary agricultural limestone (Boyd and Masuda 1994). A ph above 10 will occur in slurries containing burned lime or hydrated lime because of the presence of hydroxide. Thus,

Table 1. Neutralizing value, fineness rating, percentages of calcium and magnesium, ph of non-equilibrium slurries, and correct product identification for five classes of liming materials in Thailand. Sample no. Neutralizing value (%) Fineness rating (%) Ca (%) Mg (%) ph of slurry Identification Sold as ground calcium carbonate 1 98 96 36.9 1.9 9.5 Ag a. limestone (calcitic) 2 96 100 36.8 1.4 9.4 Ag. limestone (calcitic) 3 98 100 38.0 1.2 9.9 Ag. limestone (calcitic) 4 99 96 37.7 1.4 9.7 Ag. limestone (calcitic) 5 99 98 37.4 1.4 9.5 Ag. limestone (calcitic) 6 101 95 37.6 1.3 9.7 Ag. limestone (calcitic) 7 98 100 36.4 1.9 9.6 Ag. limestone (calcitic) 8 99 65 36.0 1.2 9.4 Ag. limestone (calcitic) 9 81 55 33.4 1.1 9.3 Ag. limestone (calcitic) 10 99 100 32.2 5.3 9.7 Ag. limestone (ordinary) Sold as ground dolomite 11 107 96 21.4 12.0 9.2 Ag. limestone (dolomitic) 12 105 97 21.5 12.4 9.3 Ag. limestone (dolomitic) 13 106 99 21.0 13.1 9.2 Ag. limestone (dolomitic) 14 107 99 20.8 13.2 9.4 Ag. limestone (dolomitic) 15 106 99 20.8 13.3 9.2 Ag. limestone (dolomitic) 16 108 96 21.8 13.0 9.3 Ag. limestone (dolomitic) 17 105 100 21.8 12.8 9.4 Ag. limestone (dolomitic) 18 104 98 26.0 9.7 9.5 Ag. limestone (ordinary) 19 105 97 23.7 11.1 9.6 Ag. limestone (ordinary) 20 104 98 24.8 10.2 9.5 Ag. limestone (ordinary) 21 42 48 11.0 3.8 9.5 Ag. limestone (ordinary) 22 41 51 9.5 6.1 9.4 Ag. limestone (ordinary) 23 98 45 29.8 6.1 9.4 Ag. limestone (ordinary) 24 53 45 10.9 6.6 9.5 Ag. limestone (ordinary) 25 107 99 20.8 5.5 11.3 Lime (ordinary) 26 107 100 21.1 13.7 9.4 Ag. limestone (dolomitic) Sold as marl 27 94 77 35.7 0.0 9.4 Ag. limestone (calcitic) 28 71 63 29.1 2.7 9.8 Ag. limestone (ordinary) 29 88 56 34.9 1.3 9.3 Ag. limestone (calcitic) Sold as ground seashell 30 72 45 26.0 1.6 12.4 Lime (calcitic) 31 103 64 41.1 1.2 12.4 Lime (calcitic) 32 82 60 31.0 1.2 12.3 Lime (calcitic) 33 97 57 38.8 1.0 12.4 Lime (calcitic) 34 90 59 29.7 5.3 12.3 Lime (ordinary) Sold as lime 35 133 86 48.8 1.8 12.6 Lime (calcitic) 36 138 100 48.4 1.9 12.2 Lime (calcitic) 37 124 85 41.2 1.6 12.3 Lime (calcitic) 38 157 98 30.7 18.5 12.5 Lime (dolomitic) 39 130 85 41.3 3.9 12.4 Lime (ordinary)

Table 1. (continued) Sample no. Neutralizing value (%) Fineness rating (%) Ca (%) Mg (%) ph of slurry Identification 40 124 90 44.6 2.3 12.5 Lime (ordinary) 41 124 85 41.7 2.9 12.5 Lime (ordinary) 42 100 97 38.4 0.6 12.4 Lime (calcitic) 43 104 99 38.3 1.4 12.6 Lime (calcitic) 44 109 59 39.5 1.9 12.6 Lime (calcitic) 45 108 71 43.5 1.9 12.3 Lime (calcitic) a Ag. ¼ agricultural. all of these samples should be identified as agricultural limestone. Neutralizing values ranged between 55 and 100%, and eight samples had neutralizing values above 95%. Calcium concentrations ranged from 32.2 to 38.0%, and all samples contained more than 1% magnesium. Pure calcium carbonate has a neutralizing value of 100% and contains 40% calcium. None of the products were pure calcium carbonate, but all samples, other than Sample 10, can be considered calcitic agricultural limestone because they contain less than 2% magnesium. According to Jones (1974) good quality, calcitic, or ordinary agricultural limestone should have a fineness rating and a neutralizing value greater than 85%. Samples 8 and 9 had low fineness ratings, and Sample 9 also had a low neutralizing value (Table 1). The ph of the non-equilibrium slurries of all but one sample designated as dolomite ranged from 9.2 to 9.6 indicating that they consisted of calcium or calcium and magnesium carbonates (Table 1). Sample 25 was ordinary lime, because it had a slurry ph of 11.3 revealing that it had been burned. Neutralizing values ranged from 41 to 108%, calcium concentrations were between 10.9 and 28.0%, and magnesium varied from 3.8 to 13.3% (Table 1). Pure dolomite has a neutralizing value of 108.5%, contains 21.7% calcium, and 13.2% magnesium. Limestone with 12% or more magnesium is considered dolomitic (Jones 1979). None of the samples were pure dolomite, but eight samples with magnesium concentrations of 12.45 13.34% can be considered dolomitic agricultural limestone. The remaining samples should be considered ordinary agricultural limestone, for the magnesium concentrations are too high to allow them to be classified as calcitic agricultural limestone. The dolomite samples had neutralizing values above 100% and fineness ratings above 95% suggesting that they are of high quality. Three of the ordinary limestone samples were of good quality even though they were identified incorrectly as dolomite by vendors. It was not possible to determine if the products marketed as marl were actually made from marl, but they all could be classified as agricultural limestone (Table 1). Three of the samples were of poor quality because of a low fineness rating, low neutralizing value or both. Thus, only Sample 26, which was composed primarily of dolomite, was of good quality. If limestone burning is complete, the final product will be calcium or calcium and magnesium oxides, but limestone will remain in the final product if burning is

incomplete (Wingate 1985). Oxides react with water to form hydroxides, and if hydration of burned lime is complete, the product will be hydrated lime. Burned lime initially containing only oxides will begin to hydrate if stored in a moist place. Thus, burned lime may represent a variety of compositions to include: (1) calcium or calcium and magnesium oxides; (2) a mixture of calcium or calcium and magnesium oxides and hydroxides; (3) calcium or calcium and magnesium hydroxides; (4) a mixture of limestone with one of the other three possibilities. It is conventional to refer to any of the above four combinations of substances as lime. Burned lime or hydrated lime prepared from pure calcium carbonate have calcium concentrations of 71.43 and 54.05%, respectively. If pure dolomitic limestone is the source, burned lime will contain 41.53% calcium and 25.24% magnesium while hydrated lime will have 30.23% calcium and 18.37% magnesium (Ca:Mg ratio of 1.64). Pure burned lime made from calcium carbonate is calcium oxide, and it has a neutralizing value of 178.5%. Pure burned lime made from dolomite is a 1:1 mixture of calcium and magnesium oxides and has a neutralizing value of 207.8%. The neutralizing value of pure calcium hydroxide is 135% and for pure calcium magnesium hydroxide is 151%. Five of the samples (Table 1) were sold as ground seashell. These samples had calcium concentrations of 26.0 41.1% and magnesium concentrations of 1.1 5.3%, which indicated that the major component of the seashell was calcium carbonate. The ph of distilled water slurries was 12.3 12.4, so the ground seashells had been burned, and the products should be classified as lime. The highest neutralizing value was 103%, which is low for lime, and fineness ratings of 45 64% also were low. Products sold as burned lime, hydrated lime, or lime (Table 1) had from 30.8 to 48.8% calcium and from 0.6 to 18.5% magnesium. All samples represented stronger bases than ground limestone because the ph of distilled water slurries was 12.25 12.60. All products were correctly identified by vendors as lime. Good quality lime should have a neutralizing value of 120% or more and a fineness rating of 85% or larger (Jones 1979). Samples 35 41 had neutralizing values of 120% or greater. The sample with the highest neutralizing value (Sample 38) was made from dolomitic limestone as stated by the vendor. From the analysis, it is impossible to determine if it was incompletely burned dolomite or completely burned dolomite that was not completely hydrated. These seven samples also had fineness ratings of 85% or better. The remaining samples had low neutralizing values, and all but two had low fineness ratings. The poor quality samples all contained high percentages of calcium which suggests that the limestone raw material was neither burned nor ground properly. Thirty-three of the 45 samples (73%) were correctly labeled, and 23 or roughly half of the samples were both correctly labeled and of good quality (Table 2). Lime was more expensive than the other products, ground dolomitic limestone was more expensive than ground calcitic limestone and seashells, and marl was the least expensive material (Table 2). In general, there was no relationship between product quality and cost. Agricultural limestone is the preferred product to use in aquaculture ponds for most purposes (Boyd and Tucker 1998). Lime is highly caustic and causes a high

Table 2. Number of samples, samples labeled correctly, samples labeled correctly and of good quality, and price for five classes of liming materials in Thailand. Label n Label correct Label correct and quality good Price range (US$=kg) Ground calcitic limestone 10 9 a 8 0.035 0.045 Ground dolomite limestone 15 7 b 7 0.05 0.09 Marl 4 4 1 0.01 0.02 Ground seashells 5 0 0 0.04 0.05 Lime or burned lime 13 13 7 0.10 0.14 Total 47 33 23 a All were sold as ground limestone. b Fourteen were sold as ground limestone. ph if applied to pond soils or waters at doses greater than 50 or 100 kg ha 1.The recommended use of lime is for disinfecting the bottoms of ponds or pond waters by applying 1000 2000 kg ha 1 during pond preparation (Boyd and Tucker 1998). There is little benefit in dolomitic agricultural limestone over ordinary or calcitic agricultural limestone unless there is a magnesium deficiency. The typical quality dolomite had a neutralizing value of 105% and the typical calcium carbonate product had neutralizing value of about 98% (Table 1) a difference of about 7%. In Thailand, the typical dolomite costs about twice as much as the typical calcium carbonate. Obviously, the dolomite does not have enough added value in neutralizing acidity to justify its much greater cost. Marl is quite cheap in Thailand compared to the other agricultural limestone products, but other than Sample 26, all marl samples were from products of low quality. Nevertheless, the marl is so cheap that it may be less expensive per unit of neutralizing value than the other materials. The liming rate for ponds is expressed in terms of pure calcium carbonate of 100% fineness rating. The equivalent amount of a product with a different neutralizing value and fineness rating may be computed as follows: Amount of product ðkg ha 1 Þ¼ Liming rate ðkg ha 1 as CaCO 3 Þ ðnv=100þðfr=100þ where NV is the neutralizing value (%) and FR is the fineness rating (%). For example, suppose two products are available: (1) marl with a neutralizing value of 88%, a fineness rating of 56%, and a cost of US$ 0.015 kg 1 and (2) ground calcitic limestone with a neutralizing value of 90%, a fineness rating of 92%, and a cost of US$ 0.04 kg 1. The liming rate is given in terms of pure calcium carbonate with a neutralizing value and fineness rating of 100%. Thus, it would require 2029 kg of the marl or 1208 kg of the ground calcitic limestone with costs of US$ 30.43 and 48.32, respectively, to be equivalent to 1000 kg of pure, finely ground calcium carbonate. Of course, the extra cost of transportation and handling of the larger amount of marl will increase its cost relative to higher quality agricultural limestone.

Fish and shrimp farmers in Thailand would obviously benefit greatly if liming materials were labeled as to neutralizing value, fineness rating, and calcium and magnesium contents. This would allow them to make more informed decisions about which product to purchase based on intended use, quality, and cost. Acknowledgments Partial funding for this project came from the Pond Dynamics=Aquaculture Collaborative Research Support Program (PD=A CRSP) funded by USAID Grant No. LAG-G-00-96-90015-00 and by contributions from the participating institutions. The CRSP accession number is 1251. The opinions expressed herein are those of the author(s) and do not necessarily reflect the views of the US Agency of International Development. References Boyd C.E. 2000. Water Quality: An Introduction. Kluwer Academic Publishers, Boston, MA, USA, 330 pp. Boyd C.E. and Hollerman W.D. 1982. Influence of particle size of agricultural limestone on pond liming. Proceedings Annual Conference Southern Association of Fish and Wildlife Agencies 36: 196 201. Boyd C.E. and Masuda K. 1994. Characteristics of liming materials used in aquaculture ponds. World Aquaculture 25: 76 79. Boyd C.E. and Tucker C.S. 1998. Pond Aquaculture Water Quality Management. Kluwer Academic Publishers, Boston, MA, USA, 700 pp. Jones U.S. 1979. Fertilizers and Soil Fertility. Reston Publishing Company, Reston, VA, USA, 368 pp. Wingate M. 1985. Small-Scale Lime Burning. Intermediate Technology Publications, London, UK, 185 pp.