Comparison between THE oven and THE Karl Fischer methodS for THE determination of the moisture content of maize (Zea mays L.) and soya (Glycine max (L) Merrill) seeds

COMPARISON BETWEEN THE OVEN AND THE KARL FISCHER METHODS FOR THE DETERMINATION OF THE MOISTURE CONTENT OF MAIZE (Zea mays L.) AND SOYA (Glycine max (L) Merrill) SEEDS

M.A.A. TILLMANN¹; S.M. CICERO²

¹Depto. de Fitotecnia/UFP, C.P. 354 - CEP: 96010-900, Pelotas, RS.

²Depto. de Agricultura-ESALQ/USP, C.P 9, CEP: 13418-900, Piracicaba, SP.

ABSTRACT: The objective of the present paper was to compare the seed water content of maize and soya, obtained by the oven method at different temperatures and the Karl Fischer method. The water contents used were 11.2 %, 12.8 % and 16.2 % for maize and 10.7 %, 12.4 % and 16.0 % for soya. Samples were dried for a sufficiently long time in order to obtain the same results in both methods. The results indicate that the official method for seed moisture determination (105^oC oven method) should be revised in order to obtain more reliable results. The drying time in the oven varied according to the seed water content and the temperature used.

Key Words: seed, moisture content, Karl Fisher

COMPARAÇÃO ENTRE OS MÉTODOS DA ESTUFA E DE KARL FISCHER PARA A DETERMINAÇÃO DO TEOR DE ÁGUA EM SEMENTES DE MILHO (Zea mays L.) E DE SOJA (Glycine max (L) Merril)

RESUMO: O objetivo da pesquisa foi comparar os teores de água de sementes de milho e de soja obtidos pelos métodos da estufa e de Karl Fischer. Os teores de água utilizados foram 11,2% 12,8% e 16,2% para sementes de milho e 10,7%, 12,4% e 16,0% para sementes de soja. Para as determinações dos teores de água nas sementes foi utilizado o método da estufa com diferentes temperaturas, sendo as amostras secadas por um período suficiente para obter os mesmos resultados obtidos pelo método de Karl Fischer. Os resultados indicaram que o método oficial para a determinação do teor de água em sementes (método da estufa, a 105^oC) precisa ser revisado para a obtenção de resultados com maior precisão. O tempo de permanência das sementes na estufa, para a determinação da umidade, varia conforme o teor de água das sementes e da temperatura utilizada.

Descritores: semente, umidade, teor de água, Karl Fisher

INTRODUCTION

Seeds often have to be submitted to seed moisture testing during the period between preharvest and selling in order to meet requeriments of quality control programs. These testing methods need to be precise, since seed water content is one of the factors that determines the quality of the product.

In Brazil, the Seed Analysis Rules (SAR) were first elaborated and published by Oswald Bacchi, in São Paulo, in 1956. At that time the laboratories were not equipped sufficiently to follow the international rules, therefore, a methodology was established according to the conditions and possibilities of the laboratories, as mentioned by Bacchi (1963) in his foreword to the SAR. Beginning in 1956, the official method for seed moisture determination used in Brazil for seeds of all species was the oven method without forced ventilation, at 105^oC during 24 hours, using whole seeds. The SAR were revised in 1963, 1967, 1976 and, most recently, in 1992. From 1976 on, the methods from the International Seed Testing Association (ISTA) were adapted as official methods (Brasil, 1976).

There is a great diversity in official methodologies for the determination of seed moisture content of the same species, and no agreement exists, between countries, of which one is the most indicated (Hunt & Pixton, 1974; Sasseron, 1978; American Society of Agricultural Engineers, 1992).

The oven method of seed moisture determination is classified as a basic method, but results depend on applied temperature and drying time. A single oven procedure cannot be used with the same accuracy for all species. The appropriate test conditions have to be determined for each species by calibration with standard methods whose results are independent of temperature and drying time (Grabe, 1990).

Researches conducted in Brazil to evaluate the efficiency of the artificial methods used for the determination of seed water content are rare. The most frequently used method (oven at 105 ^oC), introduced in 1956, remains without evolution until today.

Based on these considerations, the objective of the present paper was to compare the seed moisture content of maize and soya, obtained by the oven method and the Karl Fischer method.

MATERIALS AND METHODS

The research was done with three seed moisture contents: 11.2 %, 12.8 % and 16.2 % for maize and 10.7 %, 12.4 % and 16.0 % for soya. These moisture contents were determined by the 105 ^oC oven method.

Moisture content determination by the oven methods: The studied water contents were 11.2 % for maize seeds and 10.7 % for soya seeds. For the determination of the seed moisture contents of whole seeds of maize and soya, the 105ºC oven method and the 130 ^oC oven method were used. For ground maize seeds, the 130^oC oven method was used, while the 103 ^oC oven method was applied for ground soya seeds.

In all determinations, the samples were dried for a sufficiently long time in order to obtain the same results as those obtained by the Karl Fischer method.

In each drying period, the water content was calculated at previously established time intervals. Three sample repetitions were done, located at the lower, middle and upper oven shelf.

Moisture content determination by the Karl Fischer method: For this determination, the seeds were ground and sieved through a 0.84 mm sieve (20 mesh). Three repetitions of about 10 grams of ground seeds were put in three hermetically closed jars.

One gram of each sample was placed in a glass jar and 50 ml of ethanol was added. The jar was closed and left at room temperature for 24 hours for water extraction (Grabe, 1987).

The water content was determined by the Karl Fischer method at the beginning of each series of oven tests.

Moisture content determination by the 130^oC oven method: This method, in which whole maize and soya seeds were used, was applied to evaluate the other moisture contents for the two species: 12.8 % and 16.2 % for maize seeds and 12.4 and 16.0 % for soya seeds. The samples were dried for a sufficiently long time in order to obtain the same results as those obtained by the Karl Fischer method.

In each drying period, the water content was calculated at previously established time intervals.

Statistical analysis: The drying curves were determined by curve fitting using polynomial regression, relating seed water content and drying time. For each temperature and moisture content functions of the type y = x/a + bx and y = x²/ax² + bx + c were fitted for two seed species.

RESULTS AND DISCUSSION

Drying curves for maize and soya seeds demonstrate the effect of temperature and drying time on the water content measured by the oven method (figures 1 to 8). Seed water contents determined by the Karl Fischer method are indicated by a straight line in each graph.

Figure 1
- Moisture contents of whole maize seeds, obtained in a 105ºC oven and by the Karl Fisher method (KF)

The oven-drying time necessary for the moisture content to be the same as that obtained by the Karl Fischer method can be observed at the intersection of the drying curve with the straight line corresponding to the value of Karl Fischer, according to the proceeding established by the Seed Moisture Committee of the ISTA 1989-1992 (Grabe, 1992).

Drying curve values greater than those obtained by the Karl Fischer method are apparently due to the liberation of volatile substances (Benjamin & Grabe, 1988). According to Grabe (1989), the weight loss during drying is due to the loss of water and other volatile compounds, addition of water by chemical decomposition, and reduction of dry matter by decomposition. Grabe (1989) emphasizes that drying times in official methods need to be chosen to leave a certain quantity of water in the seed to compensate the weight of lost volatile compounds.

The water content of 11.2%, result obtained by the official 105^oC oven method and used for maize seeds, corresponded to 13.3% when using the Karl Fischer method after 240 hours, using whole seeds (figure 1). Tillmann et al. (1991) obtained values equal to those from the Karl Fischer method after 72 hours when using the 105 ^oC oven method with whole maize seeds at 17.4% moisture, and at 12.0% all water was removed not even after 264 hours. These results are in agreement with those found in the research, as the water content was removed more quickly from seeds at high moisture content.

At 130 ^oC, the results of the oven method matched with those of the Karl Fischer method in 29 hours for whole seeds (figure 2) and in 51 hours for ground seeds (figure 3). However, it should be noted that the seed-grinding may have introduced errors in the determination by the oven method of ground seeds, as an excessive heating was observed within the seed mill and the excessive heating of some mills can modify the moisture content during the grinding (Hunt & Pixton, 1974; Grabe, 1989). Williams & Sigurdson (1978) tested 4 mills and observed that all samples lost water when their water content was about 10 %. Water losses of 0.38 % and 0.57 % were observed when grinding maize seeds with moisture contents of 13.7 % and 17.6 %, respectively (Hart & Golumbic, 1963). The grinding process is one of the major sources of error and can cause a variation of 1 % in the moisture content (Henderson, 1991).

Figure 2
- Moisture contents of whole maize seeds, obtained in a 130ºC oven and by the Karl Fisher method (KF)

Figure 3
- Moisture contents of ground maize seeds, obtained in a 130ºC oven and by the Karl Fisher method (KF)

This study shows that the official 105^oC oven method is not adequate for maize seeds, as it underestimates the moisture content. The results of the 130^oC oven method became equal to those obtained by the Karl Fischer method in less time, which makes it more desirable for determining seed water content in maize.

The 10.7% water content obtained by the 105^oC oven method, corresponded to 10.6% by the Karl Fischer method in soya seeds. The moisture contents obtained by the 105^oC oven method were equivalent to the Karl Fischer method after 14 hours and 40 minutes (figure 4). It was not possible to estimate an ideal exposure time within 26 hours when using ground seeds in a 103^oC oven (figure 5). Using whole seeds at 130^oC oven, moisture content reached those obtained by the Karl Fischer method in 2 hours (figure 6).

Figure 4
- Moisture contents of whole soya seeds, obtained in a 105ºC oven and by the Karl Fisher method (KF)

Figure 5
- Moisture contents of ground soya seeds obtained in a 130ºC oven and by the Karl Fisher method (KF)

Figure 6
- Moisture contents of whole soya seeds obtained in a 130ºC oven and by the Karl Fisher method (KF)

Tillmann et al. (1991) using soya seeds with a moisture content of 17.4% and 9.6%, obtained equal results by the 105^oC oven method and the Karl Fischer method after 8 hours and 40 minutes and 26 hours and 40 minutes, respectively. These results are in agreement with those of this study, i.e., the water was removed more easily from seeds with a higher water content. It should be emphasized again that excessive heating of the seed mill may have influenced results obtained with ground seeds.

According to the results for soya seeds, it should be noted that the exposure time in the 105^oC oven can be reduced, although the determined water content after 24 hours differed by 0.1% from that obtained by the Karl Fischer method. Besides of this temperature, recommended by the SAR, the water content of these seeds can also be determined at 130^oC.

The 130^oC oven method with whole seeds was the most convenient for maize and soya seeds. An advantage of drying at 130^oC is that the oven can be used more times each day so more determinations can be made (Benjamin & Grabe, 1988). According to Grabe (1990), methods using whole seeds should eventually be evaluated when time is not a limiting factor.

The water contents of 12.8% and 16.2% for maize seeds and 12.4% and 16.0% for soya seeds, as determined by the official 105^oC oven method, corresponded to 15.0% and 17.1% for maize and 12.4% and 14.6% for soya by the Karl Fischer method.

For whole maize seeds, using the 130^oC oven method, the water contents of 13.3%, 15.0% and 17.1% equalized to the Karl Fischer method in 29 hours, 17 hours and 22 minutes and in 3 hours and 42 minutes, respectively (figure 7). Using whole soya seeds, these times were 2 hours, 1 hour and 48 minutes and 1 hour, respectively, for water contents of 10.6%, 12.4% and 14.6% (figure 8).

Figure 7
- Moisture contents of whole maize seeds, obtained in a 130ºC oven and by the Karl Fisher method (KF) for three moisture levels.

Figure 8
- Moisture contents of whole soya seeds, obtained in a 130ºC oven and by the Karl Fisher method (KF) for three moisture levels.

Fron figures 7 and 8 it was observed that water is removed more rapidly from seeds with a higher moisture content. Similar results were obtained by Grabe (1987), who verified that the drying time at 130^oC needed to get results equal to those of the Karl Fischer method for low (4-7%), medium (7-11%) and high (12-16%) moisture contents were 6, 3 and 2 hours for Lolium perenne L., 3, 2 and 0.7 hours for Dactylis glomerata L. and 3, 3, and 0.5 hours for Poa pratensis Hurds. According to the author, it has to be accepted, as a method-inherent failure, that it is impossible to choose an oven drying time equally accurate for all moisture contents of the same species.

The exposure time of the seeds to the oven temperature varied with the moisture content. However, from a practical point of view, it will be impossible to establish more than one drying time for each species, as one should know the moisture content before determining it.

The results of this study allowed to conclude that the drying time reduces when the moisture content is higher. Apparently, this fact is due to the occurrence of high water losses at the beginning of the drying process in seeds with a higher water content.

The water contents of 11.2%, 12.8% and 16.2% for maize seeds and 10.7 %, 12.4 % and 16.0% for soya seeds, as obtained by the official 105^oC oven method, corresponded to those obtained by the Karl Fischer method, but at the highest moisture level the oven method superestimated the moisture content.

The official and most used seed water content determination method in Brazil (105^oC oven method) indicates the same drying time for all species. However, the results obtained for maize and soya showed that the same procedure cannot be applied with the same accuracy for seeds of all species, confirming the results of Hart et al. (1959) and Grabe (1987, 1989).

CONCLUSIONS

The results obtained in the present study allowed to conclude that the official seed moisture content determination method (105^oC oven method) should be revised in order to obtain more precise results. The oven-drying time varied according to the water content and the applied temperature.

Recebido para publicação em 16.05.95

Aceito para publicação em 15.06.95

• AMERICAN SOCIETY OF AGRICULTURAL ENGINEERS. Moisture measurement in ground and grains and seeds. In: Standards, 1992: standards engineering practices and data. 39.ed. St. Joseph, 1992. p.404.
• BACCHI, O. Regras para análise de sementes. Săo Paulo: Secretaria da Agricultura, 1963. 53p.
• BENJAMIN, E.; GRABE, D.F. Development of oven and Karl Fischer techniques for moisture testing of grass seeds. Journal of Seed and Technology. v.12, n.1, p.76-89, 1988.
• BRASIL. Ministério da Agricultura. Departamento Nacional de Produçăo Vegetal. Regras para análise de sementes. Brasília, 1976. 188p.
• GRABE, D.F. Report of the Seed Moisture Committee 1983-1986. Seed Science and Technology, v.15, n.2, p.451-62, 1987.
• GRABE, D.F. Measurement of seed moisture. In: STANWOOD, P.C.; McDONALD, M.B. Seed moisture. Madison: CSSA, 1989. p.69-92.
• GRABE, D.F. Report of the Seed Moisture Committee 1986-1989. Seed Science and Technology, v.18, n.1, p.87-93, 1990.
• GRABE, D.F. Report of the Seed Moisture Committee 1989-1992. Seed Science and Technology, v.20, sup., p.77-89, 1992.
• HART, J.R.; GOLUMBIC, C. Methods of moisture determination in seeds. Proceedings of International Seed Testing Association, v.28, n.4, p.911-33, 1963.
• HART, J.R.; FEINSTEIN, L.; GOLUMBIC, C. Oven methods for precise measurement of moisture content of seeds. Washington: USDA/Agricultural Marketing Service, 1959. 16p. (Marketing Research Report, 304).
• HENDERSON, S. Sources of variation in the determination of moisture content of cereal grains and oilseeds by ovendrying methods. Postharvest News and Information, v.2, n.5, p.335-9, 1991.
• HUNT, W.H.; PIXTON, S.W. Moisture, its significance, behavior, and measurement. In: CHRISTENSEN, C.M., ed. Storage of cereal grains and their products. St Paul: American Association of Cereal Chemists, 1974. p.1-55.
• SASSERON, J.L.. Umidade e temperatura dos grăos armazenados. Viçosa: Centro Nacional de Treinamento em Armazenagem, 1978. 80p.
• TILLMANN, M.A.A.; PIANA, Z.; CICERO, S.M. Comparaçăo entre a percentagem do grau de umidade em sementes de soja e milho pelo método da estufa e o Karl Fischer. In: CONGRESSO BRASILEIRO DE SEMENTES, 7., Campo Grande, 1991. Resumos. Informativo ABRATES, v.1, n.4, p.115, 1991.
• WILLIAMS, P.C.; SIGURDSON, J.T. Implications of moisture loss in grains incurred during sample preparation. Cereal Chemistry, v.55, n.2, p.214-29, 1978.

Publication Dates

• Publication in this collection
  26 Feb 1999
• Date of issue
  Jan 1996

History

• Accepted
  15 June 1995
• Received
  16 May 1995