ENVERONMENTAL CONDETEQNS AND PHY‘Q'MAL FRQé’ERTEfl WHICH PRGDUCE FISSU'JRES {N RICE Thais {‘00 the Degree of DB. D. MlCHIGAN WT‘E UNFVERSITY Otto Robert Kunze 3&64 LIB p. A R Y [VIN i." *1: J 5.23:3 U19“. sv‘«*:zs§ty I'lllclj HUI! U le 911L101! Iljfllflfllfl Iljllfljfllglilflflll This is to certifg that the thesis entitled Environmental Conditions and Physical Properties Which Produce Fissures in Rice presented by Otto Robert Kunze has been accepted towards fulfillment of the requirements for PhoD- degree in A. E. @mam Major professor Date July 28, 1964 0-169 ABSTRACT ENVIRONMENTAL CONDITIONS AND PHYSICAL PROPERTIES WHICH PRODUCE FISSURES IN RICE ' by Otto Robert Kunze Grains of rice may fissure during the process of moisture adsorption from the atmosphere. This phe- nomenon has been postulated but no research has been conducted under controlled conditions to verify the theory. Research was conducted with brown rice and concerns the parameters of l) damage by fissuring of the grain, 2) moisture adsorption from a more humid atmosphere and 3) the time interval after exposure. Six varieties and two ages of rice were used. Compressed air was bubbled through flasks (connected in series)-containing saturated salt solutions to pro- duce desired relative humidities. Grains were initially equilibrated at each of three temperatures and three relative humidities. Otto Robert Kunze Samples of 50 grains were used to determine 1) fissured grains, 2) total fissures and 3) weight of moisture ad- sorbed at preset time intervals after exposure to a more humid atmosphere at a given temperature. After grains had equilibrated in the more humid atmosphere their dry matter weight was determined and used to calculate equi— librium moisture contents at the previous conditions of equilibration. Equilibrium moisture content lines were superimposed on a psychrometric chart by extrapolation ‘of the calculated points. There was no detectable difference in EMC values for grains of different ages. Different grain varieties had different EMC values. Considering all varieties and ages, there was no consistent indication that age made grains more susceptible to fissuring. Some grain varie- ties fissured more readily than others. The varieties aligned themselves in the following order of least to most resistant: l) Fortuna, 2) Zenith, 3) Bluebonnet 50, 4) Belle Patna, 5) Rexoro and 6) Century Patna 231. The Fortuna, Zenith and Century Patna 231 varieties rather consistently assumed the first. second and sixth positions. Otto Robert Kunze respectively, in the grain fissuring evaluations. Vari- ations among the other three were small and inconsistent. When rice kernels were subjected to a more humid condition. the individual.grains deve10ped damage at dif— ferent time intervals after exposure. The response times of the grains were observed and were found to have a nor— mal distribution. Mathematical models of 1 e -1~5(X)2 i=5? were developed to represent the distribution of the grain response intervals. ‘Areas (representing percent grain damage) under the standardized nOrmal curve were deter- mined and plotted as cumulative percentage (grain damage) polygons. Rice varieties more susceptible to damage incurred fissures quicker than did less susceptible varieties. When adjusted mean response times for the respective grain varieties were added for all conditions in which 30 or more grains per sample fissured, the varieties aligned themselves in the following order of smallest (least resistant) to largest (most resistant) sum of Otto Robert Kunze adjusted mean response times: 1) Fortuna, 2) Zenith, 3) Rexoro. 4) Bluebonnet 50, 5) Belle Patna and 6) Cen- tury Patna 231. Differences in adjusted mean response times and standard deviations were found to be as great as 100 percent between the Fortuna and Century Patna 231 (least and most resistant, respectively) varieties. Fortune and Century Patna 231 rice in equilib- rium at 38°F and 59.6 percent RH adsorbed moisture more than ten times as fast over a 23-hour period when sub- jected to a 0.030 psi vapor pressure increase as did other grains of the same varieties equilibrated at 11.2 percent RH and subjected to a 0.027 psi vapor pressure increase. Approvedm Q24! Mr Major Pro essor ENVIRONMENTAL CONDITIONS AND PHYSICAL PROPERTIES WHICH PRODUCE FISSURES IN RICE BY Otto Robert Kunze A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Engineering 1964 _ ,3 (2 to g 3 (L) en La: ACKNOWLEDGMENTS The author wishes to express his sincere appre— ciation to all who voiced an interest and gave encourage- ment to this research. Particular thanks are expressed to Dr. Carl W. Hall (Professor and Chairman, Department of Agricultural Engineering) for his time and patience in guiding the work to make it interesting, educational and rewarding. Thanks are also extended to Dr. F. H. Buelow (Agricultural Engineering), Dr. G. E. Mase (Metallurgy, Mechanics and Material Science) and Dr. J. F. Hannan (Statistics) for serving on the guidance committee. The author wishes to express his deepest grati- tude to the National Science Foundation for the financial assistance which it gave in making this doctorate program possible. Further thanks in this respect are extended to the Ford Foundation and to Dr. A. W. Farrall (Depart- ment of Agricultural Engineering). ii The assistance of Mr. James Cawood and his staff in the Agricultural Engineering research laboratory is gratefully acknowledged. The Texas A&M University and in particular the Department of Agricultural Engineering are due grateful acknowledgement for granting the author sufficient leave of absence to complete this work. Recognition is also given to the Rice Pasture Experiment Station at Beaumont. Texas, for supplying the rice. This dissertation is gratefully dedicated to my wife. Alice. and children, Glenn. Allen and Charles. who consistently gave their encouragement and support to this venture. O. R. K. iii VITA Name: Otto Robert Kunze Born: May 27, 1925 Place: Warda, Texas Academic Degrees 0 1. Bachelor of Science. Agricultural Engineering Texas A&M University. 1950 2. Master of Science. Agricultural Engineering Iowa State University. 1951 3. Doctor of Philosophy, Agricultural Engineering Michigan State University. 1964 Academic Awards and Achievements 1. National Science Foundation Science Faculty Fellow 2. Recipient of "The Agricultural Faculty Award of Merit," Texas A&M University 3. High School Valedictorian. 64 students Teaching and Research Experience 5 years: 2/3 teaching. 1/3 research Agricultural Engineering Department. Texas A&M University Publications: Primary author ----- 8 Co-author -------- 3 Membership - Professional and Honor Societies 1. American Society of Agricultural Engineers 2. Phi Kappa Phi ‘ 3. Sigma Xi 4. Tau Beta Pi 5. Alpha Zeta 6. Gamma Sigma Delta Commercial and Industrial Experience 5 years: Industrial and Agricultural Engineer. Central Power and Light Company. San Benito. Texas Family Status: Married. 3 children Church Affiliation: Lutheran. Missouri Synod Military Service: 22 months total. 17 months European Theater EAME Campaign Ribbon with 2 Bronze Stars. 76th Division. Third Army iv ABSTRACT . . . . ACKNOWLEDGMENTS. VITA . . . . . . LIST OF TABLES . LIST OF FIGURES. TABLE OF CONTENTS LIST OF APPENDIX TABLES. . . . . . . . . . . . . . ABBREVIATIONS AND SYMBOLS. INTRODUCTION . . . . . . . . . . . . . . . . Objective C O I O O O O O O O O O O O 0 Statement of the Thesis Problem . . . . REVIEW OF LITERATURE . . . . . . . . . . . . I. 1.1 1.2 II. 2.1 2.2 2.3 2.4 2.5 2.6 III. THE Physical Dimensions of the Grain. . . . Rice Milling. . . . . . . . . . . . . . Research Recommendations. . . . . . . . Drying of the Grain . . . . . . . . . . Equilibrium Moisture Content Conditions Moisture Adsorption by the Grain. . . . INVESTIGATION. . . . . . . . . . . . . . Preliminary Research. . . . . . . . . . 3.1.1 Wooden humidity containers . . . 3.1.2 Plastic humidity containers. . . V Page ii iv vii viii xi xiii Web 12 16 18 26 29 40 41 47 51 Table of Contents-—Continued 3.2 Selection of Rice Varieties and Types . 70 3 3 Selection and Development of Equipment 73 0 3.3.1 Dynamic system for relative humidity . . . . . . . . . . . 77 3.3.2 Observation and inspection chamber. . . . . . . . . . . . 86 3.4 Standard Procedure. . . . . . . . . . . 88 3.4.1 General. . . . . . . . . . . . . 88 3.4.2 Weighing procedures. . . . . . . 94 3.4.3 Determination of grain damage. . 98 3.4.4 Determination of moisture adsorbed . . . . . . . . . . . 107 3.5 Codification of Results . . . . . . . . 109 IV. THEORETICAL CONSIDERATIONS . . . . . . . . . 114 4.1 Psychrometrics. . . . . . . . . . . . . 114 4.2 Statistical Implications. . . . . . . . 117 V. DISCUSSION AND RESULTS . . . . . . . . . . . 128 5.1 Critical Analysis of the Literature . . 128 5.2 Fissuring of the Rice Grain . . . . . . 135 5.2.1 Desorption damage. . . . . . . . 136 5.2.2 Adsorption damage. . . . . . . . 139 5.3 Evaluation of Grain Damage. . . . . . . 146 5.4 Evaluation of Moisture Adsorption . . . 171 VI. SUMMARY AND CONCLUSIONS. . . . . . . . . . . 184 6.1 Summary . . . . . . . . . . . . . . . . 184 6.2 Conclusions . . . . . . . . . . . . . . 187 REFERENCES . . . . . . . . . . . . . . . . . . . . 194 APPENDIX . . . . . . . . . . . . . . . . . . . . . 202 vi LIST OF TABLES TABLE 1. Results when fissured grains and adjusted mean response times were added for seven test condi— tions in which 30 or more grains fissured in all samples . . . . . . . . . . . . . . . . . . Results when the number of fissured grains of a given variety and age were added for the nine test conditions for which less than 50 grains per sample fissured . . . . . . . . . . . The average differences in equilibrium moisture content for the varieties as shown when Zenith was used as the standard . . . . . . . . . . . . A comparison of the rate of moisture adsorption by grains initially at different equilibrium moisture contents before being subjected to approximately the same vapor pressure increases vii Page 158 162 169 182 LIST OF FIGURES FIGURE 1. The relative humidities which the respective saturated salt solutions will produce in en- closed containers at the indicated tempera- tures . . . . . . . . . . . . . . . . . . . . A plastic freezer container (left) with sat— urated salt solution and grain tray. A top view (right) of freezer container with cover removed. (Photo neg. 641367-1, 641367-2) . . One of the controlled atmosphere chambers with the saturated salt solutions and equip— ment which was used for inspecting the grains. (Photo neg. 641367-4) . . . . . . . . . . . . Effects of moisture adsorption by brown rice from various magnitudes of relative humidity change. . . . . . . . . . . . . . . . . . . . Effect of moisture adsorption from various magnitudes of temperature and relative hu- midity change . . . . . . . . . . . . . . . . Efgect of cycling high moistuge grain between 38 F. 86.7 percent RH and 100 F. 85.6 percent RH. Grain moisture content essentially re- mained constant . . . . . . . . . . . . . . . Rice grains were moved along the path indi- cated. No fissures develOped while grains were kept at a constant equilibrium moisture content . . . . . . . . . . . . . . . . . . . viii Page 52 53 53 57 60 64 65 List of Figures—-Continued. Figure Page 8. 10. 11. 12. 13. 14. Effect of cycling low moisture grain be- tween 380F. 34.8 percent RH and lOOOF, 49.9 percent RH. Grain moisture content essentially remained constant . . . . . . . . 67 A single complete dynamic system (E) for conditioning air to a desired humidity be- fore it passed through the observation and inspection chamber (F) in which the rice grains were studied. (Photo neg. 641367-5) . 81 The observation and inspection chamber with brown rice grains in position. Conditioned air entered through the tube at the t0p of the picture and was exhausted at the opposite end of the chamber. (Photo neg. 641367-3). . 81 The equipment used for inspecting the rice grains after the hulls had been removed. (Photo neg. 641367-6) . . . . . . . . . . . . 90 Grains of brown rice after equilibrating for three weeks were separated into groups of 50 grains and placed into 1-dram vials in pre- paration for the grain damage and moisture adsorption studies. (Photo neg. 641367-7). . 90 Equilibrium moisture content lines for brown rice superimposed on a standard psychrometric chart (Photo neg. 641375) . . . . . . . . . . 106 A probability graph illustrating the accumu- lated grain damage after the indicated re- sponse time. The linear relations indicate that the respective response times followed normal distributions. . . . . . . . . . . . . 120 ix List of Figures--Continued. Figure Page 15. Points of cumulative grain damage plotted directly and a normal curve plotted from the mean and standard deviation calculated from the experimental data. . . . . . . . . . 125 16. Cumulative grain damage plots which illustrate the responses obtained from Fortuna rice when exposed to thg indicated relative humidity changes at 92 F . . . . . . . . . . . . . . . 155 17. Cumulative grain damage plots which illus- trate the responses obtained from Fortuna. Bluebonnet 50 and Century Patna 231 rice when these varieties were exposgd to the indicated humidity change at 92 F . . . . . . 156 18. The rate of moisture adsorption for a unig weight (dry matter) of Fortuna rice at 38 F . 174 19. The rate of moisture adsorption for a unig weight (dry matter) of Fortuna rice at 68 F . 175 20. The rate of moisture adsorption for a unit weight (dry matter) of Century Patna 231 rice at 38 F. . . . . . . . . . . . . . . . . 176 21. The rate of moisture adsorption for a unit weight (dry matter) of Century Patna 231 rice at 68 F. . . . . . . . . . . . . . . . . 177 APPENDIX 1. A summary of data for the A summary of data for the A summary of data for the A summary of data for the A summary of data for the A summary of data for the LIST OF APPENDIX TABLES calculations and grain damage Zenith rice variety . . . . . calculations and grain damage Fortuna rice variety . . . . . calculations and grain damage‘ Bluebonnet 50 rice variety . . calculations and grain damage Rexoro rice variety . . . . . calculations and grain damage Belle Patna rice variety . . . calculations and grain damage Century Patna 231 rice variety The maximum number of fissures which were observed in samples of 50 grains at 38°F When the indicated relative humidity changes were made The maximum number of fissures which were observed in samples of 50 grains at 68°F when the indicated relative humidity changes were made . The maximum number of fissures which were observed in samples of 50 grains at 92°F when the indicated relative humidity changes were made . xi Page 203 204 205 206 207 208 ' 209 211 213 List of Appendix Tables--Continued. Appendix Page 10. Equilibrium moisture contents which were determined after brown rice had equilibrated for three weeks or more . . . . . . . . . . . 215 11. A sample data sheet recording grain damage and total fissures. . . . . . . . . . . . . . 217 12. A sample data sheet recording moisture ad— sorption for grain damage experiments . . . . 219 13. Grain equilibrium moisture contents deter- mined at 38 F . . . . . . . . . . . . . . . . 220 14. Grain equilibrium moisture contents deter- . o mined at 68 F . . . . . . . . . . . . . . . . 221 15. Grain equilgbrium moisture contents deter- mined at 92 F . . . . . . . . . . . . . . . . 222 16. A sample data sheet recording weight gains from moisture adsorption. . . . . . . . . . . 223 17. A sample data sheet recording moisture ad- sorption weights and rates corrected to the base of a unit dry matter weight. ... . . . . 224 xii CC db ea. EMC gm gmdm 9mM hr Min ABBREVIATIONS AND SYMBOLS adjusted adjusted mean re—_ sponse time. adj x degrees Centigrade cubic centimeters damage. grains fissured dry basis each equilibrium moisture content, percent and others 2.7183 degrees Fahrenheit gram gram dry matter gram moisture hour moisture for example moisture change in moisture grams minimum or minutes MRT n ND No. psi rep RH dif Temp T dif VP .AVP wb xiii mean response time. x sample size. a number no damage. grains did not fissure number pounds per square inch roentgens equivalent physical relative humidity. percent relative humidity dif- ference, percent sample standard devia— tion sample standard error of the mean temperature Eemperature difference, F p0pu1ation mean vapor pressure. psi vapor pressure change. psi wet basis measured value of the variable, response time interval or power of magnification Abbreviations and Symbols—~Continued. R sample mean. sample mean response time y dependent variable. ordinate value Z x-u x-u EF’ or '3:- x [1 change, change in w infinity v 3.1415 % percent V‘ population standard deviation 2. summat ion xiv Q INTRODUCTION Rice is among the oldest of cultivated crops. History makes first mention of its being grown in China as early as 2800 B.C. Today nearly one-half of the world's (arable land is used for the production of cereals and about one-fifth of this is used for the production of rice. This cereal has been and continues to be the basic food for over one-half of the world population. Rice ranks third, behind potatoes and wheat. in world production and second. behind wheat only. as a food for human consumption. According to Grist (1959). Asia is the origin of practically all rices grown in America. Over 5000 varieties of paddy from various parts of the world have been intro- duced into the United States within the past 30 years. First attempts at producing rice in our nation were made in the 1680's in South Carolina. Since then the industry has grown continuously. Production for 1963 in the United States was 153 million bushels or nearly 7 billion pounds harvested from approximately 1,800,000 acres, according to figures released by the Rice Millers' Association (1963, 1964). The U. S. Department of Agriculture (1961) states that rice ranks third as a staple food grain in our nation with an annual per capita consumption of 6 pounds. The rice grain as harvested consists of the hull (lemma and palea), pericarp. germ and the endosperm. The threshing operation separates the enclosed grain (spikelet) from the panicle. Husking or shelling the grain is an addi- tional operation. The pericarp is described variously by different investigators. It is the seed coat (bran) which may have as many as seven different cell layers. The inner- most of these is known as the aleurone layer and is rich in proteins. oils and vitamin B. The embryo or germ consists of protein. fat and nitrogen free extract. It is usually removed with the bran in the milling process. The endosperm. the product remaining after the hull. bran and germ have been removed. is the primary product of rice production. This product. commonly known as polished rice, is screened to remove broken particles. Unbroken kernels (3/4 kernels or greater in size) are referred to as head rice. Broken kernels are separated into second» heads. screenings and brewer's rice. The final proportion of clean rice classified as head rice in a given lot is the main factor determining its economic value. Size of grain (brown rice) as reported by Kramer (1951b) is dependent upon the variety considered. Kernels from short—grain types may be only 0.2—inch long while long- grain types may have up to 0.3-inch lengths. Width of the grain is generally 0.1 inch or less while the thickness is seldom more than 0.08 inch. The grain is likened to a porous mass by Angladette (1964). It contains water which evaporates when the vapor pressure of the moisture in the grain is higher than the vapor pressure of the ambient air. When the vapor pressure of the moisture in the air is greater than that in the ker- nel. the rice grain will adsorb moisture. Much research has been conducted involving the harvesting. drying and milling of rice in an effort to determine the cause of broken rice grains. Some of this work has been only partially successful because of varying conditions which have existed while the research was being conducted. Different rice varieties. different equipment used. different moisture contents of the product and dif— ferent seasons under which the products were grown are only a few of the variables which were involved. A second reason why research in the past has been only partially successful is the fact that very little is known about the physical and mechanical properties of the rice grain. Hence much of the data collected are not sub- ject to a logical or full interpretation. The growers. processors and millers in the rice industry need to know the physical and mechanical prOperties of rice in order to develop equipment and processes which will lead to the high- est quality product at minimum cost. 1.1 Objective The objective of this research was to determine the relationship of environmental factors and physical grain conditions causing moisture adsorption and fissur- ing of the rice grain. 1.2 Statement of the Thesis Problem The work reported in this thesis may be divided into four parts: 1. The observation that rice grains may fissure during the process of moisture adsorption from the atmosphere and preliminary investigations which were made to definitely confirm this observation. 2. The selection of materials and the development of equipment which would produce the atmospheric conditions used to subject the rice grains to sudden environmental changes. 3. The determination of certain physical conditions of the grain and corresponding environmental conditions which cause the grains to fissure. 4. The evaluation of the data to predict the rate and extent of grain damage which may be expected when grains at a given equilibrium condition are subjected to an atmospheric change. II REVIEW OF LITERATURE During the last four decades many volumes have been written concerning different aspects of growing. harvesting, processing. storing, milling and utilizing rice. Many publications concerning chemical and physio— logical studies, breeding and selection. diseases and their control. economics of production, cultural prac- tices and other related areas of the rice industry are also in the literature. Basio, §£_§1. (1958, 1959. 1960. 1962) abstracted research articles relative to rice at the University of the Philippines College of Agriculture and published four volumes containing 4.957 abstracts divided into 24 categories. The articles had been published between 1926 and 1960. None of the 24 categories are directly concerned with the physical. mechanical or rheological properties of the rice grain. Three categories most closely related to the engineering properties of the grain may be 1) physiological studies. 2) drying and storage and 3) milling. The fact remains that little information is available on the physical. mechanical and rheological properties of the rice grain. Whatever information is available is often so diverse and widely scattered throughout the literature that the re- searcher has had little prospect of piecing it together to develop some clear concepts of the grain's engineering properties. Prior to this research a concerted effort was made to find available information pertaining to the phy- sical and mechanical prOperties of the rice grain and with this information to develop a research project which could make a contribution to the presently meager know- ledge in the area. The review of literature was conducted 1) by consulting books. journals. bulletins and other pub- lications and 2) by corresponding with the rice researchers throughout the nation. This information led to a concen- trated search in the area of moisture adsorption by the rice grain. 2.1 Physical Dimensions of the Grain The dimensions of a rice grain have an important influence on all the processes involved from harvesting to the final consumption of the grain. A study of seed dimensions was conducted by Kramer (1951 b) using samples of 35 varieties of rice harvested in 1949. Length. width and thickness measurements were made on both the unhulled and the hulled grain. A binocular microsc0pe and a stan- dard micrometer permitted readings to be interpolated to within 0.001 of an inch. All dimensions for a given var— iety were based on the measurement of 20 kernels (unspeci- fied moisture content) grown in one season and in one lo- cality. The influences of fertility. soil type. irrigation and climate on grain dimensions were not considered. Along with external dimensions. internal structure has an important influence on grain reaction during the processing Operations. The structure of the rice grain has been studied directly by microscopic and X-ray tech- niques and indirectly by observations made after cooking of the grain. Hogan. Larkin and MacMasters (1954) used X-ray techniques to examine unhulled rice for cracks. checks. insect damage and immature grains. Cracks or checks were later confirmed by making photomicrographs of halves of the same kernels. Refinements in seed sectioning made it possible for them to obtain photomicrographs which revealed in greater detail the internal structure of the grain. These authors stressed the fact that too little is known about the internal fine structure of the rice kernel and that few reports with photomicrographs have been published that contribute to an understanding of the conditions present in damaged rice. The product which remains after a rice grain has been milled is primarily the endosperm. Its cells vary in size and shape. Dachtler (1959) describes the endo- sperm as composed of cells filled chiefly with starch and it is therefore known as the sunchy endosperm. These cells (longitudinal cross-section) are bricklike in shape. having greater length than depth. except in the center of the kernel where they are polyhedral. The bricklike cells extend laterally out from the longitudinal axis. Fissuring 10 or cracking across the grain may be favored by this cell arrangement. Considerable research has been conducted at the Central Food Technological Research Institute. Mysore. India. on the specific effects that cooking has on the rice grain. Observations have been made by Desikachar and Subrahmanyan (1959) on the expansion of old and new rice during cooking. They reported that old rice increased in length. without disintegrating. more than new rice. The terminal portions of the long axis as well as the ventral segment exhibited greater expansion than the central por— tions and the dorsal segment. Cell walls of parboiled rice remained intact even after being cooked for 30 min- utes. They offer the following explanation for the ex- pansions which they observed. It may be that the cells are more free to expand toward the ventral line of fusion (suture), which is a weak line morphologically. During puffing. the rice grain bursts along this line of fusion. Also the thickness of the aleurone layer is smaller along the ventral side than along the dorsal side of the grain. Desikachar and Subrahmanyan (1959) show the cell arrangement of a transverse section (perpendicular to the longitudinal axis) of the rice grain. The cell 11 structure extending radially out from the center of the grain appears more like a honeycomb or like the scales on a fish. Cells extending from the dorsal to the ventral side tend to be parallel. tubular and elongated. These cells lie in the plane of the cross-section used by Dachtler (1959) from which he described the cell structure to be bricklike and radiating out from the central axis of the grain. These researchers also suggest that more detailed investigations of the cellular structure of fresh and older rice be made with thin microtome sections in order to reconfirm and extend their findings. Long-grain rice. which is dry and fluffy when properly cooked. has the greatest consumer preference in the United States. The demand for short and medium- grain types is considerably less. Halick and Keneaster (1956) pointed out that long-grain varieties now being grown in this country vary greatly as to cooking behavior. Processors have experienced difficulty in consistently securing supplies of rice having the specific properties required for their processes. 12 In general, Indica varieties vary more in physi- cal properties and chemical composition than do Japonica varieties, according to Bienvenido, Bautista. Lugay and Reyes (1964). Japan is generally the origin of the Ja— ponica varieties while the Indica varieties originated in other parts of Asia. Types most common in the United States are of the Indica variety. These authors explain that cooking causes the gelatinization and swelling of the starch granules in the rice endosperm. Although rice starch. when being cooked, may increase 60 times in volume. the kernel in- creases no more than four times its original volume. Non-starch kernel constituents tend to suppress this swelling. 2.2 Rice Milling Milling is one of the processing operations through which rice must pass before it is ready for the domestic market. The milling process includes cleaning. hulling. poliShing and screening operations. According to Aten and Faunce (1953) good paddy stock (rough rice) will usually yield the following products. 13 Percent Milled and coated rice (head rice) 51 Broken and brewers rice 18 Total yield of edible grain 69 Bran ‘ 10 Rice polishings 1 Total by products 11 Husks (including invisible milling shrinkage) _29 100 Slusher (1952) reported that the yield of fancy rice (head rice) may vary from less than 25 percent to over 90 percent of the total clean rice. After milling rice is divided into various classifications according to the extent of breakage. Fancy rice includes whole grains and broken rice three-fourths of a grain or larger; second heads include broken rice of one-half to three-fourths of a grain; screenings include broken rice less than one-fourth of a grain in size. Of the total clean rice. the proportion of fancy rice largely determines the value of the grain. 14 Many rice kernels are broken or cracked in the hull before the rice is milled. When the hulls are re— moved the pieces fall apart. Rice remaining uncut in the field too long experiences sun-checking. according to Smith and McCrea (1951). Breaking of rice kernels by milling really starts in the first break huller or cone. More rice is broken in this operation than in any other in the milling process. Observations by Autrey. §g_§1. (1955) indicate that about 20 percent of the breakage occurs during re- moval of 75 percent of the bran. Breakage in milling was found to occur only in the stone sheller. the first- and second-break hullers and the brush. For optimum yields the mill room should be maintained at 70 to 80 percent relative humidity. Rough rice entering the mill should be at the temperature of the mill room. For milling small samples, the McGill rice sheller and the McGill miller are used. A sample of 125 grams rough rice is required. Hulls are removed with the sheller. The brown rice is weighed before it is milled for 2.5 min- utes in the miller. The milled rice is then removed and 15 placed in a paper bag or envelope until it cools to room temperature. This prevents the checking that would occur if the samples were spread out immediatahion a table. as related by Adair (1952). In his book, Rigs. Grist (1959) states that the milling quality of paddy depends on the size and shape of the grain. i.e., on the variety. conditions under which it was grown, the degree of ripeness and amount of exposure to the sun. Over-ripeness and excessive exposure cause cracks in the grain. leading to excessive breakage in milling. Age, moisture content and conditions under which the rice grain has been dried and stored also materially affect the milling quality. Roberts. et. al. (1954) conducted research with parboiled rice. It is made commercially by steeping rough rice in warm-water. steaming the soaked rice to gelatinize the starch in the endosperm. drying and fi- nally milling the grain. From the milling standpoint. head rice yields from parboiled rice are substantially higher than from untreated rice. The increased costs 16 of processing are largely offset by the increased yield of head rice, since whole grains sell for about 50 per- cent more than broken grains. 2.3 Research Recommendations As recently as February. 1963. the Rice Research and Marketing Advisory Committee met in New Orleans. Louisiana. dations. l. and among others made the following recommen- Fundamental research should be expanded on the properties of the starch, proteins and minor components, such as enzymes. lipids and non—starch carbohydrates. and on the detailed structure of the kernel as related to processing and eating char- acteristics of different varieties of rice initially. during aging and after various treatments. The committee recommends ex- peditious compilation of known research facts on the chemical and physical proper- ties of rice and the immediate publication of such material. In developing cost studies on drying and handling of rice consideration should be given to the decrease in value of rice because of decreased quality. which would include both reduction in head and total yield. and reduction in grade because of damage. The committee recommends that more work be done on the breakage of paddy 17 and bulk milled rice caused by conven- tional field dryer and mill handling equipment, and the development of methods‘ or devices to reduce this breakage. The same sentiments are expressed by the rice researchers throughout the nation. Hogan (1962) comments that little information is available on the physical. mechanical and rheological properties of the rice kernel. In contrast to wheat and corn. basic data on the physical and chemical properties of rice are scanty. The princi- pal reason for this is that rice has traditionally been utilized as a whole grain cereal. In the past. engineer- ing information has not been needed. However. with the increased industrial use of rice and rice products in food processing operations. a need has arisen on the part of chemists and engineers for fundamental physical data dealing with the technological properties of rice and its products. Such data would furnish essential in— formation for use in both process control and design. Rivenburgh (1962) expresses similar thoughts to the effect that very little information is available on this subject for which little or no research is being done at this time. 18 According to Houston (1962). the relationships between moisture. temperature and physical dimensional changes in the rice kernel which cause cracking and re- sulting economic loss are still incompletely understood. 2.4 Drying of the Grain Research has shown that moisture removal from rice can be most effectively achieved with multiplepass drying. The time intervals between passes are considered as tempering periods during which moisture equalizes in the kernel and stresses are relieved. Head rice yield generally is used as the criterion for determining the .most effective drying method. Rice grains. described by Aten and Faunce (1953) are thin and brittle. Consequently. the drying process must be carried out slowly and at low temperatures to prevent cracking and checking of the kernel which would occur from rapid contraction if high temperatures were used. Hogan. Larkin and MacMasters (1954) point out that proper control of the handling. drying and processing operations is necessary in order to minimize the degrad- ation of the product through formation of cracks and 19 checks in the endosperm. Improper drying methods are one of the most important causes of checks and cracks which result from stresses induced in the kernel by ex- cessive moisture gradients and possibly from thermal changes.- To date there is no adequate method for mea- suring or determining this grain damage in spite of its economic importance. Evidence collected by Kramer (1951a) indicated that rapid field drying of rice. immediately before har- vest in drained fields during periods of low humidity. lowered head rice yield. Research conducted by Henderson (1957) on the effects of drying-air temperature and humidity on the quality of milled rice showed that yields were signifi- cantly reduced as air temperatures were raised or as humidities were lowered. When the drying air was humidi- fied head yield was increased but the drying time was ex- tended. Studies by Langfield (1957) on both long and short grain types revealed that delayed harvesting. accompanied by decreased moisture content led to increased breakage of rice grains. Early harvesting was therefore recommended. 20 Research is being conducted in virtually all the rice growing countries in the world relative to the har- vesting, processing and storing of rice° Angladette (1964), Inspecteur General, Institut de Recherches. Agronomiques Tropicales et des Cultures Vivrieres, Paris. France, has compiled a complete review of literature concerning the principles and techniques of rice drying used throughout the world. This compilation is particularly interesting because it contrasts our (United States) methods and tech- niques with those in other nations. Many nations of the world do not have the mechanization which exists in the United States. but in these nations research is being com- ducted on a level which is commensurate with their present methods of rice production. The comments which follow are abstracted from Ang- ladette's work but the original documents are cited as references. Montgrand (1958a. 1958b) developed curves for naturally dried Bellardone rice which illustrate that an increase in moisture content from 13 to 28 percent gave a corresponding decrease in head rice from 66 to 48 percent. 21 Similar research was conducted by Umalo, Silverio and Santos (1956) in the Philippines with the following results. Moisture at milling Commercial rice percent percent 20.0 55.1 18.0 56.1 16.0 62.7 14.2 63.7 12.0 64.0 10.0 66.5 8.5 64.1 These data indicate that the lower moisture con- tent grain gave a higher yield of head (commercial) rice. Langfield (1957) recommended early harvesting at high moisture content for optimum yields of head rice. Umalo. Silverio and Santos (1956) as well as Montgrand (1958a. 1958b) found that rice dried naturally and then hydrated to higher moisture contents yielded less head rice. Di- verse results of this type are one reason why rice research 22 has made such slow progress. Other reports indicate that different varieties of rice, grown in the same area from year to year and after identical drying periods, yielded different percentages of split and broken kernels. The differences are noted but no explanation for them is given. Reports by Angladette (1956, 1960) and Dobelmann (1955. 1961) from the Lake Alaotra and Marovoay Stations in Madagascar attributed a very high increase in brokens in 1955 to rain at the end of the ripening period. Research conducted by Ten Have (1959. 1961) in Surinam also showed considerable variation in split ker- nels for the same variety of rice grown in 1956 and 1957. The percentage of splits in 1956 was low but for 1957 was high. according to Ten Have. owing to the very hot dry weather in the second half of the growing period. Other observations were that a relatively low percentage of split kernels developed prior to the 145th day of paddy age. After the 145th day. the percentage of splits generally increased quite rapidly. Dobelmann (1955, 1961) conducted studies on the effect of drying methods (before threshing) on milling 23 yield and quality of milled rice. The rice was cut, bundled and stacked as indicated in the following table. After the grains were fully dry. the milling yield was determined. ==—==r Milling yield Time of stacking as (Total Brokens related to harvesting percent percent Immediately 60 22 After 1 night on the ground 60 37 “ 2 nights " " " 60 38 " 3 " " “ " 59 39 " 4 " " " " 59 38 " 5 " " " " 61 47 " 6 " " " " 61 48 " 7 " " " " 60 52 " 8 " " " " 59 55 " 9 " " " " 60 52 There was considerable difference in milling results with immediate and delayed stacking. i.e.. a 15 percent increase of brokens after a delay of one night. No mention is made 24 of temperature or humidity conditions. Similar research results are given by Coyaud (1950) who worked in Cambodia. Experiments in which the paddy was dried in the sun and in the shade are reporumtby Angladette (1956, 1960) and Dobelmann (1955. 1961). Results revealed highly signifi- cant differences in percentages of cracked grain between drying in the shade and in the sun. The broken percentage was much lower when paddy was dried in the shade. Similar results were obtained in Guinea and Mali. The foregoing observations would indicate that paddy should be dried under shelter and in the shade. Protection from dew and rain is also desirable although not always possible. Experimental results reported by Coyaud (1950) indicated that dew had little effect on the percent of brokens in milling. There was, however. a marked difference when paddy was cut in the over—ripe stage. Occasional wetting proved to be of no importance if the paddy was dried in the shade. When dried in the sun, the percent brokens increased considerably. When rice was harvested in the over-ripe stage, the percentage of brokens was very high even if the rice was dried in the 25 shade. Grain dried quicker in the sun (12 percent in 2 to 4 days) but the brokens percentage also increased greatly. In summarizing the experimental work conducted under tropical conditions relative to drying. Angladette (1964) made the following statements. As a general rule. the grain should be dried in the shade and, as far as possible, sheltered from the rain. When the grain is wetted by rain. it should be dried in the shade. For quicker moisture removal. sun drying is preferable. but the quality of the milled product is impaired. Burmistrova. §£_§1. (1956) in Russia made some pertinent observations on the harvesting and drying of rice. They reported that some spikelets were obviously broken during threshing. A certain percentage. however. had cracked grain hidden by undamaged hulls. Cracked rice grains were not only found after machine threshing but already when the plants were still in windrows and even in the standing crops. According to Burmistrova. factors which affected the amount of cracking other than the working parts of harvesters were 1) fluctuations in 26 water regime during the growth period. 2) water tempera— ture, 3) variations of air temperature and humidity dur- ing harvest and drying and 4) the length of windrow dry- ing. Extended stalk drying in windrows abruptly increased percentage of cracked grain. especially in plants in upper parts of the windrows. Over a 4-day period of windrow drying. cracking in the lower part of the windrows in- creased from 11 — 78 percent and in their upper part from 14.3 - 83.6 percent. 2.5 Equilibrium Moisture Content Conditions Rice which has been harvested will reach an equi- librium condition with the ambient air provided the ambient conditions remain constant over an extended period of time. This is seldom the case in an open atmosphere and conse- quently rice grains in a single layer thickness are cone tinuously adsorbing or desorbing moisture. Constant con— ditions of relative humidity and temperature in limited environments can be achieved by various means. These may then be used to determine the equilibrium moisture contents of the grain. 27 Coleman and Fellows (1925) were the first indi— viduals who attempted to establish equilibrium moisture content conditions for rice. Their preliminary studies indicated that it required from six to eight days for equilibrium moisture content to be reached. Tubes with grains (60 grams) were weighed. every 24 hours and were kept in position until two 24—hour weighings were identi- cal. Drying was done in a water oven at 1000C for 120 hours at which time the weight of the grains essentially had become constant. Commercial classes of rice were used. but the term commercial is not further defined. The sensitivity of their balance was not reported. Their work was done at 80°F and over a span ranging from 15 to 100 percent relative humidity (RH). Further research on equilibrium moisture contents was done by Karon and Adams (1949). They worked with field dried and artificially dried rough rice at 77°F and reported results at 10 percent relative humidity in- crements between 10 and 90 percent. Salt solutions were used to produce equilibrating atmospheres. The samples generally attained hygrosc0pic equilibrium within a period 28 of three weeks with the exception of those equilibrated over solutions of very high or very low relative humidi- ties. Their data indicate lower equilibrium moisture content values for corresponding temperature. relative humidity conditions than reported by Coleman and Fellows for commercial rice. Additional work was done by Hogan and Karon (1955). They equilibrated rough rice at 80. 94 and lllOF and reported results in 2 percent moisture content intervals from 12 to 22 percent. Values for 11 percent equilibrium moisture content (EMC) were also given. The following observations were made: 1) the EMC of rough rice at a given relative humidity was higher at lower temperatures and 2) as the moisture content of the grain increased in the range of 11 to 22 percent (dry basis) the effect of temperature on the EMC decreased. Data from Karon and Adams are in good agreement whereas the data from Coleman and Fellows generally show higher moisture content values for a given relative humidity. Houston (1952) and Houston and Kester (1954) did similar work with whole grain edible forms of rice at 25°C (77°F). Brown rice had a slightly higher (0.3 percent aver- age, wet basis) EMC than polished rice and about a 1.6 per- cent higher EMC than rough rice. 2.6 Moisture Adsorption by the Grain The rice grain is a living organism. Rice either adsorbs or desorbs moisture as temperature and relative humidity conditions change. After the grain is dried to an equilibrium moisture content for a specific ambient condition, the grain will adsorb moisture whenever the ambient condition changes to such that the EMC for the grain is higher. For single grains, once dried. the pro- cess of moisture adsorption is equally as common as the process of moisture desorption if the grain on the aver- age is to remain at its original (dried) moisture content. Many studies have been conducted on the drying (desorption of moisture) of rice but only a secondary interest has been exhibited in the wetting (adsorption of moisture) process. One of the earlier observers who manifested an interest in the adsorption process was Stahel (1935) in Surinam. He reported that two Japanese researchers. Kondo and Okamura (1929) discovered that moistening of dry paddy resulted in cracking of the grains. Stahel was unaware of their work but came to the same conclusion in 1932. 30 His Observation was that if paddy was exposed over night or was moistened by any other means. fine cross-wise cracks developed in the grain. The cracks became visible shortly after water adsorption started and remained when the rice finally dried. Before 1930 it was generally accepted that these cracks were due to rapid drying in the sun. hence the name sun cracks. But when paddy was suddenly dried in a thin layer in the sun to below 10 percent moisture, no sun cracks were formed. This was confirmed by putting paddy dried in this way into a sealed flask and no cracks developed. On the other hand a sample dried in this manner before being sub- jected to moist air showed severe cracking quite rapidly. The point at which cracking began when dried grain was remoistened was determined by harvesting 20 to 24 per- cent moisture rice (wet basis) at 10:00 to 11:00 a.m. This rice was put on frames and dried in the sun. During the drying process. portions were removed at regular in— tervals and their moisture contents were determined. Each portion was stored in a closed flask. This process was continued until 7:00 or 8:00 p.m. at which time the samples 31 of different moisture content were put into basins of water for one and one-half hours before being spread on the frames again to dry during the following days. After drying had progressed to below 10 percent mois- ture content (wb), the grains were processed through a Smith Shelling Device and the resultant broken grains were observed. When the initial moisture content was from 15 to 24 percent. remoistening had no influence on the breakage. Graphs with percent whole grain plotted on the ordinate and percent moisture, before wetting. plotted on the abscissa reveal that breakage started to occur be- tween 14 and 15 percent moisture (wb) and the percent of Whole grains after milling dropped rapidly when the pre- wetting moisture content of the grain was below 14 percent. After the work of Stahel the question of rice damage by moisture adsorption remained dormant for nearly two decades. Swanson (1943) conducted studies on the effects of moisture on the physical properties of wheat. He observed the test weights of non-weathered wheat were 4 pounds heavier than those of weathered wheat. The 32 explanation given was that swelling caused by wetting disturbed the internal compact condition and after the water had escaped, vacuoles were left which decreased the specific gravity and hence the test weight of the grain. Research related to moisture adsorption by maca— roni was conducted by Earle and Ceaglske (1949). The modulus of rupture, modulus of elasticity, coefficient of thermal expansion and coefficient of moisture shrink— age of macaroni products were determined. They found that thermal stresses were small compared to moisture stresses. Plastic flow (relaxation) occurred at high mois— ture levels and thus relieved the stresses which accom— panied the formation of a moisture gradient. Towards the end of drying, the moisture gradient set up tensile stresses in the interior and compressive stresses at the outer surface which were sufficient to cause checking. Checks originated at the points of highest tensile stress. Intensive investigations have been made concerning the fissuring in wheat caused by weathering of the ripened grain. Milner, Shellenberger, Lee and Katz (1952) conducted 33 such a study (using radiographic methods) at the Kansas Agricultural Experiment Station. The radio-graphs dis- closed that some seemingly sound samples of grain pro- duced fine radiographic shadows which indicated the ex- istence of cracks or fissures oriented at right angles to the longitudinal axis of the kernel. Other samples did not produce these shadows. Closer investigation re- vealed that samples which exhibited the fissures were weathered. Thus the radiographs offered visual proof of the belief that loss of density experienced by wheat during weathering was caused by the formation of vacuoles in the endosperm of the grain. The radiographic method, originally develOped for the detection of insect infes- tation of grain, was also found to be effective for de- tection of cracked or broken kernels in rough rice prior to milling. Following the work of Stahel (1935). Kik (1951) observed cracks in nonnstained rice grains by placing them on the stage of a stereoscopic microscope and shining a light through them. Observations of the number of cracks 34 were made before staining because the moistening and drying in the staining process caused additional cracks to develop. Henderson (1954) made a brief exploratory study of the causes and characteristics of rice checking. To date this remains the most extensive study on record. His major efforts were directed at checking which re— sulted from the drying process. Fractures originated at the center of the kernel and developed along the minor axis (perpendicular to the longitudinal axis) towards the outside circumference. Caloro rice, a short-grain type. was used. Dew was found to produce only a small increase in checks. Rice in a sealed plastic bag subjected to an increase in temperature from 700 to 1380F for 10 minutes showed an increase in half. full and total checks. Fast drying rates would be expected to cause failure to progress from the outer surface toward the center. according to Henderson. X-rays. however, showed this process to be non—existent. Therefore. checking during fast drying was ascribed to the increase in tem- perature which produced fast drying rather than to the 35 decrease in moisture in the surface portions. Although not conclusive, the evidence indicated that checking re- sulted from a moisture or temperature increase. His con- clusions were that internal faults developed from 1) a fast increase in temperature which resulted from high drying air temperatures or intense sunshine or 2) a fast increase in moisture content such as would be experienced at night under heavy dew. Later work by Henderson (1957) revealed that the checking of rice increases at a faster rate as drying progresses at any given humidity. Grains (Caloro short- grain type) were originally harvested at 24 percent mois— ture and then stored in sealed containers at 38°F until the tests were run. Test samples consisted of approxi- mately 1800 grams of grain dried in l4-l/2—inch square and l-inch high wire mesh containers. Grains were dried to approximately 13.5 percent moisture content. During the removal of the first 4 percent moisture at 1300F, 12 percent of the grains checked while 45 percent checked during the last 4 percent moisture removal. 36 The work by Stahel (1935) was noticed by Breese (1955) who studied the hysteresis in the hygroscopic equilibria of rough rice at 25°C (77°F). Equilibration by adsorption was found to be extremely rapid at relative humidities above 50 percent. More than 96 percent of the eventual water uptake occurred within the first four days. The desorption rate was not so rapid and the equilibrium moisture content for desorption was higher. The difference between adsorption and desorption equilibrium moisture con- tents was 1.5 percent from 50 to 70 percent RH and generally over 1 percent in the span between 20 and 80 percent RH. The most recent observations of crack damage from moisture adsorption were made by Desikachar and Subrahman- yan (1961). Their work concerned the formation of cracks in rice (soft medium and hard fine grain) during wetting and its effects on the cooking Characteristics of the cereal. Observations were made by dropping milled raw or parboiled rice into water at different temperatures. The formation of cracks was accelerated by increasing temperatures in the case of parboiled rice. A retarding effect was noticed when water above 70°C (1580F) was used 37 with raw rice. In raw rice at high temperatures (900C or 194°F) the few cracks became cemented within a short period because of the rapid gelatinization of the starch. When the rice was put into boiling water, the formation of cracks was minimized and a longer time was required for cooking. Only 15 minutes was necessary to develop cracks to a maximum extent in milled raw rice whereas 45 minutes were required for the same occurrence in parboiled rice. Adsorption of the soak water made the grain opaque and consequently suggested a method for studying the mode and extent of water penetration into the grain. A dark line first developed near the germ tip and along the dor- sal line of fusion. These spread inward and then after a periodthe transverse cracks developed as dark lines. The occurrence of cracks during moisture adsorption is little understood. Research needs to be done on the fundamental aspects which cause the cracks to develop. Desikachar and Subrahmanyan concluded. Similar work has been done with wheat in recent years. Naumow (1959) used water with 0.025 percent fuch- sin (a bluish red dye) for immersing the wheat. After 38 specified timb periods, grains were removed and sectioned l . . to determine moisture penetrations. Results showed that waterj(25°C or 77°F) entrance through the seed coat into the endosperm began in the vicinity of the germ. then at the middle of the kernel and finally at the brush end. The time for water to penetrate throughout the endosperm was three to seven hours. When kernels were heated (38 to 40°C or 100 to 104°F) and immersed into water at room temperature, the moisture penetration proceeded nearly twice as fast. When kernels were cooled (4 to 6°C or 39 to 43°F) before immersion. the moisture penetration was only one-third to one-fourth as fast. Grosh and Milner (1959) did some excellent work relative to water penetration and internal cracking in tempered wheat grains. They found that transverse and radial cracks could be noted in grains within 30 minutes after wetting. After eight hours or longer these cracks disappeared because swelling of the endosperm decreased the intercrack spaces to dimensions below the resolving power of the X—rays. In some cases a proliferation of secondary cracks arising from the radial and transverse 39 primary cracks was noted. These tended to disappear quite readily as the grains continued to adsorb moisture. Fur- ther work is needed on the mechanical properties of the wheat endosperm and the mechanisms which cause cracking. according to these researchers. III THE INVESTIGATION One of the first problems encountered and perhaps the first decision to be made was that of working with in- dividual grains or with a mass of grains. There are prob— lems which are difficult to solve in either case. Experi- enced researchers had no strong recommendations for proceed— ing in either direction. Individuals who had worked with grain masses suggested the individual grain approach. Those who had previously encountered the problems of working with the individual grains suggested the mass approach. Both possibilities were considered and the individual grain approach was selected on the basis that less work had been done in this area. The feasibility of working with rough. brown. parboiled or milled rice was also deliberated. Brown rice appeared to be the most suitable and hence was se— lected. Therefore. any reference to the rice grain in this research refers to brown rice unless the grain is specified to be otherwise. 40 41 3.1 Preliminary Research Before doing extensive work with any engineering material. a person must devise some means of holding or handling it. i.e.. before tensile tests can be run. a means must be devised by which the material can be suit- ably held. The first objective in this research was to find some satisfactory means of holding the grain. A long-grain type was selected since the grain shape appeared more amenable to the solution of the problem. Numerous ways of holding the grain were tried. One of these was to remove a section of plastic insulation from a No. 14 electrical conductor. cut the plastic tubing into 1/10- inch lengths. inject adhesive into the ring core and then insert. the ends of the rice grain into these-rings. While individual grains were being cemented into the insulation rings. the observation was made that the rice grains tended to fissure perpendicular to their longi- tudinal axis as a result of the cementing process. This cracking was in itself a very basic physical character- istic of the grain which had not been observed before in this manner. Consequently it was a phenomenon which 42 suggested further investigation. The observation was pursued until the conclusion was reached that moisture adsorption by the grain from the adhesive had caused the grain to fissure. The observation was further veri- fied by the following: 1. Cementing a kernel to the end of a thread and lowering it into a beaker with water until the grain end touched the water surface. 2. Filing off either end of a rice grain to give a flat surface and then setting the grain upright on a wet fine-textured sponge or some other moist surface. 3. Dropping grains (brown or rough rice) into water and making periodic observations of them. In all three cases the grains could be made to fissure readily. Some grains fissured from the droplet of adhesive (water soluble) used to attach the thread to the grain end. Those which did not fissure from this pro- cess fissured readily when put into contact with the water 43 surface. The significance of the above observations was that they established a cause and effect relationship within a measurable time interval. Stated otherwise. the three parameters involved in the phenomenon were 1) moisture adsorption. 2) grain damage and 3) time. These observations caused the comments by Breese (1955) relative to Stahel's work to be reinvestigated and prompted the concentrated search of the literature relative to mois- ture adsorption. The research by Stahel (1935) and that of Desikachar and Subrahmanyan (1961) presented factual evidence that relatively dry rice grains would crack if they were made to adsorb moisture by submerging them in water for a period of time. Stahel observed that paddy exposed over night also developed fissures and thereby inferred that moisture adsorp- tion by any means would cause rice grains to crack. Hender- son (1954) commented that his evidence was not conclusive but his data indicated that checking resulted from a mois- ture or a temperature increase in the rice grain. The information of the foregoing researchers and this author's observation that fissures developed in the 44 grain from the cementing process. together precluded the possibility of doing extensive research concerning the mechanical and rheological properties of rice at differ- ent temperatures and moisture contents without first making an intensive investigation of the physical char— acteristics of the grain in relation to different envi- ronmental conditions. The preliminary investigations of this research develop some very fundamental information in this respect and form the basis for development of the final thesis problem. The terminology in the literature is mixed and indefinite when reference is made to rice damage due to separation or partial separation of the endosperm of an individual grain. Terms which generally have a synonymous meaning are fissures. faults. checks. cracks. splits. vacuoles and partial fractures. These imply that a par- tial separation has occurred between two portions of the endosperm. The separation is not complete. however, and the grain remains a single unit. The terms 1) breaks. 2) brokens and 3) complete fractures generally refer to grains in which the endosperm is separated into two parts. 45 In the case of brown rice, where the bran layers tend to hold the endosperm together whether it is broken com— pletely or not. the distinction between the two is diffi- cult tO make. Since this research is concerned primarily with brown rice and since the grain is damaged and weakened when only a small fissure occurs. no major distinction will be made between the two modes Of damage. The first (fis— sures) is simply a degree of the second (breaks). Both are objectionable and will lower head rice yield when the grain is milled. Brown rice is the product which remains after the rice hulls are removed. This operation was performed manually for this research. After hulling. the grains were inspected by placing them into a petri dish which was held over a light. The next Observation of signifi— cance was that the grains did not have to be in direct contact with water to produce the fissuring phenomenon. Grains left exposed in an Open container inside the Agri- cultural Engineering Building for a day or more would occasionally fissure. Temperature conditions were rather constant but humidity conditions were subject to change in the building. 46 Rough (unhulled) rice was subjected to the same exposure treatments. Cracks across the unhulled grains were someWhat less frequent but otherwise developed just as had been observed in the hulled grains. A pen flash- light was modified to supply a concentrated beam (1/10- inch diameter) of light under the petri dish for inspect- ing the grains. The narrow light beam could be concen- trated on any part of a grain and could also be moved from one end Of the grain to the other. Some unhulled grains were placed into water and were permitted to pass through the cracking and swelling cycle before the grains were dried again. Cracks which resulted in nearly complete fractures from the moisture adsorption process reappeared after drying. Small cracks did not reappear to the extent that they could be observed. Implications of the foregoing work were as follows: 1. The cells on the surface of the rice grain expanded when the kernel adsorbed moisture and thereby caused the fissures to develop in the grain. 2. The fissures developed under essentially constant temperature conditions. 47 3.1.1 Wooden humidity containers The foregoing experiments suggested the creation of specific humidity conditions within a refrigerated space. Hence a 68°F walk-in chamber was securedl Satur— ated solutions of different salts were used to produce various relative humidities at the given temperature. The text by Hall (1957) was consulted for this informa- tion. Numerous experiments were conducted using enclosed wooden containers as humidity compartments. A small elec- tric motor mounted on the outside turned a fan on the in— side, thus circulating the air over an enameled pan con- taining the saturated salt solution. Wooden frames covered with plastic screen served as trays on which the grain samples were set. Some of the problems encountered were 1) the com— partments were never completely air or vapor tight. 2) the exposed wood inside the compartments had to be brought to equilibrium with the humidity condition before the de- sired humidity could be achieved and 3) the same compart- ment could not be used for one humidity one day and for 48 another the next day. A hygrothermograph set inside one of the compartments showed that four or more hours were necessary to achieve the desired humidity after the com— partment had been opened momentarily. The time interval required for developing the desired humidity in the enclosed compartments also mani- fested itself in other ways. Rice grains taken from ambient conditions and placed into 89.2 percent RH at 68°F for three hours adsorbed more moisture than did other similar grains exposed to a similar saturated salt solution atmosphere for four hours but where the grains were removed for inspection approximately four minutes every hour. Grains in the saturated salt solution at- mosphere for three successive hours (compartment kept closed) fissured. Grains receiving the hourly inspection never did crack. General observations from this work (68°F) with the wooden compartments are listed below. 1. In every case when grains (Rexoro) were moved from a 33.6 percent RH atmosphere into an 89.2 percent RH atmosphere, fissur- ing occurred. Moisture was being adsorbed by the grains. 49 In no case did fissuring occur when grains were moved from 75.5 percent or 89.2 per— cent RH atmospheres into a 33.6 percent RH atmosphere. Moisture was being desorbed by the grains. For moves from 33.6 percent RH into 89.2 percent RH atmospheres. five cracks per grain was the mode. In the foregoing work. the surface area of the grains was very small compared to the surface area of the salt solution. This relationship was desirable since too many grains exposed in the enclosed atmosphere could have adsorbed moisture faster than the saturated salt so- lution could have produced it and thereby could have fur- ther extended the time interval required to produce the desired relative humidity. The foregoing procedure was unsatisfactory for the following reasons. 1. Grains could not be subjected to sudden humidity changes because a period of several hours was required for the desired relative humidity to develop. 50 2. No conclusive statement could be made concern- ing the exact conditions which caused the grains to fissure after they were subjected to a suf- ficiently large humidity change. There was no simple. reliable and relatively economical method Of maintaining a continuous record of conditions within the relative humidity chamber. 3. 1Compartments were too large to be conveniently handled in a small controlled atmosphere room. 4. Relative humidity could not be accurately con- trolled in an enclosed chamber. Even with the salt solution removed. the closed chamber would tend to develOp the humidity to which it had been previously subjected because the wooden surfaces had come to equilibrium under that condition. The research thus far demonstrated conclusively that relatively dry rice grains fissured when they were exposed to high relative humidities for an extended period of time. The problem remaining was to determine exactly the conditions which would cause grain fissures to deveIOp. 51 3.1.2 Plastic humidityicontainers Salts were selected whose saturated solutions would produce relative humidities as shown in Figure 1. Saturated solutions were made and placed into l/2-gallon. heavy duty. plastic freezer containers (Fig. 2) which were vapor tight. Grille sectiOns. 4-1/2 x 4-1/2-inch square with 12 compartments. from fluorescent light fix- tures were used to construct trays into which different varieties and ages of rice grains could be placed. A plastic screen was cemented over the openings on one side. With a little machining of the grille-section edges, the trays (1 inch deep) were made to fit into the wedge—shaped freezer containers with 1 inch Of clear- ance above the top of the tray. A glass plate was cemented to another 4-1/2 x 4-1/2-inch square grille section to pro- duce a tray for inspecting the grains. The plastic freezer containers and the inspection equipment are illustrated in Figure 3. The small metal trough (A) was used to transfer rice grains from one con- tainer to another. A special inversion tray (B) with a screen bottom and corner clips was used as an intermediate 52 100- "-'-'-' Extrapolated data 90" p ' h t ) _____ RECEO _(_otaSSium c roma e 80L NaCl (Sodium chloride) 70— NaN II I "‘ "‘ 4J§°93W 4..) ;:60- .__“‘ M 8 I‘- (No ) ~6H o ‘50— f.” , _____ __ r . ,§ K2903 chtassium carbonate) E40- . s g M Cl 3 ‘r - - — —— Wm chloride) 330- . 3 g ------ 322§392._5£°E§§§ium acetate) 20— ‘\ 1 ______ 1.19:1. rheumatism O- 30 40 50 60 70 80 90 100 Temperature. oF Fig. l.--The relative humidities which the respective satur- ated salt solutions will produce in enclosed con- tainers at the indicated temperatures: 53 .3 :i u}. 'i i g. - , “.44“: . Fig. 2.--A plastic freezer container (left) with saturated salt solution and grain tray. A top View (right) of freezer container with cover removed. (Photo neg. 641367-1, 641367-2) a”... p 1 Fig. 3.—-One of the controlled atmosphere chambers with the saturated salt solutions and equipment which was used for inspecting the grains. (Photo neg. 641367-4) 54 between the conditioning and inspection trays. Thus with two inversions the grains assumed the same posi- tion in the inspection tray as they initially had in the conditioning tray. A hygrothermograph (C) was used periodically to Obtain continuous records of temperature and relative humidity. The 2—dram bottles in which the rice grains were kept after hulling and before testing are shown on shelf "D". The NO. 6 dry cell batteries (E) were used with the pen flashlight since this light was used for hours at a time. The plastic containers (F) are shown as they were stored on shelves in the condi- tioned chamber. Additional conditioning trays are stacked behind "G". The stand for inspecting the rice grains is located in the center of the picture. Although not perfect this procedure and equipment had the following advantages. 1. The freezer containers gave compact relative humidity chambers whose walls were non-adsorbent and vapor tight. 2. A small amount of saturated solution and hence only a small amount of salt was necessary for each container. 3. 55 The systems were inexpensive and convenient to Operate. Some undesirable features which remained were: 1. The saturated solution required an unde- termined time interval to develop the desired humidity after the container was closed.) The grains could not be observed While in the humidity chambers, hence the time required for the initial fissures to occur could not be determined. The characteristics Of the system made frequent inspection of the grains impossible without defeating the purpose Of the system. The grains were usually subjected to a lower relative humidity and hence lost moisture whenever removed for inspection. The grains could not be accurately weighed to detect weight changes over certain inter- vals Of time. 56 Numerous experiments were run with the plastic freezer container humidity chambers. A typical diagram- matb summary of only one of these is shown in Figure 4. From these experiments the Observation was made that rice at equilibrium with 68°F and 50 percent RH can be exposed generally to 10 percent RH increases at 24- hour intervals without causing fissuring of the grain. Some varieties quite subject to checking may require in- crements as small as 8 percent RH to prevent damage. Responses by the grain to a given increment of humidity Change are not the same at different humidity levels. For example a 22 percent RH change between 11 and 33 percent produced some damage whereas a 21 percent change between 65 and 86 percent RH produced no damage. Surface cracks and internal fissures were observed in the treated grains. Surface cracks. which seldom occurred, develOped in the desorption process. These cracks originated at the surface and extended some dis- tance into the grain but generally not through it. The cracks were usually very small and seldom appeared tO extend over more than one-eighth the cross-sectional area Diagrammatic Summary 30 grains ea. -- §varieties -- 2 ages 1 day at each - 43.9J 54.9. 65.3 apd 75.5%.RH 1 week - 86.6% RH, 68°F 10 grains 6 varieties 10 grains 6 varieties 10 grains 6 varieties 23.2% RH. 68°F 43.9% RH. 68°F 65.3% RH, 1 week 1 week 1 week 0 86.6% RH, 68 F 4 days Shattered Shattered 43.9% RH. end end 5 days phase I phase II 65.3% RH. 3 days Note: This experiment was 43.9% RH. planned to run through the 4 days initial adsorption. desorp- tion and adsorption cycle. 75.5% RH, The anticipated cracking 2 days occurred in two phases of the experiment but not in 43.9% RH. the third. Hence the third 5 days phase was continued in a sOmewhat random manner. 75.5% RH, 2 days 43.6%.RH. 5 days 75.5% RIII 2 days 68°F 68 F 68 F 68 F 68 F 68 F 68 F 84 F 84 F end phase III Fig.4.--Effects of moisture adsorption by brown various magnitudes Of relative humidity Notes Adsorbing Equilibrating Desorbing - no crack damage observed Adsorbing - crack damage observed Equilibrating Adsorbing - practically no damage Desorbing — damage Adsorbing - some damage Desorbing - no damage Little addi- tional damage beyond first cycle Desorbing Adsorbing — more damage recorded rice from change. 58 of the grain. Their most common occurrence was at points where the bran layers had been damaged or removed from the grain. This implied that the grain dried more rapidly at these exposed areas and thereby set up tensile stresses on this surface which caused the grain cells to separate. These surface cracks were seldom if ever Observed in a perfect grain Of brown rice. The internal (usually cross—sectional) fissures appeared to start at the center of the grain and then de- veloped toward1ie outside perimeter. These were adsorp- tion cracks since their occurrence was caused by moisture adsorption. Where numerous fissures develOped in a grain. a single longitudinal crack often occurred. This crack was usually erratic because it intersected the fissures across the grain which normally developed first. Grains which successfully resisted damage through several adsorption processes between 11.2 and 33.6 percent RH at 68°F had a slight tendency to crack When they were cycled through thesame relative humidities at 84°F. This indicated that a greater vapor pressure difference was influential in causing grains to fissure. 59 As a consequence. experiments were developed which investigated the particular effects of vapor pres- sure changes. The concepts Of vapor pressure will be elaborated upon in the theoretical considerations and discussion sections of this thesis. By varying temper- atures and relative humidities. vapor pressure differences of various magnitudes were developed. The procedure for one of these experiments is diagrammatically illustrated in Figure 5. Relative hu- midities are plotted on the right hand ordinate and vapor pressures are plotted on the left hand ordinate. Temperatures are indicated above the relative humidity lines. A vertical line may be drawn on any part of the diagram to determine the atmospheric conditions at that time. The assumption was made that the rice grains reached equilibrium conditions within the 5-day intervals between which moves were made. This is not necessarily true but is probably quite close since Breese (1955) re- ported more than 96 percent Of the eventual water uptake to occur during the first four days during which rough rice was exposed to relative humidities above 50 percent. 60 T I I | T" | (6 grain varieties, 2 ages, 10 grains of each) 3.838;... .6838... | i I— I .80... I | I I -80 ' I | I [_98° .70_ I I I _.70 I I I I IEM EM? 17% 6C I I I I IT dif=0 - __ l | I IIRH dif= .60 I I II + 3'].- 7% I I IVP dif= I I I I : + .284 '6' 'SC" I I l P55- —50 . I . I I. 3‘ _.-§-8.°.F. L.- 6.8. °8_cI L.-8-8 °F_cI : I 3.8m- mc=11%0 ~40 g EMC= —Equilibrium moisture content: ND m Of grain. dry basis :T dif = + 30°F 3 - ————— Relative humidity :RH dif = _ 43 2% %-3C_n =Vapor pressureEMC=2°%O vp dif = + 095 _-30 > D = Grain damage occurred D psi ND = No damage occurred: T dif = 0 I RH dif = + 42. 7% .ZCP :V Pdi f = + .144 psi .120 I I ”ND °T dif = + 30°F .lC , RH dif = - 42.7% _ 10- I I 5T dié = VP dif = + .054 psi = 0 RH dif = + 42.7% VP dif = + .046 psi L I l 1 L_ J 0 0 5 10 15 20 25 30 Time, days Fig. 5.—-Effect of moisture adsorption from various magni- tudes of temperature and relative humidity change. Relative humidity. percent 61 The vapor pressure differences are those for the different atmospheric conditions and not necessarily those which existed between the grain and the indicated atmospheric condition. The equilibrium moiSture contents which are shown were developed during a later phase of this research. The letters "D" (damage) and "ND" (no damage) indicate if the grains did or did not fissure during the interval of exposure. Grain moves were made along the vertical broken lines. The temperature. relative humidity and vapor pres- sure differences listed are those Which existed between the different atmospheres indicated by the circular points on the graph. The alternating "D" below the vapor pressure lines indicates that additional damage developed as the grains were subjected to greater vapor pressure changes which caused moisture adsorption by the grain. This can be readily observed from the increased equilibrium mois- ture contents for the conditions under which the fissur- ing occurred. Vapor pressure and temperature increases which together caused desorption did not produce fissur- ing of the grain. 62 The chart (Fig. 5) illustrates that atmospheric vapor pressure increases occurred with every grain move. temperature increases occurred with every other move and the relative humidity alternated between 43.7 i .3 and 86.6 i .1 percent except for the final move which was only up to 75.1 percent RH. The small plus and minus variations in relative humidities result from the fact that a saturated salt solution in an enclosed volume will not necessarily produce the same relative humidity over a wide range of temperatures. (Fig. l). The final move to 75.1 percent RH caused a vapor pressure change nearly twice as great as any previously attempted in the experi- ment. ' Experiments performed with the plastic containers indicated that fissures would develOp at 84°F from the increment Of vapor pressure increase between 22.3 and 32.9 percent RH. Only a few grains develOped cracks. Additional damage occurred from the increment of vapor pressure change between 20.4 and 32.3 percent RH at 100°F and also from the increment of vapor pressure change between 19.1 and 31.8 percent RH at 111°F. The fissures usually developed within 63 48 hours after the grains were exposed to the atmosphere which caused the moisture adsorption. The equilibrium moisture content of the grain appeared to be an influential parameter in producing grain damage. Therefore two experiments were developed with the Objective of keeping the grain moisture content constant while moving the grain from one temperature. vapor pressure and relative humidity to another temper- ature. vapor pressure and relative humidity. One of these experiments (Fig. 6) shows the atmospheric condi- tions through which the grains were moved. These moves are further illustrated with a psychrometric chart (Fig. 7). The grains were initially subjected to a slow mois— ture adsorption process at 68°F until a moisture content of approximately 20 percent was reached. Thereafter the grains were subjected to the conditions Of particular interest. No damage was produced in cycling the grains be- tween the 38°F. 86.7 percent and the 100°F. 85.6 percent RH conditions. Brown rice at equilibrium with these con- ditions has a moisture content of approximately 20 per- cent (db). The atmospheric vapor pressure changed from 64 Experiment I. Outline (6 varieties. 2 ages. 10 grains each) Notes 68°F 43.9% RH 1 day Assumed ambient equili— brium of grain 0 Each 68 F 54.9. 65.3. 1 day Slowly adsorbing mois- 75.5% RH ture - no damage 68°F 86.6% RH 1 day NO damage 0 0 Cycle 38oF 86.7% RH 5 days NO damage EMC approxi- l 100 F 85.6% RH 5 days NO damage mately constant Cycle 1 was repeated 3 times with no damage to grains. (End planned part of Experiment -— Random moves follow.) 38°F 100% RH 1 day NO damage 0 0 38°F 11.2% RH 6 days No damage EMC approxi— 100 F 32.3%.RH 2 days NO damage mately constant 38°F 11.2% RH 5 days 100°F 43.4% RH 5 days Developed 11 cracks in 120 seeds 38°F 11.2% RH 5 days No damage 100°F 49.9% RH 6 days DevelOped 2 more cracks. expected more 38°F 11.2% RH 5 days NO damage 100°F 61.9% RH 1 day Developed 77 more cracks Fig. 6.—-Effect of cycling high moisture grain between 38°F, 86.7 percent RH and 100 F. 85.6 percent RH. Grain moisture content essentially remained constant. 65 40 45 50 55 60 65 70 75 80 85 90 95 100 Temperature °F (dry bulb) Fig. 7.--Rice grains were moved along the path indicated. No fissures developed while grains were kept at a constant equilibrium moisture content. 66 0.097 to 0.810 psi or nearly seven and one—half times the initial vapor pressure. A temperature change of 62°F was involved while the relative humidity remained essentially the same. Brown rice at equilibrium at 38°F and 11.2 per- cent RH has a moisture content of approximately 9 per- cent. The initial move from the 38°F, 11.2 percent RH position to the 100°F, 32.3 percent RH position maintained the grain at about the same equilibrium moisture content even though the relative humidity changed more than 20 percent. No damage developed. Succeeding moves were made to atmospheric conditions for which the equilibrium moisture content of the grain was higher. Figure 6 indi- cates the amount of kernel damage which resulted. In the latter part of Experiment I (Fig. 6). the 38°F. 11.2 percent RH condition was used as a base from which the grains were moved to higher temperatures. rela— tive humidities and consequently also ambient vapor pres- sures. The second experiment (Fig. 8) was started in a similar manner. Grains were cycled between 38°F. 34.8 percent RH and 100°F and 49.9 percent RH. .NO damage 67 Experiment II, Outline (6 varieties. 2 ages, 10 grains each) Notes 68°F 43.9% RH 1 day Assumed ambient equili- brium Of grain 68°F 54.9, 65.3, Each Slowly adsorbing mois— 75.5% RH 1 day ture - no damage 68°F 86.6% RH 2 days NO damage 38°F 34.8% RH 5 days Desorbing - no Cycle 0 damage EMC approx- 1 100 F 49.9% RH 5 days No damage imately constant Cycle 1 was repeated 3 times with no damage to grains. (End planned part of experiment -- Random moves follow.) 38°F 44.0% RH 2 days NO damage 100°F 49.9% RH 6 days No damage 38°F 59.6% RH 2 days No damage 100°F 49.9% RH 5 days No damage 38°F 68.5% RH 2 days Developed 60 cracks 100°F 49.9% RH 5 days Desorbing — no damage 38°F 75.0% RH 4 days Developed 12 more cracks 100°F 49.9% RH 5 days Desorbing - no damage 38°F 86.7% RH 1 day Developed 105 more cracks Fig. 8.——Effect Of cycling low mgisture grain between 38°F. 34.8 percent RH and 100 F. 49.9 percent RH. Grain moisture content essentially remained constant. 68 resulted even though the relative humidity was changed by 15.1 percent. The 100°F. 49.9 percent RH condition was then used as a base from which the grains were cycled to increasingly higher relative humidities at 38°F. Re- sulting grain damage is indicated in Figure 8. Some Of the fundamental Observations made in the preliminary research are summarized as follows: 1. Rice grains tended to fissure perpendicular to the longitudinal axis from the process Of moisture adsorption resulting from: A. Cementing the grain into a plastic ring. B. Placing the grain into contact with a moist surface. C. Subjecting relatively dry grains to a humid atmosphere. 2. Rough rice grains fissured less frequently but similarly to brown rice when they were subjected to a humid atmosphere. 3. Little or no grain fissuring was produced in relatively high moisture content grains (whole. unscarred) subjected to the desorption process in low humidity atmospheres. 69 Saturated salt solution in an enclosed con- tainer requires several hours (about four) to produce the equilibrium humidity after the container has been Opened momentarily. Rice at equilibrium with 22.3 percent RH and 84°F develOped fissures when subjected to 32.9 percent RH at the same temperature. More damage developed when the grains were subjected to essentially the same relative humidity Changes at higher temperatures (100°F and 111°F). Relatively dry rice (9 percent moisture, dry basis) at 68°F may be subjected to humidity increases ranging from 8 to 10 percent with- out causing fissuring of the grain. Rice subjected to temperature, relative humid- ity and vapor pressure changes within the range of ambient conditions will not fissure as long as the equilibrium moisture content Of the grain remains the same. 70 3.2 Selection of Rice Varieties and Types Rice varieties generally are classified into long. medium and short grain types. Some researchers attempt to draw a finer distinction and use the classifications of short. medium, long, long slender and medium slender grain types. These different classifications imply that all rice grains do not have the same physical dimensions and as a consequence the reactions of the grains can be expected to be different when they are subjected to vary- ing atmospheric conditions or to different mechanical processing Operations. The grains are not only differ- ent in physical dimensions but are also reported to have different chemical properties. In their studies on the physicochendcal properties of rice. Bienvenido, et a1. (1964) found that Indica varieties varied more in physi- cal properties and chemical composition than did the Ja- ponica varieties. Other researchers have made similar observations. As a consequence research must be con- ducted on numerous varieties to determine the character- istics of the grain in general. 71 For this research. 12 varieties (6 of which were used) of rough rice were secured from the Rice Pasture Experiment Station at Beaumont. Texas. Samples were Ob- tained soon after the harvest which usually occurred in September or October. Grains were secured from the 1961 and 1962 growing seasons. The exact history of the rice prior to its arrival at Michigan State University is un- known. For shipment from Beaumont. the grains were first placed into seed envelopes before they were securely packed into a cardboard container. After arrival at Michigan State University the grains were kept in the shipping container but were placed into a conditioned air chamber where the temperature was maintained at 68°F and the rela— tive humidity variation was generally limited between 40 and 70 percent. Humidity changes within the chamber occurred at a slower rate than under ambient conditions. Since the grains remained well packed. the relative humid- ity changes tO which they were subjected were of a smaller span and occurred at an even slower rate. The physical dimensions. as reported by Kramer (1951b), of five of the six selected grain varieties are listed below. 72 Variety Type Length, Width. Thickness. inches inches inches Zenith medium 0.244 0.099 0.069 Fortuna long 0.293 0.092 0.071 Bluebonnet 50 long 0.298 0.086 0.067 Rexoro long slender 0.285 0.080 0.064 Century Patna 231 long slender 0.285 0.081 0.065 Belle Patna ____________________ Many authors and researchers do not distinguish between long and long slender grain types. The distinc- tion may or may not be justified. The above data show a distinct difference between Bluebonnet and the long slen- der grain varieties. There is. however. nearly an equal amount of difference between the Fortuna and the Blue- bonnet varieties which are both in the long grain classi— fication. The Zenith variety (medium grain) has a thick— ness approximately the same as the other grain varieties but is otherwise shorter and somewhat wider. All rough rice was processed through an air sep- aration device which removed empty hulls. immature and light grains. In most samples less than 5 percent Of 73 the grains were removed. The 1962 samples of Rexoro and Fortuna were of poor quality. In many cases the hulls were nearly empty and in others there was only a poorly develOped grain. Nearly half Of the Rexoro grains and possibly a third (by volume) Of the Fortuna grains were removed in this manner. The 1962 grains for all varieties in general were smaller and lighter in weight than the 1961 grains. All rice varieties. as received. contained some cracked grains. These could not be used and were dis— Icarded. The Zenith and Fortuna varieties contained more cracked grains initially than were present in the other varieties. 3.3 Selection and Development of Equipment The preliminary research produced much fundamen- tal information but even the plastic container system had the undesirable features listed on page 55. Desirable measurements for determining rates of moisture adsorption. rates of grain damage. rates at which fissures developed and periods over which fissures develOped could not be 74 made. These measurements would yield fundamental infor- mation for a basic insight into the problem. Lord Kelvin once said, "When you can measure what you are speaking about. and can express it in numbers. you know something about it; and when you cannot measure it. when you cannot express it in numbers, your knowledge is Of a meager and unsatisfactory kind; it may be the beginning of knowledge. but you have scarcely in your thoughts advanced to the stage of a science." The preliminary research was crude in many re- spects and needed much refinement in order to confirm any theories developed from the Observations which were made. Barkas (1948) in his book, The Swelling of Wood Under Stress. summarizes these sentiments in the follow- ing words. ”It is safe to say that a theory only appears to be correct so long as the crudity of our experimental technique is sufficient to mask its short comings." With the advice and the precepts of these eminent researchers. this writer set out to produce and refine an experimental technique which would permit the desirable measurements to be made with sufficient accuracy to make the measure- ments meaningful. 75 An ideal system which would yield the desired basic information relative to moisture adsorption by the rice grains would have the following features: 1. Would instantly produce the specified rela- tive humidity and would thereby permit the grains to be subjected to sharp humidity changes. 2. Would permit frequent or continuous obser— vations Of the grains while they are exposed to the different atmospheres in order to: A. Study the nature and development of the fissures in the grain. B. Determine the rate at which the grains fissure. C. Determine the rate at which the fissures develop in the grains. 3. Would permit the grains to be weighed to de- termine the moisture adsorbed without remov- ing the grains from the system. Each of these features was unique in that the problems were not commonly encountered in other research 76 techniques and procedures. The problem of instantly developing the desired humidity was given considerable study. Basically the humidity condition is produced at the surface where the saturated salt solution is in contact with the air. Therefore producing a greater solution surface area for the given volume of air would have been one approach while circulating the limited volume Of air over the saturated solution surface would have been another approach. The latter was used in the wooden humidity compartments described earlier and was unsat- isfactory. A greater solution surface area could have been produced by: l. Vibrating the container with the saturated salt solution. 2. Circulating the saturated solution through a pump and spraying it back into the con- tainer. 3. Vigorously agitating the solution. 77 All three of the listed methods would have caused a greater saturated solution surface to contact the air but all the systems also would have presented problems in keeping water drOplets from becoming suspended in the air and eventually contacting the grains in the enclosed container. Other problems would have been caused by the mechanical mechanisms involved. Generally all the systems would have been too large. cumbersome and inflexible for the grains and facilities available for this research.. 3.3.1 Dynamic system for relative humidity A review of literature was attempted concerning saturated salt solutions but little more information was found than could be secured from the text by Hall (1957). Coleman and Fellows (1925) referred to a procedure devel— oped by Wilson (1921) who Obtained air of a desired rela- tive humidity by means of sulfuric acid solutions. Wil- son's method was basically one in which the air was bubbled through several bottles (in series) Of acid solution before it was passed through a filter and then through the material to be equilibrated. He found that equilibrium for many 78 materials could be reached within 18 to 96 hours which was much faster than could be Obtained by the method of exposing the sample over a sulfuric acid solution Where the rate of approaching equilibrium was quite slow due to the slow rate Of moisture diffusion in still air. When the rate of flow was more than about 100 cc per minute. a special form of bubbler (to give good contact between the liquid and gas) was recommended. Otherwise broken glass or beads could also be used in the acid bottles for distorting and breaking up the bubbles. With such a system, substantial equilibrium was reached by Wilson with only two bottles when air flows up to two liters per minute were used. In case two to ten liters were desired. three bottles were recommended. The size Of bottles varied from 500 cc for the slow rates of flow and up to two liters for the higher rates. Bottles were filled only one-half full of acid solution. A precaution pointed out by Wilson was to use a tube of tightly packed glass wool or similar material to remove entrained particles Of sulfuric acid from the air stream or otherwise these would be deposited on the material 79 to be equilibrated. Appreciable amounts of acid solution can be carried over unless this precaution is Observed. The foregoing reference supplied the basic infor— mation forckweloping a saturated salt solution system for producing a specified relative humidity in an air-flow system. Glass flasks (500 cc) were secured and fitted with two—hole NO. 8 rubber stoppers. The flasks were approximately seven inches tall and three inches wide (outside diameter). Flint glass tubes (7 mm outside diameter) were inserted into the stOpper. One section was eight inches long and the other was two inches. All the ends on the cut glass tubing were fired to eliminate sharp edges. The short tube section was inserted just far enough to be flush with the stopper on its insert-end. One end of the 8—inch tube was fired until only a capil- lary Opening remained. This tube was inserted into the stopper until the capillary Opening was only 0.25 inch from the bottom of the flask when the stOpper was securely seated. The capillary opening was used to break up the air stream into small bubbles. Other devices were tried 80 but none served the purpose any better. Five flasks (in series) were prepared in this manner. They were connected with rubber tubes and mounted in a rack (E) as shown in Figure 9. A sixth flask with water was generally used ahead of the series. The systems seemed to operate more successfully if the entering air con- tained some water vapor. Saturated salt solutions were prepared and the flasks were filled until 1.25 or 1.50 inches Of clearance remained between the top Of the solution and the bottom of the stopper. This caused the 8—inch glass tube to be submerged a minimum of 4.25 inches. With five flasks in series, the air was bubbled through approximately 21 inches of saturated salt solution before leaving the system. The rate of air flow ranged between 250 and 400 cc per minute and provided a brisk continuous stream of bubbles passing through the system. Glass tubes which tended to produce large bubbles (capillary Openings too large) were removed and replaced with tubes having suf- ficiently small capillary holes to give the desired small bubble size. A similar system was built for each salt solution which was used. 81 Fig. 9.--A single complete dynamic system (E) for con— ditioning air to a desired humidity before it passed through the observation and inspection chamber (F) in which the rice grains were studied. (Photo neg. 641367—5) Fig. lO.--The observation and inspection chamber with brown rice grains in position. Conditioned air entered through the tube at the top of the picture and was exhausted at the Opposite end of the chamber. (Photo neg. 641367—3) 82 In some cases the systems occasionally would tend to clog. Salt crystals seemed to form at the inside en- trance of the capillary opening thus stopping the air flow. Two remedies generally solved the problem. One was to remove the rubber connecting tubes from the tops of the long glass tubes. This Operation was started with) the last flask in the series since otherwise the back pressure in succeeding flasks forced the saturated solu- tion up and out through the disconnected glass tube. The long tube sections above the saturated solutions were then filled with distilled water. When the system was restarted. the distilled water tended to rinse or flush out the salt solutions and thus prevented salt crystals from forming in the tube which could plug the capillary hole. The second remedy was that of increasing the air pressure. The systems Operated with a minimum of 3 psi pressure. Clogging was less likely when pressures Of 4 psi were used. The regulation was also better and less sensitive at the higher pressure when several systems (4 or more) were Operated simultaneously. The systems 83 were tested up to 6 psi before the stoppers were forced from their seats or the rubber tubes started to pull Off the glass tube ends. Rubber bands were wrapped around the rubber tube ends to provide a secure fit that was vapor tight and would not pull Off. Water was used as a lubricant in securely seating the rubber stoppers. The series of bottled saturated salt solutions operated very well and satisfied the objective Of devis- ing a system which would produce the desired humidity instantaneously. Grains which had equilibrated in a plastic freezer container (static system) with a satur- ated salt solution neither lost nor gained weight when they were subjected to the atmosphere produced by the series of bottled saturated salt solutions (dynamic sys— tem). Maintaining the saturated solutions in the dynamic system was no problem since salt crystals were visible at the bottoms of the flasks as long as the solutions remained saturated. A single complete system is illustrated in Figure 9. The compressed air entered the system through the pressure hose (A) and then passed through a pressure 84 regulation valve immediately to the right of "A". A pressure gage (B) was connected into the system to in- dicate the static air pressure at which the system was operating. A series Of six control valves (C) follow. The pressure available to the network of systems was adjusted with the pressure regulation valve while the pressure for any individual system was adjusted with a specific control valve. A l/4-inch (inside diameter) rubber hose connected the control valve to the flask with water which was in turn connected to the series Of flasks (E) previously described. The batteries (D) supplied the power for the pen flashlight and the Gra- Lab Timer (G) was used to measure time intervals between counts of damaged grains or total number of fissures which had developed. After being exhausted from the last flask containing a saturated salt solution, another length of rubber hose carried the conditioned air to a filter and a final length carried it into the observation and inspection chamber (F). The author is shown position- ing rice grains in the chamber which is better illustrated in Figure 10. 85 The filter was necessary as Wilson (1921) had cautioned. Some trial runs were made by exposing grains to 86.6 percent RH without the filter. The grains were weighed after 24 hours and then moved into a static sys— tem (plastic freezer container). After 24 hours in the static system. the grains actually showed a loss Of weight. This indicated that droplets had passed through the system and finally attached themselves to the grain. The filter consisted of a 6-inch length of glass tubing (8 mm outside diameter) with a 2 to 3-inch plug of absorbent cotton. The fact that drOplets were en- trained in the air was further verified by the yellow discoloration which appeared on the upstream side of the filter plug in the potassium chromate (yellow solu- tion) system. The cotton was recognized to be an absorb- ent material, hence the systems were operated for 48 hours or more to equilibrate the cotton before any tests were run. The practice of Operating several days before run- ning tests is highly recommended because some salt solu- tions require more salt to produce a saturated solution at a high temperature than at a low one. A system moved 86 from a 92°F atmosphere with few salt crystals remaining in the saturated solution may have too much crystalized salt in a 38°F atmosphere. Adjustments of this type must be allowed for when moving the system from one temperature to another. 3.3.2 Observapgon 2nd 1p§pect10n chgmpgg The observation and inspection chamber (Fig. 10) was a vapor tight container constructed from a 4-1/2 x 4-1/2-inch section of a fluorescent light diffusing grille. It was enclosed on top and bottom with single thickness plate glass. The bottom glass plate was partitioned into ‘two 50 compartment sections by means Of some thin plastic strips. Individual rice grains were placed into these compartments. The two dark extensions in the bottom of the picture are screw caps from l-dram vials. The tops of these caps were completely removed (by sanding) to give threaded cylinders which were cemented into holes drilled into the side of the plastic grille section. One— dram vials with grains could be screwed onto these caps. Then by positioning the chamber so the vials were inverted. 87 the grains were made to fall into the chamber. The vials were removed and by means of a wire with a small loop on one end. the grains could be positioned in the respective compartments. Alignment of the grains was necessary for determining the number Of grains which were damaged and also for counting the total number of cracks in the grains. Alignment was not necessary if only weight measurements were made. After the grains were inserted into the cham- ber and the vials were removed. masking tape was used to seal these Openings. In the center of the frame between the two cemented caps, a small hole was drilled into each of the two grain chambers. This hOle was also covered with masking tape which was punctured with a needle to provide a bleeder hole for exhausting the conditioned air. To test the air tightness of the system a person could place his finger over the bleeder hole and watch the masking tape over the vial caps bulge from the pressure build up in— side the chambers. Another simple way was to brush a soap film across the bleeder hole and watch a bubble develOp. 88 This Observation and inspection chamber met the requirements of being able to see and study the grains while they were being subjected to an atmospheric change. It also permitted counts to be made of the damaged grains and Of the total number Of fissures as they developed in the grains. The grains could not be weighed in the system but a single weight measurement required that the grains be out Of the system for only one and one-half minutes. This was done by removing the tape over a bottle cap Opening and attaching a l-dram vial. The chamber was inverted to make the grains fall into the vial which was then re- moved. capped. taken to the scales and weighed (by a pro- cedure to be described later) and finally brought back to reinject the grains into the system. 3.4 Standard Procedure 3.4.1 General After the rice grains (6 varieties. 2 ages) had been processed through an air separation device. the hulls were removed by hand. This was not the fastest way of 89 removing the hulls but it was a way which would not in- jure the bran layers (pericarp) surrounding the grain. The attachment-end Of the grain was held between the thumb and forefinger Of the left hand. The style- end of the grain was placed on the forefinger of the right hand and the ends of the lemma and palea were broken Off with the right~hand thumb—nail. The unhulled grain was then pinched (attachment-end) between the left- hand thumb-nail and forefinger causing the grain of brown rice to separate and to slide out Of the style-end of the hull. Occasionally the germ Of a grain was damaged from the pinching process causing the kernel to be discarded. The grains of brown rice were then inspected with the equipment shown in Figure 11. The inspection stand (A) accommodated a plastic conditioning tray and a petri diSh. For the initial inspection the petri dish was used. The modified pen flashlight (B) was held under.the petri dish while the grains were viewed through the magni-focuser (C) and the individual grains were manipulated for inspec- tion with the tweezers (d).’ Imperfect grains may have been deformed. immature. cracked or discolored. Other causes 90 Fig. ll.——The equipment used for inspecting the rice grains , after the hulls had been removed. (Photo neg. 641367—6) Fig. 12.--Grains of brown rice after equilibrating for three weeks were separated into groups of 50 grains and placed into l—dram vials in prepar- ation for the grain damage and moisture absorp- tion studies. (Photo neg. 641367-7) 91 for rejecting a grain were a damaged germ, bran coat or a non—vitreous (chalky) appearance. Grains with config- urations that did not conform to the general shape of the variety were also discarded. Two inspections were made of the grains. The first removed all grains to be discarded while the second separated grains which were acceptable but which had minor imperfections. The nearly perfect grains were designated as first class kernels while those with the minor imper- fections were labeled as second class grains. Both classes Of grains were stored in 2—dram glass vials (vapor tight) with plastic screw caps. Before any tests could be run, the grains had to be brought to equilibrium with a given atmosphere. This was done by removing the grains from the 2-dram vials. inspecting them in the petri dish and making an approxi- mate count Of the grains required for the tests. Each of the 12 samples (6 varieties, 2 ages) was then placed into the conditioning tray for placement into a plastic freezer container with the appropriate saturated salt solution. After placement the grains were permitted to 92 equilibrate in the limited atmosphere for three weeks before any tests were run. Grains in the 2-dram vials were stored at 68°F and usually were at equilibrium with a 40 to 50 percent RH. Therefore. grains which were to be equilibrated at a lower relative humidity could be placed directly into the equilibrating atmosphere. Grains which were to be equilibrated at a higher relative humidity were initially placed into the 43.9 percent RH container and were then moved at 24-hour intervals and 10 percent RH increases until the desired humidity was reached. This procedure was necessary tO prevent grain damage from moisture ad- sorption. The test procedures required grains to be equili- brated at the temperatures and relative humidities as listed below: Tempgrature, Relative humidity, Min. grains F percent per sample 38°F 11.2 150 38°F 59.6 50 38°F 68.4 50 68°F 11.2 150 68°F 54.9 100 68°F 65.3 50 92°F 11.2 150 92:F 51.1 100 92 F 62.7 50 93 The number of grains listed were the minimum re— quired per sample for a test of grain damage by moisture adsorption. These figures add up to 850 which when multi— plied by 12 samples yield 10.200 grains. Any grain damage during the conditioning period would have reduced the num- ber of whole grains to below the required test number. Also the possibility existed of making a wrong move with a test-group of 50 grains. This would have nullified the test and would have required a re—run. Because of these eventualities. 100 extra grains were placed into each compartment. thus adding 10,800 more grains to the total. After the equilibrating period and 18 hours or more before a test. individual grain samples were removed from the conditioning tray before being inspected and sep- arated into test-groups of 50 grains. This inspection and separation required five minutes per test-group. Dur- ing this period the grains were exposed to the ambient atmosphere and consequently tended to gain or lose mois- ture. Because of this factor. the conditioning tray with the remaining samples was always replaced into the plastic container while an inspection and count were being made. 94 The enclosed saturated salt solution Obviously did not regenerate the specified relative humidity while an in- spection and count were being made. but the container limited the atmosphere to which the grains in the tray could be exposed. Each compartment in a conditiOning tray had suf- ficient space to hold three l-dram vials. Figure 12 i1— lustrates such a tray with two l—dram vials per compart- ment. Therefore all the grains required for a given test sequence could be inspected and separated in one Operation. Most grains were inspected and separated several days before being subjected to test conditions. 3.4.2 Weighingyprocedures The procedure was to remove a test-group of grains in a l-dram vial from a plastic humidity container and to immediately place a screw cap on the vial to seal the grains from the atmosphere. The grains were weighed on a 0 to 200 gram Mettler scale calibrated to 0.0001 of a gram. The last decimal place was read from a vernier. By using the magni-focuser. an estimate could be made 95 of the nearest 0.00001 Of a gram but this was rather fu- tile since air currents. building vibrations and other factors caused the scales to oscillate over a range of this magnitude or more. Hence. readings were made to the nearest 0.0001 of a gram. Since the samples consisted of only 50 grains (approximately 1 gram). the weighing procedure required as much refinement as possible to yield meaningful weight measurements. The procedure utilized a special vial (weight vial) for taking measurements. The sealer pad inside the screw cap was removed and replaced with a non- absorbent gasket material to keep the bottle at approxi- mately constant weight. The empty vial with cap attached was weighed and the weight was recorded. Both the vial with the test-group of grains as well as the weight vial were then opened, positioned mouth to mouth and inverted to transfer the grains to the weight vial which was imme- diately capped and.placed on the scales. A second weight measure was taken and recorded. The difference between the two measurements was the weight of the grains. An individual measurement of this type was reproducible within 96 T 0.0001 of a gram. Usually. however. a series of weight measurements were taken and the empty weight vial was re— weighed after every two or four weighings depending on the grains that were being handled. High moisture grains placed into the weight vial tended to increase the weight of the empty vial. The grain weight resulting after such a weight measurement was correct but the grains after they were removed from the vial were lighter by the amount of moisture adsorbed by the vial surfaces. The weight vial had to be reweighed in such a case before another weight measurement was made. When grains at equilibrium with one humidity were weighed. the weight vial usually reached a near equilibrium condition after the first four weights. If weights were taken of grains subjected to different relative humidities. the variation may have been as much as : 0.0002 of a gram. In many cases this total varia- tion may not have been completely due to the scales but actually may have been caused by a combination of 1) human error. 2) scale reproducibility and 3) true weight variation. 97 Three weight measurements were made in most of the tests for which the rate of grain damage was observed. the total number of fissures counted and in which the total grain damage developed within 24 hours. Between four and five humidity systems were usually operated simultaneously thus calling for eight to ten weights to be taken at the pre-selected times. In these cases all grain groups were removed from the Observation chambers. weighed and replaced. The time out for the grain groups was approximately two minutes per vial. Thus for eight vials the time out would have been 16 minutes and for 10 vials 20 minutes. After being replaced. the grains had to be repositioned. This will be discussed further in the section. "Determination of grain damage." When four or five systems were Operated simultaneously. different moisture content grains resulted which required that the tare weight Of the weight vial be checked after every two weights were taken. In the moisture adsorption studies. the weighing .procedure was somewhat different. Only 50 grains (1 group) ‘Were removed at a time. These were weighed and replaced 98 before another group was handled. This required a stag- gered weighing period but reduced the time out to one and one-half minutes. The procedure for this Operation was to make the zero adjustment on the scales and secure the tare weight of the weight vial before removing the grains from the conditioned air chamber. The 50 test- grains were deposited directly into the weight vial which was immediately capped, weighed and the grains returned to the conditioned chamber. all within one and one-half minutes. In these measurements both the time of weighing and the time out were critical factors. A Gra-Lab Timer with an alarm was used to maintain the desired time inter- vals. Further details of related procedures will be given in the section. "Determination of moisture adsorbed." 3.4.3 Determination of grain damage The 12 samples (6 varieties, 2 ages) were subjected to the relative humidity changes at the specified tempera- ture as shown on the following page: 99 Tempegature Initial equilibrium Conditions to which F condition, % RH grains were subjected 38°F11.2 34.8% RH 38°F 11.2 59.6% RH 38°F 11.2 75.0% RH 38°F 59.6 100.0% RH 38°F 68.4 Water 68°F 11.2 33.6% RH 68°F11.2 54.9% RH 68°F11.2 75.5% RH 68°F 54.9 86.6% RH 68°F 54.9 100.0% RH 68°F 65.3 Water 92°F11.2 32.6% RH 92°F 11.2 51.1% RH 92°F11.2 75.4% RH 92°F 51.1 86.1% RH 92°F 51.1 100.0% RH 92°F 62.7 100.0%.RH Test-groups of 50 grains were used. Counts were made Of 1) the number of grains which had developed fis- sures at a specified time and 2) the total number of fis- sures in the group of grains. 100 These counts could be made simultaneously by using a Veeder Root mechanical counter in one hand to record the number of fissured grains while mentally enumerating the total number of fissures as the grains were being inspected. After all the grains had cracked. the mechanical counting device was used to enumerate the total number of fissures. thus giving a more Objec- tive count. Intervals at which the counts were made were dependent upon the conditions to Which the grains were subjected. If complete grain damage developed within six hours. counts were made every 15 minutes. If fis- sures developed Over a period of 48 hours or more, hourly counts were made during the peak cracking period and thereafter on a 12 or 24 hour basis until the fissuring was complete. The procedure was to remove the test-groups in the l-dram vials from the static humidity containers. These were immediately capped. Exposure time to the ambient atmosphere was only five seconds per vial re- moved. Usually only two vials were removed from a given 101 container but on occasion as many as six were extracted. This would have caused the final vial to have an exposure of 30 seconds. The moisture adsorption or desorption within this period of time was considered negligible. Eight or ten grain groups were usually used during a test period. These were weighed as previously described and then injected into the conditioned chambers at a spe- cific time which was recorded. Injecting the grains in- volved removing the cap from the sealed vial containing the grain and almost simultaneously removing the masking tape from the vial cap cemented to the conditioned chamber. The vial was attached to the cap. Total time requirement was 10 seconds per vial or 100 seconds for 10 vials. The conditioned airchambers were inverted to receive the grains Which had to be arranged in an orderly manner. This was done by removing the vial. Then by means of a thin wire with a loop on one end. the grains were positioned in the individual compartments (time requirement per compartment. 5 minutes). During this period the conditioned air chamber was exposed to the ambient atmosphere to the extent pro- vided by the cap Opening. Total capacity of the conditioned 102 air chamber was 300 cc. While a system was in normal operation. the conditioned air inside a chamber was ex- changed on the average of once each minute. While the grains were being aligned in a section Of the chamber the majority Of the conditioned air passed through this section. The effects of this exposure are believed to have been small and were neglected. The weight measurement taken after the first 10 hours of exposure followed the aforementioned procedure. The situation was different. however. from the standpoint that grains were being removed from an atmosphere in which they were adsorbing moisture to an atmosphere Of no change. Diffusion of moisture inside the grain continued but the potential which would add additional moisture was tempor- arily removed. Grains were in the sealed vials for 16 to 20 minutes depending upon the number of groups in the se- quence. Thereafter. they were returned to and positioned in the conditioned air chambers as previously described. Generally this procedure was very satisfactory for grains which developed their full extent of damage within 12 hours.v Grain groups which were approaching 103 their maximum rate of grain damage and fissuring at this period showed a definite decline in rate Of fissuring and grain damage for the next few hours. This indicated that the time out definitely influenced the rate of damage and fissuring of these grains. Hence the lO-hour weight measurement was discontinued. Only the initial weight measurement (before insertion into the dynamic system) and the weight measurement after removal from the dynamic system were continued. Grains which were very slow in fissuring (Tables 1 - 6, Appendix) presented other problems. The experi- ment was set up on the basis of keeping the grains in the dynamic system for 24 hours. Some grains required more than 72 hours to develop their full extent of damage. In these cases the grains were preconditioned for approxi- mately 10 hours in a static system. subjected to the dy- namic system for 24 hours and then replaced and Observed in the static system at 24-hour intervals until no further damage developed. In these cases the lag of the static system was less important than in cases where the grain damage develOped over a period of only several hours. 104 After the grains had been subjected to the humid— ity change in the dynamic system for 24 hours, they were returned to a static system having the same equilibrium relative humidity. Grains were permitted to equilibrate in these atmospheres for three weeks before being removed. weighed and dried in a vacuum oven at 100°C and 28.5 inches of mercury vacuum for a period Of 120 hours in order to Obtain dry matter weights. Grains equilibrating at 68°F. 100 percent RH de- veloped mold growths before the end of the 3-week period. The mold growths caused all 12 samples to be removed. weighed. placed into the vacuum oven and dried to obtain dry matter weights. These samples had equilibrated be- tween 15 and 21 days. The grains as well as the vials in which the grains were equilibrating in the 100 percent and the 86.1 percent RH atmospheres at 92°F were treated with 5 x 106 rep (roentgens equivalent physical) from a Gen- eral Electric Electron Beam Generator in an effort to kill any mold producing fungus. The samples in the 86.1 percent RH atmosphere responded excellently and develOped 105 no mold growths during the 3-week equilibrating period. The groups in the 100 percent RH responded oppositely and developed such severe growths in 12 to 17 days that the resulting dry matter weights were affected enough to produce erratic moisture contents for the previous conditions Of equilibrium to which the grains had been subjected. The equilibrium moisture contents calculated from the dry matter weights are recorded in Table 10. Appendix. These average values with resulting equi- librium moisture content lines are also plotted on a psychrometric chart (Fig. 13). The procedure for the determination of grain damage may be summarized in the following steps. 1. Grains were equilibrated at given relative humidity conditions and were weighed upon removal from these conditions. 2. Grains were subjected to higher relative hu- midities at the same temperature. Observa- tions were made Of grain damage and total fissures which developed. Grain weights were also taken after certain time intervals. 106 Amend-v0 .moc ouocmv .unwno ofluuoeonzueamm oumocmum a so oomomsauomsm moan 5.6.3 now won: ucoucoo ousumfloe ESHHQHHHSUMIIHH u l ugh-(zulluh. 0.50 >80 H . .l 1 a s w .m- n m A m a w. m a A w a w .. a m a w H .23.... 2:23.: 0233 on. WES-.0 o_mk¥om6>ma mmwweozw ighgmod mo >._.m_oom . a!“ . UN AUG 81 8311 IOWA UBIVM $0 $81 107 3. Grains were permitted to equilibrate in the higher humidity atmosphere before being re- moved. weighed and dried to determine dry matter weights. 4. Dry matter weights were used to calculate moisture contents at the previous equilibrium conditions. 3.4.4 Determination of moisture adsorbed The procedure followed for this phase of the re- search was similar tO that used in determining grain dame age. Only two varieties, Fortuna and Century Patna 231. ‘harvested in 1961. were used. Each Of the varieties was subjected to the atmospheric changes as follows. Tempgrature Initial equilibrium Conditions to which F condition. %.RH grains were subjected 38:F 11.2 34.8% RH 38°F 11.2 59.6% RH 38oF 11.2 75.0% RH 38°F 59.6 86.7% RH 38 F 59.6 100.0%.RH 68:F 11.2 33.6% RH 68°F 11.2 54.9% RH 68oF 11.2 75.5% RH 68°F 54.9 86.6% RH 68 F 54.9 100.0% RH 108 Grains were initially equilibrated for three weeks. Prior to the test period. they were separated into groups Of 50 and placed into vials. Two groups of 50 kernels were used for each test. Grains were weighed prior to the tests as previously described. Both sets of 50 grains were subjected to the higher humidity atmosphere at the same time. The inter- vals between weighings were staggered as follows: 68°F, 11.2 to 33.6 percent RH Time 9:00 10:00 11:00 12:00 3:00 5:00 9:00 8:00 a.m. p.m. a.m. Sample 1 W* W W W W Sample 2 w w w w w *Times when weights were taken The next set of two grain samples was subjected to another condition. i.e., 68°F. 11.2 to 54.9 percent RH at 9:10 a.m. Likewise three more sets were started at 10 minute intervals until at 9:40 a.m. five sets were under test. The staggered schedule was maintained for a 23-hour period. The Gra-Lab Timer with an alarm was used to main- tain the pr0per time intervals. At 47 hours after exposure 109 a final weight was taken. Thereafter the samples were placed into a vacuum oven at 100°C and 28.5 inches of mercury vacuum for 120 hours to determine their dry matter weight. Initially both groups of grains were inserted simultaneously. Thereafter only one group at a time was removed for weighing. For this purpose the weight vial and a dummy vial with a closed tOp were used. Both vials were attached to the Observation chamber at the time for a weight. The weight vial received the grains to be weighed while the dummy vial closed the second cap opening. If the dummy vial was not used. the rice grains would contact and stick to the masking tape covering the second vial cap. This caused difficulty when the next weight was taken. 3.5 Codification of Results The grain damage studies utilized six grain var- ieties and two ages of grains. These grain samples were observed under three temperatures as they were subjected to six different environmental conditions at each temper- ature, except at 38°F where there were only five conditions. 110 These combinations added up to the 204 studies which were made. Rapid identification of a particular study became a problem. As a result a code was developed to represent the different test conditions. Each test involved five different factors. These were: 1. Grain variety 2. Temperature 3. Relative humidity for the initial equilibrium moisture content of the grain 4. Relative humidity into which the grains were moved 5. Year of harvest of the grains Data sheets were developed (Tables 11. 12. 16 and 17. Appendix) for recording information on both ages of grain for a given variety at a given temperature as these grains were moved from one atmospheric condition to an- other. The year Of harvest information was common to each sheet and therefore caused no distinction. Variable factors were grain varieties. temperatures. relative hu- midities for initial grain equilibrium conditions and relative humidities to which the grains were subjected. 111 These variables suggested a four number code which could readily be used to identify a particular sheet. The first number was selected to represent the variety of grain. The six varieties were codified in the following manner. Variety Type Code Zenith medium 1 Fortuna long 2 Bluebonnet 50 long 3 Rexoro long slender 4 Belle Patna long slender 5 Century Patna 231 long slender 6 The second number was selected to represent the temperature directly with only the Fahrenheit abbrevia- tion (°F) omitted. Temperatures generally varied less than : 2°F from the indicated number. Thus the second number in a code should be 38. 68. or 92. since the studies were conducted at these temperatures. The third number represents the relative humid- ity percentage at which the grains were initially equili- brated when the temperature was that represented by the I! 112 second number in the code. Decimals were omitted from the relative humidity percentages and the nearest whole number was used. Fractions Of 0.5 percent were dropped. The fourth number (or word) represents the rela- tive humidity percentage or other condition to which the grains were subjected. In two cases (38°F. 68.4 percent RH and 68°F. 65.3 percent RH) the grains were taken from the equilibrating atmosphere and placed directly into water. In these cases the word "water" was used to re— place the fourth number in the code. Thus a code number of l - 38 - ll — 60 represents the Zenith rice variety (1) under test at 38°F with an initial equilibrating relative humidity of 11.2 percent and a 59.6 percent RH in the atmosphere to which the grains were subjected. In this research. tables or graphs which repre— sent a specific variety of rice and year of harvest Of the grain. definitely specify this information and then use an abbreviated code consisting Of the last three numbers. i.e.. Figures 18 - 21 and Tables 1 - 6, Appendix. 113 Other graphs (Figs. 14 — 17) use the full code followed by the year of harvest in parentheses. This code per- mitted a rapid. concise and definite but simple method of designating test conditions. IV THEORETICAL CONSIDERATIONS 4.1 Psychrometrics The term psychrometrics refers primarily to the state Of the ambient air with particular reference to moisture. The relationships which exist among tempera- tures (wet bulb, dry bulb and dew point). relative humid- ities. vapor pressures. atmospheric pressures, specific volumes, specific heats and possibly other characteris- tics Of the ambient air are Often included in a discussion of psychrometrics. Data for standard atmospheric pressure are often plotted in graphical form to yield a psychro- metric chart. For most agricultural research. the devi- ations from standard conditions are not sufficiently great to warrant the construction of separate charts for the variations in atmospheric pressure. The stan- dard psychrometric chart is a basic reference for all processing Operations which are influenced by ambient conditions. 114 115 Much can be written about the utility of a psy- chrometric chart. A specific temperature and relative humidity are sufficient to define any ambient condition. Various thermal processes can be plotted on the chart to diagrammatically illustrate the mechanics of the proce- dures. Humidifying and dehumidifying processes also can be illustrated. A psychrometric chart for standard atmospheric pressure develOped by Dr. F. H. Buelow, Department Of Agricultural Engineering. Michigan State University. is Shown in Figure 13. Equilibrium moisture content lines (db) for brown rice at the various atmospheric conditions have been superimposed as an illustration of the chart's utility. These lines establish the conditions under Which no moisture transfer occurs while the grains are moved from one point on the line to another point on the same line. Alternatively stated, the vapor pressure within the grains at any specific point on the line is equal to the vapor pressure of the ambient air at that same point. Thus there is no moisture movement. Whenever a vapor pressure difference exists between the grain and the 116 ambient air. moisture movement occurs in the direction of the lower vapor pressure. For example, assume grains to be in equilibrium at 90°F and 41 percent RH. These grains are then moved to 38°F, 80 percent RH._ An immediate observation from the psychrometric chart is that the ambient vapor pres- sure Of the moisture in the air will be lower at the latter condition. The vapor pressure within the grains will change to that corresponding with 11 percent EMC (0.034 psi) at 38°F. This is a lower vapor pressure than that Of the ambient air at 38°F. 80 percent RH. Thus moisture will flow from the air into the grain until the grain finally comes to equilibrium at the new ambient condition; Similar analyses may be made When grains equilibrated at one condition (any point) are moved to any other point on the chart. Some general observations which can be made from the psychrometric chart are l) for a fixed relative hu- midity, the EMC decreases as the temperature increases. 2) for a fixed vapor pressure. the moisture content de— creases as the temperature increases and 3) for a given 117 relative humidity increment, the vapor pressure changes are greater at higher temperatures than at lower temper- atures. These are only a few of many Observations which could be made. Henceforth. only those observations which are pertinent to this research will be discussed. These elaborations will be made as necessary in the "Discussion and results" section of this dissertation. 4.2 Statistical Implications After rice grains are subjected to a change in atmosphere, they may or may not fissure. The nature and magnitude Of the atmospheric change determines if damage will occur. I The observation was made in the preliminary re— search that the cause and effect relationship relative to rice fissuring involved the parameters of 1) moisture adsorption. 2) grain damage and 3) time. In a statisti- cal study large samples are desired. However. when in— dividual grains need to be meticulously processed and conditioned prior to being tested. the size of sample 118 or the number Of grains involved becomes an important factor. For certain statistical studies. sample sizes of 30 or more are sufficiently large. When sample sizes of less than 30 are utilized a different statistical approach is suggested. Selection Of the sample size (n) was given much consideration in this research. The following factors were involved. 1. Total weight of available grain of each age and variety was 200 grams. 2. Preparation time for each individual grain kernel was from 30 to 60 seconds. 3. Counting time for determining grain damage and total number of fissures ranged from one to five minutes. 4. Facilities and equipment were necessary to conveniently and economically accommodate the sample size selected. From a theoretical viewpoint. the larger sample yields the more accurate and meaningful data. From a .practical viewpoint. the larger sample requires more time to make counts on damaged grains and total fissures. 119 thus making the data representing a particular time less accurate. In cases where damage developed over long periods (24 hours or more). the time to make the counts was Of less importance. However. when fissures and grain damage developed completely in 12 hours or less. the counting time became a critical factor. Following these considerations the decision was reached to use 50 grains per sample. Preliminary investigations indicated that points resulting from plotting the cumulative percent of grains damaged versus response time on probability graph paper generally formed a straight line (Fig. 14). This implied that the response time at which damage occurred had a normal distribution. The cumulative percent of total fissures was also plotted against response time on a probability graph. Results indicated that response time initially had a normal distribution but failed to follow this distribu- tion after most cracks had developed. Cracking continued after a period when fissuring should have been complete according to the normal distribution theory. This same 120 0.1 99.9 , I I l l 1 1 0.2L -99.8 0.5- _ 1 _99 2 — . A o -98 8 5 5 a“? O)A _ a. {j _95 \— u.- f.” .9) 10 +- ’,\ ,4 0°) 0 .490 S’. 7' ., 4.) (‘1’. ‘37; Nb. 4"; g 20 — 8 .' ,8 1.. 8 ‘3 8 o A; ’9 o 3 o 30 -. “to -—70 9* a o . g g 40 - —60 ..—. "-4 13) c ~r-l 60 __ o _ E :5 . 40 ll ‘ 70 L A «30 ‘ P o '6 Q) o g 80 I— #20 fit 5. . g g 90 - O _.10 m 0 c 2 O .-. 7; 95 — - 5 1'; H (O A (D o 98 _ q 2 99 _ .— 9908- .7 005 99.9_. -m 0.2 I l l l . I l , O 1 2 3 4 5 6 O 1' Grain response time. hours Fig. 14.+—A probability graph illustrating the accumulated grain damage after the indicated response time.- The linear relations indicate that the respective response times followed normal distributions. 121 tendency (less pronounced) was occasionally Observed in the cumulative grain damage versus response time plots discussed earlier. These Observations suggested appli- cation of the normal distribution theory to the grain damage phenomenon. A discussion Of this theory follows. The normal curVe represents a continuous distri- bution. It can be approximated by histograms having narrower and narrower classes. After sufficient Classes are considered in an adequately large sample, the mean of the histogram will approach the mean of the continuous distribution from the same population. Likewise, the stan— dard deviation of the histogram will approach the standard deviation Of the continuous distribution. The normal curve is bell shaped and extends an infinite distance in either direction from the mean. The curve is completely defined by its mean and standard devi- ation. The mathematical equation for the curve is _ l -% x—u)2 y- o e V]??? V‘ where u = mean standard deviation j H 122 A normal curve may be standardized by performing a change of scale such that -H .1.) m H E 5 50_. o 8; on G) g 40- 'U m c T.‘ u 30.. 0 O 0 20.— 10.. 0 O o - - l I I I J l l I I J i l 2 3 4 Grain response time. hours Fig. 15.——Points Of cumulative grain damage plotted directly and a normal curve plotted from the mean and stan- dard deviation calculated from the experimental data. 126 mean response time Of the population. Statistical methods permit an assertion to be made that the mean response time may be within a specified interval. This is called inter- val estimation. An interval may be estimated by letting H be the mean of a random sample of size n. The theoretical sam- pling distribution of R has the mean u and the estimated standard error Of the mean s_ = s. The formula for Z x (n then becomes: The mean (u) and the standard deviation (T) are popula- tion parameters which are unknown for rice. According to Freund (1960). the theoretical sampling distribution of H can be approximated if n is large (30 or more). If a 95 percent confidence interval is to be determined for u. the Z values must be between i 1.96. By substitution into and by apprOpriate transposition of the equation for Z, the following expression is evolved. s - s - O < < + . x 1 96 f3' u Ix 1 96 [fi— 127 When evaluated for the data plotted in Figure 15. this expression yields the following 95 percent confidence interval (in minutes) for u.. 140.2 < u < 155.0 This states that the mean response time would be eXpected to fall within this interval in 95 out Of 100 replications of the experiment. The statement cannot be made for any one experiment asserting that the mean re- sponse time is within the interval calculated. Further statistical manipulations can be made with the data collected. The research procedure was de- signed from the preliminary investigations to yield basic information concerning the reaction of the rice grain to moisture adsorption. This work itself was unprecedented and consequently the data were not necessarily collected with particular statistical analyses in mind. Hence the analyses which were made of the data were limited to the calculation of the sample mean response time (R). the sample standard deviation of the response time (s) and the sample standard error of the mean response time (8;). Further discussion will be made in the next section. DISCUSSION AND RESULTS 5.1 Critical Analysis of the Literature The literature contains many comments about the factors which cause grains to fissure. There is nearly unanimous agreement that fissures result from rapid dry- ing. Kramer (1951a) stated that the milling quality fac- tor most influenced by drying was pounds of head rice per barrel and that usually an artificially dried sample was of lower quality than air-dried samples but occasionally there was no decrease but even some increase in quality. This statement implies that drying is probably harmful but not necessarily so. Rapid drying with heated air and forced ventilation is accepted as a cause for grain damage and will receive no further consideration in this dissertation. When grains are dried in the field or even in the windrow the cause for grain damage is not so precise. 128 129 Here the grains are subjected to temperature. humidity and moisture content changes. Even though grains in the windrow tend to dry. the drying process is not necessarily a continuous one. Instead the grains may dry during the daylight hours (sunshine) but actually may re-adsorb some moisture during the night. If the adsorption is less than the desorption. the net result after several days is a lower moisture content grain. The data (see page 23) by Dobelmann (1955. 1961) illustrate the necessity of considering more than the single parameter of "stacked" or "not stacked" in attempt- ing to determine causes for grain damage. Grains remain- ing unstacked for one night experienced a 15 percent in— crease in brokens over those Which were stacked immediately. Grains which remained unstacked for four nights experienced only a 16 percent increase. But grains Which remained un— stacked for five nights experienced a 25 percent increase in brokens over those which were stacked immediately. The data show that one night on the ground may be more effective in producing grain damage than another night. One of the ambient conditions which may have varied from night to 130 night is the relative humidity. The research reported herein proves the theory that humidity increases of suf- ficient magnitude will cause relatively dry grains (14 percent or lower. db) to fissure. Angladette (1964) recognized that kernels may fissure during moisture regain. He stated that splitting could be attributed to the lack of elasticity of the sur— face layer of the grain which no longer expanded under the effect of moisture regain. The work of Henderson (1954) is in direct contrast with this reasoning. The research reported in this dissertation indi- cates that grain fissures generally developed from the inside toward the outer perimeter. No external surface cracks were observed during periods of moisture regain. Some minor surface checks developed when high moisture grains were suddenly subjected to static low moisture atmospheres. These small checks seldom occurred but usually developed in areas where the bran coat had been damaged. Larger external checks were more prominent in high moisture grain which was artificially dried with heated air (160°F) but internal fissures were still far more predominant. 131 Moisture regain can be accomplished in rice if the grains are not subjected to relative humidity increases Of more than 8 to 10 percent at 24-hour intervals. Some observations indicated that lZ-hour intervals were suffi— cient. Minimum interval times were not determined. The work by Earle and Ceaglske (1947). reported earlier. is in agreement with the foregoing observations. The possibility of producing fissures by extreme drying is a plausible one and consequently was watched for in this research. Damage of this type was most likely in grains which were equilibrating at 11.2 percent RH. Only Fortuna grains at 92°F and 11.2 percent RH developed fis- sures during the equilibrating period (3 weeks or more). The time when these fissures develOped was not observed. In separating the equilibrated grain mass into groups of 50 for the grain damage tests. 26 grains with fissures (1961 harvest) were observed in a total of 228 while 16 fissured grains (1962 harvest) were observed in a total of 227 grains. The other 10 samples (5 varieties. 2 ages) equilibrating in the same tray and exposed in the same container developed no damage from drying in the static 132 low humidity atmosphere. Figure 13 reveals that the grains at 92°F, 11.2 percent RH had an EMC of 6.6 percent (db). At 68°F the EMC was 7.8 percent and at 38°F it was 9.4 percent. The results are logical since fissures developed in the variety most sensitive to damage and also under the condition where the grains were dried to the lowest EMC. Henderson (1954) reported that moisture or temper- ature increases caused grains to fissure. X—rays showed that fissures developed from the inside of the grain toward the outer perimeter rather than from the outside of the grain towards the center. To increase temperature without causing drying is difficult to achieve. On the psychro- metric chart (Fig. 13) the constant moisture lines move up and to the right with increasing temperature. Hence the grains will dry unless the heated ambient air is hu— midified. This is generally not the case since heated air usually has a lower relative humidity than the ambient air. To overcome this difficulty. Henderson sealed grains in a plastic bag before heating them. X-rays were taken before and after with the results that counts showed an increase in half. full and total checks. 133 Similar preliminary experiments were performed in this research. Grains were placed into vials with screw caps and tightly sealed before being subjected to 160°F air. The atmosphere was limited for the grains in the container. As the temperature of the grains in- creased, the vapor pressure within the grains also in— creased causing desorption which rapidly humidified the limited atmosphere to 100 percent RH. No damage resulted. The implication was that rice kernels could be moved through temperature changes without developing fissures provided there was no moisture transfer. The theory that a sudden temperature increase without moisture adsorption or desorption will not cause fissuring in raw rice is indirectly supported by the work of Desikachar and Subrahmanyan (1961) who reported that the formation of fissures was retarded when grains were submersed in water at temperatures of 70, 80 and 90°C (158. 176 and 194°F). More fissures developed and at a faster rate when grains were submersed at lower tempera- tures. If both moisture adsorption and sudden tempera— ture increases cause fissuring, then their combination 134 should tend to accelerate the process. This appeared to be the case when grains were submersed at temperatures of 30, 40, 50 and 60°C (86. 104, 122 and 140°F) but the trend reversed itself at the higher temperatures. Researchers who have commented on the phenomenon of moisture adsorption are Kondo and Okamura (1929). Stahel (1935). Kik (1951). Henderson (1954) and Desik- achar and Subrahmanyan (1961). Henderson's work has al— ready been elaborated upon. The works of the other re- searchers will now be discussed. The research of Kondo and Okamura (reported in a foreign journal) is not available but is described by Stahel. They found that the drier the paddy, the greater and quicker was the formation of cracks after a rise in moisture content. After exposure the paddy was shelled by hand before the cracks were counted. The means used to cause moisture adsorption was not stated, neither was the time at which crack counts were made. Stahel's work was similar to that of Kondo and Okamura. His procedure is reported in the review of literature. Even though Stahel observed fissures develop 135 after grains were exposed to moist atmospheres, all his data are based on grains which were submersed in water. dried and milled before observations on damage were made. Kik made crack counts before and after staining rice grains. He observed that the moistening and drying in the staining process caused additional fissures to develOp. Desikachar and Subrahmanyan submersed parboiled and polished rice in water and watched the fissures de— velop. The two parameters of interest to them were 1) time of response and 2) temperature of the soak water. No work was done with brown or paddy rice. In summary. the results reported by 1) Stahel, 2) Kik and 3) Desikachar and Subrahmanyan are all based on placing the grain into direct contact with the liquid to be adsorbed. Henderson's work was primarily concerned with fissures produced by the desorption process. His research relative to moisture adsorption was limited to exposing a single sample of grain to dew over the period of one night. 136 Thus the phenomenon of grain damage resulting when rice kernels are submersed directly in water at moderate temperatures need no longer be considered as a theory but should generally be accepted as a fact. The effect of exposing relatively dry rice grains (paddy or brown) to moist air under ambient conditions has not been so well established. The phenomenon has been pos— tulated by numerous researchers but none has performed experiments under controlled.oonditionsin.which the rel- ative humidity or vapor pressure was the only variable that could cause fissures to develop. This was the chal— lenge for the research undertaken. 5.2 Fissuring of the Rice Grain 5.2.1 Qg§orption damage Rice grains dried in single layers at high tem— peratures developed both external and internal fissures. These occurred in no definite pattern and any attempt to evaluate such damage by counting the cracks became 23 rather hopeless task. External cracks developed on the 137 surface of the grain and were generally of such small magnitude that they were visible only on the surface where they occurred. Since they could not be observed through the grain even with an inspection light, checks of this nature could have been missed easily in making counts of total damage. Surface fissures extending across the dorsal or ventral lines of fusion were gen- erally visible from both sides of the grain. The exter- nal checks observed which developed from heated air dry- ing, did not develop in any set pattern. Neither did they develop to any set size. Surface checks were of dimensions just observable with a 3.5 x magnification to about 0.1 inch lengths. A very few were observed which actually enveloped the small perimeter of the grain. Occasionally a grain crazed thus producing conditions impossible to evaluate in terms of crack counts. Evalu- ations could have been made in terms of crazed grains but extent or degree of crazing would have presented another problem. Internal fissures resulting from artificial dry- ing (desorption) had a different nature than fissures 138 which resulted from moisture adsorption. Fissures re— sulting from drying generally followed a jagged pattern whereas fissures resulting from moisture adsorption were more smooth. Montgrand (1958) described cracks which resulted from drying as having broken edges. The nature of the fissures resulting from moisture adsorption will be elaborated upon in the next sub-section. In this research some grains with approximately 15 percent moisture content were dried in 160°F air simply for the sake of observing the nature of the cracks which developed. All the experimental rice used in the grain damage tests was dried for 120 hours at 100°C (212°F) under 28.5 inches of mercury vacuum. After drying a final evaluation was made of the fissure damage. Kernels which fissured severely during moisture adsorption suffered little additional damage during the drying process. Grains which developed little or no damage from moisture adsorp- tion did develop damage from the final drying procedure. The condition where the grains were moved from 11.2 to 34.8 percent RH at 38°F are of particular inter— est (Tables 1 - 6. Appendix). Only five fissures (3 in 139 Fortuna, 2 in Zenith) developed in the 600 grains in- volved in the experiment. No fissures developed in the other 10 samples during moisture adsorption. After dry- ing to dry weight. the samples on the average had two fissures per grain with exception of the Century Patna 231 variety which averaged 1.3 cracks. Only four grains in the total of 600 failed to crack during the drying process. Three of these were in the two Century Patna 231 groups. Grains for the experiment were initially at 9.42 percent EMC (Fig. 13) before being moved to an atmosphere with 11.39 percent EMC from which they were finally dried. Similar results were achieved in the experiments conducted at 68 and 92°F. The results illus- trated the development of kernel damage When grains were dried to dry matter contents by artificial means. 5.2.2 Adsorptign damggg The evaluation of grain damage from moisture ad- sorption can become quite complex especially if dry grains (9.0 percent moisture or less) are subjected to large hu- midity increases. Therefore in developing a method of 140 evaluating the grain damage. small humidity changes were desirable. When the moisture adsorption conditions were such that just a few cracks (internal fissures) developed in a grain sample. these cracks were smooth. clean and distinct. There was no problem in counting and the ac- curacy was very good. The following accuracies were gen- erally achieved when counts were made on samples of 50 grains. I+ 100 fissures 1 percent l+ 200 fissures 2 percent I + 300 fissures 4 percent Counts on grains having up to an average of four cracks per kernel could be made quite rapidly. These fissures were usually perpendicular to the longitudinal axis. When more than two or three cracks per grain occurred. the development of a longitudinal crack was quite common in the Zenith (medium-grain type) and Fortuna (long-grain type) varieties. The dimensions of these varieties showed these grains to have a greater width and thickness than any of the other grains studied. Zenith was by far the shortest grain in the entire group. When more than three 141 or four fissures developed in the remaining varieties a partial or complete longitudinal crack often appeared. The longitudinal fissure usually developed after several cross-sectional fissures had evolved. Hence this longitudinal crack was intersected by the cross- sectional fissures. As a consequence this fissure (lon- gitudinal) could develop in the following alternative ways. 1. Completely across the length of the grain in a smooth line. 2. Completely or partially across the length of the grain in a broken or segmented line because the intersecting cross-sectional fissures permitted the longitudinal crack to develop in a discontinuous manner. These alternatives presented problems in making the counts. The solution was to count a partial. a dis- continuous segmented or a whole longitudinal crack as a single fissure. This facilitated the counting procedure and helped to maintain better accuracy. A semicircular crack was quite prevalent in the Zenith and Fortuna 142 varieties. This fissure started immediately after the germ on the ventral side. extended slightly more than half-way across the width of the kernel and then termi- nated on the ventral side before reaching the style-end of the grain. The stresses for two cross-sectional cracks and the longitudinal crack probably combined to produce a fissure of this configuration. After formation of the longitudinal fissure. partial cross-sectional fissures were possible. Diagonal cracks between the cross-sectional fissures also tended to develOp. These conditions added complications to the counting procedure. At this stage of grain damage fur— ther modifications were necessary in the counting pro- cedure. Hence only those fissures extending to the dor— sal line of fusion were counted with the one additional count being permitted for the horizontal crack. At this stage each grain usually had incurred six or more fissures. A further proliferation of cracks between the major fis- sures presented a hopeless problem and made counting at- tempts quite futile. 143 From a milling point of View, a grain with two or three cross-sectional fissures has lost its commercial value and is destined for the flour mill. Therefore. the. most desired information was the determination of the time interval required for the first fissure to develop in a grain. This information was obtained as accurately as feasible. The adsorption fissures just discussed are inter- nal cracks which developed from the center of the grain to the outside, as Henderson (1954) postulated. Grains having equilibrated at 11.2 percent RH had 9 percent moisture (db) or less. These were dry. hard and flinty. Individual cracks developed to their full extent within a fraction of a second. When grains were initially equi— librated at relative humidities above 50 percent before being subjected to a higher moisture atmosphere. the ker- nels appeared a little more elastic and occasionally a fissure could be seen as it started at the center and gradually developed into a fully grown crack. Generally the development was still sudden, complete and completely unpredictable. 144 When a fissure develOps. over what length does it relieve stresses? The answer is dependent on the rate at which the fissures are developing in a grain. This in turn is dependent upon the relative humidity change to which the grain is subjected. Therefore,under conditions where an average of two fissures per grain developed. the stresses in the grain were relieved sufficiently to prevent further damage. When grains were subjected to large rela— tive humidity changes and experienced five or more cracks. the stresses were not relieved over great lengths. For example a grain 0.30-inch long with five cross-sectional cracks would be subdivided into six parts each approxi- mately 0.05 of an inch in length. Tables 7. 8 and 9 in the Appendix show that in many cases more than five cracks per grain developed. In the aforementioned tables, the fissure counts recorded are the largest that were observed in the respec- tive grain groups. When grains were moved from an equi- librium relative humidity into water, fissures developed which were readily visible. After the grains remained in the water for some time the smaller fissures at the 145 grain ends tended to swell shut and disappear. There- fore the 1ast count taken on the grains was not always the largest one. The deviation was not great and all the cracks generally develOped before the smaller cracks van- ished. Similar results. but less pronounced. were ex- perienced where grains were subjected to high relative humidities. Grains subjected to large humidity changes ex- perienced separation of the bran layers. This was espe- cially prevalent at the germ tip but not restricted to that part of the grain. The separation line often ap— peared as a crack and in many cases was superimposed di- rectly over one. Close observation with the inspection light was necessary to make the distinction. A damaged grain transmitted light up to the crack line but not across it. Thus the light focused on either side of a crack left the grain section on the Opposite side un- illuminated. If only bran separation had occurred and the light technique was applied, then the en- tire grain was always illuminated. Bran separation 146 lines were generally not smooth but instead curved and waved across the grain surface. As many as three separ- ated layers were Observed on an individual grain tip. If viewed in daylight. the areas of separation were vis- ible as slight discolorations on the grain surface. The bran separation occurred simultaneously with the fissur- ing of the grains. For large humidity changes at high temperature (i.e., 11.2 to 75.4 percent RH, 92°F) the observed crack pattern was different. Under conditions of small humid- ity change, the cracks which develOped were large (major) and extended over the grain cross-section. In cases of large humidity changes. the impending damage by major cracks was often preceded (approximately 15 minutes) by small cracks at the grain tips. 5.3 Evaluation of Grain Damage The data relative to grain damage were collected as previously described and condensed into tabular form as shown in Tables 1 - 9. Appendix. The utility of these data will now be discussed and illustrated. Each of the 147 first six tables shows the data collected for one variety of grain at two different ages. The code (column 1) al- ready has been explained. The vapor pressure differences. .AVP, (column 2) are those experienced by the grains as they were moved to the higher relative humidity at the specified temperature. The number of grains which fis- sured (sample of 50) are recorded in the third column. The data show numerous conditions under which all grains did not fissure. The number (n) used in making the sta- tistical calculations was restricted to the grains which fissured. According to Freund (1960). "n" should not be less than 30 for calculating the sample standard error of the mean. Therefore calculations were made only for samples in which 30 or more grains fissured. The sample mean response time is an average re- sponse time calculated by summing the response times of all the grains which fissured and then dividing by the total number of fissured grains. Sample mean response time _ §;individual grain response times number of grains which responded 148 An adjusted sample mean response time is recorded in the fourth column and not the sample mean response time. The adjustment which was made will be discussed later. The sample standard deviation. 8, (column 5) is calculated as follows: . 2 n 2': x. S = i=1 n Where x = a particular time response interval. 1. 2. 3, etc. n = the total number of fissured grains in a sample. It is a sample parameter which indicates the spread of the grain response times in a normal distribution. The percent- age of grains which can be expected to respond within one. two or three sample standard deviations from the mean was given earlier. The sample standard error of the mean. 8;. (column 6) has the same relationship to a sample mean response time as the sample standard deviation has to the response 149 time of an individual grain. It is calculated as follows: For a given sample mean response time. 68.26 per— cent of the grains in that sample can be expected to re— spond within one sample standard deviation from the mean. Likewise, if the entire population were subdivided into samples for which the respective means were calculated. a population of means would result. These means would also have a normal distribution and the sample standard error of the mean would specify that 68.26 percent of these means would fall within one sample standard error of the mean from the population mean. This is a measure of the dispersion of the means. Columns 7. 8. 9 and 10 are similar to columns 3. 4, 5 and 6 but record information for grains harvested in 1962. The adjustment of the mean time (column 4) will now be discussed. The data sheet of Table 11, Appendix. serves as an example to illustrate that no response oc- curred during the first two or three time intervals at 150 which readings were taken after the grains were exposed to the higher humidity. In conventional practice of cal- culating a mean, these intervals would be omitted and the calculation would begin at the time of the first response and end at the time of the last response. The mean re- sponse time would then consist of two parts: 1. The mean time beginning at the time of the first response and then calculated by conven- tional means. 2. The time increment between initial exposure of the grains and the first response. Thus there was a period of no response prior to the development of the normal distribution of grain re- sponses. The period of no response was different for each variety of grain for a given relative humidity change. Likewise, when a different temperature was used with the same relative humidity, different initial response times were obtained. Each sample had a response time of its own for every test condition. The experimental procedure required that counts be made every 15, 30 or 60 minutes depending upon the time required for the total grain 151 damage to develop. The initial response time was not determined any closer than the intervals at which the counts were made. For example, assume counts at 30 min- ute intervals. Further assume that no response occurred within the first two intervals (60 minutes) of exposure. Now if the first response occurred 61 minutes after ex— posure the damage would not have been observed until after the third interval (90 minutes). An error of 29 minutes would have resulted. This possible error in the time of initial response could not be avoided readily. In the final analysis, the time of initial response could be extrapolated more accurately from the other data col- lected on the particular sample. The foregoing conditions suggested that: l. The initial time of exposure be used as the point from which to measure response time. 2. The time of initial response be extrapolated from the other collected data whenever this information was desirable. Calculations of the sample mean, the sample standard de- viation and the sample standard error of the mean were 152 made according to conventional practices with the time intervals beginning when the grain was first subjected to the more humid atmosphere. This procedure produced a sample mean response time which was greater than the sample mean response time observed from plotting the experimental data (Fig. 15). The sample standard devia- tion and the sample standard error of the mean remained unchanged. The problem then was to adjust the theoreti- cal curve to fit the experimental points for each test. A graphical approach was used to determine the necessary adjustment. A cumulative percentage of grains damaged versus time of response chart was prepared for each sample at the 38 and 68°F temperature levels where- ever the response was sufficiently great to permit such plots to be made. These experimental points aligned themselves in the sigmoid shape of a normal distribution curve. Theoretical curves with the same time response scale were plotted on an overlay chart which was super- imposed on the plot of experimental data. The theoreti- cal cumulative distribution curve giving the best fit to the experimental points thus yielded a sample standard deviation and a sample mean time of response. 153 Evaluation of both the mathematical and the graph- ical approaches suggested that an adjustment be made in the mathematical approach to yield the solution obtained by the graphical means. This was done by adding a total of 41 mathematical calculations cm' the means and compar- ing this sum to that of the 41 graphical determinations. Results showed that the mathematical determinations were exactly 10.0 percent greater for grains harvested in 1961. The same procedure was applied to 41 samples of 1962 grains with the mathematical calculations being 9.5 percent greater. The half percent difference was no greater than other ex- perimental errors in the procedure and hence was disregarded. An adjustment consisting of a 10 percent reduction in the calculated mean response time was made. This is the in- formation recorded in columns four and eight of Tables 1 — 6, Appendix. This tabular information may now be used to recon- struct the cumulative percent damage versus time of response curves. The reader is cautioned to observe that the sample size (n) was restricted to the grains which fissured and that these results were obtained when grains were exposed in a single layer. 154 Figures 16 and 1? illustrate the use which can be made of the tabulated data. The curves in Figure 16 illustrate the responses achieved when a single variety and age of grain was exposed to different humidity changes at a temperature of 92°F. The theoretical data for these graphs are tabulated in Table 2, Appendix. The first numeral in the code on the curves is 2, identifying For- tuna rice. Observations from the tabulated data further reveal that all 50 grains incurred damage when exposed to the 92-11-75 and 92-11—51 conditions while only 48 of the 50 grains suffered damage when subjected to the 92-11—33 condition. The curve representing the 2-92-11—33 grain re- sponse in Figure 16 shows that hourly observations were made for 14.5 hours. The last observation plotted was made at 23.5 hours. Observations could not be made every hour on the hour by one individual over extended periods of time. Therefore, in making the statistical calculations, the grain damage which occurred during the interim was distributed according to the trends estab- lished by the observations before and after the interim period. .mon um woodman muflowesn m>HumHmu topmofipcw on» on pmmomxw cos? moan mcsuuom Eonm pmcflmuflo mmmcommwu map mumuumsHaw sows? muon mmmemp Cwmnm m>flumanfiso .oH musmflm munon .mEHu mmcommmu awake 155 om ma OH m o _ . _ a m. ‘1. - o 0 c .1 OH L 8 m o . m. .u. unsaom mama 0 G c I om P came Umumsncm m Quflz m>u5u Hmuaumnomna e - .1 cuss . a . P . . o o t . l . ommw . m \e o ,_ m. G C. I, 00 w... »\ o .L. «CM». . T. M a % L on d ). 0 T. a 296 ( m r 1 om.u u file- 6 o .. om . 1 OS cumulative percent Grains damaged. 100 90 80 7O 6O 50 40 30 20 10 I 156 O ’J ‘ ’9 0 (3 O A R? w 0) N 8' :3 .' m I 3 S.’ m ’ 9 (\ I .17 g, G .1 ¢V . , curve With adjusted~mean o A 9 Data points 9 0 0 I E l I 1 l n Ifi 1 I 1 l 2 3 Grain response time. hours Figure l7.——Cumulative grain damage plots which illustrate the responses obtained from Fortuna. Bluebonnet 50 and Century Patna 231 rice when these varieties we e exposed to the indicated humidity change at 92 F. 157 The curves in Figure 17 represent the responses of three grain varieties to a single humidity change at a given temperature. The code specifies the varieties to be Fortuna, Bluebonnet 50 and Century Patna 231. Tabu- lated data (Tables 2, 3 and 6, Appendix) for these 1962 grains show that all 50 grains incurred damage in each of the samples. These same curves represent typical re- sults in that Fortuna generally was most sensitiVe and Century Patna 231 generally was the least sensitive to damage. Zenith usually was the second most sensitive and the other three varieties did not align themselves in any particular order. If the tabulated data in Tables 1 — 6, Appendix. pare.studied, they reveal that 30 or more grains fissured in every sample of all varieties and ages when the grains were subjected to seven test conditions (Table 1). Two quantities involving grain response are of particular interest. These are 1) the adjusted mean re— sponse time and 2) the total number of grains which in- curred damage. The adjusted mean response time is the most informative since it reveals how rapidly a particular 158 OOHIHmINm mhIHHINm amIHHINm I I I I I mhIHHImm mmIHHIwm I I I I I mhIHHImm omIHHImm «wum3 mcofiuwpcoo umwu 038* mv.mma mom H5.mo mmm an.sm mmm Hmm maumm manuamo mH.H~H boo mm.mm mwm om.hm men mcumm mHHmm oh.¢aa chm H5.mm 0mm mo.om mmm onoxmm hm.mHH hgm mo.mm mum «n.0m mum om umsconwsam mm.m> mmo mm.hm V mmm mm.¢m mwm manuaom l mm.hm wwo ¢0.Hm Nwm m~.m¢ mum fiuwcmN am: now mcwmum “damn am: now mcwmum uwamm am: now mcflmum uwam mo 85m «0 Eom mo 85m mo Esm mo 85m mo Esm muwwnm> mAmuouImocmnsoo Mama lemm "H“ .mmamfimm Add as Umusmmww mcwmum muoe no om £0a£3 Cw h.emcofiuflméoo ummu Gm>mm How Umwom mum3 mmeu wmcommmu some Umumsnvm 0cm mcwmum vGMSmmHm con? muazmmmll.a manna sample (one 159 variety and age) responded by fissuring. Small adjusted mean response times indicate high sensi- tivity to damage. If the adjusted mean response times of a given variety and conditions. age of grain are added for the seven selected this sum provides a basis for comparison with the same variety of a different age as well as with other varieties of grain. This was done in Table 1. Several observations can be made. 1. The sum of the adjusted mean response times for older grains of a given variety is smaller than for fresher grains (columns 3 and 5). This indicates that older grains on the aver- age are more sensitive to damage and hence will fissure sooner after exposure to a more humid atmosphere. When the sums of the adjusted mean response times for both ages of a grain variety are added (column 7), the varieties align them- selves in order of increasing resistance to 160 fissuring damage as shown below. When de- creasing numbers of split grains are consi- dered. the order is not the same. Adj MRT Split grains Combined total Combined total A. Fortuna Belle Patna (least resistant) (least resistant) B. Zenith Fortuna C. Rexoro Zenith D. Bluebonnet 50 Rexoro E. Belle Patna Century Patna 231. F. Century Patna 231 Bluebonnet 50 (most resistant) (most resistant) The adjusted mean response time represents the better method of evaluating grain resistance to damage since this method was not limited by time as long as grains continued to respond. The combined total of split grains is not a good index because of imposed experimen- tal limitations. Each sample was limited to 50 grains. This created the hypothetical situation that all 50 grains of the most sensitive variety could have fissured while only 25 of the least sensitive incurred damage. To get 161 30 damaged grains in the second sample would have re- quired a greater humidity change which could no longer affect the first sample. The entire test condition would have been eliminated in the foregoing evaluation because the required 30 grains would not have fissured in the one sample. To make the split grain values meaningful, results of all the test conditions (Tables 1 - 6, Appendix) in which less than 50 grains in a sample incurred damage were added and then compared. Table 2 shows these results. The varieties align themselves in order of increasing re- sistance to fissuring as follows: A. Fortuna (least resistant) B. Zenith C. Bluebonnet 50' D. Belle Patna E. Rexoro F. Century Patna 231 (most resistant) These results compare favorably with those observed by summation of the adjusted mean response times. The Fortuna. Zenith and Century Patna 231 varieties assume 162 the same position in both evaluations. The total dif- ference among the other three varieties in the summed adjusted mean response times (Table l) is only 6.42 hours. The difference in total fissures (for conditions under which less than 50 grains per sample fissured) is only 14 (Table 2). Table 2.—-Results when the number of fissured grains of a given variety and age were added for the nine test conditions* for which less than 50 grains per sample fissured. Variety ‘ Year of barges; Combined 1961 1962 total Zenith 201 204 405 Fortuna 260 223 483 Bluebonnet 50 130 139 269 Rexoro 123 132 255 Belle Patna 127 132 259 Century Patna 231 _Zgl ._69 1134 Total 915 890 1805 / *38-11-35 38e11-60 38-60-100 68-11-34 68-55-87 68-55-100 92-11-33 92-51-86 92-63-100 163 Angladette (1964) reported that Montgrand (1958) observed splits perpendicular to the major axis of the grain after an interval of 15 days or more following moisture adsorption. This research did not reveal such occurrences. Fissures were slow in developing for the 11 to approximately 33 percent RH changes at all tempera- tures. After grains were in the dynamic system for 24 hours they were placed into a static container with the same relative humidity and inspected at 24-hour intervals until no further fissures developed. No additional fis— sures were observed after 96 hours of exposure. The final fissure count made after three weeks of equilibrating was almost always within the estimated counting tolerances mentioned in the sub-section on "Adsorption damage." Some general observations which can be made rela- tive to the data in Tables 1 - 6, Appendix, are: l. The adjusted mean response times and the stan- dard deviations of the response times become smaller as the temperature increases and essen- tially the same relative humidity changes are maintained. The vapor pressure differences vary inversely with the response times. 164 The grains submersed in 62°F water (68°F am- bient air) responded much faster than grains submersed in 38°F water (38°F ambient air). The adjusted mean response time and the stan— dard deviation for the grains at 62°F were approximately one-third those at 38°F. This observation is in agreement with those of Desikachar and Subrahmanyan (1961) which in- dicated that water temperature increases at levels below 1400F hastened the development of grain damage. Grains in equilibrium at 92°F and 51 percent RH have a moisture content of approximately 12.5 percent (db) or 11.2 percent (wb). This is only slightly below the moisture content at which rough rice is held in storage. If brown rice at these conditions is exposed (in single layers) to a 100 percent RH atmos- phere, approximately 50 percent of the grains will fissure within a period ranging between 0.8 to 1.6 hours after exposure depending on 165 the variety of rice. The response at this temperature and moisture content is surpris- ingly fast. Counts of fissures (Tables 7, 8 and 9, Appendix) were made simultaneously with the grain damage counts. Each table reports the maximum number of fissures which were observed in a sample of 50 grains at the specific temperature for the specified relative humidity change. A sample data sheet (Table 11, Appendix) illustrates how the cumulative grain damage and numbers of fissures were recorded. Table 12, Appendix, is a sample data sheet of weights which were taken of the grain samples. The ini— tial weight was determined after grains had equilibrated for more than three weeks at 11.2 percent RH and before they were placed into the 51.1 percent RH. The second weight was taken after nearly 24 hours in the dynamic system and before being moved into a static system of the same relative humidity. A third weight was determined after the grains equilibrated for three weeks or more and before the grains were placed into a vacuum oven to be 166 dried for 120 hours at 100°C and 28.5 inches of mercury vacuum. The final dry matter weight was determined after removal from the vacuum oven. These weights permitted calculation of the equilibrium moisture content of the grains at the two relative humidities to which the grains were subjected. The complete cycle for a grain sample under test required a minimum of six weeks without con- sidering the initial preparation time of the grains. Approximately 72 samples were run at each of the three temperatures. The equilibrium moisture contents determined from the foregoing tests are recorded in Table 10, Appendix. Each figure in the table is the average of two to six determinations. The EMC measurements which were possible can be visualized from the tabulated data on page 99- Breese (1955) found the hysteresis in the hygro- scopic equilibria of rough rice at 25°C (77°F) to be 1 percent or more throughout the interval from 20 to 80 percent RH. Rice grains for this research were initially at about 12 percent EMC. Therefore, the EMC determina- tions at lower moisture contents may tend to be a little 167 higher whereas those at higher moisture contents may tend to be a little lower than determinations which could be made by approaching the lower EMC conditions from extremely low moisture contents and the higher EMC conditions from extremely high moisture contents. Differences in equilibrium moisture content between 1961 and 1962 grains appeared negligible (Tables 4 13, 14 and 15, Appendix). The equilibrium moisture con— tents for the two ages of any variety at the three tem— peratures for the various humidities are tabulated in these tables. If age influenced equilibrium moisture content, then a difference should have appeared between the sums of the equilibrium moistures for the two age groups at a given temperature. In the 92°F tests (Table 15, Appendix) the differences in the sums between the two age groups of any variety was less than 0.4. This small variation suggested that there was no difference in the EMC for the grains of different ages. Do different grain varieties have different equi— librium moistures? Table 15, Appendix, indicates that differences probably do exist but they are small. The 168 greatest variation between the totals of the EMC percent- ages exists between the 1961 Zenith grains (66.59) and the 1962 Belle Patna grains (64.93). Assuming that age of grain is not a factor, then the Belle Patna variety' consistently had a lower EMC than did the Zenith variety. If the difference of the sums is divided by five (No. of RH conditions), the average difference is found to be 0.33 percent. This can be interpreted to mean that 1961 Zenith grains on the average had a 0.33 percent higher EMC at 92°F than did the 1962 Belle Patna grains. The same type of analysis can be made with any of the other varieties and ages of grain. Similar observations can be made at 38°F and 68°F. All grains equilibrating at 100 percent RH at 92°F developed heavy mold before the end of the three week equilibration period. These data were omitted from Table 15, Appendix. Some mold also developed at the 68°F. 100 percent RH level causing erratic data which were omit- ted from Table 14, Appendix. No mold growths were observed at the 38°F level. 169 The Zenith variety generally exhibited a higher EMC than did the other varieties. Therefore, it was used as a standard for comparison. The EMC percentages for a given variety, age and temperature were added. The sum— mations for the two ages of a variety were then added and divided by two to give an average value for the five hu- midity conditions. Values from other varieties were sub— tracted from the standard and then divided by five (No. of RH conditions) to give an average difference of the equilibrium moisture content. The values in Table 3 were developed by this procedure. Table 3.--The average differences in equilibrium moisture content for the varieties as shown when Zenith was used as the standard. .._.-- -.- -__-. _.‘_ ___._~_- __... --__....._—.__.--_—_.___.-_________—____-____.——__—_._ _. ..__._-_.i,‘_.-.._...._- _.__. .._- Variety Temperature (Over-all as“. 680F 920F‘fiw average Zenith Standard Standard Standard ----- Fortuna -0.084 —0.306 0.000 -0.l30 Bluebonnet 50 -0.004 -0.056 -0.064 ~0.04l Rexoro +0.014 -0.060 -0.088 —0.045 Belle Patna -0.258 -0.278 -0.260 -0.265 Century Patna 231 —0.290 -O.342 -0.228 -0.287 170 Results in Table 3 are not necessarily true for a particular temperature and relative humidity. Instead they represent the average differences observed at a par- ticular temperature. The over-all average represents an average for all three temperatures. To get a single EMC value to represent all varie- ties and ages of grains at a given temperature and rela— tive humidity, the EMC values for the respective varieties and ages were added and divided to obtain an average value. For example, the final average value of 9.42 percent EMC (Table 10, Appendix) at 38°F and 11.2 percent RH is an average of six determinations of each of the six varie- ties (3 determinations for each age group) or an average of 36 determinations. These averages were used to develop the EMC lines (db) which are plotted on the psychrometric chart (Fig. 13). EMC values determined in this research for brown rice are about 1 percent higher than those determined for rough rice at the same temperatures and relative humidities by Hogan and Karon (1955). Only a few of their points could 171 be checked since their work was at temperatures of 80. 94 and lllOF. The implication is that rice hulls hold less moisture per unit dry matter weight than the rice grain itself. This is confirmed by the work of Karon and Adams (1949) who reported that When the relative hu- midity at 25°C (77°F) was raised from 10 to 90 percent. the moisture content of rough rice increased from 4.4 to 17.6 percent (wb), of polished rice from 5.2 to 18.8 percent, of bran from 5.0 to 18.0 percent and of hulls from 3.7 to 15.3 percent. The work of Houston (1952) and that of Houston and Kester (1954) also showed brown rice to have a higher EMC than rice in any other form. Thus the results of Figure 13 are in prOper relationship to the work performed by other researchers. 5.4 Evaluation of Moisture Adsorption Only two varieties of grains having the same age were studied at the temperature levels of 38 and 68°F. The procedure was previously explained. A sample data sheet for determining the moisture adsorbed within the specified time intervals is illustrated in Table 16, 172 Appendix. Two samples (50 grains each) of a variety were used in each determination. Time intervals after which weights were taken varied for the two samples as the data sheet indicates. Since no two grain samples (50 kernels per sample) weighed the same at a particular EMC, some common means of evaluating the results was necessary. The unit dry matter weight (1 gram) was selected for this purpose. After the dry matter of a sample was determined. the previously collected data were adjusted to give the equivalents for 1 gram dry matter weight. These adjusted values for a pair of samples are tabulated in Table 17. Appendix. The first column indicates the time interval during which a certain weight of moisture per gram dry matter was adsorbed. Weights were taken to four decimal places while the correction to the unit dry matter basis was recorded to six. The corrected values (columns 2 and 4) were divided by the time interval (hours) to give an average rate of moisture adsorption. These were recorded to five decimal places (columns 3 and 5). 173 The rate of moisture adsorption per unit weight dry matter for the respective relative humidity changes was plotted against time (hours). This was done in Figs. 18, 19, 20 and 21. The reader is cautioned to observe the change in the ordinate scale in Figures 19 and 21. Each rate of moisture adsorption is plotted at the mid- point of the time interval which it represents, i.e.. the adsorption experienced during the initial hour of exposure is plotted at the one—half hour point and the adsorption experienced during the six to twelve hour interval is plotted at the nine hour point. Weight changes from adsorption within a time in- terval were small thus creating the possibility for large errors in weight measurement. For example, grains which were moved from 11.2 to 33.6 percent RH at 68°F experi- enced initial rates of moisture adsorption which were. smaller than 0.0008 gram per hour. Two weights (before and after time interval) were involved in determining the moisture adsorbed. With an accuracy of : 0.0001 gram in each weight, the total error could have been 25 percent. This would have been a large and normally 174 5.0 P Variety: Fortuna (Brown) Year of harvest: 1961 u E "' Time when fissures began o H x g: m psi AVP=0 . 045 [E Rate of moisture adsorption, . D \ . 6 ° :WQmw ‘ . K . '— — . .‘ .—.0072 1.0 L. Wpss .. 38-11-35 Avp, to.027 0 I V v . 11“ M o 5 10 15 . 20 Time, hrs Figure l8.—-The rate of moisture adsorption for a unit weight (dry matter) of Fortuna riCe at 38 F. -3 175 10.0 — Variety: Fortuna (Brown) Year of harvest: 1961 gmM x 10 gmdm A-u-Time when fissures began Rate of moisture adsorption. P81 51 2.0 +* AV 0 ps0,“ 1.0 _. 38'11‘55 P=0.150 6 - - O 5 4 l O 5 10 15 20 Time. hrs Figure l9.—-The rate of moisture adsorption for a unit weight (dry matter) of Fortuna rice at 68°F. 3 (9 176 5.0”' Variety: Century Patna (Brown) Year of harvest: 1961 A--Time when fissures began 4.0 I H O": HI x 3.0 2.5 as a ‘ Psi I‘.O45 N O T, Gammon I" O 38-1 - l 60 A—L——DN-P=e.055 "-L :3 38-11-35 0 5 10 . 15 20 Rate of moisture adsorption, Time. hrs. Figure 20.——The rate of moisture adsorption for a unit weighg (dry matter) of Century Patna 231 rice at 38 F. O -3 gmM x 10 Rate of moisture adsorption. 177 1.000 '- Variety: Century Patna (Brown) 9.0 _ Year of harvest: 1961 D "‘ Time when fissures began .9-.. Fissures occurred later 1'0 " WW O 5 10 15 20 Time, hrs Figure 21.--The rate of moisture adsorption for a unit weight (dry matter) of Century Patna 231 rice at 68 F. 178 unacceptable error. However, when successive weights were taken, a negative error in the first weight de- termination would have made the succeeding weight a greater positive value. Therefore the errors were not cumulative but instead tended to compensate each other. In the final analysis the error in total moisture ad- sorbed should not have been greater than : 0.0002 gram. A positive 0.0002 gram error would have been possible if the initial grain weight determination would have been negative and the final determination would have been positive by 0.0001 gram. If the errors had been reversed, the total adsorption weight would have been 0.0002 gram too small. The smallest gain experienced by a sample in a 47-hour period was 0.0040 gram. Thus a total error of 5 percent was possible in the total moisture adsorbed. Successive Opposing errors in mois— ture adsorption weights should have caused a scattering of the rate of moisture adsorption points When these were plotted with time. Some of this tendency is appar- ent but generally smooth curves resulted. 179 The moisture adsorption determinations were some- what critical in only the—11.2 to approximately 33 percent RH grain changes. For the other four test conditions the adsorption rates were greater, thus making the possible errors less important. The moisture adsorption tests were made to deter- mine if crack damage could be related to adsorption rates. No grain damage observations or crack counts were made. Instead, the time when fissuring began was determined from comparable grain damage tests and is indicated with an enclosed triangle On each of the graphs (Fig. 18, 19, 20 and 21). The highest adsorption rates occurred imme- diately after the grains were exposed to the more humid atmospheres. Fissuring did not start until after the period of peak adsorption, thus indicating that there was a lag between the highest rate of moisture adsorption and grain damage. For grains initially at a low EMC (9 percent or lower, db) fissures developed the fastest (lag was shortest) when the vapor pressure change to which the grains were subjected was the greatest. Grains with a higher initial moisture content (13 percent, db) developed fissures faster with a smaller vapor pressure change. 180 The rate of moisture adsorption increases with increasing vapor pressure changes, i.e., grains in equi- librium at 11.2 percent RH being moved into 34.8, 59.6 and 75.0 percent RH atmospheres (Fig. 18 and 20). Simi- lar observations can be made from Figures 19 and 21. When grains are in equilibrium at different ini- tial RH conditions at a given temperature and then are subjected to essentially the same vapor pressure increases. the grains with the higher moisture content will adsorb moisture much faster. This is strikingly illustrated by the figures and also in Table 4. This observation can be interpreted to mean that grains with a moderate EMC (approx. 14.5 percent, db) Will adsorb moisture very rapidly and respond fast by fissuring whenever they are subjected to a very humid atmosphere. They will either develop fissures within a very short period of time or they will not fissure at all. This statement is illustrated with the higher moisture con— tent grains in Figure 19. It is further illustrated in Tables 1 - 6, Appendix, where the adjusted mean response times and standard deviations are very small for all 181 grains initially in equilibrium at approximately 51 per- cent RH and 920F before being subjected to 100 percent RH at the same temperature. These observations are in agreement with those made by Breese (1955) who reported that equilibration by adsorption was extremely rapid at relative humidities above 50 percent. Also earlier in this thesis the ob- servation was reported that the crack pattern which de— veloped when dry grain (9.4 percent EMC or below) was subjected to a humid atmosphere (75.0 percent RH) was different from that which developed in grains having a higher initial EMC. The implication is that the mechan- ism within the grain which controls the rate of moisture adsorption can operate more effectively at the higher grain moisture. Close inspection of the plotted data points in Figure 18 shows that the fastest rate of moisture adsorp— tion occurred immediately after the grains were exposed to the more humid atmospheres. The rate then decreased un- tilthetmm>or four hour points were reached but then showed a tendency to increase before starting a smooth continuous 182 Table 4.——A comparison of the rate of moisture adsorption by grains initially at different equilibrium moisture contents before being subjected to ap- proximately the same vapor pressure increases. Initial Ratio Variety Initial Final AVP H20 adsorbed RH‘% EMC(db) RH %. psi in 23 hr, gm Temperature 38°F 7 Fortuna 59.6 15.2 86.7 0.030 0.0388 . Fortuna 11.2 9.4 34.8 0.027 0.0037 10.5 to Century Patna 231 59.6 15.2 86.7 0.030 0.0324 Century Patna 231 11.2 9.4 34.8 0.027 0.0028 11.6 to Fortuna 59.6 15.2 100.0 0.045 0.0552 Fortuna 11.2 9.4 59.6 0.055 0.0149 3.7 to Century Patna 231 59.6 15.2 100.0 0.045 0.0506 Century Patna 231 11.2 9.4 59.6 0.055 0.0118 4.3 to Temperature 680E Fortuna 54.9 13.9 86.6 0.107 0.0682 Fortuna 11.2 7.8 33.6 0.077 0.0074 9.2 to Century Patna 231 54.9 13.9 86.6 0.107 0.0522 Century Patna 231 11.2 7.8 33.6 0.077 0.0056 9.3 to l Fortuna 54.9 13.9 100.0 0.151 0.1230 Fortuna 11.2 7.8 54.9 0.150 0.0286 b 4.3 to 1 Century Patna 231 54.9 13.9 100.0 0.151 0.0971 Century Patna 231 11.2 7.8 54.9 0.150 0.0198 4.9 to l 183 decline. The same observation but less pronounced can be made from the other graphs. The occurrence is too consistent to attribute to errors in weight measurement especially since all weight measurements for a specified time interval were at least 10 minutes apart. This phe- nomenon remains for further investigation. VI SUMMARY AND CONCLUSIONS 6.1 Summary Preliminary research indicated that rice grains may fissure during or after the process of moisture ad- sorption from the atmosphere. These indications were verified by using saturated salt solutions in closed containers to produce desired relative humidities at a given temperature. Rice grains (brown rice) in equili- brium with ambient conditions fissured when they were subjected to higher relative humidities in the enclosed containers. Fissures which developed were observed with 3.5x optical lenses. Equipment which instantly produced the relative humidity specified for a particular saturated salt solu— tion was developed by using systems of five flasks (500 cc) connected in series and by bubbling compressed air 184 185 through them. Bubbling rate was from 250 to 400 cc per minute. Conditioned air was filtered before being passed through an air-tight observation and inspection chamber in which the test grains (50 per sample) were arranged in an orderly manner for making counts of damaged kernels and total fissures. Grain samples of six varieties and two ages of rice were hulled, inspected and equilibrated (three weeks) at each of three temperatures and three relative humidities prior to performing any grain damage tests. Grain samples from the equilibrated conditions were weighed before being subjected to five or six (depending on the temperature) higher relative humidity conditions at each temperature. The number of kernels which fissured in samples of 50 grains and the total number of fissures which developed were observed at appropriate time intervals after the grains were exposed to the higher relative humidity con- ditions. Grain samples equilibrated (three weeks) at the higher humidities before they were removed, inspec- ted, weighed and placed into a vacuum oven to obtain dry matter weight. Equilibrium moisture contents were deter- mined at the initial and final relative humidity conditions. 186 These EMC values were plotted on a psychrometric chart and equilibrium moisture content lines (db) from 4 to 28 percent were extrapolated from these points. Grain samples of two varieties and one age were hulled and inspected before being equilibrated at each of two temperatures and three relative humidities before any moisture adsorption studies were made. Grains from the equilibrated conditions were subjected to five higher relative humidities and were weighed at predetermined intervals after exposure. Following 47 hours in the higher humidity atmosphere, the grains were removed and placed into a vacuum oven to obtain dry matter weight. Weights of moisture adsorbed during the 47 hours after exposure to the higher humidity were corrected to a basis of 1 gram dry matter weight. The rates of moisture ad- sorption during the intervals between which weights were taken were determined and plotted on a "Rate of moisture adsorption per unit weight dry matter" versus "Time after exposure" graph. An analysis was made of the collected data and the results are reported in this thesis. 187 6.2 Conclusions The conclusions reported are drawn from research conducted with brown rice. The author feels that the re- sults are applicable generally to a wide variety of rice grains. The more pertinent conclusions follow. 1. Rice kernels equilibrated at a given relative humidity had lower equilibrium moisture contents at higher temperatures, i.e., 11.2 percent RH, 38°F yielded 9.42 percent EMC grain. 11.2 percent RH, 68°F, yielded 7.80 percent EMC grain, 11.2 percent RH, 92°F yielded 6.62 percent EMC grain. 2. Differences in equilibrium moisture content between grains harvested in 1961 and 1962 of a given va- riety were inconsistent and appeared negligible. When equilibrium moisture contents for a given age and variety at a given temperature but for five different humidity conditions were added and compared to the same sum for the same variety of different age, the greatest differ- ence between the sums of the two age groups in a variety was less than 0.40 at 92°F. At 68°F, fresher grains con- sistently showed slightly higher sums while at 38°F these grains consistently showed slightly lower sums for all varieties. 188 3. Different grain varieties have different equi- librium moisture contents. When average differences were determined for the temperature 38, 68 and 920F, the vari— eties aligned themselves from highest average to lowest average EMC values as follows: 1) Zenith, 2) Bluebonnet 50, 3) Rexoro, 4) Fortuna, 5) Belle Patna and 6) Century Patna 231. On the average Century Patna 231 grains had a 0.29 percent lower equilibrium moisture content than did Zenith grains. 4. More (about 20 percent) older grains of the Fortuna and Century Patna 231 varieties developed fissures than did fresher grains when all sets in nine test condi- tions for which less than 50 grains per sample fissured were considered. In all other varieties, more fresher grains (about 4.5 percent more) developed fissures. No conclusion (covering all varieties) was reached stating that one age of grain was more subject to fissuring damage than another. 5. Some grain varieties are more subject to de- veloping grain damage than others. When damaged grains of both ages of a variety were added for all sets of test 189 conditions in which less than 50 grains per sample fis- sured, the grains aligned themselves in the following order of least to most resistant: l) Fortuna, 2) Zenith, 3) Bluebonnet 50, 4) Belle Patna, 5) Rexoro and 6) Cen— tury Patna 231. The Fortuna, Zenith and Century Patna 231 varieties rather consistently assumed the first. second and sixth positions, respectively, in the grain fissuring evaluations. Variations in grain damages among the other three were small and inconsistent. 6. Grains in equilibrium at 92°F and 51 percent RH had a moisture content of approximately 12.5 percent (db) or 11.2 percent (wb). This is only slightly below the moisture content at which rough rice is held in storage. In this research when brown rice at these conditions was exposed (in single layers) to a 100 per- cent RH atmosphere, approximately 50 percent of the grains fissured within a period ranging from 0.8 to 1.6 hours after exposure depending on the variety of rice. 7. Rice grains after being subjected to a more humid atmosphere at a constant temperature develOped 190 fissures. The response (fissuring) was not immediate but occurred sometime after the grain had been exposed. When a group of grains was subjected to a more humid condition, the individual grains responded (developed damage) at different time intervals after exposure. The response times of the grains were observed and were found to have a normal distribution. 8. Rice varieties more susceptible to grain damage incurred fissures quicker after exposure than did less susceptible varieties. The time required after exposure for 50 percent of the grains in a sample to in- cur damage was designated as the adjusted mean response time and was used as a measure of grain resistance to fissuring. For a given relative humidity increase, the smaller adjusted mean response time meant little resis- tance whereas the larger adjusted mean response time meant more resistance to grain damage. When the ad- justed mean response times for the respective grain varieties were added for both ages of grain and for the seven sets of test conditions in which 30 or more grains fissured, the varieties aligned themselves in 191 the following order of smallest to largest sum of adjusted mean response times: 1) Fortuna, 2) Zenith, 3) Rexoro, 4) Bluebonnet 50, 5) Belle Patna and 6) Century Patna 231. 9. The initial response time and the adjusted mean response time were less and the rate of response was faster (smaller standard deviation) for rice grains of a given variety equilibrated at a given temperature and relative humidity (i.e., 92°F, 11.2 percent RH) when these grains were subjected to increasingly higher rela- tive humidities at the same temperature. 10. The adjusted mean response times and the standard deviations for different grain varieties equili- brated at the same initial condition were different when these grains were subjected to a more humid atmosphere. Differences in adjusted mean response times and standard deviations were found to be as great as 100 percent be- tween the Fortuna and Century Patna 231 (least and most resistant, respectively) varieties. 11. Rice grains Which had equilibrated at the same temperature but at different relative humidities responded differently to essentially equal vapor pressure 192 increases. Fortuna and Century Patna 231 rice (harvested in 1961) in equilibrium at 38°F and 59.6 percent RH ad- sorbed moisture more than ten times as fast over a 23- hour period when subjected to a 0.030 psi vapor pressure increase as did other grains of the same varieties equili- brated at 11.2 percent RH and subjected to a 0.027 vapor pressure increase. Similar results, somewhat less pro- nounced, were experienced for greater vapor pressure in- creases with the same grain varieties and base conditions. Analysis of the weight adsorption data at 68°F showed essentially the same type of response. 12. Rice grains in equilibrium at a given tempera- ture but at different moisture contents responded differ- ently in terms of grain damage when the kernels were sub— jected to essentially the same vapor pressure increases. Century Patna 231 in equilibrium at 38°F, 11.2 percent RH (9.4 percent, EMC) incurred 100 percent damaged grains when subjected to a vapor pressure increase of 0.072 psi. The same variety of grain at equilibrium with 68°F, 11.2 percent RH (7.8 percent EMC) incurred only 7 percent damage when subjected to a vapor pressure increase of 0.077 psi. 193 Therefore, grains at equilibrium at a given humidity but at a higher temperature (lower EMC) were capable of with- standing essentially the same vapor pressure changes with less grain damage. For the preceding conditions Fortuna rice also suffered 100 percent grain damage when subjec- ted to the 0.072 psi vapor pressure increase at 38°F and 77 percent grain damage when subjected to the 0.077 psi vapor pressure increase at 68°F. Studies of the other varieties showed results which fell between the two ex— tremes. REFERENCES Adair, Roy C. 1952 The McGill Miller method for determining the milling quality of small samples of rice. The Rice Journal. 55(2):21. Angladette, A. 1956 Examen des pratiques culturales exercant une influence sur l'emmagasinage et le traitement du riz en Afrique Tropicale Francaise. Groupe de Travail sur l'emmagasinage et l'usinage du riz - Commission Internationale du Riz. Cal— cutta. p. miméo.‘ Angladette, A. 1960 Examen des modalités de récolte et de séchage exercant une influence sur la qualité du riz et le rendement a l'usinage en Afrique Tropi- cale d'expression francaise. Groupe de tra— vail de la mécanisation de la production, de l'emmagasinage et de l'usinage du riz - Com- mission Internationale du riz - Saigon. IRC/WP/60/SP18 - miméo. Angladette, A. 1964 Agricultural Engineering, rice drying principles and techniques. United Nations Food and Agri- culture Organization. Informal Working Bulletin No. 23. Rome, Italy. 73 pp. Aten, A. and A. D. Faunce 1953 Equipment for the Processing of Rice. Food and Agricultural Organization of the United Nations. Rome, Italy. 55 pp. 194 195 Autrey, H., et al. 1955 Effect of milling conditions on breakage of rice grains. Journal of Agricultural and Food Chemistry. 3(7):593-599. Barkas, W. W. 1948 The Swelling of Wood Under Stress. London, England: His Majesty's Stationery Office. 102 pp. Basio, Engracio, et al. 1958 Abstract Bibliography of Rice, 1951-1955. University of the Philippines College of Agri- culture. College, Laguna, Philippines, 285 pp. Basio, Engracio, et al. 1959 Abstract Bibliography of Rice, 1941-1950. Uni- versity of the Philippines College of Agricul- ture. College, Laguna, Philippines. 314 pp. Basio, Engracio, et al. 1960 Abstract Bibliography of Rice, 1926-1940. University of the Philippines College of Agri- culture. College, Laguna, Philippines, 182 pp. Basio, Engracio, et al. 1962 Abstract Bibliography of Rice, 1956—1960. University of the Philippines College of Agri- culture. College, Laguna, Philippines, 618 pp. Bienvenido, O. J., G. M. Bautista, J. C. Lugay and A. C. Reyes. Studies on the physicochemical proper- ties of rice. Journal of Agricultural and Food Chemistry. 12(2):131-138. Breese, Michael 1955 Hysteresis in thg hygroscopic equilibria of rough rice at 25 C. Cereal Chemistry. 32: 481-487. Burmistrova, M. F., et al. ’1956 Physicomechanical PrOperties of Agricultural Crops. Jerusalem: Translation IPST. 1963. 250 pp. 196 Coleman, D. A. and H. C. Fellows 1925 Hygroscopic moisture of cereal grains and flax- seed exposed to atmospheres of different rela- tive humidity. Cereal Chemistry. 2(5):275-287. Coyaud, Y. 1950 Le Riz. Etude botanique, genetique, physio- logique, agrologique et technologique appli- quée a l'Indochina. Archives de l'Office In- dochinois du Riz. (Saigon). No. 50. 312 pp. Dachtler, W. C. (editor) 1959 Research on Conditioning and Storage of Rough and Milled Rice. ARS, USDA. 55 pp. Desikachar, H. S. R. and V. Subrahmanyan 1959 Expansion of new and old rice during cooking. Cereal Chemistry. 36(4):385-39l. Desikachar, H. S. R. and V. Subrahmanyan 1961 The formation of cracks in rice during wetting and its effect on the cooking characteristics of the cereal. Cereal Chemistry. 38(4):356- 357. ' Dobelmann, J. P. 1955 Note sur l'influence of the date of récolte et du mode de séchage du paddy sur 1e taux de brisures a l'usinage. Rix et Riziculture et Cultures Vivriéres Tropicales (Paris) 3 éme Tr.:99-100. Dobelmann, J. P. 1961 Manuel de Riziculture amelioree. Tananarive. Earle, Paul L. and Norman H. Ceaglske 1949 Factors causing the checking of macaroni. Cereal Chemistry. 26(4):267-286. Freund, John E. 1960 Modern Elementary Statistics. Second Edition. Englewood Cliffs, New Jersey: Prentice Hall, Inc. 413 pp. 197 Grist. D. H. 1959 Rice. Third Edition. New York: Longmans, Green and Co. Ltd. 466 pp. Grosh, G. M. and Max Milner 1959 Water penetration and internal cracking in tem- pered wheat grains. Cereal Chemistry. 36:260- 273. Halick, John V. and Kenneth K. Keneaster 1956 The use of a starch-iodine-blue test as a quality indicator of white milled rice. Cereal Chemistry. 33(5):315-319. Hall, Carl W. 1957 Drying Farm Crops. Ann Arbor, Michigan: Edwards Brothers, Inc. 336 pp. Henderson, S. M. 1954 The causes and characteristics of rice checking. The Rice Journal. 57(5):l6,18. Henderson, S. M. 1957 Milled rice yields. California Agriculture. ll(7):6, 15. Hogan, J. T. 1962 Chemical Engineer, Food CrOps Laboratory. USDA. SURDD. New Orleans, Louisiana. Personal corres- pondence, Jan. 26. Hogan, Joseph T. and Melvin L. Karon 1955 Hygroscopic equilibria of rough rice at elevated temperatures. Agricultural and Food Chemistry. 3(10):855-859. Hogan, J. T., R. A. Larkin and M. M. MacMasters 1954 X—ray and photomicrographic examination of rice. Journal of Agriculture and Food Chemistry. 2(24):1235-1239. 198 Houston, David F. 1952 Hygroscopic equilibrium of brown rice. Cereal Chemistry. 29(1):71-76. Houston, David F. 1962 Chemist, Cereal Investigations Field Crops Laboratory. USDA, WURDD. Albany, California. Personal correspondence, Feb. 26. Houston, D. F. and E. B. Kester 1954 Hygroscopic equilibrium of whole-grain edible forms of rice. Food Technology. 8(6):302-304. Karon, M. L. and Mabelle E. Adams 1949 Hygroscopic equilibrium of rice and rice frac- tions. Cereal Chemistry. 26(1):l-12. Kik, M. C. 1951 Nutritive studies of rice. Agricultural Experi- ment Station Bulletin 508. University of Arkan- sas College of Agriculture. 48 pp. Kondo, M. and T. Okamura 1929 Der durch die feuchtigkeits zunahme verursachte guerris (Doware) des reiskorns. Ohara Inst. f. Landw. Forsch. Ber 24:163-171. Kramer, Harold A. 1951a Engineering aspects of rice drying. Agricultural Engineering. 32(1):44-45, 50. Kramer, Harold A. 1951b Physical dimensions of rice. Agricultural Engineering. 32(10):544—545. Langfield, E. C. B. 1957 Time of harvest in relation to grain-breakage on milling of rice. J. Australian Inst. Agric. Sci. 23(4):340-34l. 199 Milner, Max, J. A. Shellenberger, M. R. Lee and Robert Katz 1952 Internal fissuring of wheat due to weathering. Nature. 170(4325):533. Montgrand, P. de. 1958a Contribution a l'étude du séchage du Riz. Bulletin d'information des Riziculteurs de France. (58):5-14. Montgrand, P. de. 1958b Contribution a l'étude du séchage du Riz. Bulletin d'information des Riziculteurs de France. (59):10—26. Naumow, I. 1959 Eindringen und umsetzen der feuchtigkeit in Weizenkorn. (Entrance and movement of mois- ture in the wheat kernel). Muhle. 96(34)460. Rice Millers' Association 1963 U. S. rice acreage statistics. The Rice Journal. 66(12):15—16. Rice Millers' Association 1964 Rice production in the United States, 1963. The Rice Journal. 67(1):28-29. Rice Research and Marketing Advisory Committee 1963 Report and recommendations of the Rice Research and Marketing Advisory Committee. The Rice Journal. 66(6):ll-l4. Rivenburgh, D. A. 1962 Rice Marketing Specialist, Grain and Feed Div- ision. USDA, FAS. Washington 25, D.C. Per- sonal correspondence. April 11. Roberts, R. L., et al. 1954 Effect of processing conditions on the expanded volume, color and soluble starch of parboiled rice. Cereal Chemistry. 31:121-129. 200 Slusher, M. W. and Troy Mullins 1952 Smith, W. 1951 Rice mill yield and grade in relation to variety and method of harvest. Arkansas Agricultural Experiment Station Bulletin 526. 36 pp. D. and Walter McCrea, Jr. Where breakage occurs in the milling of rice. Rice Journal. 54(2):l4-15. Stahel, Gerold 1935 Swanson. 1943 Ten Have, 1959 Ten Have, 1961 Umalo, D. 1956 Breaking of rice in milling in relation to the condition of the paddy. Tropical Agriculture. 12(10):255-260. C. D. Effects of moisture on the physical and other properties of wheat. II. Wetting during har— vest. Cereal Chemistry. 20(1):43-6l. I. H. De invloed van het tijdstip van de oogst op de kwaliteit van enkele rijstrassen. De Surinamse Landbow (Paramaribo). 6:201-209. I. H. The testing of new promising varieties for their yield of total milled rice and breakage resis- tance. Working Party on Rice Production and Pro- tection — International Rice Commission. (New Delhi) IRC/WP/6l/RPP. 67-10 pp. L., M. C. Silverio and I. S. Santos A preliminary study of some factors affecting the milling recovery of rice in the Philippines. The Philippine Agriculturist (Laguna). 40(2): 69-77. 201 United States Department of Agriculture 1961 Seeds - The Yearbook of Agriculture. Washington 25, D. C.: U. S. Government Print- ing Office. 591 pp. Wilson, Robert E. 1921 Humidity control by means of sulfuric acid solutions, with critical compilation of vapor pressure data. The Journal of Industrial and Engineering Chemistry. 13(4):326-33l. A PPEND IX 203 Table l. - A summary of calculations and grain damage data for the Zenith rice variety. ZENITH __ 125; - £62....— Code AVP Grains Adj 5": s s- Grains Adj R s 3- psi split* hr hr min split* hr hr m 38—11-35** .027 2 ~-- --- -- 0 -- -- -- 68-11-34 .077 38 39.37 14.52 142.0 28 --' a-s‘ '—- 92-11-33 -159 43 21.79 8.97 82.8 46 20.30 10.14 90.0 38—11-60 .055 1.2. 17.81. 4.1.6 1.0.1. 1.2 22.03 6.70 62.8 68-11-55 .150 50 7.61 1.80 15.3 50 7.88 1.83 15.5 92-11-51 .295 50 4.89 0.88 7.5 50 4.97 1.33 11.3 38-11-75 .072 50 9.67 2-45 20.8 50 10.69 2.60 22.1 68-11-75 .219 50 3.47 0.61 5.2 50 3.47 0.78 6.6 92-11-75 .475 50 1.78 0.36 3-0 50 1.58 0.39 303 38-60-87 .030 -— -—- -—— —- -. -—— ——— -—- 68-55-87 .107 0 _-— -—- --- 3 --- --- --- 92-51-86 .259 43 1.80 0.56 5.1 47 1.73 0.65 5.7 38-60-100 .045 o --- ——- --- o --- —-- -- 68-55—100 .151 2 -- --- --— 4 --- --- --- 92-51-100 .363 48 1.02 0.46 4.0 50 1.02 0.40 3.4 38-68-H20 --- 48 2.60 1.15 10.0 49 2.52 1.12 9.6 68-65-H2 O -- 50 0.70 0.41 3.5 50 0.72 0.29 2.46 92'63‘100 0267 29 -‘"’ """' "'""' 3'5 1 01.0 0 068 6 09 * Out of a total of 50 grains. f ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected ’to observe the fissuring which resulted. 204' Table 2. -- A summary of calculations and grain damage data for the Fortuna rice variety. FORTUNA - pg? 11961 *‘ '19 Code AVP Grains Adj 5": s s— Grains Adj '3': s s- ' psi split* hr hr mIn split* hr hr min 38-11-35** .027 o —--* --- --— 3 --— ---' a-- 68-11-34 .077 43 26.24 11.46 105.0 34 26.03 8.70 89.4 92-11-33 .159 48 14.02 5.58 48.3 41 12.51 6.20 58.1 38-11-60 .055 47 13.10 4.36 33.4 33 15.73 4.76 46.2 68-11-55 .150 50 5.89 1.48 12.5 50 6.08 2.18 18.5 92-11-51 ~295 50 3.86 1.34 13.7 50 4.15 1.69 14.3 38-11-75 .072 50 7.02 1.67 14.2 50 7.99 2.42 20.5 68—11-75 .219 50 1-91 0.54 4:6 50 1.84 0.61 502 92’11‘75 0475 50 1 067 0 04,1 3 05 50 1 051 0 04,2 3 06 38—60-87 .030 - --- —- --- -- --— -- --— 68-55-87 .107 0 --- _-— --- 0 ... --- --- 92—51-86 .259 49 1.60 0.67 5.8 43 1-47 0.72 6.7 38-60-100 .045 7 —-- -—— --— 4 -- —.- --- 68-55-100 .151 25 --- --- --- 27 -- --— —-- 92-51-100 .363 50 0.83 0.47 4.0 50 0.69 0.52 4.4 38-68-820 -—- 50 1.06 0.48 4.1 50 1.46 1.06 9.0 68-65-100 .117 2 --- --- —-— 0 —-- -—- —-- 92-63-100 .267 41 0.97 0.53 4.9 33 1.10 0.64 6.7 * Out of a total of 50 grains- ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected to observe the fissuring which resulted. 205 Table 3. -- A summary 0f calCulations and grain damage data for the Bluebonnet 50 rice variety. BLUEBONNET 50 ‘1 6 “1 6' Code AVP Grains Adj x s s- Grains Adj 1 s s— ,_ psi split* hr hr mi. split* hr hr mi. 38-11-35** .027 o --- --- --- o -- --- --- 36-11-34 .077 26 --- --- --— 24 ' --- -- -- 92-11-33 .159 30 21.68 11.40 124.8 35 19.14 8.04 81.6 38-11—60 -055 34 28.26 5.48 56.4 33 29.12 7.72 80.6 68-11-55 .150 50 7.20 2.06 1705 50 7-72 2-27 1903 92-11-51 .295 50 4.71 1.44 12.2 50 5.72 1.95 16.6 38-11-75 .072 50 9.61 3.25 27.6 50 10.19 3.50 29.7 68-11-75 .219 50 3 .52 0 .92 7.8 50 3 ~95 1-34 13-7 92-11-75 .475. 50 1.77 0.38 3.3 50 1.85 0.56 4.7 38—60—87 .030 -- —- -- --- - -—- -- -- 68-55-87 .107 0 -.— —-- —-- 0 --- ——- --— 92-51-86 .2 59 30 l . 85 0 . 56 6 .1 27 -- ---— --- 38-60-100 .045 1 --- --- --- 1 -- --- --— 68-55-100 .151 1 —-- -—- --- 0 --- ---- -- 38-68-H20 -- 48 2.15 1.05 9.1 45 1.62 0.73 7.0 92-63-100 «267 8 -- --- -- 19 '-- '- -- * Out of a total of 50 grains. ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected to observe the fissuring which resulted. 206 Table 4. -- A summary of calculations and grain damage data for the Rexoro rice variety. REXORO , “126; l 1 6 Code 45V? Grains Adj 1 s s- ’Grains Adj 1' 3 psi split’ hr hr mi. split* hr hr 38-11—35** .027 0 --- --- --- 0 --— -—— _.- 68-11-34 .077 24 --- -- -- 20 -- -- --— 92-11-33 .159 32 15.96 7.44 78.6 38 18.60 7.44 72.6 38-11-60 .055 40 25.20 6.84 64.8 36 29.30 8.48 84.8 68-11-55 .150 50 9.94 2.61 22.1 50 9.20 3.07 26.0 92-11-51 .295 50 4.34 1.12 9-5 50 4.57 1.44 12.2 38-11-75 .072 50 9.45 3.32 28.2 50 9.25 3.11 26.4 68'11‘75 0219 50 3 060 0 071 6 00 50 3 015 0093 709 92-11-75 .475 50 2.12 0.62 5.3 50 1.99 0.60 5.1 38-60-87 .030 -- -- --- --— -- --- --- --— 68-55-87 .107 0 --- —-- —-- 0 --— --- --— 92-51-86 .259 7 --— --- --- 16 --- --- --— 38-60-100 .045 0 --— --- --- 0 --- -- --- 68-55-100 .151 0 --- --- -—- 3 --- --- -- 92-51-100 .363 48 1.40 0.58 5.0 50 1.25 0.42 3.6 68-65-Hgo --— 50 0.43 0.20 1.7 50 0.67 0.30 2.6 92-63-100 .267 20 -- --- --- 19 ---. --- —- * Out of a total of 50 grains. ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected to observe the fissuring which resulted. 207 Table 5. -- A summary of calculations and grain damage data for the Belle Patna rice variety. -_.—...—_._—.—- M—_———_..—__ M.‘ _ i..._-. _._.. i _ ._ -q— .- -._._ BELLE PATNA 1961 19’2 Code AVP Grains Adj 1 s s; Grains Adj 1: s 8- psi split* hr hr min split* hr hr mi. 38—11-35** .027 0 --- —-- --- 0 --- —-— -—— 68-11-34 .077 23 --- --- --— 19 -- --- —- 92-11-33 .159 32 20.49 8.22 87.0 35 24.06 9.42 95.4 38-11-60 .055 49 29.38 7.68 65.8 49 32.40 7.92 67.9 68-11-55 .150 50 7.13 2.26 19.2 50 8.23 2.52 21.4 92-11-51 .295 50 4.97 1.20 10.1 50 5.31 1.70 14.4 38-11-75 .072 50 9.61 2.69 22.8 50 10.53 2.93 24.8 68—11-75 .219 50 3.73 1.07 9.1 50 3.96 1.23 10.4 92-11-75 .475 50 1.99 0.50 4.3 50 1.97 0.50 4.3 38-60-87 .030 -- ——- -- —-- - ——- --- --- 68-55-87 .107 0 --— —-— -—- 0 -—- -—- -- 92-51-86 .259 13 -- —-- -—- 18 --- —-- —-- 38-60-100 .045 0 --— -—— --— 0 —- --- -- 68-55—100 .151 1 -- —-— -—- 1 -— —-- --- 92-51-100 .363 50 0.99 0.49 4.2 49 0.98 0.42 3.6 38-68-H20 --- 50 1.81 0.52 4.4 49 1.87 0.85 7.3 68-65-H20 --- 50 0.58 0.21 1.8 50 0.67 0.25 2.1 92-63-100 .267 9 --- --- -- 10 -- -- --- * Out of a total of 50 grains. ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected to observe the fissuring which resulted. 208 Table 6. -- A summary of calculations and grain damage data for the Century Patna 231 rice variety. if ' CENTURY PATNA 231 -’ ' 126; ‘ 126g - Code lXVP Grains Adj 5 s s Grains Adj i s 3- psi split* hr hr min split* hr hr .1. 38—11-35** .027 0 -- --- —- 0 --- -- --- 68-11-34 .077 5 -—- --- ——- 2 —-- --- --- 92-11-33 .159 22 --- --- -- 16 --- --- —-- 38-11-60 .055 39 24.66 7.52 72.3 37 28.22 45.76 56.8 68-11-55 .150 50 8.53 3.12 26.5 50 9.50 3.23 27.4 92-11-51 ~295 50 5'53 1.65 14-0 50 6.93 1-45 12.3 38-11-75 ~072 50 10.67 2.90 24.6 50 11.88 3.62 30.7 68-11-75 .219 50 4.96 1.21 10.3 50 5.49 1.23 10.4 92-11-75 .475 50 1.88 0.47 4.0 50 2.21 0.44 .3.8 38-60-87 .030 -- -—- --. -- -- -- -—— -- 68-55-87 .107 0 --- -— —.-‘ 0 -—— —- -— 92-51-86 .259 6 -— —-- -— 3 --- --- -- 38—60—100 .045 0 —-— --— -—- _0 -—— -—- _--- 68-55-100 .151 0 -—— --— —_— 0 --- --- --— 92-51-100 .363 40 1.48 0.47 4.4 46 1.48 0.41 3.6 38-68-H20 -- 38 2 .30 0.94 9.16 22 -—- --— --— 68-654320 -- 49 0.77 0.29 2.71 48 0.85 0.29 2.5 92-63-100 .267 2 -- ---. -- 2 -- --- --- * Out of a total of 50 grains- ** The first number represents temperature, the second represents the initial equilibrium relative humidity of the grain and the third represents the relative humidity to which the grains were subjected to observe the fissuring which resulted. 209 Table 7. -- The maximum number of fissures which were observed in samples of 50 grains at 38°F when the indicated rel- ative humidity changes were made. m 1961 l 1963 Variety Grains Total lw Grains Total cracked cracks cracked cracks 11.2 to 32.8 percent RH Zenith 2 2 0 0 Fortuna 0 O 3 3 Bluebonnet 50 0 O O O Rexoro O O O 0 Belle Patna O O O 0 Century Patna 231 O O O 0 11.2 to 59.6 percent RH Zenith 44 79 42 66 Fortuna 47 109 38 87 Bluebonnet 50 ' 34 60 33 64 Rexoro 4O 83 36 72 Belle Patna 49 112 49 102 Century Patna 231 39 72 37 67 11.2 to 75.0 percent RH Zenith 50 254 50 251 Fortuna 50 349 50 315 Bluebonnet 50 50 243 50 242 Rexoro 50 235 50 281 Belle Patna 50 222 50 236 Century Patna 231 50 214 50 222 Table 7. -- Continued. 210 1961 196 Variety Grains Total Grains Total cracked cracks ‘ cracked cracks 59.6 to 100 percent RH Zenith O 0 0 0 Fortuna 7 8 4 4 Bluebonnet 50 l l 1 l Rexoro 0 0 0 0 Belle Patna O O O 0 Century Patna 231 0 O O O 68.4 percent RB to water Zenith 48 101 49 73 Fortuna 50 200 50 135 Bluebonnet 50 48 128 45 105 Rexoro 48 164 44 118 Belle Patna 50 164 49 118 Century Patna 231 38 75 22 28 211 Table 8. - The maximum number of fissures which were Observed in samples of 50 grains at 68°F when the indicated rel- ative humidity changes were made. 1961 1962 Variety Grains Total Grains Total cracked cracks cracked cracks 11.2 to 33.6 percent RH Zenith 38 45 28 36 Fortuna 43 77 34 70 Bluebonnet 50 26 31 24 33 Rexoro 24 40 20 26 Belle Patna 23 27 19 20 Century Patna 231 5 7 , 2 2 11.2 to 54.9 percent RH Zenith 50 295 50 286 Fortuna 50 400+ 50 358 Bluebonnet 50 50 271 50 215 Rexoro 50 307 50 327 Belle Patna 50 273 50 264 Century Patna 231 50 297 50 268 11.2 to 75.5 percent RH Zenith 50 400+ 50 379 Fortuna 50 400+ 50 377 Bluebonnet 50 50 400+ 50 367 Rexoro 50 400+ 50 400+ Belle Patna 50 400+ 50 396 Century Patna 231 50 400+ 50 370 212 Table 8. -- Continued. 1961 l 1962 1 Variety Grains Total I Grains Total cracked cracks cracked cracks 54.9 to 86.6 percent RH Zenith 0 0 3 3 Fortuna O 0 0 0 Bluebonnet 50 0 0 0 0 Rexoro O O O 0 Belle Patna 0 0 0 0 Century Patna 231 0 0 O 0 54.9 to 100 percent RH Zenith 2 2 4 4 Fortuna 25 27 27 28 Bluebonnet 50 l 2 0 0 Rexoro 0 0 3 3 Belle Patna l l 1 1 Century Patna 231 0 0 0 0 65.3 to 100 percent RH Fortuna 2 2 l 0 0 65 e 3 t O Wat er Zenith 50 211 50 199 Bluebonnet 50 50 277 50 241 Rexoro 50 357 50 231 Belle Patna 50 270 50 238 Century Patna 231 49 180 48 131 213 Table 9. -- The maximum number of fissures which were observed in samples of 50 grains at 92°F when the indicated rel- ative humidity changes were made. 161 l l 6 Variety Grains Total Grains Total cracked cracks cracked cracks 11.2 to 32.6 percent RH Zenith 43 7O 46 79 Fortuna 48 111 41 102 Bluebonnet 50 30 55 35 57 Rexoro ‘ 32 66 38 78 Belle Patna 32 45 35 48 Century Patna 231 22 36 16 22 11.2 to 51.1 percent RH Zenith 50 302 50 291 Fortuna 50 364 50 276 Bluebonnet 5O ‘ 50 367 50 274 Rexoro 50 321 50 313 Belle Patna 50 330 50 318 Century Patna 231 50 310 50 265 11.2 to 75.4 percent RH Zenith 50 Shattered 50 Shattered Fortuna 50 Shattered 50 Shattered Bluebonnet 50 50 Shattered 50 Shattered Rexoro 50 Shattered 50 Shattered Belle Patna 50 Shattered 50 Shattered Century Patna 231 50 Shattered 50 Shattered Table 9. - Continued. 214- ' 1 ‘ 1 Variety Grains Total Grains' Total cracked cracks cracked cracks 51.1 to 86.1 percent RH Zenith 43 61 47 71 Fortuna 49 96 43 86 Bluebonnet 50 30 41 27 44 Rexoro 7 7 16 20 Bella Patna 13 15 18 19 Century Patna 231 6 6 3 3 51.1 to 100 percent RH Zenith 48 92 50 98 Fortuna 50 125 50 154 Bluebonnet 50 38 69 42 88 Rexoro 48 107 50 128 Belle Patna 50 129 49 134 Century Patna 231 40 81 46 85 62.7 to 100 percent RH Zenith 29 31 35 37 Fortuna 41 66 33 48 Bluebonnet 50 8 11 19 25 Rexoro 20 25 19 21 Bella Patna 9 ll 10 14 Century Patna 231 2 3 2 3 215 Table 10. - Equilibrium moisture contents which were determined after brown rice* had equilibrated for three weeks or more. 38°F 68 9 Variety RH % EMC** RE 1 EMC** RH % EMB** 7. (db) % (db) 8 (db) Zenith 11.2 9.35 11.2 7.78 11.2 6.60 Fortuna 11.2 9.44 11.2 7.90 11.2 6.73 Bluebonnet 50 11.2 9.50 11.2 7.86 11.2 6.68 Rexoro 11.2 9.43 11.2 - 7.77 11.2 6.63 B8116 Patna 1102 9044 1102 7076 1102 6055 Century Patna 231 11.2 9.38 11.2 7.74 11.2 6-53 Average 11.2 9.42 11.2 7.80 11.2 6.62 Zenith 34.8 11.43 33.6 10.69 32.6 9.57 Fortuna _ 3408 11042 3306 10065 3206 9.64 Bluebonnet 50 3408 11054 3306 10076 3206 9.68 Rexoro 3408 11043 3306 10078 3206 9°62 B6116 Patna 3408 11026 3306 10073 3206 9052 Century Patna 231 34.8 11.27 33.6 10.62 32.6 9.43 Average 34.8 11.39 33.6 10.70 32.6 9.58 Zenith 59.6 15.18 54.9 13.99 51.1 12.93 Fortuna 59.6 15.26 54.9 13.74 51.1 12.78 Bluebonnet 50 59.6 15.25 54.9 *** 51.1 12.71 Rexoro . 59.6 15.26 54.9 *** 51.1 12.80 Belle Patna 5906 15005 5409 13099 5101 12060 Century Patna 231 5906 140910 54.9 13084 5101 12051 Average~ 59.6 15.16 54.9 13.89 51.1 12.72 * Brown rice was cycled throughitwo stages of equilibrium mois- ture content before its dry matter weight was determined. Grain damage and numbers of fissures were observed immediately after grains were placed into their second equilibrating at- mosphere. ** Each recorded value is an average of 2 to 6 values determined from grains harvested in 1961 and 1962. *** Samples were equilibrated at 100 percent RH and molded before dry matter weight was determined. 216 Table 10. -- Continued. 8°? 68:; 22°F Variety END" RH %' EMC* RH EMC‘ ~ ‘ 1.16.11 1 (up) 3‘ (<12). Z On 1th --0- 0:00- -.- -_._ 62 . 7 * * Fortuna --- --- 65.3 15.98 62.7 ** Bluebonnet 50 -— -- --- -- 62.7 ** Rexoro --- --- -.. -__ 62.7 *a Belle Patna —-_ —_- --- --- 62.7 a! Century Patna 231 --- -- -- -- 62.7 ** Average 65.3 15.98 62.7 ** Zenith 7500 17086 7505 17076 7504 17006 Fortuna 75.0 18.04 75.5 17.76 75.4 17.19 Bluebonnet 50 7500 17093 7505 17074 7504 17008 Rexoro 7500 18001 7505 17077 7504 16098 36116 Patna 7500 17059 7505 17088 7504 16082 Century Patna 231 75.0 17.48 75.5 17.36 75.4 16.95 Average 75.0 17.82 75.5 17.71 75.4 17.01 Zenith --- -- 86.6 20.93 86.1 20.23 Fortuna '-- --- 8606 19056 8601 20002 Bluebonnet 50 --- --- 86.6 20.57 86.1 19.92 Rexoro -- -- 86.6 20.33 86.1 19.91 Belle Patna ——- --— 86.6 19.79 86.1 19.61 Century Patna 231 -- -- 8606 19087 8601 19082 Average 86.6 20.18 86.1 19.91 Zenith 100.0 26.74 100.0 28.75 100.0 ** Fortuna 10000 25097 10000 27038 10000 ** Bluebonnet 50 100.0 26.30 100.0 ** 100.0 ‘* Rexoro 100.0 26.48 100.0 ** 100.0 ** Belle Patna 100.0 25.92 100.0 28.53 100.0 ** Century Patna 231 100.0 26.02 100.0 29.45 100.0 ** Average 10000 26024 10000 28053 10000 ** * Each recorded value is an average of 2 to 6 values determined from grains harvested in 1961 and 1962. ** Samples were equilibrated at 100 percent RH and molded before dry matter weight was determined. 217 Table 11. -— A sample data sheet recording grain damage and total fissures. Rice Grain Damage Experiment Sheet No. 2-22-11—21 1 1961 196g Date Time Grain Total Grains Total Grain Total Grains Total , cracks % cracked % cracks % cracked % May 20 8:303:13 _0_ __0__ 0 0 ___(_)_ __9___ __Q_ __9_ " .2139.é:9; _.9. .9.. _.9. _.9. ..9. .9._ ..9. ..9. ' 10300 522: _.3. .922 , ..9.. _.4. ..9. .9.. _.9. ..9. _.1... 19139.212; _.2. .9:9 ..2.. _.9. _.9. .921 _.9_ ..;i. I 11:90 a.m. __8_ _gcg ___5_ 10 _5_ 1.8 __4_ ___8_ ' _1_3___1=o.4m-_11__4.-138__a4_ 1.1.1.1121 _.L_12=00 .2}. .922 .12. .13. .31. 11:2 .19. .39. ..1... 1 3 0 .m. .52. 13:4 .92.. .29. ..42. .1229 .JZL. .491. " .___.P__1=00 “no .92. 12.9 .12. .64. .93.. 2L8 .19. .99. " =0 .m. .81. 32.3.9 .21. .14... .92. 33:2 .35. .19. ___1'_____ 3:00 p.m. _L8_ 2652 __435 _8L 105 38.0 _4_1_ __82_ ..E___, a 0 .m. llé_. 31.2. .42. ..29. 119. 4229, .42.. .99. 218 Table 11. - Continued. ‘ — Rice Grain Damage Experiment Sheet#fio. g-gg-gl-fi; bate Timefi lGrain Tctg’ét‘raina Total Grain Tctgéérains Total w cracks % cracked 3: cracks a; cracked $ “1&0 . - 122. 41.9 .49. .29. l2. 4L8 .96. .22. " 13.39.22..- - 111. 41.9 .20.. .190. .122. 221 .97. .24. 4.1.99.9... 2.99. 21.9 .29. 190. 12.8. 27.2 .47. .24. .11.. = 0 0111- 122 24.1 .20.. 199. ll 92.9 .41. .24. .."... : .m. 119. 2.7.1 .20. 1&9. 1.91.. 96—01 .49. .26— ..1...LZ..9__: 0 01m 223. 91.3. i 199.. 18.6. 917:. .19. .29. ..'!....L.2_.:00 om- 2.12.1942 .29.lm. 2112.16.53 .29.!99. .2... 8: do 121. 12.2 .20. 1.99. 3.14. 999 .29. 2.99. ..1'....2.9