DRY MILLING CHARACTERISTICS OF SORGHUM GRAINS AND THEIR RELATIONSHIP TO PRODUCT QUALITY

V. Subramanian, R. Jambunathan and K. E. Prasada Rao International cr&s Research Institute for the Semi-Arid Tropics (ICRISAT).

Patancheru P. 0.. Andhra Pradesh, 502 324. India.

Flour particle sire distribution, when grains grobnd under similar conditions varied among 52 sorghum cultivars. Protein co,ntent was high in the coarse flour fraction while starch damage was high in fine fractions. Relationships among flour particle size index (PSI). starch damage, and food quality are reported. The potential of sorghum for alternative food uses through modification of milling has been discussed.

Sorghum [ Sorghum blcolor (L.) Moe- nch ] is considered a coarse grain and generally used for traditional foods Whole or dehulled sorghum grain is ground in village nlills to a coarse flour in India1 and in several.African countr- ies2 before being used for the prepara- tion of foods. Dry milling can vary from simple grinding of whole grain between stones to a complex system of using sophisticated roller mills. Sorghum being harder than wheat. is more difficult 10 grind, and produces coarse particles'. This characteristic i s inherited in Mende- lian fashion4. Although several methods exist for sorghum hardness measure- ment. the pearling index. and particle size index (PSI) appear to be superior to other methods that relate to sorghunl m!lling performance'. Out of the four indirect methods to determine sorghunl grain hardness. the Slenvent hardness tester is useful for rapid screening of soryhum cultivarsc.

With increased propuction, there i s a need to intensify research on alternate

uaes of sorghum. It may be possible to select cultivars with improved milling quality which will make the utilization of sorghum competitive with other cer- eals. Work on the dry milling of sorghum is currently inadequate. Wheat milling technology can be effectively used for grinding sorghum with suitable modifi- cation, although sorghun~ has different properties cornpafed to wheat '. Modi- fication such as. use of abresive or hammer wills may improve sorghum !lour quality. Gritiding to a fine particle size has been accomplished by using impact grindysi. The granulation of flour particles is important i n blending sorghuni flour ~ i : h wheat flour for bread making. Casiar e t al.Qeported that although bread production was possible with whole flour. the quality of bread improved by using sorghum flour without the bran fraciion, obtained after passing through siebe;. Badi and H o s e n e ~ ' ~ demonstrated the use of sorghum flout for making cockiej by blending it with wheat flour. 31inding up to 10% of.sor- ghum flour ,sth wheat flour did not

markedly affect the bread quality and blends u p to 45% gave acceptable biscuits1 1

Our objective was to test several sorgh- um cultivars for their mill ing performa- nce and to stcdy the distribution of par- ticle size, and starch damage after mill ing fhe grain i n a UDY cyclone mill. A UOY mill is similar to an abrasion mill ing sys- tem wherein varying layers of grain are progressivelv rubbed off against a hard stirface 1 "ariation i n flour particle size and starch damage i n 52 sorghum culti- vars varying i n their grain hardness and the relationship of the above factors t o food quality are described i n this report.

M a t e r i a l s and M e t h o d s

Gra in samples Grain harvested from 52 sorghum culti- vars o f hard and soft eridospernl types grown during the post-rainy season of 1980 at ICRISAT Center, Patancheru, India, were milled. After harvest, the grain sarnples were dried 10 unifornl moisture level (1 0.1-1 %).

Milling so rghum grains A l l a~ialyt ical tests, unless stated other- wise were made i n duplicate and mean values are given. Grains were cut in h: If and examined for the exrent of floury endosperm, and scored on a 1-5 scale ( 1 is more corneous and 5 is floury).

Grain hardness was measured w i t h 20 cjrains from each sanlple and the force (kg) required to break the grains in the Kiya hardn?ss tester was recorded and rnean values are given ' 3 .

A sample of grain (50 g) was ground for 2 min in a UDY nii l l !itled w i th a 0.4 firm screen, t o which a vacuum cleaner

was attached. Grains o f three selected cultivars (500 g in each) were grounc in three village mills (disc type). The ground material was collected and oven dried at 700C for 1 2 h. This avoided lump formation and agglome!a!ion ol flour particles. The flour was siited thro- ugh a set of sieves having openings of 355, 350, 180, and 150 ~ 7 ! ( U S A standard testing sieve - A S T M E 1 1 specification) fi l ted on a Ty!?r Model R X - 2 4 portable sieve shaker for 15 mill. Two rubber stoppers we.e kepr or1 each sieve to disperse the f lour partir!es dur- i ng the operation of the equi;llm?nt. Th? fractions retained o n top o f e x h sievz were collected, weighed and expressed as percentage o f the total initial weight of ground flour. The weight o f the fract- ions that passed through the 150 pm seive was expressed as a percentage of total fiour and was denoted as the parti- cle size index ( PSI ), as reported by Abdelrahman aud Hoseney 1 4 .

S t a r c h damage Starch damage was determined in the flour fractions by incubating a 75 mg sample w i th 25mg glucoamylase (Sig - ma grsde ) in 1.0 n i l o f 2 M sodium acetate buffer (PH 4.8) for 2 n. The glucose released was measured by the phanol sulphirric acid method of Dubois e t a l ' s .

P r o t e i n c o n t e n t Prote i i~ cor1tents of t l ~ e flour i r x t i ons were analyzed sing the Technicon Auto (4nalyj.r (Jamburlathan et a1 1 ';).

R o t i and p o r r i d g o qua l i t y The qu~ l i ! y of r o t i prepared ffon: :vhola ~ j r a i r ~ flour o f the cultivars w3s d?icrtni. ned according to hlurty and SuS:~mari- i a n l 7 . A Taste panel avaluatiorl of r o t i s

V J S conduc!ed using 12 panelists. Porr- dge was mad? cooking 10 g f loiir in j0 ml water for 7 min on a hot plate naintained at 270' C. The cooked mate- ial mas allowed to cool for 20 min and !valuated for consistency by a panel o f i members.

7esults ar id D i s c u s s i o ~ ~ /illage mills, called chakki in India, are jerlerally used for g!inding whole grain ;orghum 10 flour. A Comparison o f the >article size distribution of grain groud n a vil1ac;e mill and grain sround in a JOY mill showed that the flour yield in I fine fraclion (<I50 ? m ) was lower hen ground in the village mill(Tab1e i). *l ie partic12 s i ze distribution snowe'd ninor variations between other sieve ri~ctions (250. 180 and 150 ;1 nl) Vv'hen jrain ground in UDY mill. variations :rnong the 52 cul;iv;r~s wait. observed 'or the the dis!ribution of starch damage :ont;nt and protein content besides ~art icla size d~s t r~bu t i on (Table 2). Flour ,articles rotained o n the 355 p m sieve Nerc much lower as compared to other ;ieve fractions. Wide variations for par- . ~ c l e j retained on the 250, 180 and 150 4 m sleves were observed.

Flour parricles which passed th!ough :he 150 Y rn sieve showed appreciable variation for the cultivars used arid this m a s dofined as the particle size index (PSI) which ranged f lom 35.7 to 70.65). Several cu l t i~ars w i th a hiyh recovery o f f i q i flour vJere observed. The culti- vsrs yie!ding higher quantities of the fine frijc!ion (Table 2) car1 bz useful for bleridlng w : ~ t i wheat fo: producing b i s c ~ i ; ~ , cake;. and bread The nleihod desl::i2cd c:;,y b e useftr\ as a selection C : ~ t e r ~ j of i;rdir:s (01 specif~c iooo Lses. TI\<? Io..-,er tt-$e particle size index, the

harder is the arainr. The distribution of a higher fine f l o ~ r fraction indicated tlrat the grain was not showing resistance to grinding and could be considered soft. Pomeran at al . ' : indicated that in corn the coarse fraction contained a . substantially higher proportiori of floury particles. The distribution of finc or coarse particles would indicate relative dkgrees of resistance to grinding.

Starch damage in sorghum flour vaiied from 9.8 t o 37.493 when ground in a UDY nl i l l (Table 2;. This sugsest$d that sorghum grains b+have different than wheat w i th respect to starch damage. However, Murty et a l . lVnd ica ted that sorghum graics cjrotrnd in a UDY mi!! produced flour v:ith 10.1 to 20 9 % starch d a m a ~ e They indicated that starch damage in flour obtained from the vil lage mil; (disc) and the laboratory mi l l (Milcent mill) were higher than the same flours from UDY mills. Starzh dam- age i n flour infiueqces the abil i iy to nbs- orb water & th? quality o f bread d;pe- nds o n the ex!ent of starch dnrnage15. Dough roll ing quality o f sorghum was correlated with sta:ch damageqs.

Starcti damage was also determined using particles rerained o n different sieves. Thouah appreciable variation was observed ior starch damage between f lour fracti0r.j for the cultivars, rhe extent of damass in particles of 150 ? r n and \ASS than i 50 .: rn was higher than other fractions ( Table 2 ). Damzged starch content pro~ressively increased frorii coarse pjrtic!es to fine particles. Starch damag. va!ues in f lour particlss wh ich passed ;h:o~,;h the 150 ,m siege ranged from ;2.0 to 46.1 0/b (mean 26.4%). I t has bs$n indicated that w h a t flour should c:,r.;;;:: adequate damaged

starch t o supply sugar dur ing ferment- ation, and excess damage produced an adverse effect o n the qual i ty o f bread'g. The data indicated that i t was possible t o select sorghum cultivars w i t h a desi- red level of s!arch damage dopending upon the requirement.

The protein content of whole sorghum flours varied from 7.8 t o 14.8'5, Miche e t a l .se indicated that sorghum w i t h high protein content should be selected for pasta making. Protein distr ibut ion i n flour fractions of the 5 2 cult ivars varied from 8.4 to 19.1% (Table 2). The coarse fractions contained hisher levels o f protein. which may be due t o the pres- ence o f outer layers of endosperm, although the differences were no t large among fractions. Outer layers o f the endosperm contained more protein Z* ,?T . The fine fractions had lower protein content as they may have part- icles from the floury endosperm wh i ch have been reported to con ta in less ~ r o t e i n g . I t may be possible t o obtain large quantities of a fraction w i t h h igh protein f ~ o m sorghum w i t 11 suitable ~nodif icat ion of mil l ing proceduresqx.

Twelve cultivars which showed 1 9 5 to 34.7% damaged starch content i n the f ine practicles (< 160 um) were studied for their rot i characters (Table 3). The PSI for these flours showed a large range between 36.1 and 70.6";. The data support the observat~on of M u r t y e t e l 2 that the variation in flour particles was affected by cultivars and nrills. The texture of the rot i assessed b y a taste panel was scored between 1.3 a n d 3.2. General acceptability of r o t i was rated between 2 0 and 2.8 i n order t o study the effect o f PSI on porridge quality, cutlivars showing a narrow range o f PSI were compared for starch daniage

and porridge quality. PSI values varied from 58.8 t o 70.2"., whi le st3rch damage showed w ide variations from 19 5 :o 34.7;: (Table 4). The porridge quality score ranged f rom 1.2 t o 3.7 where 4 i s rated good and 1 as poor.

The PSI (< 1 5 0 ,m) was negatively correlated w i t h starcn damage(rable 5). R o t i texture showed negative correla- t ion w i t h starch damage. Flours wi!h more starch damage produced r o t i wi;h hard texture as evaluated by the tasre panel. However, starch damage was influenced by the quant i ty of frne parti- cles due t o mi l l ing. Flours having lower starch damage produced aood quality porridge ( r = - 0.85) (Table S).The data suggest that i t is possible to obtain fine flour w i t h least starch damge as in cultivars l ike Surat-1, BP 53, Vidhisha, elc. Such types o f sorghum can also be used for blending w i t h wheat, where low starch damage f lour is required.

The present study suggest that variation exists for particle s i t e of sorghum flour ground under similar condil ions. There is a possibil i ty t o gr ind sorghuni to the desired particle size and starch dam.age. so that i t can b e used for a variety of foods such as making breads and cook- ies, by b lending with wheat flour. Labo- ratory evaluations o f mi l l ing quality of sorghum grain w i l l be one o f the meth- ods to assess the usefulness of various cultivars for uses based on PSI and starzh damage. This w i l l be useful for development or modif icat ion of n ~ ~ l l s , SO

that new varieties cou ld be lasted for milling quality, thereby making sorghum mote conlpeti l ive w i t h other cereals.

A c k n o w l e d g e m e n t we thank C. 0. Rarnaiah. A. Laxma Reddy 6 K.A.R. Zafar for their iechnical assistance.

TABLE 1. PAR~ICLE SIZE OISTRIBUTION OF SORGHUM FLOUR OBTAINED BY A VILLAGE MILL A N D UDY M I L L

Retained o n Per cent flour particles

sievea Vil lage mil lh UDY millc

355 ! ~ rn 5.7 2.0

11.7 250 ; ~ m 16.4

180 r m 16.4 14.9

150 : ~ m 9 6 10.4

Passed through 1 5 0 :,rn 50.3 60.2

aUSA standard testing 'sieve (ASTM.E 1 1 specification) bValues based o n average from 3 cult ivars ground in 3 CValues based o n average from 5 2 cultivars

T A B L E 2. DISTRIBUTION OF PARTICLE SIZE, STARCH DAMAGE AND PROTEIN CONTENT I N FLOUR FRACTIONS OF SORGHUMJ CULTIVARS (MEAN OF 52 CULTIVARS)

Part icles retained . Flour fractions (7,:) Starch damage in Protein i n on sieve f lour f ract ions (;,J flour fractions (7;)

-- - --- -- u m Range Mean S E I Rang M e a n S i S . Range Mean SE-I

Whole flour 9.8-37.4 21.8 0.56 7.8-14.8 11,7 0.22

Grains were g round i n UDY mi l l

TABLE 3. GRAIN CORNEOUSNESS ( C O R ) , PARTICLE SIZE INDEX (PSI),:

STARCH DAMAGE b ROT1 QUALITY I N SORGHUM CGLTIVARS.

Cultivar

SURAT 7

MALDANDI LOCAL

269

BP 53

VIDISHA 60-1

PJ 18R

NARALlGUTl

BILLORA VAN1

P J 31K

AlSPURl

COR PSI Starch Rot i quality . - damage Texture General

(:/A 1 a~ceptabil i ty

MEAN 3.8 61 .O 25.5 2.3 2.4

Corneousness was measured o n a scale o f 1-5, where 1 is more corneous E 5 is floury.

- TABLE 4. GRAIN CORNEOUSNESS (Cor), PARTICLE SIZE INDEX (PSI),

DAMAGED STARCH CONTENT (DSC), AND PORRIDGE QUALITY OF SORGHUM CULTIVARS

Cullivar Cor Starch d-amage (',I,,;) PSI Porridge score6

SURAT 1 4 19.5 70 2 3.7

BP 53 4 21.9 68.0 3.2

VlDlSHA 60-1 4 22 0 65.0 3.2

NJ 1953 4 23.5 60.9 3.6

E 35 2 ' 25.1 58.8 2.8

296 4 27.2 60.3 2.8

DHOLJO 4 27.8 60.9 3.2

PJ 31K 4 27.9 67.5 3.2

AlSPURl 4 27.9 69.4 3.2

VAN1 VAN1 4 33.6 66.8 1.2

PANDORI VAN1 5 34.7 61.3 1.8

MEAN 3 .9 26.5 65.5 2.9

SE 1. 0.21 1.42 1.22 0.23

Grains were ground in UDY mil l .

.Mean porridge score as evaluated for consistency by a taste panel of five members.

Score : 4 - Good, 1 - Poor.

- .Damaged starch content is given for tne flour particles passed through 150 rm sieve.

Roti was made from whole grain flour; evaluated by a taste panel fo! texture and general acceptability on a scale o f 1 - 4 where 4 is good and 1 is poor.

References

TABLE 5. CORRELATION COEFFICIENTS BETWEEN VARIOUS CHARACTERS OF SORGHUM GRAIN ( n - 1 4 FOR ROT!, 8 11 FOR PORRIDGE )

. . 3 -

. . PSI vs Corneousness ' '

0.34

PSI vs Starch damage , -0.62

PSI vs Roti texture 0.54

PSI vs Roti acceptability

PSI vs Por.idge quality - 0.1 4

Starch damage vs Corneousness , 0.07

Starch damage vs Roti texture -0.72**

Starch damage vs Roti acceptabijity 0.1 0 - .

Starch damage vs Porridge quality -0.85' *

. . * * significant at 1% level.

Significant at 5%' level.

1. Subramanian. V., and Jarnbunathan, R., Traditional methods of processing sorghum ( Sorghum bicolcr L. Moench ) and pearl millet ( Penni- setum americanurn L ) grains. in India. Reports of the In!e:national Association of Cereal Chemistry.1885. lo, 115.

2. Vogel. S. and Graham, M , Sorghum and millets - food production . and use. Report o f workshop held in Nairobi, Kenya. 4 7 July 1978. Inte- rnational Development Research en tre (IDRC), Ottawa, Canada, 1979.'

3. Hoseney, R. C., Varriano-Marston, E. arid Dendy, D.A.V , Sorghum and Millets. In Advances in Cereal Scie- nce and Technology. Vol VI ( ed. Y. Pomeranz), American Association of Cereal Chemists, Inc., St Paul, Minnesota. 1981.

4. 8erg. S. 0.. Is the degree of grittiness of wheat flour mainly a varietal cha- rac!er? Cereal Chem.. -1 947, 24. 274.

5. Kirleis, A. W,. and Crosby, K. D., Sorghum hardness : Compdrison of methods for its evaluation. ICRISAT (Internalional Crops Research Insti- lute for the Semi Arid Tropics) 1982. Pages 231 -241 in Proc. of the Inter- national Syrn. on "Grain Uual~ty", 28-31 October 1981. Patancheru, A. P., India, 1982. '

6. Ponieranz, Y., comparison of screen- ing methods for indirect deterrnina- tion of sorghurn hardness. Cereal Cheni, 1986. 63, 36.

7. Perten. H., Specific characteristics of niillet and sorghum milling. Proc.. Sym on "Sorghum and Millets for Human Food". International Asso- ciation of Cereal Chemistry, Vienna, 11 -1 2 May 1976, Tropical Products nstitute, London, 1977.

8 Stringfellow. A. C. and Peplinski. A. J., Air classification of sorghum flours from varieties representing different hardness. Cereal Sci. Today, 1966, 11, 438.

9. Casier, J. P. J., de Peape, G., Willems, H., Goffings, G. and Noppen, H., Bread from slarchy tropical crops. 11. Bread production from pure millet and sorghum flours by using cereal endosperm-cellwall pentosan , . as a universal baking factor. Sym on "Sorghum and Millets for Human Food". lnternational Association of Cereal Chemistry, Vienna, 1 1 . 1 2 May 1976, Tropical Products Institute, London, 1977.

10.' Badi, S. M. and Hoseney, R. C., Use of pearl millet flours in bread, and cookies. Proc., Spm. on "Sor- ghum and Millets for Human Food", International Association of Cereal Chemistry,Vienna. 1 1 -1 2 Nloy 1976, Tropical Products Institute, London, 1977.

1 1 . Olatunji. O., Akinrele, I. A.,Edwards, C. C. and Koleoso, 0. A., Sorghunl and Millet Processing and Uses in Nigeria. Cereal Foods World, 1982. 27, 277.

12. Shepherd, A. 0.. Laboratory decort- icating mill I. Study of machine and cperaiing variables. II. Applications to sorghum and bul- rush (pearl) millet. Report of the Internaiional Association of Cereal Chemistry, 1950. 10. 108.

13, Murty, 0. S. and Subramanian, V.. Sorghum ro t i I. Traditional m e t h o d s of, consumption and s t a n d a r d p ro - cedures for evaluat ion- ICR ISAT < In ternat ional C r c p R e s e a r c h I n s t i -

-- . _ tute fo r the Semi-Arid Tropics,) 1982 Pages 73-78 in Proc, i n te rna t iona l Sym, on --Sorshu.n Gra in Quality", 28-31 October t 9 8 3 , Pataccheru. A+,, India. 1 9 S 2 .

. . . . 14, ~ b d e l r a h r n a n . A. A- and Hoseney,

R, C . , - Basis of h a r d n e s s in pearl

- . '&illet, grain sorghum and c o r n , Cereal Chem.. 1984,- 61. 232. - - . . .

15, Dubois, M., Gilles, K. A. ,Hamilton, - 3'- 'K:, R e b e r s . P, A- and Smith, F.,

color imetr ic rnerhod f o r d e t e r m i n a t - 'ion o f - s u ~ a r s & related substances, ~ n ~ l - y t i c a l Chern.. 195.6. 28. 350.

- - . . . -

3 6. ~ a m h u n a t h a n , R., Rao, N. S, and - Gurtu, S., Rapid methods for &sti-

mat ing portein & lys ine in sorghum ;- -*-(Sorghum bicoior L, Moench),

. - - . . .. . . - . .

- Cereal Chem., 1983, 60, 1 92- . .

7 7, Porneranz, Y - , Czuchajowska, Z, & Lai, F, S., Comparison of . methods for determination of hardness and breakage suceptibil i ty of cornnter- cia l ly dried c o r n . Cereal Chem ,

: - 1 9 8 6 , 63, 39- v

18, Murty. D.S., Rao. N-S, & Singh, U,. Evaluation of sorghum gr&in qual i ty as inf luenced by milling methods-

-. Proc, Sym, on ' -Processing of Sor- - ghum and Millets", In ternat ional

Associat ion o f Cereal Chemstry., . 1984, 11, 24-

19, Kent, N- t., Technolagy of cereals- Pergarnon Press, O x t o r d , 1 983, 95-097.1 1 7 ,

20- Miche , G . C , . AJary . H.. d e a n J e a r ~ , M,F, and Abecaiss, J.. Potential use of-sorghum grains in pasta process- ing, Proc. Sym. on --Sorghum and M i l l e t s f o r Human Food". lnt- arnat ionai Associa~ion of C e r e a l Chemistry. Vienna. 1 1 - '1 2 M a y 1976 Tropical Products institute. London, 1 9 7 7 .

2 1 - ~ormand, F, L., H o g a n . J T, and Deobald, H - 3.- Protein content of successive peripheral layers mi l led f rom wheat, barley. and grain sorg- hum. Cereal Chem. , 1965, 42 359,

23, Rooney, L-W,, Fryer, W.V. & Cater, C. M., Protein a n d amino acid contents of successive layers rerno- ved b y abrasive milling pf sorgham grains. I b i d , 7 972, 49, 399.

23. Rooney. L. W. a n d Sullins, R , W-, The structure of sorghum a n d its 'relat- ion to processing a n d nutr i t ional va iue, P r o c , . Sym- on - . S O T Q ~ U ~ and Mi l le ts for Human Food",

- In ternat ional Assoc ia t ionof C e r e a l Chemistry. Vienna- 1 1-1 2 M a y 1 9 7 6 Tropcial Products Institute, London. 9 977.

24- Murty, 0 ,s - . Patil, H . D , and House, L-R-, Studies o n f lour particle size and endosperm texture in sorghum. J,Fd- Sci. and Technol , , India, 1984, 21, 359,