SIMULATION OF HEAT AND MASSTRANSFER DURING VENTILATION OF

WHEAT AND RAPESEED BULKSS.C. Sharma

Department of Civil EngineeringUniversity of Manitoba

Winnipeg, Manitoba

W.E. Muir

Member CSAE

Department of Agricultural EngineeringUniversity of Manitoba

Winnipeg, Manitoba

INTRODUCTION

Ventilation or aeration of grain bulkswith ambient air at flow rates of 0.5-1.0m3 min-1 ton-1 cools the grain to auniform temperature (4). Moisture migration is reduced because the temperaturegradients are lower in aerated grain bulks.By lowering the grain temperature andreducing moisture accumulation, aerationdecreases the probability of the occurrence in the grain bulk of niches suitablefor the growth of grain storage pests.

Ventilation of grain bulks with ambient air or slightly heated air at flow ratesof 1.0-2.0 m min-1 ton-1 can be usedto dry grain (9). Low-temperature dryinghas a number of advantages over high-temperature drying:a (a) equipment isless costly; (b) energy requirements arelower; and (c) kernel damage duringdrying and handling is lower. But grainspoilage can occur during low-temperature drying if the system is not managedcarefully.

Because of the apparent advantages ofventilating grain bulks, a research projectwas initiated to design optimum aerationand low-temperature drying systems forthe preservation of cereal grains andoilseeds for the different climatic zonesof Canada. One of the least expensive andmost versatile (although not the mostaccurate) methods of comparing theperformance of different systems is tocarry out computer simulations of thesystems. The accuracy of the simulations

Ekstrom, N. 1972. Drying with cold andslightly heated air (temperatures normallybelow 40°C). Paper presented at Symposium on Mechanization of Grain Harvesting,Processing and Handling after Harvest.20-27 November 1972. Moscow, U.S.S.R. 9pp.

can be improved by comparing simulationresults with experimental results. Thispaper presents the development andinitial testing of a simulation model ofthe ventilation of bulks of wheat andrapeseed. Wheat and rapeseed werechosen because they are the main cerealand oilseed crops grown in the CanadianPrairie Provinces and their thermal andhygroscopic properties are quite different.

DESCRIPTION OF MODEL

thesis of the first author.b A copy ofthe computer programme can beobtained from us. Only the modifications necessary to the model areexplained below.

Properties of Wheat

To calculate the relationship amongrelative humidity of the air, temperature,and equilibrium moisture content ofwheat, the following equation of Othmerand Huang (7) was used:

RHe =Exp [(L/Ls) Ln(2.04816 ?s) +c]2.04816 ?s

where

RHe

L

(1)

relative humidity of air that is inequilibrium with the grain, adecimal fraction;latent heat of vaporization ofwater in grain, kcal kg ;latent heat of vaporization offree water, kcal kg-1;saturation vapour pressure,g cm-2; anda constant.

The ratio (L/Lj) for wheat was calculated using the equation given byGallaher (2):

L/Ls =1+23 Exp(- 0.40 M) (2)

where

M moisture content of grain, % drybasis; and C was calculated usingthe equation given by Sutherland et al. (11):

C = - 3.34 X 104M -4.0 (3)

RECEIVED FOR PUBLICATION DECEMBER27,1973

Model Choice

By analyzing weather data at Saskatoon and Prince Albert, Saskatchewan,Moysey (5) and Moysey and Wilde (6)concluded that tough and damp grain canbe cooled and dried by ventilation withambient air. Such studies of weather dataindicate the expected drying and coolingpotential of ambient or slightly heated airbut do not define performance of alow-temperature drying system under agiven set of conditions. Neither the timerequired for drying, nor expected moisture content, nor temperatures aredefined in an analysis of weather data. Toaccomplish this, computer programmeshave been developed to simulate heat andmass transfer during the ventilation ofgrain bulks. Simulation models can beused to determine the effects on theventilation system of many variablesincluding depth of grain, initial moisturecontent of the grain, thermal andhygroscopic properties of the grain, andtemperature and relative humidity of thesupplied air.

Bloome and Shove (1) developed asimulation model, based on the finitedifference method, for ventilating corn.We considered this model precise enoughto meet the objectives of our researchproject. A complete description of themodel is not presented here because it isavailable in a paper published by Bloomeand Shove (1) and is also available in the

Sharma, S.C. 1973. Simulation of heatand mass transfer during aeration ofwheat and rapeseed bulks. UnpublishedM.Sc. Thesis, University of Manitoba,Winnipeg, Man. 105 pp.

41CANADIAN AGRICULTURAL ENGINEERING, VOL. 16,NO. 1,JUNE 1974

Both equations 2 and 3 were developedfrom the equilibrium moisture contentmeasurements of Gay (3).

The specific heat of hard red springwheat was calculated using the relationship determined by Viravanichaic for thetemperature range 8.9-21.8°C:

Test

no.

1

2

3(a)3(b)4

TABLE I CONDITIONS OF EXPERIMENTAL TEST

Initial condition of grain

Moisture

Temperature content Temperature(°C) (% dry basis) (°C)Grain

Wheat

Wheat

Wheat

Wheat

Rapeseed

28.3

24.0

22.0

22.0

23.6

24.2

25.4

19.2

19.2

17.1

Condition of ventilating air

25.0

24.2

23.1

23.1

25.0

Relative Air

midity velocity V(%) (m min-1) (°C)

72 2.44 21.977 2.44 21.7

58 2.44 19.4

58 4.36 19.4

55 2.44 19.7

Cpw =0.273 + 0.0093M (4)

where

^pw : specific heat of wheat, kcalkg"1 C"1.

Te is the initial equilibrium temperature of the air and grain assuming that the conditions of theair follow a wet-bulbprocessuntil the vapour pressuresof the air and grain are equal.

Properties of Rapeseed

Strohman and Yoerger (10) developedthe following equation for equilibriumrelative humidity of a hygroscopic material by utilizing the linear form of theOthmer plots (7):

pipe. The outer cylinder was an asbestosand portland cement sewerpipe with a20-cm inside diam. The inside wall of theouter cylinder was lined with0.254-mm-thick polyethylene to preventmoisture transfer between grain andcylinder wall. The outer pipe waswrapped with a 6.25-cm layer offiberglass insulation.

The materials for the pipes werechosen with low thermal conductivities toreduce heat transfer along the pipes thatwould affect the expected vertical temperature gradient in the grain. The outerconcentric cylinder of grain, which wasventilated at the same rate and with thesame air as the inner observation pipe,was designed to act as a guard-ringbetween the observation pipe and thelaboratory room. Heat transferredthrough the insulation mainly affectedthe temperature gradient in the outerpipe, with little effect on the innerobservation pipe.b

Air, conditioned in an environmental chamber, was pulled through thegrain columns with a centrifugal blower.The flow of air through each column wasmeasured with a Brooks Model 1110glass-tube flowmeter. The accuracy of theflowmeters according to the manufacturer was ±0.00125 m3 min-1.

Temperatures were measured along thecentreline of observation pipe at 10.2-cmvertical intervals with 0.508-mm diamcopper-constantan thermocouples. Thetemperatures were recorded by a multipoint recorder with 0.28°C (0.5°F)graduations. Relative humidity of theentering air was measured with a slingpsychrometer in the environmental chamber. The mercury-in-glass thermometerson the psychrometer had 0.56°C (1°F)graduations.

At the end of each test, samples weretaken with a 1.3-cm diam probe throughsampling holes drilled in the observationpipe at 10.2-cm intervals along the pipe.Moisture content was determined by

POLYETHYLENE -LINING

"AIR SUCTION

-LEVEL OF GRAIN

-THERMOCOUPLES

RHe=Exp[fl^MLn(ps)+cetfMj (5)

where a, b, c and d are constants. Theconstants were determined for rapeseedby using the equilibrium moisture content data of Pichler (8). For rapeseed, theequation becomes:

RHe =Exp [0.2772 e~0'1087M Ln (Ps)-2.1050e-°118oM] (6)

The following regression equation forspecific heat of rapeseed was developedfrom the data of Smalld for thetemperature range 10.2-20.6°C:

Cpr =0.333 + 0.0049M

where

(7)

Cpr = specific heat of rapeseed, kcalkg-1 C_1.

EXPERIMENTAL EQUIPMENT ANDPROCEDURE

To test the simulation model, experimental tests were carried out on laboratory equipment. Grain was aerated infour columns, each consisting of twoconcentric cylinders (Figure 1). The innercylinder was a 10.2-cm diam ABS plastic

c Viravanichai, S. 1971. Effect of moisturecontent and temperature on specific heat ofwheat. UnpublishedM.Sc. Thesis,Universityof Manitoba, Winnipeg, Man. 52 pp.

d Small, D.D. 1972. Effect of moisturecontent and temperature on the specificheat of rapeseed. Unpublished B.Sc. Thesis,University of Manitoba, Winnipeg, Man. 22PP-

CANADIAN AGRICULTURAL ENGINEERING, VOL. 16,NO. 1,JUNE 1974

OBSERVATION -PIPE

AIR PLENUM •

AIR INLET •FROM —

ENVIRONMENTALCHAMBER

/

" 7ZZZZZZBF--

A

^

y -38cm-

SAMPLING HOLES

-FALSE FLOOR

TEo

00rO

1

Figure 1. Experimental ventilation column.

oven-drying whole-kernel samples at130°Cfor 19 h.

The drybulb temperatures and relativehumidities of the air for the experimentaltests (Table I) were selected as representative of atmospheric conditions forsouthern Manitoba at harvest time. Thechosen relative humidity of the air wasless than the initial equilibrium relativehumidity of the grain, so that dryingoccurred in every test.

The apparent air velocities through thegrain (i.e., air velocities through theempty column) were selected on the basisof the previous work on aeration bySutherland et al. (11), Bloome and Shove(1) and Alam.e

e Alam, A. 1972. Simulated drying ofsoybeans. Unpublished Ph.D. Thesis, University of Illinois,Urbana, Illinois. 201 pp.

42

m • • AIR VELOCITY 2.44 m min

O-O--O AIR VELOCITY 4.36 mmin-1

0 20 40 60 80 100 120 140

DISTANCE OF AIR TRAVEL THROUGH WHEAT, cm

Figure2. Measure temperature of wheatduring early hours of ventilationin test 3 (initial moisture contentof wheat, 19.2%; air temperature,23.1°C; and relative humidity,58%).

In all experimental tests, two replications were carried out. In test 2, after 90h of aeration, one of the two graincolumns was unloaded to obtain grainsamples for moisture content determination. The remaining grain column wasunloaded after 144 h of aeration. In theother tests, the two columns weresampled at the same time.

RESULTS AND DISCUSSION

Temperature

When the vapour pressure of theventilating air was below the vapourpressure of the moisture in the grain, themeasured temperature of the graindropped rapidly during the early hours ofventilation to a temperature below theinitial temperatures of the grain and air(Figure 2). As was expected, conditionsof the air approximately follow aconstant wet-bulb process until thevapour pressure of the air comes intoequilibrium with that of the grain. Duringthe initial ventilation a wet-bulb processis not followed exactly because sensibleheat is transferred between the grain andair to bring the grain to the equilibriumair temperature. Because the heat ofdesorption of the bound water in thegrain is greater than the heat ofevaporation of free water, the enthalpy ofthe outgoing air is less than the incomingair during the remaining ventilation time.Therefore, equilibrium will occur at alower dry bulb temperature because someheat energy from the air is being absorbedby the grain. This difference should be inthe order of about 0.1°C at moisturecontents of 20% and therefore cannormally be neglected. The effect willincrease as the moisture content isreduced because the ratio of heat ofdesorption to heat of evaporation of free

43

•

o

T 1 1 1 I i

SIMULATED,AIR VELOCITY 2.44 mmin"MEASURED , ..SIMULATED ,AIR VELOCITY 4.36 mmin"1MEASURED ,

23

22- V

\ •

o

\\

-

21

20-

\\

\\

o

-

• • • o19

—

1 1 1 1 10 20 40 60 80 100 120 140

DISTANCE OF AIR TRAVEL THROUGHWHEAT, cm

Figure 3. Temperature of wheat after 75 hof ventilation in test 3 (initialtemperature of wheat, 22.0°C;initial moisture content of wheat,19.2%; air temperature, 23.1°C;and relative humidity, 58%).

water, L/L5, increases. For example, intest 3 the measured temperature (19.0°C)of the grain after the initial hours ofaeration was slightly lower than theexpected value of 19.4°C. Both temperatures were above the wet bulb temperature (17.0°C) of the incoming air.

In applying the simulation model, itwas assumed that the conditions of theair followed a constant wet-bulb processto the point where the vapour pressure ofthe air was equal to that of the grain. Thisprocess was plotted on a psychrometricchart for each set of experimentalconditions to obtain the expected drybulb temperature, Te, of the air leavingthe grain column. Because the length oftime required for the grain to come to thetemperature Te is small compared to thetotal ventilation time, it was assumed thatthis occurred instantaneously at the startof ventilation.

The predicted temperatures werecloser to the measured temperatureswhen the initial temperature of the grainin the simulation model was set equal tole instead of the measured initialtemperature (Figure 3). Because thesimulated and measured temperatures arerelatively close, the simulation methodcould be used to compare the coolingperformance of different ventilationsystems under conditions similar to thetest conditions.

Moisture Content

At the bottom of the ventilatedcolumn of grain where the grain hadnearly come into equilibrium with theincoming air, the simulated moisturecontents were lower than the measuredones in all tests (Figure 4, 5, and 6). Thisdifference was probably due mainly toerrors in the relative humidity measure-

£1 1

•

1

SIMULATEDMEASURED

1 1 1 1

,AIR VELOCITY 2.44 mmin"1

20 h~

O

SIMULATEDMEASURED

,AIR VELOCITY 4.36 mmin"1 ~"

•

co '9CO

a

~

/• /o

'

I'8a9

/*/ /A

//

P

i nUJ

8 16UJ<r3

co 15o

14 _# 8

/

°/s

1 1 1 1

~

13

1

20 40 60 80 100 120 140

DISTANCE OF AIR TRAVEL THROUGH WHEAT , cm

Moisture content of wheat after

75 h of ventilation in test 3(initial temperature of wheat,22.0°C; initial moisture contentof wheat, 19.2%; air temperature,23.1°C; and relative humidity,58%).

Figure 4.

26

i i— i i i

• • • •

-25/ o/ // /

//

24

to

CO

/ /

/ /m 23 ° / -

;/

a* 22 -

1-zUJ

8-

i

iii

ii

SIMULATED, 90 hOF VENTILATION

-

% 201-

-

ii

O MEASURED, 90hOF VENTILATION

-

co

O

S .9O °l

1° i

i

SIMULATED, 144 hOF VENTILATION

• MEASURED , 144 hOF VENTILATION

-

18 -y • •/~

17-

i i l 1 1 -

0 20 40 60 80 100 120 140

DISTANCE OF AIR TRAVEL THROUGH WHEAT, cm

Figure 5. Moisture content of wheat after90 and 144 h of ventilation in test2 (initial temperature of wheat,24.0°C; initial moisture contentof wheat, 25.4%; air temperature,24.2°C; and relative humidity,77%).

ment. The equilibrium moisture contentrelationships for the certified No. 1Neepawa seed wheat used in these testsmay not be the same as those for theAustralian wheat used by Gay (3). Thesepossible differences would cause differences between the measured and predicted moisture contents.

The observed drying zone tended to bewider than the simulated drying zone,which indicates that moisture equilibrium

CANADIAN AGRICULTURAL ENGINEERING, VOL. 16, NO. 1, JUNE 1974

18 ! i i i i i

r» O O O O O-

16CO

CO<00

> 14 - —

orQ

- O -

S?- 12

i-zUJr-

o 10 -

o

UJ o -

rr / SIMULATED, 75 h? 8 / OF VENTILATION -

co

o J O MEASURED, 75h2

O J1 OF VENTILATION

6 ~° ^y -

4 i 1 1 1 1 1

-

0 20 40 60 80 100 120 140

DISTANCE OF AIR TRAVEL THROUGH RAPESEED, cm

Figure 6. Moisture content of rapeseedafter 75 h of ventilation in test 4

(initial temperature of rapeseed,23.6°C; initial moisture contentof rapeseed, 17.1%; air temperature, 25°C; and relative humidity,55%).

between the grain and air is not reachedas rapidly as assumed. This deviation ismore pronounced when the air velocity isincreased (Figure 4).

Above the drying zone, the measuredmoisture contents indicate that somewetting occurred (Figure 4, 5, and 6).Apparently, the simulation programmedoes not predict this wetting, because inall tests the predicted moisture contentsin this zone are below the measured ones.

In the practical application of thesimulation programme, an engineer wouldwant to predict the location of the dryingfront after any given time period. It wasassumed that the drying front is locatedwhere the moisture content is within 0.5percentage units of the final measured orsimulated moisture content. Thus in test3, the simulated drying front was within3.0 cm of the measured one for an air

velocity of 2.44 m/min" _1, and within 7.1cm for an air velocity of 4.36m/min~~1.

CONCLUSION

A computer simulation programmebased on the finite difference method canadequately predict heat and mass transferin ventilated bulks of wheat and rapeseedwithin the ranges of moisture content andtemperature of the available equilibriummoisture content data.

SUMMARY

The ventilation of grain bulks with ambient air or slightly heated air can be usedto reduce the possibility of grain deterioration and may be less expensive thanother methods of controlling deterioration. A research project was initiated todesign optimum aeration and low-temperature drying systems for the preservationof cereal grains and oilseeds for the different climatic zones of Canada. This paperpresents the development of a simulationmodel, based on the finite differencemethod, of the ventilation of bulks ofwheat and rapeseed. The computer simulation model adequately predicted theheat and mass transfer in small columnsof wheat (19.2 and 25.4 percent moisturecontent, dry basis) and rapeseed (17.1percent moisture content, dry basis),which were ventilated in the laboratory atnear-room temperatures.

ACKNOWLEDGMENTS

We thank Agriculture Canada and theManitoba Department of Agriculture fortheir financial assistance.

REFERENCES

1. Bloome, P.D. and G.C. Shove. 1971. Near

CANADIAN AGRICULTURAL ENGINEERING, VOL. 16, NO. 1,JUNE 1974

equilibrium simulation of shelled corndrying. Trans. Amer. Soc. Agric. Eng. 14:709-712.

2. Gallaher, G.L. 1951. A method ofdetermining the latent heat of agriculturalcrops. Agric. Eng. 32: 34-38.

3. Gay, F.J. 1946. The effect of temperatureon the moisture content-relative humidityequilibria of wheat. J. Counc. Sci. Ind.Res. 19: 187-189.

4. Hyde, M.B. and N.J. Burrell. 1973. Somerecent aspects of grain storage technology.Chapter 14, pages 313 to 341 in R.N.Sinha and W.E. Muir, eds. Grain storage -part of a system. Avi Publ. Co., Westport,Connecticut.

5. Moysey, E.B. 1969. Refrigeration of dampgrain with natural air. Can. Agric. Eng. 11:12-14.

6. Moysey, E.B. and D.H. Wilde. 1965.Drying grain with unheated air. Can.Agric. Eng. 7: 12-13.

7. Othmer, D.F. and H. Huang. 1940.Correlating vapour pressure and latentheat data. Ind. Eng. Chem. 32: 841-846.

8. Pichler, H.J. 1956. Sorption isotherms forgrain and rape. [Transl. from German byW.E. Klinner.] J. Agric. Eng. Res. 2:159-165.

9. Shove, G.C. 1973. New techniques in grainconditioning. Chapter 9, pages 209 to 228in R.N. Sinha and W.E. Muir, eds. Grainstorage - part of a system. Avi Publ. Co.,Westport, Connecticut.

10. Strohman, R.D. and R.R. Yoerger. 1967.A new equilibrium moisture contentequation. Trans. Amer. Soc. Agric. Eng.10: 675-677.

11. Sutherland, J.W., P.J. Banks, and H.J.Griffiths. 1971. Equilibrium heat andmoisture transfer in airflow through grain.J. Agric. Eng. Res. 16: 368-386.

44