Related Commercial Resources

CHAPTER 15

METHODS OF PRECOOLING FRUITS,

VEGETABLES, AND CUT FLOWERS
Product Requirements ............................................................. 15.1 Package Icing .......................................................................... 15.8
Calculation Methods ............................................................... 15.1 Vacuum Cooling ...................................................................... 15.9
COOLING METHODS ........................................................... 15.3
Selecting a Cooling Method .................................................. 15.11
Hydrocooling .......................................................................... 15.3
Forced-Air Cooling ................................................................. 15.6 Cooling Cut Flowers ............................................................. 15.11
Forced-Air Evaporative Cooling............................................... 15.8 Symbols ................................................................................... 15.11

P RECOOLING is the rapid removal of field heat from freshly

harvested fruits and vegetables before shipping, storage, or pro-
cessing. Prompt precooling inhibits growth of microorganisms that
Commercially important fruits that need immediate precooling
include apricots; avocados; all berries except cranberries; tart cher-
ries; peaches and nectarines; plums and prunes; and tropical and
cause decay, reduces enzymatic and respiratory activity, and reduces subtropical fruits such as guavas, mangos, papayas, and pineapples.
moisture loss. Thus, proper precooling reduces spoilage and retards Tropical and subtropical fruits of this group are susceptible to chill-
loss of preharvest freshness and quality (Becker and Fricke 2002). ing injury and thus need to be cooled according to individual tem-
Precooling requires greater refrigeration capacity and cooling perature requirements. Sweet cherries, grapes, pears, and citrus fruit
medium movement than do storage rooms, which hold commodities have a longer postharvest life, but prompt cooling is essential to
at a constant temperature. Thus, precooling is typically a separate maintain high quality during holding. Bananas require special rip-
operation from refrigerated storage and requires specially designed ening treatment and therefore are not precooled. Chapter 11 lists
equipment (Fricke and Becker 2003). Precooling can be done by recommended storage temperatures for many products.
various methods, including hydrocooling, vacuum cooling, air cool-
ing, and contact icing. These methods rapidly transfer heat from the CALCULATION METHODS
commodity to a cooling medium such as water, air, or ice. Cooling
times from several minutes to over 24 hours may be required. Heat Load
PRODUCT REQUIREMENTS The refrigeration capacity needed for precooling is much
greater than that for holding a product at a constant temperature or
During postharvest handling and storage, fresh fruits and vege- for slow cooling. Although it is imperative to have enough refriger-
tables lose moisture through their skins through transpiration. Com- ation for effective precooling, it is uneconomical to have more than
modity deterioration, such as shriveling or impaired flavor, may is normally needed. Therefore, heat load on a precooling system
result if moisture loss is high. To minimize losses through transpi- should be determined as accurately as possible.
ration and increase market quality and shelf life, commodities must
be stored in a low-temperature, high-humidity environment. Vari- Total heat load comes from product, surroundings, air infiltra-
ous skin coatings and moisture-proof films can also be used during tion, containers, and heat-producing devices such as motors, lights,
packaging to significantly reduce transpiration and extend storage fans, and pumps. Product heat accounts for the major portion of
life (Becker and Fricke 1996a). total heat load, and depends on product temperature, cooling rate,
Metabolic activity in fresh fruits and vegetables continues for a amount of product cooled in a given time, and specific heat of the
short period after harvest. The energy required to sustain this activ- product. Heat from respiration is part of the product heat load, but
ity comes from respiration, which involves oxidation of sugars to it is generally small. Chapter 13 discusses how to calculate the
produce carbon dioxide, water, and heat. A commoditys storage refrigeration load in more detail.
life is influenced by its respiratory activity. By storing a commodity Product temperature must be determined accurately to calculate
at low temperature, respiration is reduced and senescence is de- heat load accurately. During rapid heat transfer, a temperature gra-
layed, thus extending storage life. Proper control of oxygen and dient develops in the product, with faster cooling causing larger gra-
carbon dioxide concentrations surrounding a commodity is also dients. This gradient is a function of product properties, surface heat
effective in reducing the respiration rate (Becker and Fricke 1996a). transfer parameters, and cooling rate. Initially, for example, hydro-
Product physiology, in relation to harvest maturity and ambient cooling rapidly reduces the temperature of the exterior of a product,
temperature at harvest time, largely determines precooling require- but may not change the center temperature at all. Most of the prod-
ments and methods. Some products are highly perishable and must uct mass is in the outer portion. Thus, calculations based on center
begin cooling as soon as possible after harvest; examples include temperature would show little heat removal, though, in fact, sub-
asparagus, snap beans, broccoli, cauliflower, sweet corn, canta- stantial heat has been extracted. For this reason, the product mass-
loupes, summer squash, vine-ripened tomatoes, leafy vegetables, average temperature must be used for product heat load calculations
globe artichokes, brussels sprouts, cabbage, celery, carrots, snow (Smith and Bennett 1965).
peas, and radishes. Less perishable produce, such as white potatoes, The product cooling load can then be calculated as
sweet potatoes, winter squash, pumpkins, and mature green toma-
toes, may need to be cured at a higher temperature. Cooling of these Q = mc p ( t i t ma ) (1)
products is not as important; however, some cooling is necessary if
ambient temperature is high during harvest.
where m is the mass of product being cooled, cp is the products spe-
cific heat, ti is the products initial temperature, and tma is the prod-
The preparation of this chapter is assigned to TC 10.9, Refrigeration Appli- ucts final mass average temperature. Specific heats of various fruits
cation for Food and Beverages. and vegetables can be found in Chapter 9.

15.1
Copyright 2006, ASHRAE
15.2 2006 ASHRAE HandbookRefrigeration (SI)

Precooling Time Estimation Methods required to reduce the fractional unaccomplished temperature differ-
Efficient precooler operation involves (1) proper sizing of refrig- ence Y by half.
eration equipment to maintain a constant cooling medium tempera- The half-cooling time is independent of initial temperature and
ture, (2) adequate flow of the cooling medium, and (3) proper product remains constant throughout the cooling period as long as the cool-
residence time in the cooling medium. Thus, to properly design a ing medium temperature remains constant (Becker and Fricke 2002).
precooler, it is necessary to estimate the time required to cool the Therefore, once the half-cooling time has been determined for a
commodities from their initial temperature (usually the ambient tem- given commodity, cooling time can be predicted regardless of the
perature at harvest) to the final temperature, just before shipping commoditys initial temperature or cooling medium temperature.
and/or storage. For a specified cooling medium temperature and flow Product-specific nomographs have been developed, which, when
rate, this cooling time dictates the residence time in the precooler that used in conjunction with half-cooling times, can provide estimates
is required for proper cooling (Fricke and Becker 2003). of cooling times for fruits and vegetables (Stewart and Couey 1963).
In addition, a general nomograph (Figure 2) was constructed to
Accurate estimations of precooling times can be obtained by
calculate hydrocooling times of commodities based on their half-
using finite-element or finite-difference computer programs, but the
cooling times (Stewart and Couey 1963). In Figure 2, product tem-
effort required makes this impractical for the design or process engi-
perature is plotted along the vertical axis versus time measured in
neer. In addition, two- and three-dimensional simulations require
half-cooling periods along the horizontal axis. At zero time, the
time-consuming data preparation and significant computing time.
product temperature is the initial commodity temperature; at infinite
Most research to date has been in the development of semianalytical/
time, product temperature equals water temperature. To use Figure
empirical precooling time estimation methods that use simplifying
2, draw a straight line from the initial commodity temperature at
assumptions, but nevertheless produce accurate results.
zero time (left axis) to the commodity temperature at infinite time
Fractional Unaccomplished Temperature Difference [i.e., the water temperature (right axis)]. Then draw a horizontal line
at the final commodity temperature (left and right axes). The inter-
All cooling processes exhibit similar behavior. After an initial section of these two lines determines the number of half-cooling
lag, the temperature at the foods thermal center decreases exponen- periods required (bottom axis). Multiply the half-cooling time for
tially (see Chapter 10). As shown in Figure 1, a cooling curve the particular commodity by the number of half-cooling periods to
depicting this behavior can be obtained by plotting, on semilogarith- obtain the hydrocooling time.
mic axes, the fractional unaccomplished temperature difference Y The following example illustrates the use of the general nomo-
[Equation (2)] versus time (Fricke and Becker 2004). graph for determining hydrocooling time.
tm t t tm Example 1. Assume that topped radishes with a half-cooling time of 2.2
Y = -------------- = -------------- (2) min are to be hydrocooled using 0C water. How long would it take to
tm ti ti tm
hydrocool the radishes from 27C to 10C?
where tm is the cooling medium temperature, ti is the initial com- Solution. Using the general nomograph in Figure 2, draw a straight line
modity temperature, and t is the commodity final mass average tem- from 27C on the left to 0C on the right. Then draw a horizontal line at
perature. This semilogarithmic temperature history curve consists the final commodity temperature, 10C. These lines intersect at 1.4
of an initial curvilinear portion, followed by a linear portion. Simple half-cooling periods. Multiply this by the half-cooling time (2.2 min) to
obtain the total hydrocooling time of 3.1 min.
empirical formulas that model this cooling behavior, such as half-
cooling time and cooling coefficient, have been proposed for esti- Using nomographs can be time consuming and cumbersome,
mating the cooling time of fruits and vegetables. however. Cooling time of fruits and vegetables may be determined
without the use of nomographs by using the half-cooling time Z:
Half-Cooling Time
A common concept used to characterize the cooling process is the Z ln ( Y )
= --------------------- (3)
half-cooling time, which is the time required to reduce the tempera- ln ( 2 )
ture difference between the commodity and the cooling medium by
half (Becker and Fricke 2002). This is also equivalent to the time Values of half-cooling times for the hydrocooling of numerous
commodities have been reported (Bennett 1963; Dincer 1995;
Dincer and Genceli 1994, 1995; Guillou 1958; Nicholas et al. 1964;
Fig. 1 Typical Cooling Curve
OBrien and Gentry 1967; Stewart and Couey 1963). Tables 1 to 3
summarize half-cooling time data for a variety of commodities.

Fig. 2 General Nomograph to Determine Half-Cooling

Preiods

Fig. 2 General Nomograph to Determine Half-Cooling Periods

Fig. 1 Typical Cooling Curve (Stewart and Couey 1963)
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.3

Table 1 Half-Cooling Times for Hydrocooling of By substituting Y = 0.5 into Equation (4), which corresponds to
Various Commodities the half-cooling time, cooling coefficient C can be related to half-
cooling time Z as follows:
Commodity Half-Cooling
Commodity Size Container Time, min.
ln ( 2 j )
Artichoke None (completely exposed) 8 Z = ---------------- (5)
Crate, lid off, paper liner 12
C
Asparagus Medium Completely exposed 1.1 Cooling coefficients have been reported by Dincer (1995, 1996),
Lidded pyramid crate, 2.2 Dincer and Genceli (1994, 1995), Henry and Bennett (1973), and
spears upright Henry et al. (1976) for hydrocooling and hydraircooling (see the
Broccoli Completely exposed 2.1 Cooling Methods section for discussion of these methods) various
Crate with paper liner, lid off 2.2 commodities, as summarized in Tables 2 to 4.
Crate without liner, lid off 3.1
Brussels Completely exposed 4.4 Other Semianalytical/Empirical Precooling Time
sprouts Carton, lid open 4.8 Estimation Methods
Jumble stack (230 mm deep) 6.0 Chapter 10 discusses various semianalytical/empirical methods
Cabbage Completely exposed 69 for predicting cooling times of regularly and irregularly shaped
Carton, lid open 81 foods. These cooling time estimation methods are grouped into two
Jumble stack (four layers) 81 main categories: those based on (1) f and j factors (for either regular
Carrots, Large Completely exposed 3.2 or irregular shapes), and (2) equivalent heat transfer dimensionality.
topped 23 kg mesh bag 4.4
Cauliflower, Completely exposed 7.2 Numerical Techniques
trimmed Becker and Fricke (1996b, 2001) and Becker et al. (1996a,
Celery 2 Dozen Completely exposed 5.8 1996b) developed a numerical technique for determining cooling
Crate, lidded, paper liner 9.1 rates as well as latent and sensible heat loads caused by bulk refrig-
Sweet corn, 5 Dozen Completely exposed 20 eration of fruits and vegetables. This computer model can predict
in husks Wirebound corn crate, lidded 28 commodity moisture loss during refrigerated storage and the tem-
Peas, Completely exposed (flood) 1.9 perature distribution within the refrigerated commodity, using a
in pod 35 L basket, lid off (flood) 2.8 porous media approach to simulate the combined phenomena of
35 L basket, lidded (submersion) 3.5 transpiration, respiration, airflow, and convective heat and mass
Potatoes Completely exposed 11 transfer. Using this numerical model, Becker et al. (1996b) found
Jumble stack 11 that increased airflow decreases moisture loss by reducing cooling
(five layers, 230 mm deep)
time, which quickly reduces the vapor pressure deficit between the
Radishes Completely exposed 1.1
commodity and surrounding air, thus lowering the transpiration
Crate, lid off, three layers of 1.9
bunches, 230 mm deep
rate. They also found that bulk mass and airflow rate were of pri-
Carton, lid open, three layers of 1.4
mary importance to cooling time, whereas relative humidity had lit-
bunches, 230 mm deep tle effect on cooling time.
topped Completely exposed 1.6
Jumble stack (230 mm deep) 2.2 COOLING METHODS
Tomatoes Completely exposed 10
The principal methods of precooling are hydrocooling, forced-
Jumble stack, five layers, 11
255 mm deep air cooling, forced-air evaporative cooling, package icing, and vac-
uum cooling. Precooling may be done in the field, in central cooling
Source: Stewart and Couey (1963).
facilities, or at the packinghouse.
Cooling Coefficient HYDROCOOLING
Cooling time may also be predicted using the cooling coefficient In hydrocooling, commodities are sprayed with chilled water, or
C. As shown in Figure 1, the cooling coefficient is minus the slope immersed in an agitated bath of chilled water. Hydrocooling is effec-
of the ln(Y ) versus time curve, constructed on a semilogarithmic tive and economical; however, it tends to produce physiological and
axis from experimental observations of time and temperature pathological effects on certain commodities; therefore, its use is lim-
(Becker and Fricke 2002). The cooling coefficient indicates the ited (Bennett 1970). In addition, proper sanitation of the hydrocool-
change in the fractional unaccomplished temperature difference per ing water is necessary to prevent bacterial infection of commodities.
unit cooling time (Dincer and Genceli 1994). The cooling coeffi- Commodities that are often hydrocooled include asparagus, snap
cient depends on the commoditys specific heat and thermal con- beans, carrots, sweet corn, cantaloupes, celery, snow peas, radishes,
ductance to the surroundings (Guillou 1958). Using the cooling tart cherries, and peaches. Cucumbers, peppers, melons, and early
coefficient for a particular cooling process, cooling time may be crop potatoes are sometimes hydrocooled. Apples and citrus fruits
estimated as are rarely hydrocooled. Hydrocooling is not popular for citrus fruits
because of their long marketing season; good postharvest holding
1 Y ability; and susceptibility to increased peel injury, decay, and loss of
= ----- ln ---- (4)
C j quality and vitality after hydrocooling.
Hydrocooling is rapid because the cold water flowing around the
The lag factor j is a measure of the time between the onset of commodities causes the commodity surface temperature to essen-
cooling and the point at which the slope of the ln(Y ) versus curve tially equal that of the water (Ryall and Lipton 1979). Thus, the
becomes constant [i.e., the time required for the ln(Y ) versus curve resistance to heat transfer at the commodity surface is negligible.
to become linear]. The lag factor j can be found by extending the lin- The rate of internal cooling of the commodity is limited by the rate
ear portion of the semilogarithmic cooling curve to the ln(Y ) axis; of heat transfer from the interior to the surface, and depends on the
the intersection is the lag factor j. commoditys volume in relation to its surface area, as well as its
15.4 2006 ASHRAE HandbookRefrigeration (SI)

Table 2 Lag Factors, Cooling Coefficients, and Half-Cooling Times for Hydrocooling Various Fruits and Vegetables
Water Cooling Half-Cooling
Commodity Temperature, C Flow Rate, Crate Lag Factor Coefficient C, Time
and Size Initial Final Water mm/s Load, kg j s 1 Z, s Reference
Cucumbers 22 4 50 5 1.291 0.001 601 546.6 Dincer and Genceli 1994
l =0.16 m 10 1.177 0.001 567 592.3
d = 0.038 m 15 1.210 0.001 385 638.2
20 1.251 0.001 243 737.6
0.5 50 5 1.037 0.001 684 432.9 Dincer 1995
10 1.228 0.001 675 536.4
15 1.222 0.001 629 548.5
20 1.237 0.001 480 612.1
Eggplant 21.5 50 5 1.077 0.000 822 933.9 Dincer 1995
l = 0.142 m 10 1.109 0.000 794 1003
d = 0.045 m 15 1.195 0.000 870 1011
20 1.206 0.000 770 1143
Peaches 21 4 50 5 1.067 0.001 585 Dincer 1996
d = 0.056 m 20 1.113 0.001 201
Pears 22.5 4 1.0 50 5 1.119 0.001 434 561.6 Dincer and Genceli 1995
d = 0.06 m 10 1.157 0.001 419 591.0
15 1.078 0.001 296 592.8
20 1.366 0.001 151 873.1
2 50 5 1.076 0.001 352 Dincer 1996
20 1.366 0.001 151
Plums 22 2 50 5 1.122 0.003 017 Dincer 1996
d = 0.037 m 20 1.171 0.002 279
Squash 21.5 0.5 50 5 1.172 0.001 272 669.6 Dincer 1995
l = 0.155 m 10 1.202 0.001 186 739.8
d = 0.046 m 15 1.193 0.001 087 799.9
20 1.227 0.001 036 866.6
Tomatoes 21 0.5 50 5 1.209 0.001 020 865.4 Dincer 1995
d = 0.07 m 10 1.310 0.000 907 1062
15 1.330 0.000 800 1222
20 1.322 0.000 728 1336
4 50 5 1.266 0.000 953 Dincer 1996
20 1.335 0.000 710

Fig. 3 Schematic of Shower Hydrocooler Fig. 4 Schematic of Immersion Hydrocooler

Fig. 4 Schematic of Immersion Hydrocooler

Fig. 3 Schematic of Shower Hydrocooler (USDA 2004)
(USDA 2004)

Commodities may be hydrocooled either loose or in packaging

(which must allow for adequate water flow within and must tolerate
thermal properties. For example, Stewart and Lipton (1960) showed
contact with water without losing strength). Plastic or wood contain-
a substantial difference in half-cooling time for sizes 36 and 45 can-
ers are well suited for use in hydrocoolers. Corrugated fiberboard
taloupes. A weighted average of temperatures taken at different
containers can be used in hydrocoolers, if they are wax-dipped to
depths showed that 20 min was required to half-cool size 36 melons
withstand water contact (USDA 2004).
and only 10 min for size 45.
Hydrocooling also has the advantage of causing no commodity
moisture loss. In fact, it may even rehydrate slightly wilted product Types of Hydrocoolers
(USDA 2004). Thus, from a consumer standpoint, the quality of Hydrocooler designs can generally be divided into two catego-
hydrocooled commodities is high; from the producers standpoint, ries: shower-type and immersion. In a shower hydrocooler, the
the salable mass is high. In contrast, other precooling methods such commodities pass under a shower of chilled water (Figure 3), which
as vacuum or air cooling may lead to significant commodity mois- is typically achieved by flooding a perforated pan with chilled
ture loss and wilting, thus reducing product quality and salable water. Gravity forces the water through the perforated pan and over
mass. the commodities. Shower hydrocoolers may have conveyors for
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.5

Table 3 Cooling Coefficients and Half-Cooling Times for Hydraircooling Sweet Corn and Celery
Spray Nozzle Water Flow Airflow Rate, Cooling Half-Cooling
Commodity Crate Type Type Rate, m 3/s m 3/s Coefficient C, s 1 Time, s Reference
Sweet corn Wirebound Coarse 0.340 0 0.000 347 Henry and Bennett 1973
0.340 0 0.000 444
0.208 0 0.000 642
0.378 0 0.000 336
Medium 0.303 0 0.000 406
0.190 0 0.000 406
0.190 0.000 414
0.378 0 0.000 492
0.378 0.000 542
0.378 28 0.000 447
0.378 45 0.000 486
0.378 78 0.000 564
Flood pan 0.946 0 0.000 464
1.513 0 0.000 567
Coarse 0.378 0 2170 Henry et al. 1976
Medium 0.303 0 1730
0.378 28 1570
0.378 45 1440
0.378 78 1220
Flood pan 0.151 0 1290
Celery Vacuum-cooling 0.173 57 3710 Henry et al. 1976
0.173 119 2360
0.173 183 2310
Hydrocooling 0.173 51 1890
0.173 99 1790
0.173 142 1390
Well-ventilated 0.173 51 2170
0.173 113 1490
0.173 145 1050

Table 4 Cooling Coefficients for Hydrocooling Peaches high cost of acquiring and operating mechanical refrigeration units,
Water Cooling
they are typically limited to providing chilled water for medium- to
Fruit Temp., C high-volume hydrocooling operations.
Hydrocooling Temp., Coefficient,
Method Water Flow C Initial Final s 1 Smaller operations may use crushed ice rather than mechanical
Flood, peaches 12.2 m3/(hm2) 1.67 31.1 8.22 0.001 05
refrigeration to produce chilled water. Typically, large blocks of ice
in 26.5 L 24.4 m3/(hm2) 1.67 29.4 6.44 0.001 11
are transported from an ice plant to the hydrocooler, and then
baskets 4.44 27.8 9.28 0.000 941
crushed and added to the hydrocoolers water reservoir. The initial
7.22 27.8 9.50 0.001 44
cost of an ice-cooled hydrocooler is much less than that of one using
36.7 m3/(hm2) 1.67 32.5 4.11 0.001 83
mechanical refrigeration. However, for an ice-cooled hydrocooler
7.22 31.7 10.5 0.001 74
to be economically viable, a reliable source of ice must be available
12.8 31.2 14.4 0.001 39
at a reasonable cost (Boyette et al. 1992).
Immersion 4.54 m3/h 1.67 29.4 6.39 0.001 23
9.09 m3/h 1.67 29.4 5.56 0.001 37
Variations on Hydrocooling
4.54 m3/h 7.22 31.1 9.67 0.001 68 Henry and Bennett (1973) and Henry et al. (1976) describe
9.09 m3/h 7.22 30.0 9.33 0.001 72 hydraircooling, in which a combination of chilled water and chilled
13.6 m3/h 7.22 30.0 10.4 0.001 30 air is circulated over commodities. Hydraircooling requires less wa-
Source: Bennett (1963). ter for cooling than conventional hydrocooling, and also reduces the
maintenance required to keep the cooling water clean. Cooling rates
continuous product flow, or may be operated in batch mode. Water equal to, and in some cases better than, those obtained in conven-
flow rates typically range from 6.8 to 13.6 L/s per square metre of tional unit load hydrocoolers are possible.
cooling area (Bennett et al. 1965; Boyette et al. 1992; Ryall and Lip- Robertson et al. (1976) describe a process in which vegetables are
ton 1979). Immersion hydrocoolers (Figure 4) consist of large, frozen by direct contact with aqueous freezing media. The aqueous
shallow tanks that contain agitated, chilled water. Crates or boxes of freezing media consists of a 23% NaCl solution. Freezing times of
commodities are loaded onto a conveyor at one end of the tank, less than one minute were reported for peas, diced carrots, snow peas,
travel submerged along the length of the tank, and are removed at and cut green beans, and a cost analysis indicated that freezing with
the opposite end. For immersion hydrocooling, a water velocity of aqueous freezing media was competitive to air-blast freezing.
75 to 100 mm/s is suggested (Bennett 1963; Bennett et al. 1965). Lucas and Raoult-Wack (1998) note that immersion chilling and
In large packing facilities, flooded ammonia refrigeration sys- freezing using aqueous refrigerating media have the advantage of
tems are often used to chill hydrocooling water. Cooling coils are shorter process times, energy savings, and better food quality com-
placed directly in a tank through which water is rapidly circulated. pared to air-blast chilling or freezing. The main disadvantage is ab-
Refrigerant temperature inside the cooling coils is typically 2C, sorption of solutes from the aqueous solution by food. Immersion
producing a chilled-water temperature of about 1C. Because of the chilling or freezing with aqueous refrigerating media can be applied
15.6 2006 ASHRAE HandbookRefrigeration (SI)

to a broad range of foods, including pork, fish, poultry, peppers, and other suspended matter. Depending on the municipality, hydro-
beans, tomatoes, peas, and berries. cooling water may be considered an industrial wastewater and, thus,
As an alternative to producing chilled water with mechanical re- a hydrocooler owner may be required to obtain a wastewater dis-
frigeration or ice, well water can be used, provided that the water charge permit (Boyette et al. 1992). In addition to daily replacement
temperature is at least 5.6 K lower than that of the product to be of hydrocooling water, shower pans and/or debris screens should be
cooled. However, the well water must not contain chemicals and bi- cleaned daily, or more often if necessary, for maximum efficiency.
ological pollutants that could render the product unsuitable for hu-
man consumption (Gast and Flores 1991). FORCED-AIR COOLING
Hydrocooler Efficiency Theoretically, air cooling rates can be comparable to hydrocool-
ing under certain conditions of product exposure and air tempera-
Hydrocooling efficiency is reduced by heat gain to the water
ture. In air cooling, the optimum value of the surface heat transfer
from surrounding air. Other heat sources that reduce effectiveness
coefficient is considerably smaller than in cooling with water.
include solar loads, radiation from hot surfaces, and conduction
However, Pflug et al. (1965) showed that apples moving through a
from the surroundings. Protection from these sources enhances effi-
cooling tunnel on a conveyer belt cool faster with air at 6.7C
ciency. Energy can also be lost if a hydrocooler operates at less than
approaching the fruit at 3 m/s than they would in a water spray at
full capacity or intermittently, or if more water than necessary is
1.7C. For this condition, they calculated an average film coefficient
used (Boyette et al. 1992).
of heat transfer of 41 W/(m2 K). They noted that the advantage of
To increase hydrocooler energy efficiency, the following factors air is its lower temperature and that, if water were reduced to 1C,
should be considered during design and operation (Boyette et al. the time for water cooling would be less. Note, however, that air
1992): temperatures could be more difficult to manage without specifically
Insulate all refrigerated surfaces and protect the hydrocooler from fine control below 1C.
wind and direct sunlight. In tests to evaluate film coefficients of heat transfer for anoma-
Use plastic strip curtains on both the inlet and outlet of conveyor lous shapes, Smith et al. (1970) obtained an experimental value of
hydrocoolers to reduce infiltration heat gain. 37.8 W/(m2 K) for a single Red Delicious apple in a cooling tunnel
Operate the hydrocooler at maximum capacity. with air approaching at 8 m/s. At this airflow rate, the logarithmic
Consider using thermal storage, in which chilled water or ice is mean surface temperature of a single apple cooled for 0.5 h in air at
produced and stored during periods of low energy demand and is 6.7C is approximately 1.7C. The average temperature difference
subsequently used along with mechanical refrigeration to chill across the surface boundary layer is, therefore, 8.4 K and the rate of
hydrocooling water during periods of peak energy demand. Ther- heat transfer per square metre of surface area is
mal storage reduces the size of the required refrigeration equip-
q/A = 37.8 8.4 = 318 W/m2
ment and may decrease energy costs.
Use an appropriately sized water reservoir. Because energy is For these conditions, the cooling rate compares favorably with that
wasted when hydrocooling water is discarded after operation, this obtained in ideal hydrocooling. However, these coefficients are
waste can be minimized by not using an oversized water reservoir. based on single specimens isolated from surrounding fruit. Had the
On the other hand, it may be difficult to maintain consistent hy- fruit been in a packed bed at equivalent flow rates, the values would
drocooling water temperature and flow rate with an undersized have been less because less surface area would have been exposed to
water reservoir. the cooling fluid. Also, the evaporation rate from the product sur-
face significantly affects the cooling rate.
Hydrocooling Water Treatment Because of physical characteristics, mostly geometry, various
The surface of wet commodities provides an excellent site for fruit and vegetable products respond differently to similar treat-
diseases to thrive. In addition, because hydrocooling water is recir- ments of airflow and air temperature. For example, in a packed bed
culated, decay-producing organisms can accumulate in the hydro- under similar conditions of airflow and air temperature, peaches
cooling water and can easily spread to other commodities being cool faster than potatoes.
hydrocooled. Thus, to reduce the spread of disease, hydrocooling Surface coefficients of heat transfer are sensitive to the physical
water must be treated with mild disinfectants. conditions involved among objects and their surroundings. Soule
Typically, hydrocooling water is treated with chlorine to mini- et al. (1966) obtained experimental surface coefficients ranging from
mize the levels of decay-producing organisms (USDA 2004). Chlo- 50 to 68 W/(m2 K) for bulk lots of Hamlin oranges and Orlando
rine (gaseous, or in the form of hypochlorous acid from sodium tangelos with air approaching at 1.1 to 1.8 m/s. Bulk bins containing
hypochlorite) is added to the hydrocooling water, typically at the 450 kg of 72 mm diameter Hamlin oranges were cooled from 27C
level of 50 to 100 ppm. However, chlorination only provides a sur- to a final mass-average temperature of 8C in 1 h with air at 1.7 m/s
face treatment of the commodities; it is not effective at neutralizing (Bennett et al. 1966). Surface heat transfer coefficients for these tests
an infection below the commoditys surface. averaged slightly above 62 W/(m2 K). On the basis of a log mean air
The chlorine level in the hydrocooling water must be checked at temperature of 6.7C, the calculated half-cooling time was 970 s.
regular intervals to ensure that the proper concentration is main- By correlating data from experiments on cooling 70 mm diame-
tained. Chlorine is volatile and disperses into the air at a rate that ter oranges in bulk lots with results of a mathematical model, Baird
increases with increasing temperature (Boyette et al. 1992). Fur- and Gaffney (1976) found surface heat transfer coefficients of 8.5
thermore, if ice cooling is used, melting in the hydrocooling water and 51 W/(m2 K) for approach velocities of 0.055 and 2.1 m/s,
dilutes the chlorine in solution. respectively. A Nusselt-Reynolds heat transfer correlation repre-
The effectiveness of the chlorine in the hydrocooling water senting data from six experiments on air cooling of 70 mm diameter
strongly depends on the pH of the hydrocooling water, which should oranges and seven experiments on 107 mm diameter grapefruit,
be maintained at 7.0 for maximum effectiveness (Boyette et al. 1992). with approach air velocities ranging from 0.025 to 2.1 m/s, gave the
To minimize debris accumulation in the hydrocooling water, it relationship Nu = 1.17Re0.529, with a correlation coefficient of
may be necessary to wash commodities before hydrocooling. Never- 0.996.
theless, hydrocooling water should be replaced daily, or more often if Ishibashi et al. (1969) constructed a staged forced-air cooler that
necessary. Take special care when disposing of hydrocooling water, exposed bulk fruit to air at a progressively declining temperature
because it often contains high concentrations of sediment, pesticides, (10, 0, and 10C) as the fruit was conveyed through the cooling
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.7

Fig. 5 Serpentine Forced-Air Cooler cooling. This gives the advantage of rapid product movement
through the cooling plant, and the size of the plant is one-third to
one-fourth that of an equivalent cold room type of plant.
Mitchell et al. (1972) noted that forced-air cooling usually cools
in one-fourth to one-tenth the time needed for conventional room
cooling, but it still takes two to three times longer than hydrocooling
or vacuum cooling.
A proprietary direct-contact heat exchanger cools air and main-
tains high humidities using chilled water as a secondary coolant and
a continuously wound polypropylene monofilament packing. It
contains about 24 km of filament per cubic metre of packing sec-
tion. Air is forced up through the unit while chilled water flows
downward. The dew-point temperature of air leaving the unit equals
the entering water temperature. Chilled water can be supplied from
coils submerged in a tank. Build-up of ice on the coils provides an
extra cooling effect during peak loads. This design also allows an
operator to add commercial ice during long periods of mechanical
equipment outage.
In one portable, forced-air method, refrigeration components are
mounted on flat bed trailers and the warm, packaged produce is
cooled in refrigerated transport trailers. Usually the refrigeration
equipment is mounted on two trailers; one holds the forced-air evap-
orators and the other holds compressors, air-cooling condensers, a
Fig. 5 Serpentine Forced-Air Cooler high-pressure receiver, and electrical gear. The loaded produce trail-
ers are moved to the evaporator trailer and the product is cooled.
tunnel. Air approached at 3.6 m/s. With this system, 65 mm diame- After cooling, the trailer is transported to its destination.
ter citrus fruit cooled from 25 to 5C in 1 h. Their half-cooling time
of 0.32 h compares favorably with a half-cooling time of 0.30 h for Effects of Containers and Stacking Patterns
similarly cooled Delicious apples at an approach air velocity of Accessibility of the product to the cooling medium, essential to
2 m/s (Bennett et al. 1969). Perry and Perkins (1968) obtained a rapid cooling, may involve both access to the product in the con-
half-cooling time of 0.5 h for potatoes in a bulk bin with air tainer and to the individual container in a stack. This effect is evi-
approaching at 1.3 m/s, compared to 0.4 h for similarly treated dent in the cooling rate data of various commodities in various types
peaches and 0.38 h for apples. Optimum approach velocity for this of containers reported by Mitchell et al. (1972). Parsons et al. (1972)
type of cooling is in the range of 1.5 to 2 m/s, depending on condi- developed a corrugated paperboard container venting pattern for
tions and circumstances. palletized unit loads that produced cooling rates equal to those from
conventional register stacked patterns. Fisher (1960) demonstrated
Commercial Methods that spacing apple containers on pallets reduced cooling time by
Produce can be satisfactorily cooled (1) with air circulated in 50% compared to pallet loads stacked solidly. A minimum of 5%
refrigerated rooms adapted for that purpose, (2) in rail cars using sidewall venting is recommended.
special portable cooling equipment that cools the load before it is Palletization is essential for shipment of many products, and
transported, (3) with air forced through the voids of bulk products pallet stability is improved if cartons are packed closely together.
moving through a cooling tunnel on continuous conveyors, (4) on Thus, cartons and packages should be designed to allow ample air-
continuous conveyors in wind tunnels, or (5) by the forced-air flow though the stacked products. Amos et al. (1993) and Parsons
method of passing air through the containers by pressure differen- et al. (1972) showed the importance of vent sizes and location to
tial. Each of these methods is used commercially, and each is suit- obtain good cooling in palletized loads without reducing container
able for certain commodities when properly applied. Figure 5 shows strength. Some operations wrap palletized products in polyethyl-
a schematic of a serpentine forced-air cooler. ene to increase stability. In this case, the product may need to be
In circumstances where air cannot be forced directly through cooled before it is palletized.
the voids of products in bulk, using a container type and load pat-
tern that permit air to circulate through the container and reach a Moisture Loss in Forced-Air Cooling
substantial part of the product surface is beneficial. Examples of The information in this section is drawn from Thompson et al.
this are (1) small products such as grapes and strawberries that (2002).
offer appreciable resistance to airflow through voids in bulk lots, Moisture loss in forced-air cooling ranges from very little to
(2) delicate products that cannot be handled in bulk, and (3) prod- amounts significant enough to damage produce. Factors that affect
ucts that are packed in shipping containers before precooling. moisture loss include product initial temperature and transpiration
Forced-air or pressure cooling involves definite stacking patterns coefficient, humidity, exposure to airflow after cooling, and whether
and baffling of stacks so that cooling air is forced through, rather waxes or moisture-resistant packaging is used.
than around, individual containers. Success requires a container High initial temperature results in high moisture loss; this can be
with vent holes in the direction air will move and a minimum of minimized by harvesting at cooler times of day (i.e., early morning
packaging materials that would interfere with free air movement or night), and cooling (or at least shading) products immediately
through the containers. Under these conditions, a relatively small after harvest. Keep reheat during packing to a minimum.
pressure differential between the two sides of the containers results The primary advantage of high humidity during cooling is that
in good air movement and excellent heat transfer. Differential pres- product packaging can absorb moisture, which reduces the packag-
sures in use are about 60 to 750 Pa, with airflows ranging from 1 to ings absorption of moisture from the product itself.
3 L/s per kilogram of product. High transpiration coefficients also increase moisture loss. For
Because cooling air comes in direct contact with the product example, carrots, with a high transpiration rate, can lose 0.6 to 1.8%
being cooled, cooling is much faster than with conventional room of their original, uncooled weight during cooling. Polyethylene
15.8 2006 ASHRAE HandbookRefrigeration (SI)

Fig. 6 Engineering-Economic Model Output

for a Forced-Air Cooler

Fig. 6 Engineering-Economic Model Output for a Forced-Air Cooler

packaging has reduced moisture loss in carrots to 0.08%, although for products that are best held at moderate temperatures, such as
cooling times are about five times longer. Film box liners, sometimes tomatoes, or for those that are marketed soon after harvest.
used for packing products with low transpiration coefficients (e.g., For more information on evaporative cooling equipment and
apples, pears, kiwifruit, and grapes), are also useful in reducing applications, see Chapter 51 of the 2003 ASHRAE Handbook
moisture loss, but they also increase the time required to cool prod- HVAC Applications, and Chapter 19 of the 2004 ASHRAE Hand-
ucts. Some film box liners are perforated to reduce condensation; lin- bookHVAC Systems and Equipment.
ers used to package grapes must also include an SO2-generating pad
to reduce decay. PACKAGE ICING
To prevent exposing product to unnecessary airflow, forced-air Finely crushed ice placed in shipping containers can effectively
coolers should reduce or stop airflow as soon as the target product cool products that are not harmed by contact with ice. Spinach,
temperature is reached. Otherwise, moisture loss will continue unless collards, kale, brussels sprouts, broccoli, radishes, carrots, and
the surrounding air is close to saturation. One method is to link cooler green onions are commonly packaged with ice (Hardenburg et al.
fan control to return air plenum temperature, slowing fan speeds as 1986). Cooling a product from 35 to 2C requires melting ice equal
the temperature of the return air approaches that of the supply air. to 38% of the products mass. Additional ice must melt to remove
Computer Solution heat leaking into the packages and to remove heat from the con-
tainer. In addition to removing field heat, package ice can keep the
Baird et al. (1988) developed an engineering economic model for product cool during transit.
designing forced-air cooling systems. Figure 6 shows the type of Pumping slush ice or liquid ice into the shipping container
information that can be obtained from the model. By selecting a set through a hose and special nozzle that connect to the package is
of input conditions (which varies with each application) and varying used for cooling some products. Some systems can ice an entire pal-
approach air velocity, entering air temperature, or some other vari- let at one time.
able, the optimum (minimum-cost) value can be determined. The Top icing, or placing ice on top of packed containers, is used occa-
curves in Figure 6 show that selection of air velocity for containers sionally to supplement another cooling method. Because corrugated
is critical, whereas selection of entering air temperature is not as containers have largely replaced wooden crates, use of top ice has
critical until the desired final product temperature of 4C is ap- decreased in favor of forced-air and hydrocooling. Wax-impregnated
proached. The results shown are for four cartons deep with a 4% corrugated containers, however, allow icing and hydrocooling of
vent area in the direction of airflow, and they would be quite differ- products after packaging.
ent if the carton vent area was changed. Other design parameters Flaked or crushed ice can be manufactured on site and stored in
that can be optimized using this program are the depth of product in an ice bunker for later use; for short-season cooling requirements
direction of airflow and the size of evaporators and condensers. with low ice demands (e.g., a few tonnes a day), it may be more
economical to buy block ice and crush it on site. Another option is
FORCED-AIR EVAPORATIVE COOLING to rent liquid ice equipment for on-site production.
This approach cools air with an evaporative cooler, passing air The cooling capacity of ice is 335 kJ/kg; 1 kg of ice will reduce
through a wet pad before it comes into contact with product and the temperature of 3 kg of produce by approximately 28 K. However,
packaging, instead of using mechanical refrigeration. A correctly commercial ice-injection systems require significantly more ice
designed and operated evaporative cooler produces air a few beyond that needed for produce cooling. For example, 20 kg of broc-
degrees above the outside wet-bulb temperature, at high humidity coli requires about 32 kg of manufactured ice (losses occur in prod-
(about 90% rh), and is more energy-efficient than mechanical refrig- uct cooling, transport, and equipment heat gain; also, a remainder of
eration (Kader 2002). In most of California, for instance, product ice is required in the box on delivery to the customer). The high ice
temperatures of 16 to 21C can be achieved. This method is suited requirement makes liquid icing energy-inefficient and expensive
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.9

(Thompson et al. 2002). Other disadvantages of ice cooling include face, most water is vaporized off the surface. The heat required to
(1) mass of the ice, which decreases the net product mass in a vehi- vaporize this water is also taken off the product surface, where it flows
cle; and (2) the need for water-resistant packaging to prevent water by conduction under the thermal gradient produced. Thus, the rate of
damage to other products; and (3) safety hazards during storage. cooling depends on the relation of surface area to volume of product
These disadvantages can be minimized if ice is used for temperature and the rate at which the vacuum is drawn in the flash chamber.
maintenance in transit rather than for cooling, or by using gel-pack Because water is the sole refrigerant, the amount of heat re-
ice (often used for flowers), which is sealed in a leakproof bag. moved from the product depends on the mass of water vaporized mv
and its latent heat of vaporization L. Assuming an ideal condition,
VACUUM COOLING with no heat gain from surroundings, total heat Q removed from the
Vacuum cooling of fresh produce by rapid evaporation of water product is
from the product works best with vegetables having a high ratio of
Q = mv L (6)
surface area to volume and a high transpiration coefficient. In vac-
uum refrigeration, water, as the primary refrigerant, vaporizes in a The amount of moisture removed from the product during vacuum
flash chamber under low pressure. Pressure in the chamber is low- cooling, then, is directly related to the products specific heat and the
ered to the saturation point corresponding to the lowest required amount of temperature reduction accomplished. A product with a
temperature of the water. specific heat capacity of 4 kJ/(kgK) theoretically loses 1% moisture
Vacuum cooling is a batch process. The product to be cooled is for each 6 K reduction in temperature. In a study of vacuum cooling
loaded into the flash chamber, the system is put into operation, and of 16 different vegetables, Barger (1963) showed that cooling of all
the product is cooled by reducing the pressure to the corresponding products was proportional to the amount of moisture evaporated from
saturation temperature desired. The system is then shut down, the the product. Temperature reductions averaged 5 to 5.5 K for each 1%
product removed, and the process repeated. Because the product is of mass loss, regardless of the product cooled. This mass loss may
normally at ambient temperature before it is cooled, vacuum cooling reduce the amount of money the grower receives as well as the turgor
can be thought of as a series of intermittent operations of a vacuum and crispness of the product. Some vegetables are sprayed with water
refrigeration system in which water in the flash chamber is allowed before or during cooling to reduce this loss.
to come to ambient temperature before each start. The functional
relationships for determining refrigerating capacity are the same in Commercial Systems
each case.
Cooling is achieved by boiling water, mostly off the surface of the The four types of vacuum refrigeration systems that use water as
product to be cooled. The heat of vaporization required to boil the the refrigerant are (1) steam ejector, (2) centrifugal, (3) rotary, and
water is furnished by the product, which is cooled accordingly. As (4) reciprocating. A schematic of the vacuum-producing mecha-
pressure is further reduced, cooling continues to the desired temper- nism of each is illustrated in Figure 8.
ature level. The saturation pressure for water at 100C is 101.3 kPa;
at 0C, it is 0.610 kPa. Commercial vacuum coolers normally oper- Fig. 7 Pressure, Volume, and Temperature in a Vacuum
ate in this range. Cooler
Although the cooling rate of lettuce could be increased without Cooling Product from 30 to 0C
danger of freezing by reducing the pressure to 0.517 kPa, corre-
sponding to a saturation temperature of 2C, most operators do not
reduce the pressure below that which freezes water because of the
extra work involved and the freezing potential.

Pressure, Volume, and Temperature

In vacuum cooling, the thermodynamic process is assumed to
take place in two phases. In the first phase, the product is assumed
to be loaded into the flash chamber at ambient temperature, and the
temperature in the flash chamber remains constant until saturation
pressure is reached. At the onset of boiling, the small remaining
amount of air in the chamber is replaced by the water vapor, the first
phase ends, and the second phase begins simultaneously. The sec-
ond phase continues at saturation until the product has cooled to the
desired temperature.
If the ideal gas law is applied for an approximate solution in a
commercial vacuum cooler, the pressure/volume relationships are
Phase 1 pv = 8.697 (kNm)/kg
Phase 2 pv1.056 = 16.985 (kNm)/kg
where p is absolute pressure and v is specific volume.
The pressure/temperature relationship is determined by the value
of ambient and product temperature. Based on 30C for this value, the
temperature in the flash chamber theoretically remains constant at
30C as the pressure reduces from atmospheric to saturation, after
which it declines progressively along the saturation line. These rela-
tionships are illustrated in Figure 7. Product temperature responds
similarly, but varies depending on where temperature is measured in
the product, physical characteristics of the product, and amount of
product surface water available. Although it is possible for some Fig. 7 Pressure, Volume, and Temperature in a Vacuum Cooler
vaporization to occur in intercellular spaces beneath the product sur- Cooling Product from 30 to 0C
15.10 2006 ASHRAE HandbookRefrigeration (SI)

Fig. 8 Schematic Cross Sections of Vacuum-Producing Fig. 9 Comparative Cooling of Vegetables Under
Mechanisms Similar Vacuum Conditions

Fig. 9 Comparative Cooling of Vegetables Under

Similar Vacuum Conditions
Fig. 8 Schematic Cross Sections of Vacuum-Producing
Mechanisms
ever, iceberg lettuce is rarely prewetted. A modification of vacuum
cooling circulates chilled water over the product throughout the
Of these, the steam ejector type is best suited for displacing the cooling process. Among the chief advantages are increased cooling
extremely high volumes of water vapor encountered at the low pres- rates and residual refrigeration that is stored in the chilled water
sures needed in vacuum cooling. It also has the advantage of having after each vacuum process. It also prevents water loss from products
few moving parts, thus requiring no compressor to condense the that show objectionable wilting after conventional vacuum cooling.
water vapor. High-pressure steam is expanded through a series of
jets or ejectors arranged in series and condensed in barometric con-
Applications
densers mounted below the ejectors. Cooling water for condensing
is accomplished by means of an induced-draft cooling tower. In Because vacuum cooling is generally more expensive, particu-
spite of these advantages, few steam ejector vacuum coolers are larly in capital cost, than other cooling methods, its use is primarily
used today, because of the inconvenience of using steam and the restricted to products for which vacuum cooling is much faster or
lack of portability. Instead, vacuum coolers are mounted on semi- more convenient. Lettuce is ideally adapted to vacuum cooling. The
trailers to follow seasonal crops. numerous individual leaves provide a large surface area and the tis-
The centrifugal compressor is also a high-volume pump and can sues release moisture readily. It is possible to freeze lettuce in a vac-
be adapted to water vapor refrigeration. However, its use in vac- uum chamber if pressure and condenser temperatures are not
uum cooling is limited because of inherent mechanical difficulties carefully controlled. However, even lettuce does not cool entirely uni-
at the high rotative speeds required to produce the low pressures formly. The fleshy core, or butt, releases moisture more slowly than
needed. the leaves. Temperatures as high as 6C have been recorded in core
Both rotary and reciprocal vacuum pumps can produce the low tissue when leaf temperatures were down to 0.5C (Barger 1961).
pressures needed, and they also have the advantage of portability. Other leafy vegetables such as spinach, endive, escarole, and
Being positive-displacement pumps, however, they have low volu-
parsley are also suitable for vacuum cooling. Vegetables that are less
metric capacity; therefore, vacuum coolers using rotary or recipro-
suitable but adaptable by wetting are asparagus, snap beans, broc-
cating pumps have separate refrigeration systems to condense much
coli, brussels sprouts, cabbage, cauliflower, celery, green peas,
of the water vapor that evaporates off the product, thus substantially
sweet corn, leeks, and mushrooms. Of these vegetables, only cauli-
reducing the volume of water vapor passing through the pump. Ide-
flower, celery, cabbage, and mushrooms are commercially vacuum
ally, when it can be assumed that all water vapor is condensed, the
required refrigeration capacity equals the amount of heat removed cooled in California. Fruits are generally not suitable, except some
from the product during cooling. berries. Cucumbers, cantaloupes, tomatoes, dry onions, and pota-
The condenser must contain adequate surface to condense the toes cool very little because of their low surface-to-mass ratio and
large amount of vapor removed from the produce in a few minutes. relatively impervious surface. The final temperatures of various
Refrigeration is furnished from cold brine or a direct-expansion vegetables when vacuum cooled under similar conditions are illus-
system. A very large peak load occurs from rapid condensing of so trated in Figure 9.
much vapor. Best results are obtained if the refrigeration plant is The rate of cooling and final temperature attained by vacuum
equipped with a large brine or icemaking tank having enough stored cooling are largely affected by the commoditys ratio of surface area
refrigeration to smooth out the load. A standard three-tube plant, to its mass and the ease with which it gives up water from its tissues.
with capacity to handle three cars per hour, has a peak refrigeration Consequently, the adaptability of fruits and vegetables varies tremen-
load of at least 900 kW. dously for this method of precooling. For products that have a low
To increase cooling effectiveness and reduce product moisture surface-to-mass ratio, high temperature gradients occur. To prevent
loss, the product is sometimes wetted before cooling begins. How- the surface from freezing before the desired mass-average tempera-
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.11

ture is reached, the vacuum pump is switched off and on (bounced) Table 5 Cooling Methods Suggested for Horticultural
to keep the saturation temperature above freezing. Commodities
Mechanical vacuum coolers have been designed in several sizes.
Size of Operation
Most installations use cylindrical or rectangular retorts. For porta-
bility, some vacuum coolers and associated refrigeration equipment Commodity Large Small
have been placed on flat-bed trailers. Tree fruits
Citrus R R
SELECTING A COOLING METHOD Deciduous a FA, R, HC FA
Subtropical FA, R FA
Packing house size and operating procedures, response of prod- Tropical FA, R FA
uct to the cooling method, and market demands largely dictate the Berries FA FA
cooling method used. Other factors include whether the product is Grapes b FA FA
packaged in the field or in a packing house, product mix, length of Leafy vegetables
cooling season, and comparative costs of dry versus water-resistant Cabbage VC, FA FA
cartons. In some cases, there is little question about the type of cool- Iceberg lettuce VC FA
ing to be used. For example, vacuum cooling is most effective on let- Kale, collards VC, R, WV FA
tuce and other similar vegetables. Peach packers in the southeastern Leaf lettuces, spinach, endive, VC, FA, WV, HC FA
United States and some vegetable and citrus packers are satisfied escarole, Chinese cabbage,
with hydrocooling. Air (room) cooling is used for apples, pears, and bok choy, romaine
citrus fruit. In other cases, choice of cooling method is not so clearly Root vegetables
defined. Celery and sweet corn are usually hydrocooled, but they With tops c HC, PI, FA HC, FA
may be vacuum cooled as effectively. Cantaloupes may be satisfac- Topped HC, PI HC, PI, FA
torily cooled by several methods. Note: sweet cherries are often Irish potatoes, sweet potatoes d R w/evap coolers, HC R
hydrocooled in packing houses but are air cooled if orchard packed. Stem and flower vegetables
When more than one method can be used, cost becomes a major Artichokes HC, PI FA, PI
Asparagus HC HC
consideration. Although rapid forced-air cooling is more costly than
Broccoli, Brussels sprouts HC, FA, PI FA, PI
hydrocooling, if the product does not require rapid cooling, a
Cauliflower FA, VC FA
forced-air system can operate almost as economically as hydrocool-
Celery, rhubarb HC, WV, VC HC, FA
ing. In a study to evaluate costs of hypothetical precooling systems Green onions, leeks PI, HC PI
for citrus fruit, Gaffney and Bowman (1970) found that the cost for Mushrooms FA, VC FA
forced-air cooling in bulk lots was 20% more than that for hydro- Pod vegetables
cooling in bulk and that forced-air cooling in cartons costs 45% Beans HC, FA FA
more than hydrocooling in bulk. Peas FA, PI, VC FA, PI
Table 5 summaries precooling and cooling methods suggested Bulb vegetables
for various commodities. Dry onions e R R, FA
Garlic R
COOLING CUT FLOWERS Fruit-type vegetables f
Cucumbers, eggplant R, FA, FA-EC FA, FA-EC
Because of their high rates of respiration and low tolerance for Melons
heat, deterioration in cut flowers is rapid at field temperatures. Cantaloupes, muskmelons, HC, FA, PI FA, FA-EC
Refrigerated highway vans do not have the capacity to remove the honeydew, casaba
field heat in sufficient time to prevent some deterioration from Crenshaw FA, R FA, FA-EC
occurring (Farnham et al. 1979). Forced-air cooling is commonly Watermelons FA, HC FA, R
used by the flower industry. As with most fruits and vegetables, the Peppers R, FA, FA-EC, VC FA, FA-EC
cooling rate of cut flowers varies substantially among the various Summer squashes, okra R, FA, FA-EC FA, FA-EC
types. Rij et al. (1979) found that the half-cooling time for packed Sweet corn HV, VC, PI HC, FA, PI
boxes of gypsophila was about 3 min compared to about 20 min for Tomatillos R, FA, FA-EC FA, FA-EC
chrysanthemums at airflows ranging from 38 to 123 L/s per box. Tomatoes R, FA, FA-EC
Within this range, cooling time was proportional to the reciprocal of Winter squashes R R
airflow but varied less with airflow than with flower type. Fresh herbs
Not packaged g HC, FA FA, R
SYMBOLS Packaged FA FA, R
Cactus
A = product surface area, m2 Leaves (nopalitos) R FA
cp = specific heat of product, kJ/(kgK) Fruit (tunas or prickly pears) R FA
C = cooling coefficient, reciprocal of hours Ornamentals
j = lag factor Cut flowers h FA, R FA
L = heat of vaporization, kJ/kg Potted plants R R
m = mass of product, kg
R = Room cooling WV = Water spray vacuum cooling
mv = mass of water vaporized, kg HC = Hydrocooling PI = Package icing
p = pressure, Pa FA = Forced-air cooling FA-EC = Forced-air evaporative cooling
q = cooling load or rate of heat transfer, W VC = Vacuum cooling
aApricots cannot be hydrocooled.
Q = total heat, kJ
bGrapes require rapid cooling facilities adaptable to sulfur dioxide fumigation.
t = temperature of any point in product, C cCarrots can be vacuum cooled.
ti = initial uniform product temperature, C dWith evaporative coolers, facilities for potatoes should be adapted to curing.
tm = temperature of cooling medium, C eFacilities should be adapted to curing onions.
to = surrounding temperature, C fFruit-type vegetables are sensitive to chilling but at varying temperatures.
gFresh herbs can be easily damaged by water beating in hydrocooler.
tma = mass-average temperature, C hWhen cut flowers are packaged, only use forced-air cooling.
v = specific volume of water vapor, m3/kg
V = air velocity, m/s Reprinted with permission from A.A. Kader (2001).
15.12 2006 ASHRAE HandbookRefrigeration (SI)

Y = temperature ratio (t to)/(ti to) Fricke, B.A. 2006. Precooling fruits and vegetables using hydrocooling.
Z = half-cooling time, h ASHRAE Journal 48(2):20-28.
= cooling time, h Fricke, B.A. and B.R. Becker. 2003. Comparison of hydrocooling time esti-
mation methods. Proceedings of the 21st IIR International Congress of
REFERENCES Refrigeration: Serving the Needs of Mankind, August 17-22, 2003,
Washington, D.C. Paper ICR0432.
Amos, N.D., D.J. Cleland, and N.H. Banks. 1993. Effect of pallet stacking Gaffney, J.J. and E.K. Bowman. 1970. An economic evaluation of different
arrangement on fruit cooling rates within forced-air pre-coolers. Refrig- concepts for precooling citrus fruits. ASHRAE Symposium Bulletin SF-4-
eration Science and Technology 3:232-241. 70. Symposium on Precooling Fruits and Vegetables, San Francisco
Baird, C.D. and J.J. Gaffney. 1976. A numerical procedure for calculating (January).
heat transfer in bulk loads of fruits or vegetables. ASHRAE Transactions Gast, K.L.B. and R.A. Flores. 1991. Precooling produce: Fruits and vege-
82(2):525. tables. Postharvest management of commercial horticultural crops,
Baird, C.D., J.J. Gaffney, and M.T. Talbot. 1988. Design criteria for efficient MF1002. Kansas State University Cooperative Extension Service, Man-
and cost effective forced air cooling systems for fruits and vegetables. hattan.
ASHRAE Transactions 94(1):1434. Guillou, R. 1958. Some engineering aspects of cooling fruits and vegetables.
Barger, W.R. 1961. Factors affecting temperature reduction and weight loss Transactions of the ASAE 1(1):38, 39, 42.
of vacuum-cooled lettuce. USDA Marketing Research Report 469. Hardenburg, R.E., A.E. Watada, and C.Y. Wang. 1986. The commercial stor-
Barger, W.R. 1963. Vacuum precoolingA comparison of cooling of dif- age of fruits, vegetables, and florist and nursery stocks. USDA Agricul-
ferent vegetables. USDA Marketing Research Report 600. tural Handbook 66.
Becker, B.R. and B.A. Fricke. 1996a. Transpiration and respiration of fruits Henry, F.E. and A.H. Bennett. 1973. Hydraircooling vegetable products in
and vegetables. In New Developments in Refrigeration for Food Safety unit loads. Transactions of the ASAE 16(4):731-733.
and Quality, pp. 110-121. International Institute of Refrigeration, Paris. Henry, F.E. A.H. Bennett, and R.H. Segall. 1976. HydraircoolingA new
Becker, B.R. and B.A. Fricke. 1996b. Simulation of moisture loss and heat concept for precooling pallet loads of vegetables. ASHRAE Transactions
loads in refrigerated storage of fruits and vegetables. In New Develop- 82(2):541.
ments in Refrigeration for Food Safety and Quality, pp. 210-221. Inter- Ishibashi, S., R. Kojima, and T. Kaneko. 1969. Studies on the forced-air
national Institute of Refrigeration, Paris. cooler. Journal of the Japanese Society of Agricultural Machinery 31(2).
Becker, B.R. and B.A. Fricke. 2001. A numerical model of commodity Kader, A.A. 2001. Post Harvest Technology of Horticultural Crops. Univer-
moisture loss and temperature distribution during refrigerated storage. In sity of California, Division of Agriculture and Natural Resources.
Applications of Modelling as an Innovative Technology in the Agri-Food Lucas, T. and A.L. Raoult-Wack. 1998. Immersion chilling and freezing in
Chain, M.L.A.T.M. Hertog and B.R. MacKay, eds., pp. 431-436. Inter- aqueous refrigerating media: Review and future trends. International
national Society for Horticultural Science, Leuven, Belgium. Journal of Refrigeration 21(6):419-429.
Becker, B.R. and B.A. Fricke. 2002. Hydrocooling time estimation meth- Mitchell, F.G., R. Guillou, and R.A. Parsons. 1972. Commercial cooling of
ods. International Communications in Heat and Mass Transfer 29(2): fruits and vegetables. Manual 43. University of California, Division of
165-174. Agricultural and Natural Resources.
Becker, B.R., A. Misra, and B.A. Fricke. 1996a. Bulk refrigeration of fruits Nicholas, R.C., K.E.H. Motawi, and J.L. Blaisdell. 1964. Cooling rates of
and vegetables, part I: Theoretical considerations of heat and mass trans- individual fruit in air and in water. Michigan State University Agricul-
fer. International Journal of HVAC&R Research (now HVAC&R Re- tural Experiment Station Quarterly Bulletin 47:51-64.
search) 2(2):122-134. OBrien, M. and J.P. Gentry. 1967. Effect of cooling methods on cooling
Becker, B.R., A. Misra, and B.A. Fricke. 1996b. Bulk refrigeration of fruits rates and accompanying desiccation of fruits. Transactions of the ASAE
and vegetables, part II: Computer algorithm for heat loads and moisture 10(5):603-606.
loss. International Journal of HVAC&R Research (now HVAC&R Re- Parsons, R.A., F.G. Mitchell, and G. Mayer. 1972. Forced-air cooling of pal-
search) 2(3):215-230. letized fresh fruit. Transactions of the ASAE 15(4):729.
Bennett, A.H. 1963. Thermal characteristics of peaches as related to hydro- Perry, R.L. and R.M. Perkins. 1968. Hydrocooling sweet corn. Paper 68-
cooling. U.S. Department of Agriculture, Technical Bulletin 1292. 800. American Society of Agricultural Engineering, St. Joseph, MI.
Bennett, A.H. 1970. Principles and equipment for precooling fruits and veg- Pflug, I.J., J.L. Blaisdell, and I.J. Kopelman. 1965. Developing temperature-
etables. ASHRAE Symposium Bulletin SF-4-70. Symposium on Precool- time curves for objects that can be approximated by a sphere, infinite
ing of Fruits and Vegetables, San Francisco. plate, or infinite cylinder. ASHRAE Transactions 71(1):238.
Bennett, A.H., R.E. Smith, and J.C. Fortson. 1965. Hydrocooling peaches Rij, R.E., J.F. Thompson, and D.S. Farnham. 1979. Handling, precooling,
A practical guide for determining cooling requirements and cooling and temperature management of cut flower crops for truck transporta-
times. USDA Agriculture Information Bulletin 298 (June). tion. USDA-SEA Western Series 5 (June).
Bennett, A.H., J. Soule, and G.E. Yost. 1966. Temperature response of citrus Robertson, G.H., J.C. Cipolletti, D.F. Farkas, and G.E. Secor. 1976. Meth-
to forced-air precooling. ASHRAE Journal 8(4):48. odology for direct contact freezing of vegetables in aqueous freezing
Bennett, A.H., J. Soule, and G.E. Yost. 1969. Forced-air precooling for Red media. Journal of Food Science 41(4):845-851.
Delicious apples. USDA, Agricultural Research Service ARS 52-41. Ryall, A.L. and W.J. Lipton. 1979. Handling, transportation and storage of
Boyette, M.D., E.A. Estes, and A.R. Rubin. 1992. Hydrocooling. Posthar- fruits and vegetables. AVI Publishing, Westport, CT.
vest Technology Series AG-414-4. North Carolina Cooperative Exten- Smith, R.E. and A.H. Bennett. 1965. Mass-average temperature of fruits and
sion Service, Raleigh. vegetables during transient cooling. Transactions of the ASAE 8(2):249.
Dincer, I. 1995. An effective method for analysing precooling process Smith, R.E., A.H. Bennett, and A.A. Vacinek. 1970. Convection film coef-
parameters. International Journal of Energy Research 19(2):95-102. ficients related to geometry for anomalous shapes. Transactions of the
Dincer, I. 1996. Convective heat transfer coefficient model for spherical ASAE 13(2).
products subject to hydrocooling. Energy Sources 18(6):735-742. Soule, J., G.E. Yost, and A.H. Bennett. 1966. Certain heat characteristics of
Dincer, I. and O.F. Genceli. 1994. Cooling process and heat transfer param- oranges, grapefruit and tangelos during forced-air precooling. Trans-
eters of cylindrical products cooled both in water and in air. International actions of the ASAE 9(3):355.
Journal of Heat & Mass Transfer 37(4):625-633. Stewart, J.K. and H.M. Couey. 1963. Hydrocooling vegetablesA practical
Dincer, I. and O.F. Genceli. 1995. Cooling of spherical products: Part I guide to predicting final temperatures and cooling times. USDA Market-
Effective process parameters. International Journal of Energy Research ing Research Report 637.
19(3):205-218. Stewart, J.K. and W.J. Lipton. 1960. Factors influencing heat loss in canta-
Farnham, D.S., F.J. Marousky, D. Durkin, R. Rij, J.F. Thompson, and A.M. loupes during hydrocooling. USDA Marketing Research Report 421.
Kofranek. 1979. Comparison of conditioning, precooling, transit Thompson, J.F., T.R. Rumsey, R.F. Kasmir, and C.H. Crisosto. 2002. Com-
method, and use of a floral preservative on cut flower quality. Proceed- mercial cooling of fruits, vegetables, and flowers. Publication 21567.
ings, Journal of American Society of Horticultural Science 104(4):483. University of California, Division of Agricultural and Natural Resources.
Fisher, D.V. 1960. Cooling rates of apples packed in different bushel con- USDA. 2004. The commercial storage of fruits, vegetables, and florist and
tainers and stacked at different spacing in cold storage. ASHRAE Journal nursery stocks. Agricultural Research Service, U.S. Department of Agri-
(July):53. culture, Washington, D.C.
Methods of Precooling Fruits, Vegetables, and Cut Flowers 15.13

BIBLIOGRAPHY Gariepy, Y., G.S.V. Raghavan, and R. Theriault. 1987. Cooling character-
istics of cabbage. Canadian Agricultural Engineering 29(1):45-50.
Ansari, F.A. and A. Afaq. 1986. Precooling of cylindrical food products.
Grizell, W.G. and A.H. Bennett. 1966. Hydrocooling stacked crates of celery
International Journal of Refrigeration 9(3):161-163.
and sweet corn. USDA, Agricultural Research Service, ARS 52-12.
Arifin, B.B. and K.V. Chau. 1988. Cooling of strawberries in cartons with
new vent hole designs. ASHRAE Transactions 94(1):1415-1426. Hackert, J.M., R.V. Morey, and D.R. Thompson. 1987. Precooling of fresh
Bennett, A.H. 1962. Thermal characteristics of peaches as related to hydro- market broccoli. Transactions of the ASAE 30(5):1489-1493.
cooling. USDA Technical Bulletin 1292. Harvey, J.M. 1963. Improved techniques for vacuum cooling vegetables.
Bennett, A.H., W.G. Chace, Jr., and R.H. Cubbedge. 1969. Heat transfer ASHRAE Journal 5(1):41-44.
properties and characteristics of Appalachian area Red Delicious apples. Hayakawa, K. 1978. Computerized simulation for heat transfer and mois-
ASHRAE Transactions 75(2):133. ture loss from an idealized fresh produce. Transactions of the ASAE
Bennett, A.H., W.G. Chace, Jr., and R.H. Cubbedge. 1970. Thermal prop- 21(5):1015-1024.
erties and heat transfer characteristics of marsh grapefruit. USDA Tech- Hayakawa, K. and J. Succar. 1982. Heat transfer and moisture loss of spher-
nical Bulletin 1413. ical fresh produce. Journal of Food Science 47(2):596-605.
Beukema, K.J., S. Bruin, and J. Schenk. 1982. Heat and mass transfer dur-
ing cooling and storage of agricultural products. Chemical Engineering Isenberg, F.M.R., R.F. Kasmire, and J.E. Parson. Vacuum cooling vegeta-
Science 37(2):291-298. bles. Information Bulletin 186. Cornell University Cooperative Exten-
Burton, K.S., C.E. Frost, and P.T. Atkey. 1987. Effect of vacuum cooling sive Service, Ithaca, NY.
on mushroom browning. International Journal of Food Science & Kader, A.A., R.F. Kasmire, and J.F. Thompson. 1992. Cooling horticultural
Technology 22(6):599-606. commodities. Publication 3311. University of California, Division of
Chau, K.V. 1994. Time-temperature-humidity relations for the storage of Agricultural and Natural Resources.
fresh commodities. ASHRAE Transactions 100(2):348-353. Rohrbach, R.P., R. Ferrell, E.O. Beasley, J.R. Fowler. 1984. Precooling blue-
Chuntranuluck, S., C.M. Wells, and A.C. Cleland. 1998. Prediction of chill- berries and muscadine grapes with liquid carbon dioxide. Transactions of
ing times of foods in situations where evaporative cooling is signifi- the ASAE 27(6):1950-1955.
cantPart 1: Method development. Journal of Food Engineering 37: Smith, W.L. and W.H. Redit. 1968. Postharvest decay of peaches as affected
111-125. by hot-water treatments, cooling methods, and sanitation. USDA Market-
Chuntranuluck, S., C.M. Wells, and A.C. Cleland. 1998. Prediction of chill- ing Research Report 807.
ing times of foods in situations where evaporative cooling is signifi- Smith, R.E., G.L. Nelson, and R.L. Henrickson. 1967. Analyses on tran-
cantPart 2: Experimental testing. Journal of Food Engineering 37: sient heat transfer from anomalous shapes. Transactions of the ASAE
127-141. 10(2):236.
Chuntranuluck, S., C.M. Wells, and A.C. Cleland. 1998. Prediction of chill-
ing times of foods in situations where evaporative cooling is signifi- Smith, R.E., G.L. Nelson, and R.L. Henrickson. 1968. Applications of
cantPart 3: Applications. Journal of Food Engineering 37:143-157. geometry analysis of anomalous shapes to problems in transient heat
Flockens, I.H. and H.F.T. Meffert. 1972. Biophysical properties of horti- transfer. Transactions of the ASAE 11(2):296.
cultural products as related to loss of moisture during cooling down. Thompson, J.F., Y.L. Chen, and T.R. Rumsey. 1987. Energy use in vacuum
Journal of the Science of Food and Agriculture 23:285-298. coolers for fresh market vegetables. Applied Engineering in Agriculture
Gan, G. and J.L. Woods. 1989. A deep bed simulation of vegetable cool- 3(2):196-199.
ing. Land and Water Use: Proceedings of the 11th International Con- Woods, J.L. 1990. Moisture loss from fruits and vegetables. Postharvest News
gress on Agricultural Engineering, Dublin, pp. 2301-2308. and Information 1(3):195-199.

Related Commercial Resources