2
Drip production in meat refrigeration
The quality of fresh meat exposed for retail sale is initially judged on its
appearance. The presence of exudate or ‘drip’, which accumulates in the
container of prepackaged meat or in trays or dishes of unwrapped meat,
substantially reduces its sales appeal (Malton and James, 1983). Drip can
be referred to by a number of different names including ‘purge loss’, ‘press
loss’ and ‘thaw loss’ depending on the method of measurement and when
it is measured.
In general, beef tends to lose proportionately more drip than pork or
lamb. Since most of the exudate comes from the cut ends of muscle fibres,
small pieces of meat drip more than large intact carcasses. The protein
concentration of drip is about 140 mg ml
-1
, about 70% of that of meat itself.
The proteins in drip are the intracellular, soluble proteins of the muscle
cells. The red colour is due to the protein myoglobin, the main pigment
of meat.
The problem of drip loss is not however confined to retail packs. The
meat industry uses large boneless primal cuts, which are packed in plastic
bags, for distribution throughout the trade. These may be stored under
refrigeration for many weeks before use and during this time a consider-
able volume of drip may accumulate in the bag. Not only does this exudate
look unattractive, but it also represents an appreciable weight loss to the
user when the meat is subsequently removed from its container.
Excessive drip could have a small effect on the eating quality of meat.
Perceived juiciness is one of the important sensory attributes of meat.
Dryness is associated with a decrease in the other palatability attributes,
especially with lack of flavour and increased toughness (Pearson, 1994).
However, moisture losses during cooking are typically an order of
magnitude higher than most drip losses during refrigeration. Consequently,
small differences in drip loss will have little affect on eating quality.
The potential for drip loss is inherent in fresh meat and is influenced by
many factors. These may include breed, diet and physiological history, all of
which affect the condition of the animal before it is slaughtered. After
slaughter, factors such as the rate of chilling, storage temperatures, freezing
and thawing can all influence the drip produced.
The mechanism of drip formation has been well described by Taylor
(1972), Bendall (1974) and Penny (1974) and form the basis of this chapter.
To understand how drip occurs, it is useful to have a basic understanding
of the biochemistry of meat. This includes the structure of muscle, the
changes that take place after death and where water is held in the muscle.
The factors affecting drip production through the refrigerated cold chain
can then be quantified.
2.1 Biochemistry of meat
2.1.1 Structure of muscle
The structure of muscle has been well described by Voyle (1974) and forms
the basis of this section. Meat consists mainly of skeletal muscles which all
have a similar structure. Figures 2.1–2.4 show in diagrammatic form the
levels of organisation of the components which together form a muscle.The
gross levels of organisation can be resolved with the unaided eye, and it
may be observed that each muscle is separated from its neighbour by a
sheet of white connective tissue – the fascia. This gives support to the func-
tional components of the muscle and connects it to the skeleton through
tendinous insertions. The connective tissue consists mainly of collagen and
in some muscles includes elastic fibres.
In cross-section (Fig. 2.1) a muscle appears to be subdivided into tissue
bundles surrounded by thin layers of connective tissues. These bundles
consist of a number of very long, multinucleated cells or fibres each sur-
rounded by a thin layer of connective tissue. Each fibre is about as thick as
a hair of a young child and may be several centimetres in length. Fibres are
normally elliptical in cross-section and have blunt tapered ends (Fig. 2.2).
Fibre thickness varies between muscles within an animal as well as between
species. It is also dependent on age, sex and nutritional status. As an
example, the fibres of the eye muscle (M. longissimus dorsi) of an 18-month-
old steer are about 40mm in diameter.
Each fibre is surrounded by a typical lipoprotein membrane, the sar-
colemma, which in its native state is highly selective in its permeability to
solutes.The space within the sarcolemma is mostly occupied by smaller lon-
gitudinal elements, or myofibrils, each about 1mm in diameter. Figure 2.3
shows part of a single myofibril in longitudinal section. Figure 2.4 repre-
22 Meat refrigeration
sents a single muscle fibre in cross-section, showing myofibrils and associ-
ated structures that are referred to below.
Each myofibril is enwrapped in a thin vesicular structure the sarcoplas-
mic reticulum, which is involved in the transmission of the nervous impulse
to the contractile elements. The characteristic striated appearance of each
muscle fibre, represented in Fig. 2.2, may be observed by direct microscopy.
The finer details of structure, represented in Figs 2.3–2.4, can only be
resolved by electron microscopy.
Between the myofibrils are small particles, the mitochondria, which
provide the energy for contraction via oxidative processes. The myofibrils
are bathed in a fluid, the sarcoplasm, which contains many soluble enzymes.
These are mostly concerned with the process of glycolysis by which lactic
acid is produced in the oxygen-free post-mortem muscle. The myofibrils
occupy about 74% of the total fibre volume.
The myofibrils are packed with contractile microfilaments of actin and
myosin which, in cross-section, may be seen to be arranged in a hexagonal
lattice. The interdigitating sliding action of these filaments when stimulated
to contract is suggested by the longitudinal view represented in Fig. 2.3. A
fibril contains about 16% contractile protein and about 84% water in
which are dissolved small solutes such as adenosine triphosphate (ATP),
Drip production in meat refrigeration 23
Fig. 2.1 Diagrammatic representation of cut surface of muscle to show bundles of
fibres (source: Voyle, 1974).
Fig. 2.2 Single muscle fibre. Diagrammatic representation of morphology as seen
by direct microscopy (source: Voyle, 1974).
the fuel for contraction, but from which the larger enzyme molecules are
excluded.
The fluid within the fibrils is distributed between the microfilaments of
the hexagonal lattice. After rigor in a muscle at rest length the filament
lattice volume decreases and releases fluid into the spaces between the
myofibrils, i.e. into the sarcoplasm.The permeability of the sarcolemma also
changes after rigor, and fluid, generally referred to as ‘drip’, escapes into
the extracellular space. The extent to which this happens depends upon the
ultimate level of pH attained by the post-rigor muscle.
24 Meat refrigeration
Fig. 2.3 Part of myofibril. Diagrammatic representation to show filament-array in
longitudinal section with adjacent structures (source: Voyle, 1974).
Fig. 2.4 Cross-section of single fibre, showing myofibrils and other structures
(source: Voyle, 1974).
2.1.2 Changes after slaughter
Muscles of freshly killed mammals are relaxed, soft, extensible and flexible.
However, after a short time they become stiff, rigid and contracted. This
state is called rigor mortis.
Muscles obtain the energy they need for contraction by taking up glucose
from the blood and storing it in a polymeric form called glycogen. The
chemical fuel the muscle cells use is adenosine triphosphate (ATP), which
as well as providing the energy required to shorten muscle fibres, acts as a
lubricant during contraction preventing cross-linking. Muscles power con-
traction by hydrolysing this ATP to the diphosphate (ADP) and inorganic
phosphate (Pi) but there is only enough ATP in muscle cells to fuel a con-
traction for three seconds. For a sustained contraction, the ATP has to be
resynthesised from ADP and Pi by coupling this energetically unfavourable
reaction to the energetically favourable breakdown of glycogen to lactic
acid (Fig. 2.5).
In muscle after death, the rate of breakdown of ATP is low but still
appreciable and the muscle draws slowly on its glycogen stores. These are
not replenished because there is no longer a blood supply. The lactic acid
accumulates and the pH falls from an initial value of about 7 to a final value
of about 5.5 to 6.0.
When the breakdown of glycogen comes to a halt, the ATP concentra-
tion falls to zero and the force-generating machinery of the muscle stops in
mid-cycle causing the muscle to become rigid and inextensible. It is then
said to be in the state of rigor mortis (rigor for short).
The most important structural change in muscle tissue during the onset
of rigor is the formation of actomyosin complex caused by the cross-linking
of actin and myosin filaments and muscle contraction brought about by the
breakdown of ATP. Breakdown of ATP also contributes to the temperature
rise (0.2–2.0 °C) which is sometimes observed in the deep musculature of
pigs and beef animals during the first hour or so after slaughter, as described
by Bendall (1972) and measured by Morley in 1974.
Normal rigor sets in before glycolysis ends, i.e. before reaching the final
pH value. The time that rigor takes to develop (Table 2.1) is dependent on
muscle type, its posture on the carcass, rate of cooling and so on (Offer
et al., 1988). Temperature is particularly significant. Between 10 and 37 °C
Drip production in meat refrigeration 25
Lactic acid
Glycogen
ATP
ADP + Pi
Fig. 2.5 Reaction of ATP in muscle (source: Bendall, 1972).
the rate of rigor development increases with temperature, like many other
metabolic processes. The rate increases three to four times for each 10 °C
rise in this range.
As a result of this fall in pH a number of enzymes change their activity.
Some lose it by changing their three-dimensional structure and some
enhance their activity, i.e. especially liposomal enzymes which are necessary
for the conditioning process (Honikel, 1990). In the course of the break-
down of energy-rich compounds (shortly before they get used up) the onset
of rigor occurs which increases the rigidity of the meat, i.e. the meat tough-
ens. Conditioning reduces the toughness as the number of rigid longitudi-
nal and transversal cross-links in the myofibres are reduced by enzymic
action (Honikel, 1990).
The conditions for the onset and development of rigor have a profound
influence on the tenderness, juiciness and water-holding capacity of meats.
While factors such as species, breed, age, nature of muscle, ante- and post-
mortem treatments, and so on all have an influence, temperature is prob-
ably the most important.
Conditions of exhaustion or stress before slaughter can cause changes in
the degree of glycolysis producing detrimental effects to the meat. Animals
subjected to severe exhaustion shortly before slaughter use up their glyco-
gen reserves thus less lactic acid is formed producing high pH (6.0–6.5) dark
meat, often described as dark, firm and dry (DFD) meat. DFD problems
can occur in pork, mutton, veal and beef. By convention all pork above pH
6.0/6.2 is classified as DFD meat (Honikel, 1990). Drip losses from DFD
meat are less than from normal meat (Offer et al., 1988).
A second cause of shrinkage is protein denaturation. In life, muscle pro-
teins are stable for many days at 37 °C and pH 7. However, after death the
musculature, especially in the interior of the carcass, cools relatively slowly
and becomes acidic. Under this combination of high temperature and low
pH, some proteins especially myosin, the principal protein of muscle, slowly
denature. If sufficient myosin is denatured, the myofibrils shrink about twice
as much as usual and the meat is pale, soft and exudes drip more quickly
and in greater amounts than usual. Consumers react unfavourably against
the unattractive paleness of this pale, soft and exuding (PSE) meat.
26 Meat refrigeration
Table 2.1 Typical time for rigor onset
Type of meat Development time Range (h)
for rigor (h)
Beef 18 8–30
Lamb 12 10–20
Pork 3 0.6–8
Source: Offer et al., 1988.
With beef and lamb, provided the chilling regime is adequate, only a little
myosin denaturation occurs probably because the carcass is chilled suffi-
ciently before a low pH is reached. PSE meat is therefore not usually a
problem with these species, except sometimes in the deep muscle if the
carcass has been chilled slowly (Offer et al., 1988).
With pork, however, the pH fall is faster, especially in carcasses of stress-
susceptible animals. In these carcasses, the pH falls to below 6.0 within
45 min of slaughter when the carcass temperature is above 35°C. Myosin
denaturation may then be extensive and pig carcasses are vulnerable to the
PSE state.As well as stress, this condition may be genetically predetermined
(Honikel, 1990).
PSE is not an all-or-none phenomenon and the drip loss depends on the
extent of myosin denaturation. The drip loss can therefore be controlled to
some extent by the chilling regime. Frozen PSE meat exhibits excessive drip
loss on thawing (Honikel, 1990).
2.1.3 Water relationships in meat
In living muscle,85–95% of the total water is held within the fibres in dynamic
equilibrium with the remaining 5–15% (plasma water) outside the fibre
walls.Within the fibre, the water is held both by the contractile, myofibrillar,
filament proteins,myosin and actin,and by the soluble,sarcoplasmic proteins
which include myoglobin and the glycolytic enzymes. The water balance is
such that it allows movement of the proteins within the fibre and exchange
of metabolites in and out of the fibre, without altering the overall amount of
water held. Therefore, when a force is applied to a pre-rigor muscle, excised
immediately after the death of an animal, very little fluid can be squeezed
out. The distribution of space in muscle is shown in Table 2.2.
Calculations can be made of the diameters of the capillary-like spaces
between the filaments of the myofibril and between sarcoplasmic proteins
from which the number of water molecules between nearest-neighbour
structures can be deduced. The results are shown in Table 2.3.
Drip production in meat refrigeration 27
Table 2.2 Approximate distribution of the spaces in
excised muscle
Structure Volume as % total vol.
pre-rigor post-rigor
Extrafibre space <12 100
Intrafibre space 88–95
Extrafibrillar space 22–24 30–32.5
Intrafibrillar space 66–71 58–62
a
a
Assuming a 12% reduction in filament lattice volume post rigor.
Source: Penny, 1974.
These show the capillary spaces between the elements are very small so
that it seems reasonable that much of the water would be held by surface
tension forces. In addition, quite a large proportion of the water should be
immobilised by surface charges on the proteins.
When a muscle goes into rigor a number of important changes take
place, which affect the water balance. As a result of the loss of ATP, the
actin and myosin filaments become bonded together and tend to squeeze
water out of the filament lattice into the sarcoplasmic space, and possibly
also into the spaces between fibres. This squeezing effect is increased as
the pH falls from 7.2 in pre-rigor muscle to 5.5–5.8 in post-rigor muscle.
This is because the proteins are then much nearer the mean isoelectric
point of 5.0–5.2 at which their hydration is at a minimum and their
packing density maximal (Rome, 1968). This, no doubt, explains Hegarty’s
(1969) finding that muscle fibre diameter decreases during rigor, which also
suggests that the fibre wall has become leaky and allowed fluid to escape.
Table 2.2 gives the approximate change in the distribution of space which
would occur if the myofibrillar lattice volume was reduced by 12% (Rome,
1968).
The loss of water binding by the proteins also depends on the amount
of denaturation that has taken place in the post-mortem period. Denatu-
ration is an irreversible alteration to the structure and properties of the
proteins. Denaturation leads to extra loss of water binding and to closer
packing of the fibrillar proteins. It is a function of the post-mortem rate of
cooling and the rate of pH fall, and increases dramatically at low rates of
cooling and high rates of pH fall.
As a result of all these post-mortem changes, a considerable amount of
previously immobilised water is released by the proteins and redistributed
from filament spaces to sarcoplasmic spaces within the fibres, and also into
the spaces outside the fibres. This released water makes up most of the fluid
(drip) which can then be squeezed out of the meat.
28 Meat refrigeration
Table 2.3 Diameters of the ‘cylindrical’ capillary spaces between nearest-
neighbour elements of the fibre and the number of water molecules
accommodated between surfaces of nearest-neighbour protein molecules
Elements Diameter of capillary Number of molecules
(nm) of water
Actin–myosin overlap 21.5 42
Myosin–myosin (H-zone) 38.4 120
Actin–actin (I-zone) 45.3 67
Sarcoplasmic proteins
a
15.3 30
a
Assuming the average molecular weight (MW) = 120 000 Da and a mean diameter of
6.52 nm.
Source: Penny, 1974.
2.1.4 Ice formation in muscle tissues
In general, freezing and thawing exacerbate drip loss through damage of
the muscle structure. It is necessary to differentiate between the effects of
freezing in pre-rigor and post-rigor muscle. For most practical purposes,
meat is in the latter condition but there has been considerable interest in
the rapid freezing of ‘hot’, i.e. pre-rigor meat.
2.1.4.1 Pre-rigor muscle
The freezing of meat immediately after slaughter appears at first sight to
be an excellent method of overcoming many of the chilling, hygiene and
storage problems of conventional production methods. However, there are
two problems, ‘cold shortening’ and ‘thaw rigor’, that result in very tough
meat and that have to be overcome to make such a process viable. Thaw
rigor, or ‘thaw contractor’ as it is sometimes called, also significantly
increases drip loss after thawing.
If the meat temperature falls below 10 °C before the supply of fuel for
contraction, i.e. ATP, is used up, but freezing has not occurred, the muscle
will contract. This phenomenon called ‘cold shortening’ was first described
by Locker and Hagyard (1963) and is discussed in Chapter 3 of this book.
The protein denaturation that results from cold shortening produces a large
amount of drip (Offer et al., 1988).
If very high rates of heat extraction can be achieved, then the meat can
be frozen fast enough to stop cold shortening. However, in this case, a
more severe shortening, thaw rigor, will occur during thawing. In unre-
strained muscle up to 25% of the muscle weight will be lost in the form
of drip during thawing (Bendall, 1974). Bendall stated that the problems
associated with thaw rigor could be overcome by holding the frozen
meat at -3 to -5 °C for at least 48 h. However, such a process is not used
commercially.
2.1.4.2 Post-rigor muscle
Chemical changes after slaughter cause the acidity of the tissue to increase
and the pH falls to a level which is normally in the range of 5.5–5.7. This
compares with a pH of about 7.0–7.2 in the living tissue. One of the conse-
quences of this fall in pH is a change in the permeability of the sarcolemma
which now permits sarcoplasmic proteins and water to pass more readily
out of the cell (Voyle, 1974). When the tissue is slowly cooled below its
freezing point, this protein-containing fluid is extracted from the cell to con-
tribute to the growth of extracellular ice crystals. Loss of fluid from the cell
results in an increase in the intracellular salt concentration. This in turn
causes some denaturation of those proteins remaining within the cell. A
more rapid rate of freezing will cause the intracellular water, including that
in the actin–myosin lattice, to crystallise.
Drip production in meat refrigeration 29
2.2 Measurement of drip
Many methods have been used to measure drip loss from meat. Data
obtained using different methods can be used to determine trends but the
values obtained are not directly comparable.
The most important factor that affects the measurement of drip is the
ratio of cut surface to weight or volume. It is clear that the free water has
to move to the surface before it can drip from the meat and therefore the
more cut surface to volume there is, the less distance the water has to travel.
In 1956, Howard and Lawrie reported that drip from beef quarters, domes-
tic joints and small samples in the laboratory ranged from 0.3 to 1, 1.2 to 2,
and 4 to 10% of weight, respectively. Howard (1956) showed that pieces
with the same cross-section but 1 and 3cm thick lost 8 and 6% as drip,
respectively. The drip is also reduced if the pieces are cut along the direc-
tion of the fibres rather than across it. Pressure applied to slices or blocks
of meat increases the amount of drip and so does an absorbent material
placed on the cut surfaces because of the increase in hydrostatic pressure.
It is therefore important that an appropriate method is used in order to
obtain data that are directly applicable to a commercial situation. Weigh-
ing unwrapped samples of meat provides information on total weight loss.
However, some of the loss is due to evaporation from the surface, not drip.
One simple method is to hang the preweighed meat, using a nylon mesh to
support it. A polythene bag is then placed round the sample but not in
contact with it to prevent evaporation. The system is then kept in a con-
trolled environment and the sample reweighed after a set time.
For experimental purposes more information and better reproducibility
can usually be obtained from methods where force is applied rather than the
simple method of measuring ‘free’ drip (Penny,1974).These include the press
method of Grau and Hamm (1953) or methods depending on centrifugation.
2.3 Factors affecting the amount of drip
Some factors that affect the amount of drip are inherent in the animal and
include the breed of the animal and the position of the meat within the
animal. Treatment of the animal before slaughter, especially in the case of
pork, can influence drip production by producing DFD or PSE meat. The
conditions in and the length of the refrigerated cold chain will further influ-
ence the resulting drip.
2.3.1 Animal factors
2.3.1.1 Breed
In pigs especially, there are large differences in drip loss from meat from
different breeds. Taylor (1972) measured drip loss from leg joints from four
30 Meat refrigeration
different breeds subjected to two different chilling regimes. Drip loss was
estimated by suspending pieces of meat in sealed polythene bags and weigh-
ing the amount of free liquid that accumulated in the bag during storage at
0 °C. Since most of the drip was lost during the first two days of storage,
drip was always expressed as the weight of exudate after 2 days at 0°C.
Analysis of the mean values for each breed (Table 2.4) showed that there
was a substantial difference, up to 2.5-fold, in drip loss between breeds.
In a further comparison, four major leg muscles were excised 24 h post-
slaughter from sides of Large White and Pietrain pigs. The average levels
of drip loss from the four muscles varied by 1.65-fold for slow cooled and
just over two-fold for the quick cooled between the breeds (Table 2.5).
The rate of pH fall after slaughter was shown to be a major factor where
pig meat was concerned. Pigs with pH values below 6.1 (30 min after slaugh-
ter) tended to give meat with high drip loss, while values above 6.1 were
associated with low loss. The incidence of rapid pH change varies to some
extent with breed and the excessive drip from the Pietrain samples was
Drip production in meat refrigeration 31
Table 2.4 Drip loss after two days storage at 0 °C from
leg joints from different breeds of pig cooled at different
rates
Breed Drip loss (% by wt.)
Slow Quick
Landrace 0.47 0.24
Large White 0.73 0.42
Wessex X Large White 0.97 0.61
Pietrain 1.14 0.62
Source: Taylor, 1972.
Table 2.5 Drip loss after 2 days storage at 0 °C from four muscles from two
breeds cooled at different rates
Drip (as % muscle weight)
Cooling Semi- Semi- Adductor Biceps Combined
rate tendinosus membranosus femoris (four
muscles)
Pietrain Quick 2.82 4.40 5.52 2.69 3.86
(13 pigs) Slow 3.99 6.47 6.61 4.11 5.30
Large Quick 1.69 2.01 2.92 1.04 1.92
White Slow 1.95 3.50 5.07 2.32 3.21
(6 pigs)
Source: Taylor, 1972.
undoubtedly a consequence of their rapid pH fall (Lister, 1970; MacDougall
and Disney, 1967). The average pH
30
for Pietrains in the experiment com-
paring drip from muscles was 6.04 (range 6.80–5.60), while that for Large
Whites was 6.52 (range 6.75–6.35).
The range of ultimate pH in these experiments was very narrow
(5.5–5.7), a result of using animals which had been well rested overnight
before slaughter. It is only when pre-slaughter conditions cause an abnor-
mally high ultimate pH, that the water-holding capacity of the meat is
markedly improved and drip reduced.
2.3.1.2 Muscle type
Different muscle groups show different degrees of drip. Taylor (1972)
showed there to be a significant anatomical distribution of drip loss in pig
carcasses that was not changed by either breed, carcass weight or rate of
cooling. This general pattern is illustrated in Fig. 2.6. Three breeds of pig,
Large White, Landrace and Pietrain were used, with carcass weights ranging
from 40 to 60 kg. Twenty-two sides were chilled at a variety of cooling rates
until carcass temperature was uniformly at 0 °C after 24 h. The cooled sides
were then jointed in a chill room at 0 °C and drip estimations carried out
using the method detailed in the previous section.
32 Meat refrigeration
6
5
4
3
2
1
0
g drip/100 cm
2
cut lean surface
g drip/100 g meat
0
0.4
0.8
1.2
1.6
2.0
14
7
13 12 11
3
4
31
2
1
6
9 10
5
8
(b)
(a)
Fig. 2.6 Distribution of drip in pig carcasses expressed on basis of (a) area of cut
lean surface and (b) of weight of joint. Numbers refer to standard jointing system
(source: Taylor, 1972).
The joints with the greatest drip loss were the commercially valuable
chops, chump and leg joints. Table 2.6 shows that 84% of the gross drip loss
came from these joints which made up only 54% of the total weight of meat
on the carcass.
In beef the topside and rump regions are particularly bad in respect to
drip, in comparison with the l. dorsi (Taylor, 1972). However, the difference
between muscles may not apply to all animal species. Dawood (1995) found
that average drip losses after thawing Najdi Camel steaks from chuck,
ribeye and leg were all in a similar range, 9.77–12.34%, and unaffected by
the age (8–26 months) of the animal.
2.3.2 Refrigeration factors
The rate of change of temperature during chilling and the temperature at
which meat is stored during the cold chain influence drip loss. Freezing and
subsequent thawing substantially increase drip loss from meat.
2.3.2.1 Chilling
Rapid cooling of meat immediately after slaughter will reduce drip loss
after subsequent cutting operations.The potential for drip loss is established
in the first period of cooling, the temperature range conducive to drip is
down to about 30 °C or perhaps a little lower.
There are a number of publications showing that rapid cooling can
reduce drip production. Taylor (1972) compared two cooling treatments for
pig carcasses (Table 2.7).
In 38 out of 40 paired legs, the drip loss was less after the quicker cooling.
The difference varied between breed (Table 2.4) and ranged from ap-
proximately 1.6- to two-fold. In other studies the two cooling rates again
gave highly significant differences in drip loss from four muscles (Table 2.5).
Similar experiments were also carried out on beef using four cooling
procedures:
1 23 h at 0 °C (air at 1–2 m s
-1
) + 24 h at 0 °C (still air)
2 47 h at 0 °C (still air)
Drip production in meat refrigeration 33
Table 2.6 Drip loss from wholesale groups of joints, after 2 days at 0 °C
Joint Joint weight % of carcass Drip loss g drip/100 g
(kg) weight from joint
(g)
Leg 5.608 24.8 43.4 0.77
Loin 6.500 28.7 82.0 1.26
Fore-end 7.044 31.1 21.8 0.31
Belly 3.482 15.4 2.1 0.06
Side total 22.634 – 149.3 0.66
Source: Taylor, 1972.
3 6h at 15°C + 41 h at 0°C (still air)
4 23h at 15°C + 24 h at 0°C (still air).
Opposite sides of the same animal were used for comparison and drip was
measured in four 25.4mm thick slices of l. dorsi cut between the 9th and
10th ribs and stored at 0°C for 43h. The saving in drip gained by cooling
quickly after slaughter was clearly shown (Table 2.8). The mean loss from
the samples taken from sides cooled quickly by method 2 was 1.1%,
while that from sides cooled at the slowest rate, method 4, was 2.7%.
Other subsequent studies have shown that rapid chilling, as long as freez-
ing of the muscle or cold shortening is avoided, will substantially reduce
drip loss.
Gigiel et al. (1985) removed cylindrical samples of muscle from freshly
slaughtered beef. The curved surface and one end of the cylinder were sur-
rounded by insulation and the free end placed in contact with solid carbon
dioxide (CO
2
). Since heat was only extracted from one end this produced
a wide range of cooling rates through the length of the cylinder. After
cooling and equalisation, the cylinder was cut into discs and the drip poten-
tial of each disc measured using a centrifuge technique described by Taylor
(1982). The resulting plot of drip loss against cooling rate is shown in (Fig.
2.7). Close to the surface in contact with the CO
2
the rate of cooling was
34 Meat refrigeration
Table 2.7 Temperatures in pig carcasses during cooling quickly and slowly
Time after slaughter (h) Temperature ( °C)
Deep leg Longissimus dorsi
Slow Quick Slow Quick
5 301924 7
10 14 5 6 0
15 300
Source: Taylor, 1972.
Table 2.8 Drip loss after 43 h at 0 °C from longissimus dorsi samples removed
from slowly and quickly cooled beef sides
Cooling method Range drip (% by wt) Ratio drip,
Slow Quick Slow Quick
slow/quick
3 1 0.2–3.2 0.1–2.0 2.2
4 2 0.8–4.1 0.6–1.8 2.4
4 1 3.6–4.4 0.9–2.1 2.9
Source: Taylor, 1972.
highest but freezing occurred and the drip was high. Minimum drip poten-
tial was measured in the next region where high cooling rates were achieved
without freezing. Drip then increased as cooling time to 7 °C increased.
Hot boning, i.e. removing the meat from the carcass immediately after
slaughter, has been shown to reduce drip in pork provided chilling is strictly
controlled. Honikel (1990) showed that preventing cold shortening is the
key to reducing drip. The two muscles studied, l. dorsi and semimembra-
nosus, exhibit different pH fall rates, with semimembranosus having a
slower pH fall than l. dorsi. Consequently under the chilling conditions
used, even with the slower rate in the first experiment (Table 2.9), the hot
boned semimembranosus exhibited some shortening, thus the difference
Drip production in meat refrigeration 35
14
Drip loss
12
10
8
6
4
2
0
0.86 1.3 2.4 3.5 10.9 13.7 15.7 17.8 19.6 20.4 21.3 21.7 22.2 23.5 25
Cooling time to 7 °C (h)
% Drip loss
Fig. 2.7 Percentage drip loss from beef sample as function of cooling time at 7 °C
(source: Gigiel et al., 1985).
Table 2.9 Effect of hot versus cold boning on drip loss from pork muscles,
stored at 0–3 °C for 7 days post mortem
Expt. No. of Muscle Drip loss (%)
samples
Hot boned
a
Cold boned
b
Difference
1 19 L. dorsi 9.5 (±2.4) 11.8 (±3.1) 2.3
19 L. dorsi 9.9 (±1.5) 11.65 (±2.7) 1.75
19 Semimembranosus 6.95 (±2.9) 7.05 (±3.1) 0.1
2 22 L. dorsi 6.95 (±2.4) 7.25 (±2.75) 0.3
22 L. dorsi 7.0 (±2.4) 7.6 (±3.1) 0.6
22 Semimembranosus 7.65 (±2.6) 6.6 (±2.6) -1.05
a
Hot boned within 45 min post-mortem and chilled uniformly; in experiment 1, 7 °C within
10 h, 2 °C at 24 h post-mortem; in experiment 2, 7 °C within 6 h, 2 °C at 14 h post-mortem.
b
Cold boned after chilling carcass less uniformly to 12 °C within 15 h; then deboned and chilled
rapidly to 2 °C.
Source: Honikel, 1990.
between hot and cold boning is small. The faster chilling rate in the second
experiment induced even more shortening in the hot boned semimem-
branosus increasing drip further. Under the conditions used in the first
experiment neither cold- nor rigor-shortening occurred in the l. dorsi.
2.3.3 Chilled storage
In meat, drip loss increases with length and temperature of chilled storage
(Fig. 2.8). Drip loss from pork cubes increased substantially during 21 days
of storage at 0, 3 and 7°C. The rate of increase was greater at the higher
temperatures. In storage at 0 and 3 °C no increase in drip loss with time was
measured after 21 days. At 7 °C drip was still increasing between 21 and 28
days. At -4 °C the samples remained frozen and no drip was observed until
the samples were thawed.
In beef, Boakye and Mittal (1993) reported a different relationship
between the length of time longissimus was conditioned (aged) and drip
loss. There was small, but not significant, increase in drip over the first 8
days of ageing.A marked increase in drip was measured on day 12 followed
by a marked decrease on day 16.
The importance of effective secondary cooling after cutting is shown by
the data of L?ndahl and Eek (1986). The amount of drip loss from pork rib
after cutting when held at 10 °C was twice that of meat chilled and held at
2.5°C (Table 2.10).
Drip from offal also increases with length of storage but is very variable.
Strange (1987) cut pigs liver into 1.25 cm thick strips on the day of slaugh-
ter, stored them at 5 °C and measured the drip loss during chilled storage.
On the day of slaughter the average drip was 0.72% (range 0–2.27%), after
2 days storage it averaged 2.61% (range 0.9–4.52%), and after 4 days it was
2.9% (range 1.01–5.07%).
36 Meat refrigeration
5
4
3
2
1
0
0714 21 28
% Drip loss
Storage time (days)
–4°C
0°C
7°C
3°C
Fig. 2.8 Drip loss from vacuum-packed cubes of pork stored at -4, 0, 3 and 7 °C
(source: Lee et al., 1985).
2.3.3.1 Freezing
The above considerations apply to fresh meat, where the decrease in the
water-holding capacity is determined by the post-mortem conditions. Freez-
ing, apart from one case reported by Deatherage and Hamm (1960), always
tends to decrease water-holding capacity and hence increase drip. When
meat is frozen quickly, the water released by the fibrils as the meat has gone
into rigor, and the water which is still held are both frozen simultaneously.
Consequently, there is no change in their relative positions or amounts. At
slower freezing rates, however, the water balance is altered, the extracellu-
lar water freezing first. As freezing continues, the existing ice crystals grow
at the expense of water from the intrafibrillar space. This can result in salt
crystallisation and pH changes (van den Berg, 1964; 1966) which potentially
cause protein denaturation. This has been well documented for frozen
fish (e.g. Love, 1966) which is much more susceptible to freezing damage
than meat.
A number of scientific investigations, which can be compared to com-
mercial practice, have defined the effect of freezing rate on drip produc-
tion. Petrovic et al. (1993) stated that the optimal conditions for freezing
portioned meat are those that achieve freezing rates between 2 and 5 cmh
-1
to -7 °C. Grujic et al. (1993) suggest even tighter limits of 3.33–3.95 cmh
-1
.
They found that ‘slow freezing’ up to 0.39 cmh
-1
resulted in decreased
solubility of myofibrillar proteins, increase in weight loss during freezing,
thawing and cooking, lower water-binding capacity and tougher cooked
meat. ‘Very quickly frozen’ meat (>4.9cmh
-1
) had a lower solubility of
myofibrillar proteins, lower water-binding capacity and tougher and drier
meat. The samples were thawed after storage times of 2–3 days at -20°C
so the relationship between freezing rates and storage life was not investi-
gated. Sacks et al., (1993) found that after 2.5 months, drip loss from mutton
samples frozen using cryogenics was >2% less than in those using air freez-
ing (Table 2.11).
These results are scientifically very interesting, however, in industrial
practice most meat is air frozen in the form of large individual pieces or
cartons of smaller portions. In commercial situations freezing rates of
0.5cmh
-1
in the deeper sections would be considered ‘fast’ and there would
Drip production in meat refrigeration 37
Table 2.10 Drip loss (%) from pork ribs after cutting
when held at 10 or 2.5 °C
Storage temperature Storage time (h)
(°C)
20 45 70 90
10 1.3 2.2 2.9 3.2
2.5 0.5 1.0 1.4 1.7
Source: L?ndahl and Eek, 1986.
be considerable variation in freezing time within the meat. The samples
frozen by Sacks et al. (1993) were much smaller (77.6 g in weight) than most
commercial products. Even with such small samples there was no signifi-
cant difference in drip after 48 h between cryogenic freezing at -90°C and
a walk-in freezer operating at -21°C.
Sakata et al. (1995) investigated the freezing of samples similar in size
to a domestic joint. They found no significant difference in drip loss from
700 g samples of pork l. dorsi frozen in air at -20 or -80°C. At -20°C
samples required ca. 6 h to pass from -1 to -6 °C compared with half this
time at -80°C. Average drip losses were 3.7% at -20°C and 5.2 at -80 °C.
As described in Chapter 9, in commercial situations freezing times typically
range from tens of hours to a few days. Freezing rates are therefore outside
the values that influence drip potential.
In a number of operations, meat is ‘tempered’, i.e. partially frozen to aid
cutting, dicing, slicing and so on. This process will increase drip loss though
not to the same extent as full freezing. Irie and Swatland (1993) found that
drip loss from 3 mm thick slices of pork that had been ‘lightly frozen’ before
slicing averaged 8.0 ± 4.2% over a 4-day-period. Drip losses from samples
that had been kept in a freezer at -10 °C for 6 days had a higher drip loss
of 14.0 ± 4.3%. Drip was measured by hanging 7 slices in a bag in a refrig-
erator at 5 °C.
In other cases, partial freezing during a chilling operation may increase
drip. James et al. (1983) found that partial freezing of pork during ultra-
rapid chilling produced a four-fold increase in drip.
2.3.3.2 Frozen storage
Storage temperature has a marked effect on the behaviour of ice crystals
that could be detrimental to the ultimate quality of the meat. It has been
demonstrated that frozen tissue stored for 180 days at -20 °C has small ice
38 Meat refrigeration
Table 2.11 Drip loss (%) from 77.6 g samples of Mm longissimus lumborum et
thoracis frozen under different methods and thawed at 4 °C
Freezing conditions Freezing time Freezing rate Storage time
to -2.2 °C (cm h
-1
) at -20 °C
48 h 2.5 months
Cryogenic -90°C 15 month 6.4 3.34
a
9.49
a
Cryogenic -65°C 22 month 4.4 4.70
ab
9.72
a
Blast freezer -21°C 1.83 h 0.55 5.53
b
12.74
b
Walk-in-freezer -21°C 1.88 h 0.53 4.71
ab
13.18
b
Domestic freezer -25°C 1.96 h 0.51 5.26
b
11.72
b
Values given in the ‘storage time’ columns that have the same superscripts (a or b) are not
statistically different (P > 0.05).
Source: Sacks et al., 1993.
crystal formations of irregular shape in the extracellular spaces. Storage for
the same period at -3 °C results in the development of large rounded ice
formations with a concomitant compression of the muscle fibres (Moran,
1932). These changes are also reflected in the increased amount of drip,
which is released from frozen tissue stored at the higher temperature. It is
thus desirable that frozen meat should be stored at a sufficiently low tem-
perature to prevent growth of ice crystals in the extracellular spaces. Such
growth occurs if the temperature of the frozen tissue is allowed to rise
above its eutectic point. However, quoted values for the eutectic point of
meat range from ‘probably just below -20 °C’ (Moran, 1934) to -52°C
(Riedel, 1961).
Drip loss in frozen storage has also been shown to increase with storage
time. Calvelo (1986) states that in general, drip production increases as time
and frozen storage temperature increase. After approximately 42 and 63
days, drip from beef stored at -10 or -15°C had reached 80% and 90%,
respectively of its maximum. At -25 °C it required over 120 days to reach
the 80% value.
Storage times of 48 h and 2.5 months were used during investigations of
the effect of different freezing systems and rates on drip production from
small samples of mutton muscle (Sacks et al., 1993). In all cases drip loss
after 2.5 months was at least double the percentage measured after 48 h
(Table 2.11).
Drip loss during thawing from ground beef patties was also found to
increase with the length of time the patties had been in frozen storage
(Bhattacharya et al., 1988). For higher fat content samples, drip loss
increased from 1.8% in fresh samples to 12.5% after 20 weeks in storage.
Higher drip losses in thawing were obtained from samples stored at
-12.2 °C than those stored at lower temperatures. However, there was no
difference between storage temperatures of -23.3 and -34.4°C.
2.3.3.3 Thawing
When meat is thawed the reverse of the freezing process occurs. Water
which has been frozen is released and has to re-establish equilibrium with
the muscle proteins and salts. Obviously if the muscle proteins have been
denatured they will reabsorb less water. Since the fibres have been squeezed
and distorted by ice formation, this non-reabsorbed water will lie in wider
channels within the meat structure, thus increasing the potential drip. If cell
walls have also been damaged by freezing, even less water will be reab-
sorbed and will exude as drip.
Experiments with pig liver (Strange et al., 1985; Strange, 1987) showed
that repeated freeze–thaw cycles produced increased drip. Strips of liver
were frozen and held at -20°C for 70h (Strange et al., 1985) or 7 days
(Strange, 1987) then subjected to cycles of thawing at 5 °C for 24h followed
by freezing at -20 °C for 24 h. In the 1985 investigations the drip loss was
compared to chilled liver held at 5 °C for sequential 24-h-periods
Drip production in meat refrigeration 39
(Table 2.12). Average drip loss was higher after each freeze–thaw cycle.
However, there was considerable variability between samples and whilst
freeze–thaw cycling always produced more drip than the chilled material,
the difference was not statistically different.
After the first, second and third thawings respectively in the 1987 work
the average drip losses were 5.8% (range 1.7–10.4%), 8.3% (range
3.29–13.4%) and 11.2% (range 4.3–15.8%). Drip loss from fresh liver before
freezing was <1%.
2.4 Conclusions
The amount of drip which is exuded from meat depends on its intrinsic
characteristics, post-mortem treatment and the pH of the meat. It also
depends on the conditions of chilling/freezing, the temperature and time of
storage, the size of the pieces of meat when thawed and the conditions of
thawing.
In general:
1 Although, the potential for drip loss is predetermined to a large extent
by breed and conditions before slaughter, the realisation of this poten-
tial is influenced by the temperature/time history in the cold chain.
2 Rapid cooling substantially reduces drip production, especially over the
critical range from 40 °C to just below 30 °C.
3 During chilled storage, transport and display, drip loss increases with
time.
4 The lower the storage temperature the lower the amount of drip
produced.
5 Pre- and post-rigor muscles differ in their initial reaction to cooling
down to freezing temperature.
6 Freezing will substantially increase the amount of drip.
7 During frozen storage, drip potential increases with time of storage.
8 Within the normal commercial range the rate at which meat is frozen
or thawed has little influence on drip.
9 In commercial situations the amount of drip that appears is greatly influ-
enced by the cut surface area to sample volume ratio.
40 Meat refrigeration
Table 2.12 Drip loss (%) and standard deviation in () from liver subjected to
repeated freeze–thaw cycles or stored at 5 °C
Fresh Cycle 1 Cycle 2 Cycle 4 Cycle 5 Cycle 6
Freeze/thaw 0 3.1 (2.1) 5.5 (3.6) 8.9 (2.4) 11.4 (5.3) 10.5 (4.8)
Chilled 0 1.9 (0.9) 2.7 (2.7) 3.4 (3.7) 5.9 (4.6) 5.7 (5.6)
Source: Strange et al., 1985.
2.5 References
bhattacharya m, hanna m a and mandigo r w (1988), Effect of frozen storage con-
ditions on yields, shear strength and colour of ground beef patties, J Food Sci,
53(3) 696–700.
bendall j r (1972), Calculated post-mortem heating, in Cutting C L, Meat Chilling:
Why and How? Meat Research Institute Symposium No. 2, 12.1–12.3.
bendall j r (1974), The snags and snares of freezing rapidly after slaughter, in
Cutting C L, Meat Freezing: Why and How? Meat Research Institute Symposium
No. 3, 7.1–7.8.
boakye k and mittal g s (1993), Changes in pH and water holding properties of
longissimus dorsi muscle during beef ageing, Meat Sci, 34 335–349.
calvelo a (1986), Freezing effects on meat tissues, in Le Maguer M and Jelen P,
Food Engineering and Process Applications, Oxford, Elsevier Applied Science.
dawood a a (1995), Physical and sensory characteristics of Najdi-camel meat, Meat
Sci, 39 59–69.
deatherage f l and hamm r (1960), Influence of freezing and thawing on hydration
and charges of muscle proteins, Food Res, 25 623.
gigiel a j, swain m v l and james s j (1985), Effects of chilling hot boned meat with
solid carbon dioxide, J Food Technol, 20 615–622.
grujic r, petrovic l, pikula b and amidzic l (1993), Definition of the optimum freez-
ing rate. 1. Investigations of structure and ultrastructure of beef M. Longissimus
dorsi frozen at different rates, Meat Sci, 33(3) 301–318.
grau r and hamm r (1953), A method for measuring the water binding of muscle,
Naturwissenschaft, 6 36.
hegarty p v j (1969), Measurements of unfixed mammalian muscle fibres from pre-
and post-rigor muscles of animals at different stages of development. Proceedings
of the 15th European Meeting of Meat Research Workers, A5, 50.
honikel k o (1990), Pork and pork products, in Gormley T R, Chilled Foods: the
State of the Art, Oxford Elsevier Applied Science, 117–133.
howard a (1956), The measurement of drip from frozen meat, CSIRO Food Res
Quart, 16(2) 31.
howard a and lawrie r a (1956), Technical Paper No. 2, Spec. Rept. Fd. Invest. Bd,
CSIRO Division of Food Preservation and Transport.
irie m and swatland h j (1993), Prediction of fluid losses from pork using subjec-
tive and objective paleness, Meat Sci, 33 277–292.
james s j, gigiel a j and hudson w r (1983), The ultra rapid chilling of pork, Meat
Sci, 8 63–78.
lee b h, simard r e, laleye l c and holley r a (1985), Effects of temperature and
storage duration on the microflora, physiochemical and sensory changes of
vacuum-packaged or nitrogen-packed pork, Meat Sci, 13 99–112.
lister d (1970), The physiology of animals and the use of their muscle for food,
Physiology and Biochemistry of Muscle as a Food – II, University of Wisconsin
Press.
locker r h and hagyard c j (1963), A cold shortening effect in beef muscles, J Sci
Food and Agriculture, 14 787.
l?ndahl g and eek l (1986), Cooling of meat cuts, AGA Frigoscandia publication.
love r i (1966), The freezing of animal tissue, in Merryman H T, Cryobiology,
London & New York Academic Press, 317.
macdougall d b and disney j g (1967), Quality characteristics of pork with special
reference to Pietrain, Pietrain ￥ Landrace and Landrace pigs at different weights.
J Food Technol, 4 285.
malton r and james s j (1983), Drip loss from wrapped meat on retail display, Meat
Industry, May, 39– 41.
Drip production in meat refrigeration 41
moran t (1932), Rapid freezing. Temperature of storage, J Soc Chem Ind, 51,20T.
moran t (1934), Rapid freezing of meat, Report of the Food Investigation Board for
1933, 28.
morley m j (1974), Measurement of the heat production in beef muscle during rigor
mortis, J Food Technol, 9 149–156.
offer g, mead g and dransfield e (1988), Setting the scene: the effects of chilling
on microbial growth and the eating quality of meat, Meat Chilling, IFR-BL:
Subject Day, Meat Chilling, 23 February 1988.
pearson a m (1994), Introduction to quality attributes and their measurement
in meat, poultry and fish products, in Pearson A M and Dutson T R, Quality
Attributes and Their Measurement in Meat, Poultry and Fish Products. Advances
in Meat Research Series, Blackie Academic & Professional, 9 1–13.
penny i r (1974),The effect of freezing on the amount of ‘drip’ from meat, in Cutting
C L, Meat Freezing: Why and How? Meat Research Institute Symposium No. 3,
8.1–8.8.
petrovic l, grujic r and petrovic m (1993), Definition of the optimal freezing rate
– 2. Investigations of the physico-chemical properties of beef M. longissimus dorsi
frozen at different freezing rates. Meat Sci, 33 319–331.
riedel l (1961), On the problem of bound water in meat, K?ltetechnik, 13 122.
rome e (1968), X-ray diffraction studies of the filament lattice of striated muscle in
various bathing media, J Molecular Biol, 37 331.
sacks b, casey n h, boshof e and vanzyl h (1993), Influence of freezing method
on thaw drip and protein loss of low-voltage electrically stimulated and non-
stimulated sheeps muscle, Meat Sci, 34(2) 235–243.
sakata r, oshida t, morita h and nagata y (1995), Physico-chemical and processing
quality of porcine M. longissimus dorsi frozen at different temperatures, Meat Sci,
39 277–284.
strange e d (1987), Quantitation and characterisation of drip from frozen–thawed
and refrigerated pork liver, J Food Sci, 52(4) 910–915.
strange e d, jones s b and benedict r c (1985), Damage to pork liver caused by
repeated freeze–thaw cycling and refrigerated storage, J Food Sci, 50(2) 289–294.
taylor a a (1972), Influence of carcass chilling rate on drip in meat, in Cutting C L,
Meat Chilling: Why and How? Meat Research Institute Symposium No. 2, 5.1–5.8.
taylor a a (1982), The measurement of drip loss from meat, Meat Research Insti-
tute Internal Report.
van den berg l (1964), Physicochemical changes in some frozen foods, J Food Sci,
29 540.
van den berg l (1966), pH changes in buffers and foods during freezing, Cryobiol-
ogy, 3 236.
voyle c a (1974), Structural and histological changes associated with the freezing of
meat, in Cutting C L, Meat Freezing:Why and How? Meat Research Institute Sym-
posium No. 3, 6.1–6.6.
42 Meat refrigeration
