Journal of Dairy Research (1982), 49, 329-336 Printed in Great Britain 329

An improved method for measurement of the syneresis of curd

formed by rennet action on milk
BY RICHARD J. MARSHALL
National Institute for Research in Dairying, Shinfield, Reading RG2 9 AT, UK

(Received 1 October 1981 and accepted for publication 26 November 1981)

SUMMARY. In this method, the curd, formed in a beaker, was retained after cutting
by means of a wire grid placed on top of it. This enabled the whey to be poured off
easily at timed intervals for measurement of its volume. The method gave results
reproducible within 1-5%. Increasing the amount of rennet added to the milk
increased the rate of syneresis of the curd slightly. Adding CaCl2 to the milk, raising
the temperature and decreasing the pH, all increased the rate of syneresis. Addition
of fat, as cream, to skim-milk decreased the rate of syneresis proportionally. The rate
of syneresis was decreased if the milk was concentrated by ultrafiltration and
increased if the milk was diluted with permeate. In all instances the early stages of
syneresis followed first-order kinetics, the rate depending on the amount of whey
remaining within the curd. The later stages of syneresis appeared to depend on
hydrophobic forces. Probably, the increasing aggregation of the casein within the curd
caused the whey to be squeezed out.

Syneresis of curd formed by the action of rennet on milk comprises shrinkage with
expulsion of whey. It is an essential step in reducing the moisture of the curd, which
is an important factor in determining Cheddar cheese quality (Lawrence & Gilles,
1980). Syneresis in cheesemaking is initiated by cutting the curd, and enhanced by
stirring and by increasing the temperature and acidity of the curds and whey. When
modifications are made to the milk or processing conditions, the rate of syneresis may
be affected.
Two main classes of methods for measuring syneresis have been described. (1) Curd
draining methods involve separation of the curds and whey so that the weight, volume
or composition of one can be measured. Some of the methods are discontinuous (e.g.
Lawrence, 1959a), requiring a separate sample of curds and whey for each time
interval. This may lead to difficulties of scale, particularly in experiments involving
comparison. Other methods require repeated manipulation of the same curd (e.g.
Thome et al. 1958). This may be difficult to achieve without causing the curd to break
so that effective separation from the whey becomes impossible. (2) Tracer methods
involve progressive dilution of an added tracer as the whey volume increases (e.g.
Beeby, 1959). For small-scale use, the major problem with these is to achieve
complete mixing of all the whey with the tracer without excessive disturbance of the
curd. Probably because of these disadvantages, few of the methods have been used
by workers other than the original authors.
This paper describes an improved method for measuring syneresis rate which is
simple and precise. It is a curd-draining method but involves placing a grid over the
curd in the beaker. This enables the whey to be poured off repeatedly with little
330 R. J. MARSHALL

disturbance of the curd. The reproducibility of the method and the effect of removing
whey on the rate of syneresis were assessed. The effects of changes in pH, temperature
and rennet level and in the concentration of Ca2+, fat and total solids in the milk were
investigated. In all instances the effect of cutting time, which relates to the firmness
at which the curd was cut, was determined. The variables were selected as being ones
which could be affected by the choice of cheesemaking conditions.

EXPERIMENTAL
Materials
Raw milk was obtained from the Institute herd of Friesian cows. The rennet used
was Standard Cheese Rennet from Chr. Hansen's Laboratories, Reading, Berks, UK.
Methods
Measurement of syneresis. Up to 12 beakers (150 ml), each containing 50 ml milk,
were suspended in a water bath by a polystyrene frame so that the surfaces of the
milk were about 2 cm below that of the surrounding water. Rennet was added and
stirred in for 30 s. The rennet clotting times (RCT) of the milks were determined in
separate samples under the same conditions, using the method of Berridge (1952).
At the appropriate time, the curds were cleanly cut crosswise with a scalpel and
around the edge with a thin, straight spatula. Immediately, stainless steel wire grids
(8 wires/cm), slightly smaller in diameter than the inside of the beakers, were placed
gently on the surface of the curds. At intervals the beakers were removed from the
water bath and the volumes of whey were poured off into measuring cylinders. A
spatula, with its end bent at right angles, was used to keep the wire grids in place
and so prevent the curds from falling out. The whey was drained for 30 s and then
the beakers were replaced in the water bath. This was repeated every 10 or 15 min
and the cumulative whey volumes were measured each time.
The grids exerted a pressure of about 12 Pa on the curd surface.
Experimental conditions. Standard conditions for measuring syneresis of raw whole
milk were defined as 30 °C, 0-26 ml/1 rennet, pH 66, no added Ca or fat, normal
milk concentration.
The Ca level of the milk was increased by adding 1 M-CaCl2 at constant pH. The
pH of the milk was lowered by adding small quantities of 4-5 M-lactic acid and
allowing 15 min for equilibration. The fat content of the milk was varied by skimming
the milk and adding cream to achieve the desired fat concentration. Milk was
concentrated in a bench-scale ultra-filtration plant at 50 °C (Model DC2A, hollow fibre
filtration cartridges type H1P10, cut off MW 10000; Amicon, Woking, Surrey, UK).
All the milk was concentrated to approximately twice normal and lower concentrations
were produced by back-dilution of retentate with permeate.

RESULTS

Time course of syneresis

The rate of whey loss increased rapidly in the first 60 min after cutting (Fig. 1).
During that time, the rate of syneresis, under standard conditions, fitted (r = 1-00)
the first-order kinetic equation:
 Syneresis of rennet curd 331
50

30

20
1
10

30 60 90 120 150 180

Time from cutting, min
Fig. 1 Syneresis of curd from raw whole milk at pH 6-6, 30 °C, 0-26 ml/1 rennet. Curd cut 40 min after
renneting. Other experimental details are given in the text.

where Co is the original milk volume and C is the curd volume (milk volume minus
whey volume) after time t. (The value of C at t = o was not included because some
whey was released at the instant of cutting. From the constant 0-023 its volume was
calculated to be 2-4 ml). This indicated that the initial rate was dependent on the
proportion of whey in the curd. The results at later times deviated from this equation
indicating that other factors were also significant in determining whey expulsion. At
60 min after cutting, the water content of the curd was approximately 70% (v/v).
Reproducibility of the method
In 6 identical experiments carried out simultaneously on the same milk, the
coefficient of variation of the whey volumes at each time interval, as a percentage
of the milk volume, was less than V5%.
Effect of removing whey on the rate of syneresis
Removal of the whey by draining every 10 min increased syneresis by an average
of about 3-5 % compared with the volume of whey collected by draining once only
(Table 1).
Effect of curd firmness on syneresis
All the variables which affect the syneresis rate may also influence the rate at
which the curd firms. Therefore, in each of the syneresis experiments, curds were cut
at 4 different times in the range 2—5 x the RCT and the rates of syneresis were
measured in each instance. During this period the curd-firmness, measured by the
Ultra-Viscoson 1800 vibrating-reed viscometer (Bendix Corporation Inc., Lewisburg,
W. Virginia, USA) (Marshall et al. 1981) increased with time (Table 2). Thus, it was
possible to take account of both the effect of curd-firmness and another variable on
the rate of whey expulsion. With most of the controls, the syneresis rates were faster
from curds cut early and late than from those cut at intermediate times when there
was less influence of cutting time on syneresis rate (Fig. 2).
332 R. J. MARSHALL

Table 1. Effect of draining whey on the rate of syneresis of curd formed from normal
renneted milk under standard conditions
Whey volume, % of milk volume
Time from cutting,
min Drained every 10 min Drained once only
20 118 —
30 15-9 154
40 200 18-6
50 22-2 221
60 250 24-2

Table 2. The curd firmness of normal milk coagulated by rennet under standard
conditions at various cutting times after the rennet coagulation time (RCT)
Time of cutting relative Curd firmness* (% of
to RCT ( = x l ) firmness at 2 x RCT)
2 100
3 131
4 154
5 172
* Mean of 3 determinations at each time.

Effects on syneresis rate of pH, temperature and rennet level and concentration of Ca,
fat and total solids in the milk
For all these experiments the results are expressed as average rates of syneresis
between 15 and 60 min after the cutting time.
Decreasing the pH of the milk from 6-6 to 6-0 increased the rate of syneresis at
all cutting times, though the increase in the range pH 6-6-63 was greater than that
between pH 6-3 and 6-0 (Fig. 2a). When the temperature was raised from 25 to 35 °C,
the rate of syneresis more than doubled, but there was a tendency for the rate to fall
at 35 °C with increase in cutting time (Fig. 26). Increasing the rennet concentration
4-fold increased the syneresis rate slightly (Fig. 2c). At the higher rennet level the
rate at intermediate cutting times was less than at shorter and longer ones, the
difference being more pronounced than for the control. When 2 mM-CaCl2 was added
to the milk, the rate of syneresis was increased at all cutting times (Fig. 2d). However,
a further increase to 4 mM-CaCl2 caused an additional increase in the syneresis rate
only at the 2 shorter cutting times. Increasing the fat concentration from zero in
stages up to approximately twice that in the control progressively decreased the rate
of syneresis at all cutting times (Fig. 2/). The addition of fat reduced the amount
of whey present in the curd so that the syneresis rates were lower by 1-5, 4, and
6 % in the samples containing half-normal, normal, and twice-normal fat respectively.
The rate of syneresis was markedly affected by the concentration of the colloidal
components in the milk (Fig. 2e). Concentration of the milk by ultrafiltration caused
a large decrease in the rate of syneresis of the curd and dilution of the milk with
permeate caused a massive increase in syneresis rate. The change in rate was
approximately inversely proportional to the milk concentration.
 Syneresis of rennet curd 333
0-5 r (a) 0-4

0-4

0-3

o
0-2
'E CD
UUj
CUt1 ting ti
hey vol. as%

0-3 (c) 0-45 r (d)

0-2
fter

CD
S C
C
'E 0-75 r(e)
o

1
to 0-7
•D
c
CO
c in

Sr c
'o 0-6
0)
twe
ate

0-5

0-4 r (fl

0-2

0-1
3 4 5 2 3 4
Cutting time of curd, relative to the RCT (= 1)
Fig. 2 Effect of various parameters on the average rate of syneresis. Experimental details are given in the text.
Controls are with raw whole milk 026 ml/1 rennet, pH 6-6, 30°C. Effect of; (a) milk pH; O, control; • , pH
6-3; A, pH 60. (6) Temperature; O, 25 °C; • . control; A, 35 °C. (c) Rennet concentration; O, control; • .
104 ml/1 rennet, (d) CaCl2; O> control; 0 , +2mM-CaCl2; A, + 4mM-CaCl2. (e) Milk concentration; O, control;
• , milk x 2-2; A, milk x 0-5. {J) Fat; O, control; • , skim-milk; A, fat 37% of control; A fat 213% of control.

DISCUSSION
The method described here was simple in terms of both equipment and technique.
It had a satisfactory precision for the quantitative measurement of syneresis, being
about the same as the methods of Emmons et al. (1959) and Zviedrans (1978). The
general shape of the course and the extent of syneresis measured by this method was
similar to other methods where the curd was not stirred (Thome et al. 1958; Lawrence,
1959 a). It avoided some of the drawbacks of the other methods such as the need for
special equipment (Thome et al. 1958; Schulz & Kley, 1956), problems of scale
(Lawrence, 1959a, b), adverse effects of fat on the accuracy of measurements (Beeby,
1959) and the effects of adding extra liquid to the whey (Beeby, 1959; Lawrence &
Hill, 1974). It enabled all the variables to be easily controlled. For example, the
disturbance of the curd during removal of whey was minimal. Where the curd is
disturbed (Lawrence, 19596; Emmons et al. 1959; Patel et al. 1972; Lelievre, 1977),
the measured rate of syneresis depends on the amount of work done on the curd. This
may be difficult to control in small-scale experiments.
334 R. J. MARSHALL
It appears that the relationship observed between curd-firmness and the rate of
syneresis depends, in part, on the experimental conditions and on the methods used
(Emmons et al. 1959; Lelievre, 1977). In the present work, curd-firmness at cutting
was not measured directly, but it increased as the cutting time became later. Curds
cut shortly after the RCT had a higher rate of syneresis than those cut at intermediate
times, probably because their structure was insufficient to control whey release. The
higher rates of syneresis obtained after long cutting times were probably due to a
high syneresis pressure generated by the aggregation of the casein within the gel.
Emmons et al. (1959) found that, over a wide range of curd strengths, the maximum
and minimum rates of syneresis were associated with maximum and minimum curd
strength respectively. Lelievre (1977) found that the rate of syneresis increased as
the rigidity modulus increased at low gel strengths but decreased at high gel
strengths, although the effects were small.
It has generally been observed (Lelievre, 1977;Cheeseman, 1962;Patele<a£. 1972),
that the syneresis rate increased as the pH was lowered although the absolute size
of the effect depended on the method and conditions used. The tendency in the
present experiments for a smaller increase in syneresis rate between pH 6-3 and 60
than between pH 6-6 and 6-3 was similar to the effect of pH on curd firming rate (Kelly,
1951; Marshall et al. 1981) where the rate increased rapidly with pH from 67 to 60,
slowed to a maximum at pH 5-8 and declined at pH 5-6. Presumably both these effects
relate to the increased tendency of casein to aggregate as the isoelectric point is
approached owing to less electrostatic repulsion. However, in the range pH 4'8-4'6,
the rate of syneresis decreased as the pH was lowered (Emmons et al. 1959).
The magnitude of the effect of temperature depends on the method used to
measure syneresis. The effect observed here was greater than that found by Lawrence
(1959a) but similar to that observed by Patel et al. (1972). It appears that increasing
the temperature accelerates the initial rate of syneresis, but it may increase the final
moisture in the curd at 24 h (Schulz & Kley, 1956).
The effect of increasing the amount of rennet observed by Lelievre (1977) and
Emmons et al. (1959) was similar to that found in the present work. Increasing the
rennet concentration reduces the RCT of milk in inverse proportion (e.g. Payens,
1977) and also accelerates the rate of curd-firming (Marshall et al. 1981). These latter
authors showed that the early stages of curd-firming, measured by a high-frequency
vibrating-reed viscometer, are affected more by rennet concentration than are the
later stages, measured by a very low frequency device. This suggests that the main
effect of increasing rennet is on the initial stages of coagulation and curd-firming, and
there is less effect at later times. In accord with this, the effect of rennet concentration
on syneresis is relatively small.
Adding CaCl2 to milk increases the Ca bound to micelles and accelerates
coagulation (Green & Marshall, 1977). However, the ionic strength of the whey is also
increased on adding CaCl2. These 2 phenomena may affect the syneresis rate
differently, thus accounting for the complex effects observed. At very high levels of
CaCl2 addition, as with NaCl and KC1, syneresis is inhibited (Cheeseman, 1962) by
ionic effects.
Fat in milk inhibited syneresis more than was expected for the consequent
reduction in water content of the curd, in agreement with previous results (Schulz
& Kley, 1956; Beeby, 1959). Fat in milk also reduces the curd firmness (author's
unpublished observation).
Concentration of milk by ultrafiltration reduced the rate of syneresis, but
increased the rate of curd firming (Green et al. 1981). Emmons et al. (1959) found that
 Syneresis of rennet curd 335
syneresis was also reduced by increasing the solids content of reconstituted milk. The
increase in curd-firming rate probably resulted from the higher casein content and
the decrease in syneresis rate was probably due to there being less whey in the curd.
Lowering the pH, raising the temperature and adding Ca2+ all increase the rate
of curd firming (Marshall et al. 1981) and these factors also increased the rate of
syneresis but to a lesser degree. This suggests that curd-firming and syneresis are
different aspects of the same phenomenon. Electron microscope studies (Kimber et
al. 1974) showed that the aggregation of casein micelles to form curd is followed by
increasingly closer contact between the micelles leading to fusion and segregation of
the protein from the whey. In accord with this, Van Dijk et al. (1979) found that the
syneresis pressure of uncut curd is very small. Thus, when the curd is cut, the whey
leaks out. The process followed first-order kinetics initially because the pressure
depended on the amount of whey in the curd. Holding the curd in whey retards
syneresis by the back-pressure of the surrounding whey. Removing the whey
accelerates syneresis by allowing the forward pressure to dominate. When the curd
had been reduced to 70 % of the initial volume, syneresis became dependent on factors
other than the amount of whey present. It seems likely that the progressive
aggregation of the casein causes shrinkage of the curd so forcing out the whey. Thus,
syneresis probably results from mainly hydrophobic interactions within the casein
network. This accords with the observation that syneresis rate is increased by
lowering the pH of the milk or adding CaCl2, which enhance the hydrophobicity of
casein micelles by neutralizing part of their negative charge (Green & Marshall, 1977)
or by raising the temperature, which increases the energy available for hydrophobic
interactions (Tanford, 1973).

REFERENCES
BEKBY, R. 1959 A method for following the syneresis of the rennet coagulum in milk. Australian Journal of
Dairy Technology 14 77-79
BEKRIDOE, N. J. 1952 An improved method of observing the clotting of milk containing rennin. Journal of
Dairy Research 19 328-329
GHEESEMAN, G. 0. 1962 Syneresis of milk and caseinate solutions. 16th International Dairy Congress,
Copenhagen, B 465-472
[Link]. D. B., PRICE, VV. V. & SWANSON, A. M. 1959 Tests to measure syneresis and firmness of cottage cheese
coagulum and their application in the curd making process. Journal of Dairy Science 42 866-869
GREEN, M. L., GLOVER, F. A., SCURLOCK, E. M. W., MARSHALL, R. J. & HATFIELD, D. S. 1981 Effect of use
of milk concentrated by ultrafiltration on the manufacture and ripening of Cheddar cheese. Journal of Dairy
Research 48 333-341
GREEN, M. L. & MARSHALL, R. J. 1977 The acceleration by cationic materials of the coagulation of casein
micelles by rennet. Journal of Dairy Research 44 521-531
KELLY, L. A. 1951 The factors affecting rennet coagulation time and rennet curd tension. Thesis. University
of Wisconsin, USA
KIMBER, A. M., BROOKER, B. E., HOBBS, D. G. & PRENTICE, J. H. 1974 Electron microscope studies of the
development of structure in Cheddar cheese. Journal of Dairy Research 41 389-396
LAWRENCE, A. J. 1959a Syneresis of rennet curd. I. Effect of time and temperature. Australian Journal of
Dairy Technology 14 166-169
L A W R E N C E , A. J . 19596 Syneresis of r e n n e t curd. I I . Effect of stirring a n d of t h e volume of whey. Australian
Journal of Dairy Technology 14 169-172
LAWRENCE, A. J. & HILL, R. D. 1974 A method for measuring the syneresis of cheese-curd. 19th International
Dairy Congress, Sydney IE 204-205
LAWRENCE, R. C. & GILLES, J. 1980 The assessment of the potential quality of young Cheddar cheese. New
Zealand Journal of Dairy Science and Technology 15 1-12
LELIEVRE, J. 1977 Rigidity modulus as a factor influencing the syneresis of renneted milk gels. Journal of
Dairy Research 44 611-614
MARSHALL, R. J., HATFIELD, D. S. & GREEN, M. L. 1982 Assessment of two instruments for the continuous
measurement of the curd-firming of renneted milk. Journal of Dairy Research 49 127-135
PATEL, M. C , LUND, D. B. & OLSON, N. F. 1972 Factors affecting syneresis of renneted milk gels. Journal
of Dairy Science 55 913-918
336 R. J. MARSHALL

PAVENS, T. A. J. 1977 On enzymatic clotting processes. 11. The colloidal instability of chymosin-treated casein
micelles. Biophysical Chemistry 6 263-270
SOHULZ, M. E. & KLEY, W. 1956 [Shrinkage of the coagulum of milk.] Mikhwissenschaft 11 116-123
TANFORD, C. 1973 The hydrophobic effect: formation of micelles and biological membranes. New York: John
Wiley & Sons
THOME, K. E., AXELSSON, I. & LILJEGREN, G. 1958 [Method for the estimation of whey separation in rennet
coagulum.] Milk and Dairy Research Report No. 53, Alnarp, Sweden
VAN DIJK, H. J. M., WALSTRA, P. & GEURTS, T. J. 1979 Preliminary note on syneresis pressure in rennet curd.
Netherlands Milk and Dairy Journal 33 60-61
ZVIEDRANS, P. 1978 Measurement of syneresis. Dairy Research Report No. 5 p. 12. Northfield Research Centre:
Department of Agriculture and Fisheries, South Australia