Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol.

., Vol. 24, 2002

THE EFFECT OF LIMING CONDITIONS IN JUICE CLARIFICATION

By

W.O.S. DOHERTY1, J. GREENWOOD2, D. PILASKI2 and P.G. WRIGHT3

1
Sugar Research Institute, Mackay, Qld
2
Department of Chemical Engineering, University of Sydney, Sydney, NSW
3
SRI Associates, Mackay, Qld

[email protected]

KEYWORDS: Defecation, Lime, Saccharate, Flocculation, Clarification, Juice Quality.

Abstract
The juice clarification process used in the Australian cane sugar industry is best
described as simple defecation. However, there are many variants of this simple process
used throughout the cane sugar industry. Milk of lime is often used instead of lime
saccharate solutions, lime addition may be batch-wise rather than continuous, and the
point of lime addition may be to cold juice (35–40oC), to intermediate juice (72–76oC)
or to hot juice (100oC). Despite a wealth of practical experience, the impact that the
various liming conditions have on the physical chemistry of clarification is not well
known. This study reports some recent laboratory work to quantify the effects of liming
techniques on clarified juice quality and floc settling behaviour. The results indicate that
liming with lime saccharate gives lower turbidity but higher mud levels and slower
settling flocs than liming with milk of lime. However, the clarified juice obtained with
saccharate has a higher calcium level but a lower phosphate level than milk of lime
clarified juice. The higher calcium level in the clarified juice obtained with saccharate
may result in an increase in scale formation in the evaporators. The underlying physical
chemistry of the clarification process is used to explain the observed differences in the
performance of the various liming techniques.

Introduction

The aims of the clarification process are to remove suspended and colloidal particles,
remove non-sugar soluble impurities (e.g. phosphate and proteins) and raise the juice pH to
minimise inversion of sucrose during subsequent processing. Clarification affects the juice
filtrability, evaporator scale composition, sucrose crystallisation and the quality and quantity of raw
sugar produced. Clarification also affects the colour, crystal morphology, crystal content, ash and
polysaccharide contents of raw sugar. In the sugar refinery, poor filtration and reduced refined sugar
yields can be traced back to poor clarification of mixed juice in raw sugar mills.

The clarification process used in many cane sugar industries is simple defecation. In this
process, lime is added to the juice to raise the pH from ~5.2 to 7.5–7.8 and to react with inorganic
phosphate present in the cane juice to form calcium phosphate floc. The floc settles, sweeping
insoluble impurities from the cane juice. The lime is added either in the form of milk of lime
(M.O.L.), lime-juice mixtures, or lime-syrup saccharates. The addition of lime may be batch-wise
rather than continuous, and the point of lime addition may be to cold juice (35–40oC), to
intermediate juice (72–76oC) or to hot juice (100oC). An acrylamide-sodium acrylate copolymer is
added at the feed launder in the clarifier to enhance settling of the calcium phosphate flocs. Davis
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

(1962) suggested that clarified juice (CJ) of good quality is characterised by low turbidity (less than
15 NTU), low soluble phosphate level (less than 8–10 ppm P) and low insoluble phosphate (less
than 5 ppm P). Flocculant addition, clarifier design, and liming conditions all contribute to control
the CJ turbidity.

During the defecation process, a wide range of chemical and physical reactions takes place
in the juice. The main chemical reactions include:

• Precipitation of amorphous calcium phosphate.

• Denaturation of proteins (and other organics, such as gums, pectins and waxes).
• Inversion of sucrose due to the combined action of pH and temperature.
• Degradation of reducing sugars to organic acids due to high pH and temperature.
• Precipitation of organic and inorganic acid salts.
• Hydrolysis of starch by the natural amylase in the juice.
• Formation of colour bodies due to the polymerisation (either enzymatically or thermally) of
flavonoids and phenolic compounds.

The major physical reaction is the enhanced flocculation of micro-particles (mainly calcium
phosphate particles) by the addition of the acrylamide-sodium acrylate coplymer. The molecular
weight of the copolymer, the charge of the copolymer, the juice density, the charge and size of
particles in juice, and the juice pH and temperature are factors that affect flocculation.

The composition of the lime mixtures and the liming conditions used in the cane sugar
industries varies throughout the world. The work of Davis (1962) established that saccharate liming
gave CJ of the lowest turbidity and led to the wide adoption of saccharate liming in Australian sugar
mills. The Australian process is based on the addition of lime as lime saccharate to intermediate
juice (72–76oC) to a pH of 7.8. This is followed by secondary heating of the limed juice to about
102 to 105oC and flashing in a tank to remove air bubbles before being distributed to the clarifier. A
flocculant is added near the flash tank or in the feed launder of the clarifier. Australian sugar mills
usually have enough clarifier capacity to cope with the effects of reduced settling rate and increased
mud levels obtained with saccharate liming. Other parts of the world either use M.O.L. on hot juice
or M.O.L. on cold juice. North-Coombes et al. (1981) and Scott (1988) compared the effect of
liming hot juice (100oC) with M.O.L. to that with saccharate on South African juices. North-
Coombes and co-workers (1981) concluded that saccharate liming gave a clearer juice and
improved purity drop across the clarification station than hot liming with M.O.L. However, Scott
(1988) remarked that the improvements obtained with turbidity reductions were not carried through
to the syrup and raw sugar. All these researchers observed slower settling of mud particles and
higher mud volumes with saccharate liming.

Recent factory work in Louisiana (Eggleston, 2000a, 2000b) compared the effects of the
current USA practice of cold liming with hot liming using M.O.L. It was concluded that while the
hot liming technique gave lower calcium and dextran levels in CJ, it gave higher mud volumes.
Eggleston’s findings revealed no significant difference in clarified juice turbidity between hot and
cold liming.

Despite these publications on various liming methods, there appears to be no universally

accepted liming method. Some Australian sugar mill technologists raised the issue recently as to
whether it was necessary to use saccharate for liming instead of M.O.L. in view of the improved
highly reactive lime now commercially available. In fact, one Australian sugar mill for most of the
2001 crushing season used M.O.L. and observed no difference in juice quality (in terms of
turbidity) from the changeover. The debate over which liming method to use continues because the
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

literature has failed to examine the underlining physical chemistry of the clarification process. An
explanation of the observed differences in clarification performance is needed.

To gain further insight into the chemistry of clarification, this paper reports on laboratory
clarification experiments conducted on-site in a sugar mill. Variants of the simple defecation
process that are practised around the world were used. The floc size and structure were examined
using static light scattering techniques, and confocal microscopy was used to provide visual
representation of the calcium phosphate floc.

Laboratory studies

For each series of experiments, samples of mixed juice and primary juice were taken from
the factory. Clarification experiments were carried out using M.O.L. and saccharate solutions made
up with the same CaO alkalinity in order to eliminate any differences in hydroxyl ion concentration.
A dose equivalent to 5 ppm on juice of Separan A2130 acrylamide-sodium acrylate flocculant was
applied before settling. The methods of clarification used are listed in Table 1.

The volume of lime or saccharate solution used for each trial was recorded. The rates of
settling and final mud levels were observed in standard juice settling tubes in a SRI-designed heated
and illuminated clarification box. As the mud was unstirred, the mud levels were not indicative of
the values obtainable in commercial clarifiers.

Table 1—Details of the clarification methods investigated.

Method Details
Cold liming Mixed juice (36oC), lime saccharate (or M.O.L.) to 7.8 pH, heat to 100oC,
add flocculant and clarify.
Warm liming Mixed juice, lime saccharate (or M.O.L.) to 6.0 pH, heat to 76oC, lime
saccharate (or M.O.L.) to 7.8 pH, heat to 100oC, add flocculant and clarify.
Intermediate liming Primary juice (at76oC), lime saccharate (or M.O.L.) to 7.8 pH, heat to
100oC, add flocculant and clarify.
Hot liming Primary juice (at 76oC), heat to 100oC, lime saccharate (or M.O.L.) to 7.8
pH, add flocculant and clarify.

Samples of CJ were siphoned off for storage and subsequent analysis. Some of the
flocculated mud was also sampled and stored frozen.

Analytical procedures

The settling rate of the flocculated mud particles was determined by measuring the level of
the mud interface at 0.5, 1.0, 2.0 and 3.0 minutes. The initial settling rate was obtained from a
graphical analysis of the data.

Turbidity of the CJ absorbance (A) was measured at 900 nm in 1 cm glass cells against
distilled water. The turbidity of the clarified juice was defined as 100A.

Calcium, magnesium and silica were determined by atomic absorption spectrophotometry,

while phosphate was determined by the phosphate-molybdate colorimetric method.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

The concentrations of the soluble proteins in the juices were determined

spectrophotometrically. The procedure involved the addition of an acidic dye to the sample and the
absorbance measured at 595 nm.

The concentrations of the total soluble polysaccharides were determined colorimetrically by

phenol and sulphuric acid.

To examine the floc size and structure, the mud and CJ samples were analysed using the
static light scattering technique. The frozen CJ and mud particles obtained from the batch settling
tests were defrosted prior to use.

The fractal dimension of a floc is a measure of the looseness or compactness of the floc
structure. The theory underlying the interpretation of the light scattering data is the Rayleigh-Gans-
Debeye theory (Jung, 1997). The fractal dimension of floc can be determined from the following
relation:

I ∝ Q dF (1)

where I is the scattering intensity, Q is the momentum of the floc particles and dF is the
fractal dimension. Q is related to the refractive index of the particles, the wavelength of the light
and the angle of the detector to the incident beam. The dF value of a floc lies between 1.0 and 3.0. A
floc with a dF value of 2.5 is said to be more compact than a floc with a dF value of 2.0.

Results

The details of the conditions and results are given in Tables 2a, 2b and 2c, and the statistical
comparisons on the paired results are summarised Table 3.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Table 2a—Conditions for batch settling tests – paired samples.

Test no. Liming temp., oC Lime type Alkali vol. to 6 pH Alkali vol. to 7.8 pH
(mL) (mL)
Run 3, Morning 19 Oct., 2001
3.1 36 Cold saccharate 10.2
3.2 36 Cold M.O.L. 10.8
3.3 76 Warm saccharate 2.2 10
3.4 76 Warm M.O.L. 3 10.2
3.5 76 Inter. saccharate 12.6
3.6 76 Inter. M.O.L. 13.6
3.7* 100 Hot saccharate 7
3.8* 100 Hot M.O.L. 5.6
Run 4, Afternoon 19 Oct., 2001
4.1 36 Cold saccharate. 9.6
4.2 36 Cold M.O.L. 10.4
4.3 76 Warm saccharate 3 10
4.4 76 Warm M.O.L. 3 10
4.5 76 Inter. saccharate 13
4.6 76 Inter. M.O.L. 13
4.7* 100 Hot saccharate 2.6
4.8* 100 Hot M.O.L. 2.4
Run 5, Morning 20 Oct., 2001
5.1 36 Cold M.O.L. 11.2
5.2 36 Cold saccharate 10.6
5.3 76 Warm M.O.L. 3 10.8
5.4 76 Warm saccharate 3 10.8
5.5 76 Inter. M.O.L. 13.6
5.6 76 Inter. saccharate 13
5.7 100 Hot M.O.L. 15
5.8 100 Hot saccharate 16.6
Run 6, Morning 20 Oct., 2001
6.1 100 Hot M.O.L. 17
6.2 100 Hot Saccharate 16.6

* Juice obtained after the secondary juice heaters.

Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Table 2b—Results for batch settling tests.

Test Initial CJ Final CJ CJ turbidity CJ calcium, CJ magnes- CJ CJ phosphate,
no. settling turbid- mud brix per 100 Bx ppm on brix ium, ppm on silica, ppm on brix
rate, ity level brix ppm on
cm/min % brix
3.1 35.2 8.6 32.3 19.17 44.9 1038 1440 178 98.1
3.2 42.4 13.3 28.0 18.73 71.0 1025 1452 185 105.2
3.3 39.2 7.7 31.7 19.27 40.0 1391 1391 261 43.1
3.4 41.6 14.1 – 18.66 75.6 1313 1399 244 56.3
3.5 20 6.3 27.8 19.11 33.0 1376 1402 268 31.9
3.6 20 20.8 23.2 18.55 112.1 1364 1396 265 69.0
3.7 23.2 5.5 28.6 19.37 28.4 1766 1435 217 7.2
3.8 20.8 5.1 28.3 18.82 27.1 1812 1403 217 4.3

4.1 30.4 13.6 28.0 18.61 73.1 1795 1644 198 78.5
4.2 32.8 20.6 26.7 18.27 112.8 1664 1642 192 88.1
4.3 26.4 10.5 27.7 18.82 55.8 2147 1578 239 29.8
4.4 42.4 12.6 24.0 18.22 69.2 2009 1570 239 36.2
4.5 21.6 7.3 27.8 18.42 39.6 1933 1520 242 10.9
4.6 35.2 11.4 26.8 18.87 60.4 1770 1510 234 28.6
4.7 19.2 4.4 31.3 20.02 22.0 1688 1139 233 2.5
4.8 16.8 4 30.5 19.99 20.0 1611 1186 233 4.0

5.1 40.8 14.2 30.3 18.39 77.2 1218 1267 218 84.3
5.2 37.6 9 31.4 18.42 48.9 1189 1276 205 83.6
5.3 44.8 11.2 31.0 17.6 63.6 1489 1227 244 47.2
5.4 37.6 7.6 32.7 18.71 40.6 1555 1240 236 35.8
5.5 36.8 11.2 31.3 18.72 59.8 1239 1058 257 50.7
5.6 28 4.9 28.0 18.76 26.1 1290 1071 283 17.1
5.7 29.6 14.3 – 20.05 71.3 1426 1012 233 24.9
5.8 27.2 5.5 31.9 20.68 26.6 1456 1083 229 9.2

6.1 15.2 15.6 24.5 18.23 85.6 1289 779 256 67.5
6.2 20.8 3.5 33.7 18.12 19.3 1446 806 268 34.2
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Table 2c—Concentration of proteins and

polysaccharides in clarified juices.
Test No. Protein, Polysaccharide,
ppm on brix ppm on brix
4.1 159 6350
4.2 153 6251
4.3 171 6268
4.4 188 6611
4.5 142 6443
4.6 188 6371
4.7 117 6305
4.8 123 6239

5.1 129 6463

5.2 108 6148
5.3 164 7001
5.4 123 6581
5.5 104 6128
5.6 93 5580
5.7 116 6158
5.8 99 6239

Table 3—Summary of the mean differences in clarified juice measurements from saccharate and
milk-of-lime clarification.
Measurement CJ- Initial Alka- CJ- CJ- CJ- CJ- CJ- CJ-
turbidity settling linity calcium magnes- silica phos. protein poly-
/100 Bx velocity titre ium sacchar-
ide
Units Units per cm/min mL/ ppm on ppm on ppm ppm ppm on ppm on
100 Bx 100 brix brix on brix on brix brix brix
mL
Mean for 40.7 29.5 12.8 1510.4 1313.8 236.9 42.9 127.9 6229.7
saccharate
Mean for 78.1 34.7 13.2 1436.9 1301.1 233.4 59.8 148.9 6426.0
M.O.L.
Mean difference –37.4 –5.2 –0.4 73.5 12.7 3.5 –16.9 –21.0 –196.3
t Statistic –6.3 –2.8 –1.4 3.7 1.9 1.0 –4.5 –3.1 –1.9
Probability, 0.01 2 20 0.5 10 35 1 2 10
%(T<=t)
- two-tail t-test
(5%)
Significance of Very high High Not High Not Not High High Not
mean difference signif. signif. signif. signif. signif. signif. signif. signif. signif.

The effect of the type of liming solution

Clarified juice turbidity

A plot of all results (except secondary juices) for CJ turbidity (per 100 brix) is shown in
Figure 1. The results of the secondary juices were omitted from the plot since mills do not normally
clarify secondary juice. Saccharate liming tends to give a lower CJ turbidity than M.O.L. additions.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

As seen in Table 3, the mean value for saccharate is 40.7 vs. 78.1 for M.O.L. and from the (two
tailed) t-test on paired samples the difference is highly significant.

120.0 120.0
112.1 112.8

100.0 100.0

85.6
CJ Turbidity per 100 Bx

80.0 77.2
80.0
75.6
73.1
71.0 71.3
69.2
63.6 Saccharate
60.0 60.4 59.8 60.0
55.8 M.O.L.
48.9
44.9
40.0 40.0 39.6 40.6 40.0
33.0

26.1 26.6

20.0 19.3 20.0

0.0 0.0
3.1 3.3 3.5 4.1 4.3 4.5 5.2 5.4 5.6 5.8 6.2
Run Number

Fig. 1—The effect of the liming type on the turbidity of clarified juice.

Initial settling velocity

A plot of all results (except secondary juices) for the initial settling velocity (ISV) of the
mud is shown in Figure 2. Saccharate liming tends to give a lower initial floc settling rate than
M.O.L. additions. As shown in Table 3 the mean value for saccharate is 29.5 vs. 34.7 for M.O.L.
and from the (two tailed) t-test on paired samples the difference is significant.

50 50

44.8 45
42.4 42.4
41.6
40.8
40 39.2 40
Floc initial settling rate, cm/min

37.6 37.6
36.8
35.2 35.2 35
32.8

30 30.4 30
29.6
28
27.2
26.4 Saccharate
25
M.O.L.
21.6
20.8
20 20 20 20

15.2 15

10 10

0 0
3.1 3.3 3.5 4.1 4.3 4.5 5.2 5.4 5.6 5.8 6.2
Run Number

Fig. 2—The effect of the liming type on the initial settling velocity of the flocs.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Concentrations of impurities in clarified juices

Plots were also made to explore the effect of lime solution type on the other parameters
measured in this work. However, the main presentation here is limited to the statistical summary in
Table 3, which shows highly significant differences for calcium (+5% increase with saccharate),
soluble phosphate (–18%) and protein (+14%), and no significant differences for magnesium
(+1%), polysaccharide (–3%) and soluble silica (+1.5%). Although the data set for final mud level
was sparser, there was observed to be a significantly higher final (unstirred) mud level (+10%
increase) for saccharate liming.

The effect of liming technique

Tables 2b and 2c gives the results of the effect of liming techniques. The lowest turbidity
values are achieved when intermediate or hot juice are limed especially with saccharate, rather than
cold juice. However, the liming of intermediate or hot juice achieved lower initial settling rates than
liming cold juice especially when the liming was done with M.O.L.

Fractal dimension studies

The size and the fractal dimensions of the floc remaining in the clarified juice and the mud
floc were measured using light scattering. The 90th percentile particle size for the CJ was generally
below the 10th percentile particle size for the mud indicating that reasonable size separation has
been achieved. A limited number of measurements on the fractal dimensions of CJ and mud flocs
are shown in Table 4. There is a clear difference in the fractal dimension of the floc in the mud over
that of the floc in the CJ. This seems to indicate that the more compact floc settle faster. There does
not appear to be any trend in floc size or fractal dimension of the CJ or of the mud with liming
method. This is unexpected since visual inspection of the mud indicated larger and looser floc with
saccharate liming. In addition, previous measurements of the floc formed in a synthetic juice
indicated that the fractal dimension of the floc formed by liming with saccharate was in fact lower
than that obtained with M.O.L (Greenwood et al., 2001). However, the flocs from this present work
had been frozen and transported prior to analysis. This seems to suggest that the floc structure of the
highly hydrated calcium phosphate was altered by freezing and storage and that any difference in
floc structure that might have existed was lost.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Table 4—Results for fractal geometry measurements.

Test No. Lime type Fractal dim. (dF) Fractal dim. (dF)
clarified juice floc mud floc

3.1 Cold saccharate 1.44 2.2

3.2 Cold M.O.L. 1.66 2.03

3.3 Warm saccharate 1.33 2.11

3.4 Warm M.O.L. 1.44 2.04

3.5 Inter. saccharate 1.37 1.98

3.6 Inter. M.O.L. 1.3 2.04

3.7 Hot saccharate 1.88 1.98

3.8 Hot M.O.L. 1.98 2.05

6.1 Hot M.O.L. 1.08 2.16

6.2 Hot saccharate 1.73 2.17

Discussions and conclusions

The results from this work has confirmed the practical experiences reported by Australian
sugar mills that liming with lime saccharate gives lower turbidity but higher mud levels and slower
settling flocs than liming with M.O.L.

The differences observed in the juice and mud between those limed with lime saccharate and
those limed with M.O.L. can be explained by considering the solution chemistry of juice and the
nature of the calcium phosphate precipitate formed. The particles in juice carry a negative charge.
For coagulation of the particles to take place the overall charge on the particles has to be neutralised
to near zero charge. Calcium phosphate formed during clarification carries a positive charge which
partially neutralises the negative charge of the particles and causes aggregation or coagulation.
When liming with saccharate, because of the increased solubility of lime in sucrose, a higher
proportion of ‘free’ calcium ions are available for the precipitation of calcium phosphate than with
M.O.L. The highly concentrated calcium ion solution upon mixing with juice causes spontaneous
precipitation of amorphous calcium phosphate of very fine particles. The increased calcium
phosphate supersaturation obtained through saccharate liming accounts for the finer texture of the
calcium phosphate particles formed. The particles formed from juices limed with lime saccharate
would therefore have higher charge to mass ratio than the particles obtained with M.O.L. and hence
are more effective in neutralising the negatively charged juice particles. The overall effect of this is
the increased rate of collision between the juice particles, increasing the rate in which aggregates
form and reducing juice turbidity.

An additional explanation for the lower CJ turbidity with saccharate is connected with the
calcium level in the CJ. The CJ obtained with saccharate has a higher calcium level but a lower
phosphate level than M.O.L CJ. Calcium ions tend to foster flocculation by reducing the
electrostatic repulsive forces between particles (Kampen, 1997). This implies that juices treated
with saccharate, would enhance attraction of the particles and reduce juice turbidity due to the
higher calcium concentration.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

The higher calcium levels of the CJs obtained with saccharate may result in an increase in
scale formation in the evaporators compared to CJs obtained with M.O.L. The scale formed with
juices treated with M.O.L. will contain a higher proportion of phosphate scale that is easier to
remove. However, a disadvantage of having a high phosphate level in juice is an increase in the
phosphate content of the raw sugar which will affect the raw sugar filtrability. The higher phosphate
levels of CJs treated with M.O.L probably arise as a result of side reactions between undissolved
lime and other juice ions (e.g. carbonate ions) during the initial stages of clarification.

The fractal dimension study indicated no difference in floc size and fractal dimension with
liming method. This is probably due to the collapse of the highly hydrated amorphous calcium
phosphate floc during freezing and storage prior to analysis. Studies on calcium phosphate floc
freshly formed from synthetic juice indicated that liming with sacharate gave larger and looser floc
than liming with M.O.L, and therefore saccharate would be expected to be better for removing
impurities through physical entrapment and adsorption. The reduced settling rate obtained with
saccharate liming aid the ‘sweeping’ of suspended solids.

Best juice quality (involving turbidity, phosphate and protein levels) is achieved with
intermediate or hot liming irrespective of the lime type. However, there is an operational
compromise necessary because of the associated higher mud levels and slower floc settling rates.
The difference between the performances of the various liming techniques is related to the final
charge of the particles in the limed juice. The lower the final particle charge, the more effective is
the clarification process. Intermediate or hot liming has a lower final charge than cold liming
because some of the outermost adsorbed materials (e.g. proteins) on the juice particles that
determine the value of the charge are denaturated or dehydrated by the action of heat.

The differences observed between the various liming techniques may also be related to
differences in calcium phosphate morphology brought about by differences in the reaction pathways
during clarification. When cold juice is limed to raise the pH of juice prior to heating to 100oC, it
contains a larger proportion of unreacted calcium and phosphate ions than the intermediate juice or
hot juice (Doherty, 2000, private communication). This is because calcium phosphate is more
soluble at low temperatures. When the cold juice is then heated to boiling, there are competing
reactions amongst the unreacted calcium and phosphate, the formed calcium phosphate and juice
constituents. With intermediate or hot juice the reaction for the formation of amorphous calcium
phosphate is spontaneous and complete at pH 7.8.

Some sugar mills are set up to partially lime their juice prior to the juice entering the
primary heaters in order to reduce the corrosion of the horizontal tubular heater units. In this work
this liming method has been described as warm liming. The results in Tables 2b and 2c indicate that
this liming technique generally results in higher phosphate and protein levels in the CJs than the CJs
obtained from intermediate or hot liming. So, this practice of batch-wise addition has an adverse
effect on juice quality.

The results of this work indicate that the best juice of consistent quality is achieved with
intermediate or hot juice liming using saccharate, at the expense of reduced settling rate and higher
mud volume. If the juice to be processed is of good quality (low overall negative charge), then
liming intermediate juice with M.O.L. is sufficient. This should reduce evaporator scaling.

One of the major differences between saccharate liming and M.O.L. is in CJ turbidity. Using
cationic polymers in addition to the flocculant may eliminate this difference. It is proposed that this
study be extended to vary the flocculant level and type, and flocculation performance evaluated in
the presence of cationic polymers.
Doherty, W.O.S., Greenwood, J., Pilaski,D. and Wright, P.G. Proc. Aust. Soc. Sugar Cane Technol., Vol. 24, 2002

Acknowledgments

The financial support of the member mills of Sugar Research Limited and the Australian
Research Council (ARC) is greatly appreciated. Thanks are also due to staff of the CSR Plane
Creek factory, CASCO Australia, and SRI chemists involved in aspects of this work.

REFERENCES
Davis, C.W. (1962). Clarification and raw sugar filterability – Recent improvements in the lime
defecation process. Proc. Qd Soc. Sugar Cane Technol., 29th Conf., 73–82.
Eggleston, G. (2000a). Hot and cold lime clarification in raw sugar manufacture I. Juice quality
differences. Int. Sugar J., Vol. 102, No. 1221, 406–416.
Eggleston, G. (2000b). Hot and cold lime clarification in raw sugar manufacture II. Lime addition
and settling behaviour. Int. Sugar J., Vol. 102, No. 1221, 453 – 457.
Greenwood, J.A., Tang, P., Cheung, W.H., Doherty, W.O.S. and Raper, J.A. (2001)
Flocculation of calcium phosphate in sucrose solution. Proc. 6th World Congress of Chem.
Engineers, Melbourne (CD Rom).
Jung, S.J. (1997). A new technique for characterisation of floc structure. PhD Thesis, UNSW,
Australia.
Kampen, W.H. (1997). Improved clarification with a cationic flocculant. Int. Sugar J. Vol. 99, No.
1183, 353–360.
North-Coombes, S., Taylor, K and Koster, K.C. (1981). The practical development and
application of saccharate liming at the Pongola sugar mill. Proc. Sth. Afr. Sugar Technol.
Assoc., 55: 71–74.
Scott, R.P. (1988). Modifications to and experiences with Rapidorr clarifiers including saccharate
liming at Amatikulu. Proc. Sth. Afr. Sugar Technol. Assoc., 62: 32–35.