ArticlePDF Available

A review on acid beverage floc

Authors:
  • Leaf Resources Pty Ltd

Abstract

This paper is a review on the nature of floe that forms in acidic beverages prepared from cane and beet sugars. The floe forming compounds, the mechanism behind floe formation, and ways for testing and removing floc are described. The review is restricted to floe formation in beverages that is not a consequence of biological contamination. Formation of haze in beers is briefly described as the use of haze removing compounds in beer may be applied to the understanding and removal of floe in acidic beverages.
A Critical Review of Published Literature on
Acid Beverage Floc
Les A. Edye, April 2001
Executive Summary
This report comprises a critical review of literature on acid beverage floc (ABF) in soft
drink manufacture. The references were obtained by a literature search conducted at the
Sugar Research Institute, and copies of relevant published articles retrieved through the
search are compiled in a separate document.
The scope of the literature search was restricted to floc formation in soft drinks that is not
a consequence of microbial contamination. It is this floc, known as ABF that is of
importance to sugar manufacturers as impurities in sugar products are implicated in ABF
formation.
The soft drink manufacturer’s perspective is considered. The appearance of ABF in soft
drinks is a quality problem that is not easily managed. When ABF is manifest in soft
drinks the product is usually at the point of retail. It is recognised by soft drink
manufacturers that the most common cause for floc formation in clear soft drinks is
microbial growth. In addition to ABF resulting from impurities in the sugar ingredient,
floc may also form as a result of algal polysaccharides in the water supply. Certainly,
these possibilities should be eliminated before the sugar supply is implicated in floc
formation.
ABF may form in soft drinks when beet sugar is used as an ingredient. ABF resulting
from beet sugars consists of saponins and small amounts of high molecular weight
materials, most likely pectin. A mechanism based on interaction of negatively charged
saponins or polysaccharides and positively charged proteins as an initial step in ABF
formation has been proposed. ABF problems resulting from the use of beet sugar in soft
drink manufacture have been essentially overcome through processing modifications that
minimise the amount of saponin impurity in the sugar product.
There is consensus among researchers that ABF in cane sugar sweetened soft drinks is a
complex coacervate of silica, polysaccharides, protein, waxes and other organic
compounds. Certainly, most of the polysaccharides and waxes and some of the proteins
and silica are present as impurities in refined cane sugar (albeit at concentrations less than
0.1 %). Much of the research on ABF is concerned with the elucidation of
polysaccharide structures in the floc coacervate, and therefore, polysaccharide impurities
in refined cane sugar also are considered here. A mechanism based on interaction of
negatively charged indigenous sugarcane polysaccharide (ISP) and positively charged
proteins as an initial step in ABF formation has been proposed. While the mechanism is
prima facie believable, the evidence supporting the mechanism (viz., that a mixture of
ISP and protein in a simulated acid beverage forms floc) is by no means proof that ABF
forms in this way in authentic soft drinks.
Tests for ABF propensity of sugars are summarised. At this point in time there seems to
be no published, reliable and rapid method to determine the floc propensity of sugars.
The Coca Cola test based on observation of simulated acid beverages at high sucrose
concentrations after 10 days is the only internationally recognised test for ABF.
Recent literature on a related beverage quality issue (viz., chill haze in alcoholic
beverages) is considered. Chill haze is caused by the interaction of proteins and
polyphenols. The beer manufacturing industry uses silica to stabilise the product, and the
interaction of silicates with proteins is certainly relevant to ABF formation in softdrinks.
The adsorption of proteins onto silica hydrogel has been found to occur between pH 3
and pH 5. The adsorption capacity of the silica depends on size, surface area, porosity
and density of silanol groups. In this pH range polyphenols and polysaccharides also
were observed to bind to the silica hydrogel. Silica gel and polyphenols appear to
compete to bind haze active proteins. The significance of this work to ABF in soft drinks
is that it establishes the interaction of colloidal silicates with solutes by hydrogen
bonding, and both ionic and hydrophobic interactions. The importance of stoichiometry
of solutes in complex systems (involving more than one solute type) is also illustrated. In
the reviewer’s opinion the role of silica in ABF formation is poorly understood, but may
be pivotal in the initial mechanism of ABF flocculation.
This critical review of published literature on ABF identifies a best bet technology for the
removal of floc precursors from sugar products. The technology that has been identified
is micro or ultrafiltration (and possibly at temperatures in the region of 60 oC), although
other forms of filtration (e.g. plate and frame with Celite filter aids) at reduced
temperatures may also be effective.
1. Introduction
A floc or floccule is a small portion of matter resembling a tuft of wool or a wispy
cloud. In soft drink or acidic beverage manufacture the term is used to describe a
visible defect in the product. This visible defect may be particulate and sedimentary or
tuft-like and suspended in the beverage, and may be attributed to microbial
contamination or to water and sugar ingredients that are of unsuitable quality for
beverage manufacture. Microbial contamination of soft drinks is considered to be
outside the scope of this review. For simple tests of microbial contamination and a
review of the microbiology of soft drinks the reader is referred to ‘Chemistry and
Technology of Soft Drinks and Fruit Juices’ (Ed. Ashurst, 1998).
This review focuses on floc in soft drinks that is not a result of microbial contamination.
This type of floc is referred to by the sugar and beverage manufacturing industries as acid
beverage floc (ABF). While ABF is harmless, it is nevertheless a visible defect and
consumers reject the soft drink product for aesthetic reasons. Since in many parts of the
world soft drink manufacturers are large consumers of sugar (sucrose) and sugar is
implicated in ABF formation, the issue of ingredient quality and ABF is a major concern
to both soft drink and sugar manufacturers.
In the late 70's, 25 % of US sugar production was consumed by soft drink manufacturers,
and consequently, ABF was a major issue to the US industry. During this time
researchers at the Sugar Processing Research Institute Inc. (SPRI, then the Cane Sugar
Refining Research Project Inc.) conducted extensive research on the nature and cause of
ABF in soft drinks manufactured from cane sugar. Clarke, Godshall, Roberts, and
Carpenter are the authors of work produced by the SPRI during this period. Their
published work is a major part of literature on this subject and is given coverage here. In
the early 80's, high fructose corn syrup (HFCS) consumption by US soft drink
manufacturers increased from 0.8 to over 3 million tons, with a consequent decrease in
sucrose consumption to less than 300,000 tons (ca. 5 % of production). In November
1984, major US soft drink companies announced that 100 % HFCS would be used in
their products (Moore, et al., 1991). ABF was no longer an issue to the US sugar industry
and research on the subject, previously considered essential, effectively ceased.
Since both beet and cane sugar are reported to form ABF, the literature search was not
restricted to cane sugar. The similarities and differences between ABF from cane and
beet sugar are discussed. Where sugar is implicated in ABF formation, it is principally
through the precipitation of polysaccharide impurities in sugar that are sparingly soluble
in the acidic beverage. Therefore, reports on polysaccharide impurities in sugar are
considered to be relevant, and some are included here. Chill haze in alcoholic beverages
is also mentioned.
2. The soft drink manufacturer’s perspective
From the perspective of the soft drink manufacturer, the appearance of ABF in soft drinks
is a quality problem that is not easily managed. There is no easy, quick and reliable test
for ABF in the sugar ingredient. When ABF is manifest in soft drinks the product is
usually at the point of retail. While there is a strong link between ABF and sugar quality,
ABF is a complex coacervate containing material from the sugar, water and most likely
also the flavoured syrup ingredients. Furthermore, the mechanism of ABF formation is
still not well established. While to some extent water quality plays a role in ABF
formation, nevertheless, no floc guarantees in sugar purchase contracts are often required
by soft drink manufacturers. When ABF appears in soft drink, the sugar manufacturer is
quickly blamed and compensation is sought.
Taylor (1998) in Chemistry and Technology of Soft Drinks and Fruit Juices describes the
ingredients of soft drinks and briefly mentions an unsightly precipitate in soft drinks
resulting from algal polysaccharides and polypeptides, and humic acids in the water
supply. Taylor is an employee of the food ingredients section of Danisco, a large
European sugar company. His failure to acknowledge the contribution of sugar products
to soft drink floc may be part of the parley between beverage and sugar manufacturers.
Taylor continues to describe the contents of soft drink syrup formulations. Food grade
saponins are used to improve the foaming characteristics of cola and other formulations,
and stabilizers, such as alginates, carrageen, vegetable gums, pectin, acacia, guar,
tragacanth, xanthan and carboxymethylcellulose are used to improve mouthfeel, increase
viscosity, and stabilize natural cloudiness (and in fruit drinks to disperse fruit solids). It
is not surprising that it has been difficult to elucidate the chemistry of acid beverage floc
formation and the contribution of sugar impurities to its cause, given that polysaccharides
and saponins may be present in the syrup formulations as well as in the sugar as
impurities (albeit at mg/kg quantities in both cases).
The section on carbonated beverages (Jones, 1978) in the Kirk-Othmer Encyclopaedia of
Chemical technology (contributed by an employee of Royal Crown Cola Corp.) also lists
and describes the ingredients of soft drinks. Poor quality sugar is noted to have
extremely detrimental effects on beverage taste, odour and stability. Quality control is
discussed and algae in the water supply causing sediment mentioned. No mention is
made of ABF. However, in terms of quality control there is very little the beverage
manufacturer can do about ABF, as the beverage would likely be on the retail shelf
before the problem is manifest.
Hammond (1998) also in Chemistry and Technology of Soft Drinks and Fruit Juices
states, ‘The most common cause for floc formation in clear soft drinks is microbial
growth.’ He also attributes another cause of floc to algal polysaccharides in the water
supply. Certainly, these possibilities should be eliminated before the sugar supply is
implicated in floc formation.
3. Floc in soft drinks from beet sugars
As early as 1952, researchers at Spreckels Sugar Company, a Californian beet processor,
had isolated floc from simulated acid beverages and found the floc to contain saponins and
small amounts of high molecular weight materials (most likely pectin) (Eis, et al., 1952).
The saponins were determined to be the glycosides of the triterpene oleanolic acid.
Subsequent to this work, Ridout, et al. (1994) elucidated the structures of three beet
saponins (all were glucuronic acid glycosides of oleanolic acid). Walker and Owens (1953)
isolated a precipitate from acidified ABF positive beet sugar and found it to contain
saponins, fats, carbohydrates, colloidal carbon particles from the manufacturing process
and silica.. Walker and Owens believed that saponins were primarily responsible for floc
formation, and that the surfactant saponin carried fats into the refined sugar and the floc.
Silica and colloidal carbon particles were believed to be swept up by the floc aggregate,
rather than play active roles in the floc formation mechanism.
Clarke, et al. (1997) was able to form ABF like flocs in simulated beverages containing
non-flocculating beet sugars by the addition of a methanolic extract of sugar beets or a
beet saponin plus a protein. They proposed a mechanism based on interaction of
negatively charged saponins or polysaccharides and positively charged proteins as an
initial step in ABF formation. While the mechanism is prima facie believable, the
evidence supporting the mechanism (viz., that a mixture of a methanolic extract of sugar
beets or a beet saponin plus a protein forms floc) is by no means proof that ABF forms in
this way in authentic soft drinks.
ABF problems resulting from the use of beet sugar in soft drink manufacture have been
essentially overcome by minimising the amount of saponin impurity in the sugar product.
Since saponins are primarily in the leaves rather than the roots of beets, attention to
topping during harvest significantly reduces saponins in beet processing streams.
Diffusion at lower pH, carbonatation, ion exchange resins and activated carbon
treatments all to some extent remove saponins from beet processing liquors.
4. Floc in soft drinks from cane sugar
Stansbury and Hoffpauir (1959) isolated ABF from cane sugar sweetened soft drinks.
Analysis of the floc revealed the presence of starch, lipids (wax), particles of activated
carbon from refining, protein, and ash (principally silica). They concluded that ABF
formation was initiated by adsorption of solutes onto the carbon particles, but later this
was shown by Cohen, et al. (1970) not to be the case.
Cohen, et al. (1970) surveyed the impurity composition of raw and refines cane sugar,
and investigated the nature of ABF in cane sugar. ABF was found to contain protein,
inorganic material and polysaccharides. No correlation was found between floc and
inorganic matter or polysaccharide concentrations. They also reported that all ABF
forming sugars also developed a haze on addition of ethanol, but not all alcohol haze
forming sugars would develop ABF.
Miki, et al. (1975, 1980, 1984) isolated ABF from beverages made from Australian,
Philippine, Cuban and South African raw sugars. All flocs contained polysaccharides
(23.7 to 56.4 %), silicates (24.8 to 43.2 %) and protein (5.6 to 25.7 %). The
polysaccharide component contained glucose, mannose and galactose (and some
arabinose, xylose and rhamnose). They postulated the existence of a
galactoglucomannan, or a mixture of a galactomannan and a glucomannan from sugar in
the ABF complex. Treatment of the polysaccharide component with starch and dextran
hydrolysing enzymes reduced the glucose content, and after gel permeation
chromatography a galactomannan was obtained. Miki (1984) proposed two possible
structures for the galactomannan, viz., a backbone of 1-6 and 1-2 mannopyranosyl
structure with single 1-2 galactosyl side chains, or a backbone of 1-6 mannopyranosyl
structure with 1-2 mannopyranosyl side chains terminating with either 1-2
mannopyranosyl or 1-2 galactopyranosyl units. The isolation of the galactomannan
does not preclude the existence of a galactoglucomannan or a glucomannan in the floc
complex.
Roberts and Carpenter (1974) isolated floc from simulated beverages and found it to
contain silica (63.5 %), fats or waxes (5.26 %), starch (5.50 %) and other organic
compounds including protein and other polysaccharides. The polysaccharide fraction
was further purified, hydrolysed and analysed. The polysaccharides consisted of
arabinose (0.63 %), rhamnose (0.48 %), xylose (0.69 %), mannose (1.21 %), galactose
(0.58 %) and glucose (14.4 % - all numbers based on % of original floc material). The
amino acid composition of proteins in ABF also is reported.
Morel du Boil (1997) reviews published work on ABF from cane sugar and identifies
types of floc (viz., alcohol haze in sweetened alcoholic beverages, precipitation of
polysaccharides and proteins, silicate floc from water supply and microbial
contamination). Morel du Boil’s commentary with 48 references is a summary of the
work of others, rather than a critical review. She subscribes to the mechanism proposed
by Clarke, et al. (1977) described below, and seems to disregard the work of Miki, et al.
(1975, 1980, 1984).
There is consensus among researchers that ABF in cane sugar sweetened soft drinks is a
complex coacervate of silica, polysaccharides, protein, waxes and other organic
compounds. Certainly, most of the polysaccharides, waxes and some of the proteins
and silica are present as impurities in refined cane sugar (albeit at concentrations less
than 0.1 %). Much of the research on ABF is concerned with the elucidation of
polysaccharide structures in the floc coacervate. This is not surprising given that it is
most likely that a sparingly soluble high molecular weight polysaccharide is pivotal to
initial floc forming mechanisms. For this reason polysaccharide impurities in refined
cane sugar are considered part of the scope of this review, and are summarised below.
5. Polysaccharides impurities in refined cane sugar
The polysaccharides impurities in cane sugar may be either indigenous to the cane plant
or result from microbial activity in the field or the factory. The types of polysaccharides
present in cane sugar and their sources have been well researched and are described in
several texts (e.g., Gratius, et al., 1995; Kitchen, 1988). The polysaccharide impurities of
cane sugar and their origins are summarised in table 6.1.
Table 5.1 Summary of polysaccharide impurities in cane sugar.
Polysaccharide Origin Chemical nature
Starch Cane plant Amylose - linear 1-4 glucopyranosyl polymer;
Amylopectin - 1-6 branched, 1-4 glucopyranosyl
polymer
Dextran Microbial infection 1-4branched, 1-6 glucopyranosyl polymer
Cellulosans Cane plant Soluble fragments of cellulose (1-4 glucopyranosyl
polymer)
Hemicelluloses Cane plant Soluble fragments of xylans, mannans, galactans and
arabinoglucuronoxylans
Indigenous sugarcane
polysaccharide (ISP) Cane plant Polymer of arabinose, galactose and glucuronic acid
Sarkaran Unknown, possibly
microbial linear 1-4 and 1-6 glucopyranosyl polymer - a
pullulan
Pectin Cane plant 1-4 glacturonic acid polymer
Levans Microbial infection 2-6 fructofuranosyl polymer
Galactoglucomannan Unknown Either a galactoglucomannan or a mixture of a
glucomannan and a galactomannan
The most well known and well researched polysaccharide impurity is dextran (e.g.,
Sutherland, 1960, Leonard and Richards, 1969, Hidi, et al., 1974, Charles, 1984).
Dextran in sugar was first identified by Nicholson and Liliental (1959). Bruijn (1966)
observed it in deteriorated cane and attributed its presence to Leuconostoc mesenteroides
infection. Bruijn used globular proteins as standards in gel permeation chromatography
and underestimated the molecular weight of dextran. Covacevich, et al. (1977) elucidated
the structure of native dextran by methylation analysis and noted its similarity to that
from L. mesenteroides NRRL B512-f.
Indigenous sugar cane polysaccharide (ISP) was first isolated by Roberts, et al. (1964),
and later identified as an arabinogalactan (Roberts, et al., 1976a). In subsequent work
they reported the presence of glucuronic acid groups on the polysaccharide (Roberts, et
al.,1978a). Blake and Clarke (1984a) isolated ISP from Australian canes and compared it
to samples of Roberts’ Louisiana ISP. They found much lower levels of glucuronic acid
in both the Australian and Louisiana ISP isolates, and disputed the existence of
glucuronic acid in the ISP structure. To the best of the reviewer’s knowledge this
remains an unresolved issue.
Sarkaran has been identified in raw sugar from stale (South African, Bruijn, 1966) and
stand over (Australian, Blake and Clarke, 1984b) sugar canes. Bruijn noted the
differrence between the structure of the stale cane polysaccharide and pullulan, and
therefore named it sarkaran. Blake and Clarke obtained the sarkaran from stand-over
cane, but were not able to obtain it from yeast or fungal cultures isolated from the same
stand-over cane. However, neither Bruijn nor Blake and Clarke seemed aware that
Catley, et al. (1966) had noted minor structural features of pullulans (viz., pullulan
contains both maltotetrose and maltotriose subunits polymerised by 1-6 linkages) that
would encompass the sarkaran structure. Sarkaran is therefore a pullulan and most likely
is from microbial sources.
6. Tests for acid beverage floc
Alcian blue or Basacryl orange dye tests are summarised by Clarke, et al. (1997). They
are sensitive to the presence of negatively charged polysaccharides. They were never
good predictors of floc formation and the carcinogenic dyes are now banned substances.
Stansbury and Hoffpauir (1959) reported that floc tests used by the beet sugar industry
did not work for cane sugars. They also investigated a test used by Pepsi Cola Co. in the
late 50's, that was based on measurement of turbidity after coagulation of negative
charged colloids with a quaternary amine and found the procedure failed to differentiate
ABF positive and ABF negative sugars. They were unable to make any marked
improvement on the Coca Cola 10 day floc test.
Liuzzo and Wong (1982) developed a protein-dye-binding colorimetric method to
measure trace amounts of protein in refined sugars. They compared the protein-dye-
binding method to the Kjeldahl method and tested three floc forming sugars and three
non-flocculating sugars. Floc forming sugars had protein concentrations between 0.3 %
and 0.4 %, whereas non-flocculating sugars had protein concentrations between 0.004 %
and 0.006 %. Liuzzo and Wong concluded that there was a clear difference in protein
concentrations between ABF positive and ABF negative sugars. In this study on a
limited number of samples, they did not describe or quantify the ABF from the three floc
forming sugars. No follow up study on a larger sample of sugars has been reported, and
to the best of the reviewer’s knowledge this protein-dye-binding method is not use by
either soft drink or sugar manufacturers.
At this point in time there seems to be no published, reliable and rapid method to
determine the floc propensity of sugars. The Coca Cola test based on observation of
simulated acid beverages at high sucrose concentrations after 10 days is the only
internationally recognised test for ABF. There are several variants of the Coca Cola test
(see Clarke, 1997). In 1970, the International Commission for Uniform Methods of
Sugar Analysis tentatively adopted the 10 day Coca Cola test (McGinnis, 1970). The SRI
and the Australian sugar industry uses a variant of this test which is based on semi-
quantitative measurements at 7, 14, 21 and 28 days; a copy of this method is provided
(Anon, 1982).
7. Mechanism of ABF formation in cane sugar sweetened soft drinks
Roberts, et al. (1976 b) investigated the effect of silica and polysaccharide interactions in
ABF formation. While simulated beverages containing floc negative sugar, indigenous
sugar cane polysaccharide (ISP) and silica formed a floc, those containing indigenous
sugar cane polysaccharide and silica did not. They concluded that the sugar contained a
factor that interacted with the (ISP) and down played the role of silica. In later work
(Roberts and Godshall, 1978a, 1978b) colloidal ISP was found to carry a negative charge
and floc was obtained in simulated beverages upon the addition of ISP and a protein
fraction isolated from cane sugar.
Clarke, et al. (1977, 1978) summarised the findings of the SPRI research efforts and
proposed the a mechanism for ABF formation. This mechanism is shown in figure 1. In a
1997 paper on beet sugar ABF, Clarke, et al. (1997) again outlines the mechanism for
ABF formation in cane sugar sweetened soft drinks (viz. ISP, a polysaccharide with
glucuronic acid residues from plant cell walls, and protein from the plant or a processing
additive form a coacervate that then interacts with neutral polysaccharides such as
dextran and other compounds to form a floc). The assumption that the initial step is
based on the interaction of ISP and protein is based on the ability to form a haze and then
a floc when these compounds are mixed in a simulated beverage. While the mechanism
is prima facie believable, the evidence supporting the mechanism is by no means proof
that ABF forms in this way in authentic soft drinks. Clarke, et al. (1997) also make brief
mention of a regional specific floc; presumable they are referring to the work of Miki, et
al. who isolated a floc with a different polysaccharide composition.
Figure 7.1 Mechanism of ABF formation proposed by Clarke, et al. (1977)
Larsson, et al. (1999) studied effect of simple electrolytes (e.g., MgCl2, Na2SO4, KBr) on
the flocculation of nanosized silica particles (size 3 to 5 nm) and linear or branched
cationic polymers. The linear polymer was polyacrylamide and the branched polymer
was a diethylaminoethyl linked amylopectin. Their interest in these reactions was
founded in the use of these types of polymers in the dewatering step of paper
manufacture. The significance of this work to the mechanism of ABF formation is it
establishes that silica and charged polysaccharides (albeit positively charged) can form
floc. It also is interesting that the stoichiometry of silica to polyelectrolyte, and the nature
of the polyelectrolyte (branched or unbranched) affected the structure and size of the floc
complex. This recent work is strong evidence for a more pivotal role of silica in ABF
formation, and is supported by the work on chill haze in alcoholic beverages.
8. Chill haze in alcoholic beverages
Many alcoholic beverages, and especially beers manufactured from malted barley and
hops, may form a haze on chilling to temperatures less than 5 ΕC. This haze, caused by
the interaction of proteins and polyphenols, is well described in the literature (e.g.,
INDIGENOUS SUGARCANE POLYSACCHARIDE PROTEIN+
MOLECULAR AGGREGATES
FLOC PARTICLES
NETWORK
ACID BEVERAGE FLOC
DEXTRANS
COLLOIDAL &
SOLUBILIZED
SPECIES
STARCH & OTHER
POLYSACCHARIDES
SILICATES
COALESCE
Siebert and Lynn, 1999, Asano, et al., 1982, McMurrough. et al., 1999). Some of this
literature has implications to ABF formation in non-alcoholic acid beverages. For
example, the beer manufacturing industry uses silica to stabilise the product, and the
interaction of silicates with proteins is certainly relevant to this review.
The mechanism that leads to chill haze involves the oxidation of polyphenols and
subsequent interaction of these oxidised polyphenols with proteins at low temperatures
(Fernyhough, et al., 1994). This chill haze disappears on heating. Oxidation of
polyphenols leading to polymerisation may also produce an irreversible haze (O’Rourke,
1994). McMurrough, et al. (1999) also attribute chill haze to colloidal instability of the
protein-polyphenol complex, but note that hazes may still form by precipitation of
polysaccharides or calcium oxalate.
It has been observed that the addition of silica hydrogel (a solution of colloidal silica (~30
%) and water (~70 %), with a narrow particle size distribution) is effective in decreasing
the amounts of polyphenol sensitive proteins in beer (Fernyhough, 1994). The properties
of the silica hydrogel particles such as the pore diameter and the specific surface area
influence the removal of proteins from the beverage. Chill haze forming proteins are best
removed by silica hydrogel particles with a mean pore diameter in the range of 30-120 Å.
Silica surfaces bare residual valence electrons that react with water at acidic pH to form
silanol (SiOH) groups. Polar organic molecules containing oxygen or nitrogen atoms (i.e.,
bearing a free pair of electrons) can hydrogen bond to the silanol groups. In this way
proteins are adsorbed onto the silica surface. Proteins may also be adsorbed onto the silica
surface by ionic bonding through quaternary ammonium ions, and by hydrophobic
interactions. Fernyhough, et al. (1994) have shown that proteins bind most strongly at their
isoelectric point, and in some cases the pH range of binding may be very narrow. In beer,
the adsorption of proteins onto silica hydrogel has been found to occur between pH 3 and
pH 5. The adsorption capacity of the silica depends on size, surface area, porosity and
density of silanol groups. In this pH range polyphenols and polysaccharides also were
observed to bind to the silica hydrogel.
Studies by Siebert and Lynn (1999) have shown that silica gel can remove about 25 % of
haze active proteins in apple cider and about 85 % of haze active proteins in beer. To
explain this difference, they propose that the silica gel and polyphenols compete to bind
haze active proteins. Apple cider contains a high concentration of polyphenols compared
to protein, so there is little adsorption of proteins onto silica. Conversely, beer contains a
low concentration of polyphenols compared to proteins, so more proteins are free from
ployphenol interaction and can adsorb onto the silica surface.
The significance of this work to ABF in soft drinks is that it establishes the interaction of
colloidal silicates with solutes by hydrogen bonding, and both ionic and hydrophobic
interactions. The importance of stoichiometry of solutes in complex systems (involving
more than one solute type) is also illustrated (i.e., the proportions of solutes must be
optimal in order to achieve significant adsorption onto the silica surface). In fact,
stoichiometric effects are also observed in solute-solute (i.e., protein-polyphenol)
interactions. Siebert’s model (1999) to explain the protein-polyphenol interaction
proposes that if the amounts of protein and polyphenol molecules are about the same,
then there will be maximum haze development, and if the stoichiometry is not balanced
there will be smaller particles and less haze.
9. Technology options to reduce Floc formation
Dunsmore, et al. (1978) investigated the removal of ABF forming constituents in refining
processes. Specifically, they investigated the use of enzyme treatments and clarification
and filtration processes. Proteolytic enzymes were not suitable because the pH and
temperature optimums restricted their use in refining. While it was observed that
carbonatation appeared to produce a refined sugar that did not floc, the same observation
could not be made for phosphatation (with and without flocculants). Syrups that were
filtered (0.45 to 1.20 µm) at temperatures below 60 oC did not floc. Filtration (0.45 µm)
at higher temperatures (up to 80 oC) also produced syrups that did not floc. At the time
industrial scale micro and ultrafiltration was considered to be too costly, so the authors
investigated as range of diatomaceous earth filter aids as a means of removing floc
precursors. Filtering through Celite at temperatures below 60 oC also produced syrups
that did not floc. The report appears to provide practical solutions to prevention of floc
formation, i.e., carbonatation clarification, and/or filtering with Celite at temperatures
below 60 oC.
The conclusion of Dunsmore, et al. that carbonatation clarification removes ABF
precursors but phosphatation clarification does not, appears to be based on a single
sample of raw sugar. However, it is supported by the observations of Roberts, et al.
(1978c) that polysaccharides are better removed by carbonatation than by phosphatation.
Roberts et al. measured total polysaccharide content of process streams throughout a
carbonatation refinery and a phosphatation refinery, and speculated that charged
polysaccharides (i.e. containing carboxyl groups) rather than neutral polysaccharides
were being removed in the clarification process. The work of Roberts, et al. has been
frequently cited (e.g. Chen and Chou, Eds., 1993, Chou, Ed., 2000) with the unfortunate
error of reporting Roberts’ speculation (on charged polysaccharide removal) as fact.
However, since both charged and neutral polysaccharides are found in the ABF, the
removal of any polysaccharides (charged or neutral) by clarification should at least
reduce the extent of ABF formation. While the owners of carbonatation refineries may
feel secure in the knowledge that their refined sugar product is unlikely to form ABF, the
extent of the ABF problem for phosphatation refineries is not enough to justify the cost of
changing to a carbonatation process.
Dunsmore, et al. (1978) refer to the ABF formation mechanism of Roberts and Clarke
and speculate that filtering (and the use of filter aids) removes impurities that are
occluded in the ABF coacervate. Furthermore, they speculate that it is these impurities
that render ABF visible and that an invisible floc still forms in beverages made from
filtered sugar solutions. They do not provide an explanation for the effect of filtration
temperature on the propensity of sugars to form ABF, but it is clear from the presented
data that filtration at temperatures of 60 oC or lower removes ABF precursors. Since the
ease of further impurity removal from process streams decreases with each unit operation
of refining, Dunsmore, et al. recommend Celite filtration be carried out as early as
possible in the refining process.
For phosphofloatation refineries, the practical solution the ABF problem is micro or
ultrafiltration. While membrane filtration of raw sugar melts generally has been
demonstrated to be uneconomic, many investigations report almost quantitative removal
of polysaccharides by this process. From an economic standpoint it is clear that the best
approach to controlling ABF is to membrane filter only the sugar destined for beverage
manufacture (as a liquid refined sugar and immediately before it is used in beverage
manufacture), rather than at the raw sugar production stage.
10. Conclusions
While there is consensus among researchers that ABF in cane sugar sweetened soft drinks
is a complex coacervate of silica, polysaccharides, protein, waxes and other organic
compounds, the mechanism of ABF formation is still not completely elucidated.
Certainly, most of the polysaccharides, waxes and some of the proteins and silica are
present as impurities in refined cane sugar (albeit at concentrations less than 0.1 %), so
the implication of poor quality sugar ingredients in ABF formation is reasonable. The
proposed mechanism of ISP and protein interaction as an initial step in ABF formation is
based upon the observation that a mixture of ISP and protein in a simulated acid beverage
forms floc. This is by no means proof that ABF forms in this way in authentic soft
drinks, and in fact the presence of glucuronic acid groups (imparting a negative charge to
ISP that is essential to the mechanism) remains in dispute.
Literature on the related beverage quality issue (viz., chill haze in alcoholic beverages)
establishes the interaction of colloidal silicates with solutes (such as proteins and
polysaccharides) by hydrogen bonding, and both ionic and hydrophobic interactions. In
the reviewer’s opinion the role of silica in ABF formation is poorly understood, but may
be pivotal in the initial mechanism of ABF flocculation.
This critical review of published literature on ABF identifies a best bet technology for the
removal of floc precursors from sugar products. The technology that has been identified
is micro or ultrafiltration (and possibly at temperatures in the region of 60 oC), although
other forms of filtration (e.g. plate and frame with Celite filter aids) at reduced
temperatures may also be effective.
11. References
Anon (1982) Excerpt from - Analysis of raw, refined and affined sugars and refined
liquors for floc potential by beverage floc test, SRI internal publication.
Ashurst, P. R. (Ed.) (1998) Chemistry and Technology of Soft Drinks and Fruit Juices,
Sheffield Academic Press, Sheffield.
Asano, K., Shinagawa, K. and Hashimoto, N. (1982) Characterisation of Haze-Forming
Proteins of Beer and Their Roles in Chill Haze Formation, J. Amer. Soc. Brewing
Chemists, 40, 147-154.
Bruijn, J. (1966) Deterioration of sugar cane after harvesting, Int. Sugar J., 68, 356-358.
Blake, J. D. and Clarke, M. L. (1984) Observations on the structure of ISP, Int. Sugar J.,
86, 295-299.
Blake, J. D. and Clarke, M. L. (1984) A water soluble polysaccharide from stand-over
cane, Int. Sugar J., 86, 276-279.
Catley, B. J., Robyt, J. F. and Whelan, W. J. (1966) A minor structural feature of
pullulan, Biochem J., 100, 5 - 6.
Charles, D. F. (1984) Polysaccharides in refined and raw sugar, Int. Sugar J., 86, 105-
109.
Chen, J. C. P. and Chou, C. C. (Eds.) Cane Sugar Handbook: A manual for cane sugar
manufacturers and their chemists, 12th Ed., John Wiley & Sons, Inc., New York.
Chou, C. C. (Ed.) (2000) Handbook of sugar refining: A manual for the design and
operation of sugar refining facilities, John Wiley & Sons, Inc., New York.
Clarke M. A., Roberts, E. J., Godshall, M.A. and Carpenter, F.G. (1977) Beverage floc
and cane sugar, Proc. 16th ISSCT Congress, 2587-2598.
Clarke M. A., Roberts, E. J., Godshall, M.A. and Carpenter, F.G. (1978) Beverage floc
and cane sugar, Int. Sugar J., 80, 197-201.
Clarke, M. A., Roberts, E. J. and Godshall, M. A. (1997) Acid beverage floc from beet
sugars, Zuckerindustrie, 122, 873-877.
Cohen, M. A., Dionisio, O. G. and Drescher, S. J. (1970) The isolation and
characterisation of certain impurities responsible for quality problems in refined cane
sugar, Proc. 29th Ann. Meeting Sugar Ind. Technol., 123-165.
Covacevich, M. T. and Richards, G. N. (1977) Studies on dextrans isolated from raw
sugar manufactured from deteriorated cane, Int. Sugar J., 79, 3-9.
Dunsmore, A., Heal, M. J., Matic, M. and Runggas, F.M. (1978) A practical solution to
the acid beverage floc problem, Proc. 37th Ann. Meeting Sugar Ind. Technol., 514-527.
Eis, F. G., Clark, L. W., McGinnis, R. A. and Alston, P. W. (1952) Floc in carbonated
beverages, Ind. Eng. Chem., 44, 2844-2848.
Fernyhough, B., McKeown, I.. and McMurrough, I. (1994) Beer Stabilisation with Silica
Gel, Brewers’ Guardian, 123, 44-50.
Gratius, I., Decloux, M., Dornier, M. and Cuvelier, G. (1995) The determination of
polysaccharides in raw cane sugar syrups, Int. Sugar J., 97, 296-300 & 339-343.
Hammond, D. A. (1998) Analysis of soft drinks and fruit juices, in Chemistry and
Technology of Soft Drinks and Fruit Juices, Ed. Ashurst, P.R., Sheffield Academic Press,
pp 166-196.
Hidi, P., Keniry, J. S., Mahoney, V.C. and Paton, N.H. (1974) Observations on the
occurrence and nature of polysaccharides in sugar canes, Proc. 15th ISSCT Congress,
1255-1265.
Jones, M. B. (1978) Kirk-Othmer Encyclopedia of Chemical Technology, 3rd Ed. John
Wiley & Sons, New York, 4, pp 710-725.
Kitchen, R. A. (1988) Polysaccharides of sugarcane and their effects on sugar
manufacture, Chemistry and Processing of Sugarbeet and Sugarcane, Eds. Clarke, M. A.
and Godshall, M. A., Elsevier Science Pub., Amsterdam, pp 208-235.
Larsson, A., Walldal, C. and Wall, S. (1999) Flocculation of cationic polymers and
nanosized particles. Colloids and Surfaces A: Physicochemical and Engineering Aspects,
159, 65-76.
Leonard, G. J. and Richards, G. N. (1969) Polysaccharides as causal agents in production
of elongated sucrose crystals from cane juice, Int. Sugar J., 71, 263-265.
Liuzzo, J. A. and Wong C. M. (1982) Detection of Floc-Producing Sugars by a Protein
Dye-Binding Method, J. Agric. Food Chem., 30, 340-341.
McGinnis, R. A. (1970) Referees Report Subject 19 - Characteristics of white sugar,
Report on the proceedings of the 15th session of ICUMSA, pp 200-213.
McMurrough, I., Madigan, D., Kelly, R. and O’Rourke, T. (1999) Haze Formation Shelf-
Life Prediction for Lager Beer, Food Technol., 53, 58-62.
Miki T., Saito, S. and Kamoda, M. (1975) Composition of polysaccharides in carbonated
beverage floc, Int. Sugar J., 77, 67-69.
Miki, T., Saito, S., Ito, H. and Kamoda, M. (1980) Composition of ploysaccharides in
carbonated beverage floc from raw cane sugar, Proc. 17th ISSCT Congress, 2751-2762.
Miki, T. (1984) A galactomannan in carbonated beverage floc from raw cane sugar.
Carbohydr. Res., 129, 159-165.
Moore, W. and Buzzanell, P. (1991) Trends in U.S. Soft Drink Consumption - Demand
Implications for Low-Calorie and Other Sweeteners, Sugar and Sweetener Outlook
Report, USDA SSRV 16N3.
Morel du Boil, P. G. (1997) Refined sugar and floc formation, a survey of the literature,
Int. Sugar J., 99, 310-314.
Nicholson, R. I. and Liliental, B. (1959) Formation of a polysaccharide in sugarcane,
Aust. J. Biol. Sci., 12, 192-203.
O’Rourke, T. (1994) The requirements of beer stabilisation, Brewers’ Guardian, 123, 30-
33.
Ridout, C. L., Price, K. R., Parkin G., Dijoux, M. G. and Lavaud, C. (1994) Saponins
from Sugar Beet and the Floc Problem, J. Agric. Food Chem., 42, 279-282.
Roberts, E. J. and Carpenter, F. G. (1974) Composition of acid beverage floc, Proc. 1974
Tech. Session Cane Sugar Refining Research, 39-50.
Roberts, E. J. and Godshall, M. A. (1978a) Identification and estimation of glucuronic
acid in indigenous polysaccharides of sugar cane, Int. Sugar J., 80, 10-12.
Roberts, E. J. and Godshall, M. A. (1978b) The role of charged colloids in floc formation,
Int. Sugar J., 80, 105-109.
Roberts, E. J., Clarke, M. A., Godshall, M. A. and Carpenter, F. G. (1978c) Removal of
some polysaccharides in refineries, Sugar J., Feb 1978, 21-23.
Roberts, E. J., Godshall, M. A., Carpenter, F. G. and Clarke M. A. (1976a) Composition
of soluble indigenous polysaccharides from sugar cane, Int. Sugar J., 78, 163-165.
Roberts, E. J., Godshall, M. A., Clarke M. A. and Carpenter, F. G. (1976b) Some
observations on acid beverage floc, Int. Sugar J., 78, 326-328.
Roberts, E. J., Jackson, J. F. and Vance, J. H. (1964) Progress in research on the soluble
polysaccharides of sugarcane, Proc. Tech. Sessions Cane Sugar refining Res., 76-84.
Siebert, K. J. and Lynn, P. Y. (1997) Mechanisms of Adsorbent Action in Beverage
Stabilisation, J. Agric. Food Chem., 45, 4275-4280.
Siebert, K. J. (1999) Protein-Polyphenol Haze in Beverages, Food Technol., 53, 54-57.
Sutherland, G. K. (1960) An investigation of the polysaccharides present in sugar mill
syrups, Aust. J, Biol. Sci.,13, 300-306.
Stansbury, M. F. and Hoffpauir, C. L. (1959) Composition of “Floc” Formed in Acidified
Sirups from Refined Granulated Cane Sugars, J. Agric. Food Chem., 7, 353-358.
Taylor, R. B. (1998) Ingredients, in Chemistry and Technology of Soft Drinks and Fruit
Juices, Ed. Ashurst, P.R., Sheffield Academic Press, pp 16-54.
Walker, Jr., H. G. and Owens, H. S. (1953) Beet Sugars: Acid insoluble constituents in
selected samples, J. Agric. Food Chem., 6, 450-453.
... Occasionally, these products develop flocs (large gauzy-appearing structures that float in the product). A number of causes have been associated with these, but most authors have attributed this to associations between positively charged proteins and negatively charged polysaccharides that form either under acidic conditions or in products containing ethanol (Clarke et al., 1978;Foong et al., 2002;Morel du Boil, 1997). ...
Article
Beverages such as beer, wine, clear fruit juices, teas, and formulated products with similar ingredients are generally expected by consumers to be clear (free of turbidity) and to remain so during the normal shelf life of the product. Hazy products are often regarded as defective and perhaps even potentially harmful. Since consumers are usually more certain of what they perceive visually than of what they taste or smell, the development of haze in a clear product can reduce the likelihood of repeat purchasing of a product and can have serious economic consequences to a producer. Hazes are caused by suspended insoluble particles of colloidal or larger size that can be perceived visually or by instruments. Hazes in clear beverages can arise from a number of causes, but are most often due to protein-polyphenol interaction. The nature of protein-polyphenol interaction and its effect on haze particles, analysis of haze constituents, and stabilization of beverages against haze formation are reviewed.
Article
In three weekly tests in 2018, samples of raw sugar, refinery liquor and refined sugar were collected from a Tucuman sugar factory/back-end refinery and analyzed for starch, polysaccharides, proteins and silicates. The standard 10-day acid floc behavior, which is a sample being either floc positive or floc negative, was correlated against the levels of the four impurities. Of these impurities, starch had statistically the most significant effect on floc formation, both in the refined sugar set as well as in the raw sugar/refinery liquor set of samples. Despite the use of some α-amylase at the mill, the starch levels in refined sugar were relatively high. The effects of polysaccharides, proteins and silicates on floc formation were statistically less significant, although the proteins were on average higher in floc-positive than in floc-negative sugars.
Article
The main aim of this paper is to discuss the current state of knowledge on the chemical composition of floc isolated from acidified sugar solutions and explain the mechanism of its formation. The literature data shows that depending on the raw material used for the production of white sugar, i.e. sugar beet and sugarcane, the chemical composition of floc is different. The presence of polysaccharides i.e. dextran or levan in floc is a consequence of microbial activity. Therefore, it cannot be explicitly assumed that the origin of individual ingredients of floc is only the raw material, i.e. beets or sugarcane. Taking into account the literature data concerning the chemical composition of floc, it seems reasonable to link the direct impact of microbial biofilms present in sugar industry installations to the tendency of the final product to create floc. Currently, the basis for assessment of the proneness of sugar to create floc is the result of a 10-day test (ICUMSA Method GS2/3-40A). From the perspective of both producers and consumers of sugar, there is a need to develop and validate quick methods for assessing the tendency of sugar to create floc. This task is only possible to complete, when the mechanism of floc formation is fully explained.
Chapter
Fruit juices are of great commercial importance in their own right, as well as for direct use as ingredients in food and beverage products. The beverage technologist has a wide range of fruit types to choose from. This chapter investigates some of the procedures associated with the processing of these to produce fruit juices commercially. To achieve a year-round supply of sound, good quality fruit, a highly efficient international quality management system is required to operate throughout the whole procedure of growing, harvesting, ripening, storing and handling of the fruit. The characterisation of fruit type or variety is reflected in the flavour profile of its volatile components. Aroma volatiles recovered during evaporative concentration of juices provide a prime source of natural flavouring. Legislative and regulatory adjustments are designed to clarify where there has been confusion, to maintain product standards, and to protect the consumer from being misled by descriptive ambiguity.
Chapter
This chapter covers the methods of analysis for different key elements of a soft drink formulation or a fruit/vegetable juice (e.g. sweeteners, acids and preservatives, etc.). It discusses the sensory assessment of soft drinks and fruit juices. In a soft drink, or a reconstituted fruit juice from concentrate, the quality of water used is an essential element. Two compounds that are used in some soft drinks formulations for specific purposes are caffeine and quinine. Vitamins are generally assayed in foods and beverages by either a specialist microbiological route or by a chemical procedure. The presence of colours in the beverage can interfere with the analysis of the vitamins. Diode array detection (DAD) or mass spectrometric (MS) detection is quite critical for this analysis, to be sure the correct material is being evaluated and that the peak is not co-eluting with any interferences.
Article
The major proportion of polysaccharides entering the mill or the refinery are generated by the sugarcane plant. The remaining polysaccharides are formed by microbiological contamination, which can occur either in the field or during processing. Recycling of these polysaccharides occurs during processing, and in refineries can result in a 15 fold increase in content from the raw sugar to the final molasses. All of these polysaccharides have a detrimental effect on sugar manufacture, and this effect is not easily diminished as very few of the processing steps remove polysaccharides.
Article
Acid beverage floc (ABF), a flocculated turbid material that can form in sugar sweetened, acidified, carbonated beverages after several days standing, is a customer problem to beverage bottlers and their suppliers of sugar. ABF from beet sugar has been reported (Eis, 1952; van der Poel, 1966) to be caused by a saponin from the beet plant. Recent work has shown the presence of several saponins in sugarbeet (Massiot, 1994; Ridout, 1994). Investigations at SPRI have confirmed that isolation and test procedures for saponins, as reported in the literature, are actually for oleanolic acid. In this paper, observations on ABF and its causes are reported. ABF from beet sugar is proposed to have a two factor basis: a negatively charged component and a positively charged component interact at acid beverage pH, forming a coacervate and subsequently coagulating into a floc. The negatively charged factor can be oleanolic acid, any of the saponins that contain a glucuronic acid moiety, or beet cell wall polysaccharide containing uronic acids. The positively charged component can be protein or peptide, with isolectric point above the beverage pH value of 2.5-3.0. ABF can be made readily by adding these components to non-floccing sugars. Acid beverage floc and its causative factors can be removed by intense or tight filtration. The negatively charged components can be removed to a great degree by formation of their calcium complexes (at pH above 8.5) and subsequent filtration.
Article
A literature survey on floe problems, shows a consensus that both protein and a 'charged' polysaccharide are necessary for floc formation in acidified sugar solutions. A mechanism incorporating sugar cane protein and indigenous sugarcane polysaccharide (ISP) has been postulated. The participation of ISP can be regarded as analogous to the behaviour of beet sugar saponin in the formation of acid beverage floc (ABF) with beet sugar. Floe formation occurs as a result of colloid de-stabilisation and the relative concentrations of protein and polysaccharide are critical.