Note: Descriptions are shown in the official language in which they were submitted.
CA 02274242 1999-OS-28
WO 98/28401 PC'>r/GB97/03281
AGGLGMERATED SILICAS
Field of the Invention
The present invention relates to a novel granular composition. The present
invention more particularly relates to agglomerated synthetic amorphous
silicas and
their use in stabilising fermented beverages against haze formation during
storage.
t3ackground to the Invention
Alcoholic fermented beverages, for example beers, have a tendency to produce
haze which can be of biological or physico-chemical origin, and a number of
products
and processes are used for the removal of haze-forming constituents. Whilst
gross
haze effects are resolved by filtration, flocculation, or centrifugation,
secondary haze
develops during storage due to interactions between certain polypeptides and
polyphenofs which coagulate and precipitate. This haze therefore becomes
apparent
only at a stage when the beverage is being prepared for consumption and when
removal is impractical. A number of organic and inorganic substances can be
used to
remove the polypeptide and polyphenol haze precursors prior to packaging and
so
stabilise the beverage; such as tannic acid, polyvinylpolypyrrolidone,
bentonite, active
carbon and silicas.
Amorphous silica hydrogels and xerogels, and blends thereof, selectively
remove
polypeptide haze precursors without impairing properties such as body,
flavour, colour
and head formation; see for example Hough, J.S., "Silica Hydrogeis for Chill
Proofing
Beer", MBAA Technical Quarterly vol. 13) No. 1, pp 34-39 {1976); Halcrow,
R.M., "Silica
Hydrogels", The Brewers Digest, pp44 {August 1976 and Hough, J.S., and Lovell,
A.t-.,
"Recent Developments In Silica Hydrogels, for the Treatment and Processing of
Beers",
MBAA Technical Quarterly, vo1.16, No.2, pp90-100 (1979).
Amorphous silica hydrogels are non-dusty, easy to handle powders. in contrast,
amorphous silica xerogels, while being effective beer stabilisers, are
extremely dusty
and pose difficult handling problems.
The present invention is aimed at delivering low dusting, highly stabilising
silica
' gels for use in stabilising beverages. Although the invention is directed to
the
treatment of beers, which term includes lager, Pllsner, Dortmund and Minato
beer, as
well as top fermented beers such as ale) porter and stout, it is applicable to
other
fermented liquids which are liable to generate haze on storage, as well as to
non-fermented beverages that may form haze during storage, such as fruit
juices and
iced teas.
Prior literature shows that silicas either in the form of hydrogels, xerogels
and
precipitated silicas, and blends thereof, with surface areas ranging from 200
to 1100
m2/g, pore volumes 0.35 to 2.5 cc/g and particle sizes in the ranges 3 to 30
microns
1
CA 02274242 1999-OS-28
WO 98/28401 PCT/GB97/03281
can be used to remove protein from beer, and so stabilise the beverage against
haze
formation during storage. More specifically the use of silica hydrogels with
surface
areas greater than 700 m2/g are disclosed in 681215928 for the treatment of
beer.
Fine particle sized xerogels and precipitated silicas as disclosed in EP-A-
0683222,
US-A-4515821, EP-A-0287232, US-A-5149,553, have one pronounced disadvantage
and that is their excessive dusting when introduced into the dosing equipment
normally
employed in the brewery. Silica hydrogels, or blends of hydrogels and xerogels
containing 30% to 60% by weight of SiOz are practically non-dusting, but these
materials are extremely cohesive and as a consequence they require specially
designed equipment for them to be handled in bulk.
The use of agglomeration to reduce the dustiness of precipitated silicas, or
siliceous materials, is disclosed in US-A-3646183, US-A-4336219, GB-A-2013165,
JP 56314/1982, DE-A- 2150346 and DE-A-1807714, US-A-4052334, GB-A-1543576 and
GB-A-1365516. These materials find use either as rubber fillers or catalyst
supports
and as such do not have pore structures suited to beer stabilisation. No prior
art
discloses the use of agglomerated silicas for beverage stabilisation
application.
Standard Procedures
The agglomerates, and their component particles, of the invention are defined
in
terms of their physico-chemical properties. The test methods used to determine
these
properties are:-
i) Surface Area and Pore Volume.
Surface area and pore volume was measured using an automatic BET specific
surface measuring apparatus; Micromoritics ASAP 1400 in accordance with the
BET method based upon nitrogen adsorption.
Measurement was taken by making reference to the following literature, S.
Brunauer, P.H. Emmett and E.Teller, J.Am.Chem.Soc., 60,309 (1938). Samples
were outgassed under vacuum at 270°C for 1 hour before measurement at
about
'196°C.
ii) Mean Pore Diameter (MPD)
The mean pore diameter was calculated in accordance with a cylindrical
mode! using:
MPD = 4,000 x PV/SA nm
Where PV is the pore volume (cc/g) and SA the specific surface area (m2/g) to
nitrogen as determined in (i) above.
iii) Total moisture content
The total moisture content was determined from the loss in weight of a
silica when ignited in a furnace at 1000°C to constant weight.
2
CA 02274242 1999-OS-28
WO 98/28401 PCT/GB97/03281
iv) Weight Mean Particle Size
The weight mean particle size of the sificas before agglomeration was
determined using a Malvern Mastersizer Model X, made by Malvern Instruments,
' Malvern, Worcestershire with MS15 sample presentation unit. This instrument
uses the principle of Fraunhoffer diffraction, utilising a low power HelNe
laser.
The particuiates are dispersed ultrasonically in water for 7 minutes to form
an aqueous suspension and then mechanically stirred before they are subjected
to
the measurement procedure outlined in the instruction manual for the
instrument,
utilising a suitable lens in the system.
The Malvern Particle Sizer measures the weight particle size of the
particulate. The weight mean particle size (d50) or 50 percentile, the 10
percentile
(d10) and the 90 percentile (d90} are easily obtained from the date generated.
v) Beer Stabilisation
The proteins in beer which precipitate in the presence of tannic acid, the
so called sensitive proteins -see Chapon) L., J.lnst.Brew.) vo1.99, 49 {1993) -
are
considered to be important poiypeptide haze precursors.
The sensitive protein content of degassed beer was determined using a
Tannometer supplied by Bearwell International Systems) 125 St. Mary's Road,
Market Harborough, Leicestershire, LE16 DT, England.
Measurement was made at 25°C using a 0.01 % w/v tannic acid solution
(Brewtan
C., supplied by Omnichem n.v., Cooppallaan 91, B9230, Wetteren, Belgie} by
following haze formed in EBC units. The haze formed at a concentration of
l0mg/1
added tannic acid was recorded. The results are expressed as the percentage
reduction in this haze value following treatment of the beer with silica at
30g/hi
either following overnight contact or following 10 minutes contact time.
vi) Agglomerate strength
Agglomerate strengths were measured using a Microson XL2020 Sonicator
programmable ultrasonic liquid processor, manufactured by Misonix Inc.,
Farmingdale, New York and supplied in the UK by Labcaire Systems Ltd., Avon.
The Microson XL2020 Sonicator ultrasonic processor has a maximum of 550 watts
output with a 20kHz convertor and is fitted with a 3/. inch tapped horn. The
processor has variable amplitude control and a microprocessor controlled
digital
timer integrated with a Pulsar cycle timer with power output and elapsed time
displays.
The piezoelectric convertor transforms electrical energy to mechanical energy
at a
frequency of 20KHz. Oscillation of piezoelectric crystals is transmitted and
focused by a titanium disruptor horn that radiates energy into the liquid
being
3
CA 02274242 1999-OS-28
WO 98/28401 PCT/GB97/03281
treated. A phenomenon known as cavitation, the formation and collapse of
microscopic vapour bubbles generated by the strong sound waves, produces a
shearing and tearing action. Almost all of the activity takes place just in
front of
the probe tip.
The generator provides high voltage pulses of energy at 20 kHz and
adjusts for varying load conditions, such as viscosity and temperature. It
senses '
impedance change and increases or decreases power to the probe tip
automatically.
The 3/. inch probe is a medium intensity horn for processing volumes
between 25 and 500m1. The maximum ampf itude at the tip of the probe is 60
microns.
Hence, sonicator processors operating at output control setting 10 have 60
microns of amplitude (peak to peak amplitude of the radiating face of the tip)
at the
tip of the probe.
Therefore, there is a linear relationship between the output control knob (or
amplitude adjustment knob) and the amplitude at the tip of the probe, i.e. 6
microns
of amplitude per control knob setting. The generator draws energy accordingly
to
maintain a constant amplitude at the tip for a given output control setting.
This is
displayed on the % output power meter and is power in Watts (i.e.: output -
%/100
550 watts available = x watts delivered).
A paper given by Mr. S. Berliner, (Director, Technical Services, Heat
Systems Ultrasonics Inc.) at the 9'" Annual Technical Symposium of the
Ultrasonic
Industry Association, entitled "Application of Ultrasonic Processors (Power vs
Intensity in Sonification"" provides further detailed information of the
principles
involved in this experimental technique.
Procedure
A 250mI Pyrex beaker is insulated and fitted with a lid with a 3/4 inch hole
in the centre to accommodate the ultrasonic probe and a 1 /8 inch hole to the
side
to accommodate a temperature probe.
Into the insulated beaker weigh the desired amount of deionised water,
and the desired amount of inorganic granule to obtain a final weight of 200 g.
A
magnetic stirrer bar is introduced into the beaker and the beaker is placed on
a
magnetic stirrer hotplate equipped with a temperature sensor (Heidolph MR3003
magnetic stirrer hotplate with a stainless steel PT-100 temperature sensor and
rpm
stirrer speed) obtainable from Orme Scientific,
4
CA 02274242 1999-OS-28
WO 98/28401 PCT/GB97/03281
Manchester). The beaker contents are stirred on setting 3 (-300 rpm)) the
ultrasonic probe is immersed to a depth of 5/8 inch into the liquid and the
temperature sensor is inserted into the liquid to continuously monitor
temperature.
~ The Sonicator ultrasonic processor is switched on and information on
processing time and pulsed mode programmed, as required. Cavitation is
introduced to the system by turning the output control knob to the desired
amplitude setting, whilst the temperature profile is closely monitored. The %
power
output required to maintain the amplitude at the tip is also recorded,
according to
the setting. When the cavitation process is complete, the stirrer is switched
off and
the magnetic stirrer bar is removed. Manual stirring is continued with a
spatula to
maintain dispersion.
+45 micron Wet Sieve Test Method
The inorganic particle dispersion is poured through a 45 micron sieve.
Any residue in the beaker is washed through the sieve, using half the amount
of
initial water. The sieve is then dried to constant weight in an oven at
150°C. The
residue which remains on top of the 45 micron sieve is then weighed and
expressed as a percentage of the initial weight of inorganic granule. The
greater
the amount retained on the sieve, the stronger the agglomerate strength of the
granule and the more difficult it is to breakdown. An optimum product will
have no
residue remaining on the sieve.
!t has been found that, for a granule to satisfactorily breakdown in
beverage application, it will have less than 5%) preferably less than 2%, most
preferably less than 1 % by weight) residua on a +45 micron sieve after
ultrasonification on setting 10 (60 micron amplitude) for a period of 7
minutes.
vii) Particle Size Distribution by Sieve Analysis
An accurate measure of the true particle size distribution of the granular
composition is done using sieve analysis.
1008 of the sample is placed on the top sieve of a series of BS sieves, at
approximately 50 micron intervals to cover the particle size range of the
granule.
The sieves are arranged in order with the finest at the bottom and the
coarsest at
the top of the stack. The sieves are placed in a mechanical vibrator e.g.
Inciyno
Mechanical Sieve Shaker by Pascail Engineering Co. Ltd., covered with a lid
and
shaken for 10 minutes. Each sieve fraction is accurately weighed and the
results
calculated:
% residue = Wt. Of residue * 100
Wt. Of sample
Brief Descril~tjon of the Invention
5
CA 02274242 1999-OS-28
WO 98/28401 ~ PCT/GB97/03281
According to the present invention there is provided a process for the
treatment
of a fermented alcoholic beverage which comprises contacting the beverage with
a
granular composition comprising 45 to 98 % w/w on dry basis of a water
insoluble
particulate wherein more than 75% by weight on dry basis, preferably more than
90% '
by weight on dry basis, of the water insoluble particulate is made from an
amorphous
silica) having a weight mean particle size of from 5 to 30 microns, preferably
from 10 to '
30 microns) a pore volume of more than 1 cc/g, and a mean pore diameter of
more than
60 Angstroms, the granular composition having a particle size, as measured by
dry
sieve analysis, such that more than 75% by weight of the granular composition,
preferably more than 95% by weight, has a particle size of more than 45
microns.
Preferably, the granular composition is sieved at 125 and 600 microns so that
more than 90% w/w of the granular composition has a parcile size between 125
and
600 microns.
Preferably, the granular composition has a granular strength such that less
than
5%, more preferably less than 2%, most preferably less than 1 % by weight,
residue is
left on a 45 micron wet sieve after ultrasonification on setting 10 (60 micron
amplitide)
for a period of 7 minutes.
Preferably also, the granular composition has a total moisture content of less
than
40%, preferably less than 30%, more preferably less than 20%.
Amorphous silicas which can be used in the present invention may be prepared
via an acid gel route or an alkaline precipitate route.
Siaecific Description of the Invention
Examples of the preparation of agglomerated synthetic amorphous silicas will
now
be gmen to illustrate but not limit the invention.
Preparation of 8gaiomerates
Silica gels having the properties shown in Table 1 were agglomerated
individually
at 2008 powder batch size (laboratory scale) with deionised water using a
Sirmon CV6
mixer, supplied by Metcalfe Catering Equipment Ltd., 8laneau Ffestiniog,
Wales. The
resulting wet agglomerates were then dried in an oven at 150°C for 4
hours, gently
forced through a 600 micron screen and sieved at 125 microns to collect the
greater
than 125 micron fraction. Each silica 1 to 5 was agglomerated this way to
produce
agglomerates A to E.
As an alternative route to prepare the agglomerates, approximately 1 kg of the
silica powder was dry compacted using an Alexanderwerk roller compactor
WP50/N75
with a feed rate setting at number 3, supplied by Alexanderwerk AG-D-42857,
Remscheid-Kippdorfstrasse 6-24. The strength of agglomerate prepared in this
way
depended on the pressure applied to the rollers. The silica cited as silica
number 1 in
6
CA 02274242 1999-OS-28
WO 98!28401 PCT/GB97/03281
Tabte 1 was agglomerated in this way at 15 bar (code F) and 50 bar (code G).
Example H was also produced with silica number 1 and was made even harder by
multiple passes at 50 bar. The prepared agglomerates were classified and
sieved as
~ described for the wet route structures.
Aaalomerate strength
The strength of the prepared agglomerates was determined as outlined earlier.
Table 2 records these strengths as the energy required to reduce 50% of the
structure
to less than 45 microns particle size. The results of Table 2 show that a wide
range of
strengths is attainable, the strength being a function of the physical
characteristics of
the silica particles making up the structure, as well as pressure applied and
number of
passes made through the aggfomerator.
deer Stabilisation
Agglomerated silicas of greater than 45 micron agglomerate size will only be
effective in beer stabilisation if they break down to deliver particles of 20
micron or
less. The results shown in Table 3 for an adjunct lager, serve to illustrate
this effect.
Thus samples coded G and H were measured for their effectiveness at removing
sensitive protein from a beer, before agglomeration (i.e. as large particle
agglomerates), and following de-agglomeration by dispersion in a beer sample.
The
samples were prepared from a common feedstock (i.e. silica example 1 in Table
1 )
under conditions designed to deliver different agglomerate strengths. The
results of
Table 3 show that when agglomerated to large particle size the stabilising
effectiveness
of the silicas is lost. Agglomerate G gives better performance than H due to
its weaker
structure (cf. Table 2) partially breaking down during addition to the beer.
Application
of shear to the system (i.e. magnetic stirrer bar) serves to deaggiomerate
sample G
such that it is once again able to deliver effective protein removal. The
strength of
sample H is such that the gentle shear forces applied are insufficient to
return the
primary particles and hence it fails to deliver the pre-agglomerated protein
removal.
Table 4 shows, on a beer different from the bear tested on Table 3, that an
input
of sufficient energy to reduce the agglomerates to small particle size (cf.
Table 2} is
effective in delivering the beer stabilising performance of the pre-
agglomerated
particles. Example O (prepared from silica example 4 of table 1 ) is a poor
stabiliser in
both pre and de-agglomerated forms due to not having a suitable pore structure
for
beer stabilisation, i.e. having a mean pore diameter of less than 60 Angstroms
and a
pore volume of less than 1 cm'/g. All of the remaining samples are good
stabilisers
and recover their performance following de-agglomeration.
7
CA 02274242 1999-OS-28
WO 98/28401 PCT/GB97/03281
Table 1 : Physical larol~erties of the silicas i~rior to
~,galomeration
EXAMPLE 1 2 3 4 5
TVM 7.5 8.6 8.6 22.4 7.5 '
Surface Area (m(lg)766 295 319 662 304
Pore Volume (cc/g)1.3 1.8 1.1 0.4 1.2
Mean Pore Diameter68 244 138 24 158
(A)
Particle Size
(microns)
D10 5.3 6.1 5.1 2.8 4.8
D50 11.9 11.5 10.7 8.7 8.8
D90 25.5 20.8 20.6 19.5 15.6
Table 2 ~ Agatomerate strenr~th - Ener4v to reduce 50% Qf structure
t-o less than 45 micros
A B C D E F G H
Energy (Joules)1,500 3,000 800 550 14,000850 900 18,900
Table 3 ~ Effect of Particle Size on S nc~tivp prot in Removal
Pre-agglomeration Agglomerated De-agglomerated
G H G H G H
Particle 11.9 11.9 125-600 125-60032.5 238
Size
(microns)
Reduction 62 65 30 12 60 22
in
Sensitive
Protein
Table 4 ~ Beer Stabilisation % Redmction in Sensitive Protein
A B C D E F G
Pre-a lomeration73 62 71 10 67 73 73
~De-agglomerated74 ~ 60 70 ~ 8 65 76 72
~ ~
r
8
