CA1051711A - Stabilized ice cream - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
A stabilizer system has been developed that is particularly useful in ice cream requiring more than usual stabilization. The system comprises microcrystalline cellulose with one or more of carboxymethyl cellulose and a galactomannan gam. A preferred system consists of micro-crystalline cellulose, carboxymethyl cellulose and locust bean gum or tara gum.

Description

~`~ c~731 S~
The invention relates to ice cream and to a st~biliser *yst~lu iol llse, l`or ins-tancc, ~ ic~ cream.
The way in which ice cream beha~e~ on ex~oPure to nor~al room temperature is important for -the consumer. I~ ~ product behaves too atypically, for instance if a product melts too rapidly or separates into a ~atty phase and a clear aqueous phase on melting, -then the product will be unacceptable. In the ice crea~ industry methods have been developed ior measuring suoh properties, ~or instanee melt-down &nd stand-up. These are described later.
It is known that such properties can be afieeted by the use of stab~lisers, o~ten called thickeners. A problem that arises is that the stabiliser~ deleteriously af~ect the feel of the ice cream in the mouth; a eloying, ~ummy or even greasy ~eel can oocur. ~his problem is acute in ice creams that require more than usual stabilisation.
~hat is desired is a stabiliser system that is good or at least adequate with respect to all aspects of stability. This is difficult to achieve ~or normaliee creams and particularly so for ice creams that require more than usual stabilisation.
A stabiliser ~ystem has now been ~ound that is surprisingly e~ective in stabilising ice cream without gi~ing an unacceptable mouth-feel. The stabiliser system i9 microcry~talline cellulose in combination with one or more o~ carboxymethylcellulose and galaoto~annan gums. PreYerably the stabiliser system oonslst~ o~ microcrystalline cellulose, ~ carboxymethylcelluloqe and a galactomannan gum. Example3 ; ~of galactomannan gums are guar gum, locu~t bean gu~ and tara gum.
The lnvention there~ore~ provides an ice cream ~ontaining a stabilising~amount o~ microcrystalline cellulose and one

2 - ~

cQ731 5~ ~3 ~
or luole of c~lrbo.~ymeltlylcclllllose and galactoman~tc~ uu~s, ~alactom~lltlan gum is p~fer.lbl~r pres~ntO
Tho nluoullt of microcrystalline cellulose IS preferably at least O.Olp, particularly preferably at least O.l~oO
Prefer~bly not more -than 0.8~o will be used~ particularly prcferably not more than Oo~S/7cost is a factor.
pre~erred range is 0015qo to 004~0, par-ticularly to 0,~/0.
The amolmt o~ total car~o~Yymethyl-cellulose (cQlclllated as sodium carbo~ymethylcellulose) and galactomannan gums is pre~erably not more than 15~, particu~arly preferably not more than 0.5~0 and preferably is in the range OolS~/O to 0035/0. The lo~er limit for carbo~Yyme-thylcellulose is preferably 0.01%9 The lolrer limit for galactomannQn gu~s, in particular for locust bean gum, is preferably O.OS,0~
Of course not more than one compone~t should be at or near their lol~er limits, in any one stabiliser sys-tem.
- The weight ratio of microcrystalli~e cellulose to total carboxymethylcellulose (calculated as sodium carbox~mæthyl-cellulose) alld galactomannan gume is pre~erably ill the~ -range 4:1 to 1:~1, particularly preferabl~ in the ranga 2:1 to 1:3. Total stabliser is preferably in the range 0015 to 1-0~ particularly pre~erably 0.30 to O~S~O The weight ratio of microcrystalline cellulose to carbo~Ymethylcellulose is preferably no-t more than 3:1 alld preferably is not less than 1:2.
As emphasised above the stabilisei- system is particularly use~ul in ice creams that require more than usual stabilisation A proble~l~with con~ention~l ice creams is tha-t at deep , Xree~e temperatures, eg. -20C, they cannot be served or eaten as readily as when they are a-t normal eating temp-eratures, eg. -10 C. The consumer canIIot treat them ~ven aRpro~imately ln -the normal ma~lner im~ediately whell t~l;en from the deep ~roeze. In some cases cnnventional icc crea~s ¢annot even be scooped out with a spoon at -20C, i.e. are ::
not, spoon~lble. Re~ormulatibn to ensure that such ~ o~731 ~S~73L~

properties, eg. spoonability at deep freeze temperatures are approxima-tely those expected at normal eating temp-eratures is comparatively simple. Methods are outlined later. The dif~iculty is that such reformulation leads to products that do not have acceptable properties, in particular stability, a$ normal eating temperature~. It has 3esmed impussible to get an ice cream that ha~ at both deep freeze and normal eating temperatures even approxi~ately the serving and eating properties conventionally e~pected at normal eating temperatures and that is sufficiently stable. The present inve~tion provides a stabiliser ~ystem with which this can be achieved.
It will be appreciated that the product-characteristics required ~or a conventional ice cream will depend on the personal tastes oi the consumer and ice creams are formulated to meet a variety o~ such ta~te~; the ~or~ulation o~ any one conventional ice cream will depend on the taste~ oi the consu~er~ concerned. In thi~ context a conventional ice cream is one prepared by a process involving ~reezing and hardening to tem-peratures in the order of -20C to -40C.
One important characteristic oi ice cream particularly in relation to scoopability is the log C, as de~ined later, o~ the ice crea~. In the UK ~or instance, an ioe crea~
can normally only be called a conventional ice cream i~
its log C at -20C aiter hardening i3 in the range 2.9 to

3.7; usually -the log C oi a UX non dairy ice crea~ at _20C ai~ter hardening will be in the range 3.3 to 3.7; ~or a dairy ice cream the range is 2.9 to 3.3. In other countries values for log C wlll be comparable but can be different, often higher, a~d indeed even within a country

- 4 -. , ~

~Q731 ~ 5~
various conventional ice creams will vary in -th~ir log C
values (A technique for measuring C~ the penetrometer value, and hence log C is described la-ter in the speciiication).
Whatever the conventional ice cream used, its properties at deep-freeze temperatures can be approximated to those expected at normal ea-ting templerature by adding ~reezing-poin-t depressants such as monosiaccharides and low molecular-weight alcohols, preierably polyalcohols and in particular glyoerol and sorbitol. It has been found that nor~ally sufiicient oi such freezing-point depressants should be added to the iormulation oi a co~ventional ice cream, eg.
at expense o~ wa$er, to lower the log C at -20C by between 0.25 and 1, pre~erably by 0.4 to 0.75. ~he notional replace-men$g eg. o~ ~reezing-point depressant ~or sugar~water, should be such that the product has the desired (by the consumer) sweetness as well as the desired log C, or spoonability, at -2QC.
As indicated above, a problem ~acing ice crea~ manu-facturers is that in general ice creams ~ormulated to have the conventional eating temperature properties at -20C, in particular to be spoonable at -20C, have unacceptably poor properties, eg. stand-up and meltdown, at nor~al eating temperatures~ For the ice cream to be spoonable at -20C
it has been found that its log C should preierably be less than 2.8, particularly pre~erably less tha~ 2.5; a corre-lation has bee~ iou~d to exist between spoo~ability and log C.
An e~pecially important aspect oi the present invention is an improvement in an ice cream whose log C (C being it5 penetrometer value) at -20C has been lowered by between 0.?5 and 1 by use o~ ~reezi~g-point depressants, the improve-~ent oonsisting o~ the use o~ the stabiliser system o~ ithe . . - . . . . . .

cQ731 ~I:)S~iL7~
present invention.
Microcrystalline cellulose is a well-known industrial product.
Its use at comparatively high levels in low calorie products, including low-calorie ice creams, is described in sritish patent specification 961~398. Processes for its preparation are well-known and are for instance described in US 3,157,518. One problem with microcrystalline cellulose is its dispersability.
Methods for overcoming this are well-known; a particular technique involves the use of carboxymethylcellulose. Microcrystalline cellulose is sold under the trade mark Avicel by FMC Corporation and a readily dispersible form containing sodium carboxymethyl-cellulose is sold as Avicel RC-591. It is stated to be a colloi-dal form of microcrystalline cellulose which has been blended with sodium carboxymethylcellulose and dried. The amount of sodium carboxymethylcellulose lS ~ , by weight of micro-crystalline cellulose. Microcrystalline cellulose is fully characterised in for instance GB 961,398 and US 3,157,518 but briefly can be stated to be cellulose crystallite aggregates with a level-off D.P. Level-off DP is the average level-off degree of polymerisation measured in accordance with the paper by O.A.
Battista entitled "Hydrolysis and Crystallisation of Cellulose"
Vol. 42 (1950), Industrial and Engineering Chemistry, pages 502 to 507. As stat d in GB 961j398 suitable microcrystalline celluloses have average level-off DP's in the range 125 to 375, particularly 200 ~o 300; the particle siæe of the aggregates of microcrystalline cellulose will usually be in the range 1 to 300 microns.
Galactomannan gums are well-known material and are ~ 6 --cQ7~'l ~CI 5~7~
- describecl for instance hy 1~l. Glick~tan in "Gum Techno'lo~y in the Food Industry", ~cademic Press, 1969. Pre~erred galactomannan gu~s ~or use in -the in~ention are locust bean gtllU and tara gum. Carboxymethylcelluloses are standa1tl S industrial products.
In this specification, including the claims, percen-tages are by weight and in particular are by l~eight o~ ice cream e~cept l~here the conte~t reqnires otherwiseO
Other than in the use of s~ icie~t ~reezing point ' \ ' ' :
- \ . ~
- \ , - , ' ' :~

\
: , \ ' , ~``

\
'' , .

': ' \ ~ :. ' \

cQ731 ~LC35~7~L

depressant for the preferred aspect of the invention and in the use o~ a thickening agent compri~ing particular components no especial insight is required in the formulation or processing o~ ice creams accord:ing to the invention. Details o~ conventional ~ormulations and processing conditions for ice cream can be ~ound in the usual trade publications and text books. Particularly useful in this respect is Arbuckle, "Ioe Cream", 1972 ~2nd Edition), AVI Publishing Corp., Westpoint, Conn.

The invention will now be illustrated further by the ~ollowing examples.
The properties of the stabili~er system are most surprising when compared with the properties of the separate componentsO This is illustrated in the examples but it will be appreciated that the stabiliser system is also useful in products other than ice cream.

.
An ice cream was prepared by conventional proces~ing techniques to the following ~ormulation:

In~redient Made~up skimmed ~ilk (32.5% solids) 27 Sucrose ~3 ~lucose syrup 2 Liquid oil blend 9.5 Monoglyceride emul~i~ier 0.45 Colour and flavour 0.03 ( Locust bean gum 0.15 Thick-ening Avicel RC 591~* 0.2 agents SQdium carboxymethyl cellulose* 0.15 Salt 0 05 Glycerol 3.0 Water to 100 *Supplied by ICI as powder B600 **trade mark ~ cQ731 ~35~

* Supplied by FMC and believed to contain by weight il%
sodium carboxymethyl cellulose.
The presence of the thickening agents can be detected analytically in such a product. The product itself is an excellent ice cream resembling conventional UK ice cream in eating properties but being spoonable at -20C.
E~AMPLE 2 An ice cream mix was prepared frDm the Yollowing ingredients, in parts by weight:

Palm oil 505 Stearic monoglyoeride 0.15 Spray-dried milk powder 10.0 Sucrose 14.0 Microcry~talline cellulose 0.4 (containing 11~ of sodiam carboxymeth~lcellulose) Sodium carboxymeth~loellulose* 0.2 Locust bean gum 0.22 Trisodium citrate 0.3 Water 64 The stearic monoglyceride was di~persed in the p~lm oil to give a fat pha~e. The ~ilk powder was di3persed in the water and to the dispersicn waæ added the remaining ingredients, giving an aqueous pha~e. ~he ~at and aqueous phase at 65 were mixed, homogenised at a pressure of 2000 psi and the emul~ion ~ormed was pasteurised at 70 ~or 20 minutes and cooled at 5 at which it ~ad pH 605. Aiter ageing for 2 hour~ at 5C 6 part~ of a concentrated orange ~uice, 0.04 parts of coloaring a~ent, and 3 parts o~ a 33~0 by weight aqueous solution of citric acid were mi~ed with the emulsion. The re~ulting emulsion o~ pH 3.5 w~s converted to ~ 5~
*Supplied by ICI as/powder B600 .
_ 9 _ :::

cQ731 " .~
~ L~S~t~
an ice cream by cooling and whipping at -4, and the ice cream was blast frozen to -20 and stored.
This example shows the use of the stabiliser system in stabilising an acid ice cream, a type of ice cream that requires more stabili~ation than an average ice cream.
The stabiliser system is particularly useful in an acid ice cream, i.e. an ice cream with a pH in the range 3.0 to 5.2. The pH should, as well as being within this range, preferably be sufficiently below the isoelectric point of any acid-precipi-table protein present in substantial amount for that protein tobe present substantially uncoagulated. Alternatively whey protein preferably purified by reverse 05mosis, can be used;
whey protein is not acid precipitable. Such ice creams axe described and claimed in our co-pending German OS 2361658, open for public inspection before June 4, 1975.

EXAMPLES 3~ to 14 and COMPARISONS A TO F
Ice cream mixes were prepared conventionally to the following formulation. Further details are given in Table 1 immediateIy before claims which also shows results obtained with ice cream prepared conventionally from the mixtures. A standard U.K. non-dairy ice cream differs from this formulation in containing no glycerol and 1.4~ by weight more sugar. 3%
glycerol is roughly equivalent in sweetness to 1.5% sugar.
~Spray dried milk powder 9.5 Sugar 13.5 Maltodextrin 40 DE* (Glucose syrup) 1.7 Palm oil ~ 9.5 Monoglycerdie from palm oil Q.5 Glycerol 3.Q

' ~ cQ731 ~S~
Salt 0~05 Flavour and colour O.1 S-tabiliser~ Table l**

~ater to 100 * DE = dextrose equivalent ** ~he SCMC used was powder B600 supplied by ICI ~imited.

The log C values at -20C o~ Examples 3 to 14 and Comparisons A to F were in the range 2.S and 2.9 and averaged 2~7. The log C of the standard ice cream mentioned above was in the ra~ge 3.2 to 3.3.

Test methods . _ .
Melt-Down Test and Shape Retention A rectangular block o~ ice cream o~ length 13.6 cm, height 400 cm and width about 8.8 cm which has been stored at -20C i9 placed on a wire gauze (10 ~ires per inch) in a~ atmosphere maintained at 15C. ~rrangements are made ior collection o~ the liquid drained fro~ the gauze. The time ~or the collection o~ the iirst 10 ml o~ liquid is noted. The volume o~ liquid collected i~ each subsequent 10 ~inute period is measured and the slope o~ the ~raph obtained by plotting volume collected against time i~ taken as the melt-down ~mls/hr). A~ter 4 hours thawi~g photo-graphs o~ the residue o~ the brick are take~, and the degree oi shape retention assessed as bad~ poor~ -fair, good or very good.
~tability to Temperature ~y~
This was carried out o~ a~ appro~i~ately cuboid ~ gallon block oi ice cream in a plastic container. A~ter storage in a deep-~reeze it ~as tra~s~erred to ambient (20C~ ~or 12 hours and then to~a re~ridgerator at -10~ Next day the block was subjected to ~urther temperature shock cycling by being taken , .

c~731 3~5~7~L~
out o~ the re~rigerator and le~t at ambient ~or 1 hour.
~his (each day 1 hour at ambient) was repeated to a total of six times and then the block was returned to the deep-~reeze for assessment the next day. ~he total test -took, allowing ~or a weekend, not more than ten days~ Product stability was assessed as ~ollows:
Bad : total breakdown Poor : ~20~o~ product converted to serum Fair : 5 - 20% of product converted to serum Good : C 5% 0~ product converted to serum C and Lo~ C
To determine C and hence log C the iollowing method is used:
Principle The hardness o~ ice cream is measure~ by allowing a standard cone to penetrate a sample ior 15 seconds using a cone penetrometer. ~he C-value can be calculated ~ro~ the penetration depth.
~ 7 - . : .~ ,. . .. .

cQ.73~

~53L~

Appara$us:
Ebonite cone With an apex angle of 40 +10 and the tip blunted by a few strokes on fine abrasive paper to give a flat 0-3 _0.03 mm in diameter. '~otal weight of cone and sliding penetrometer shaft 80 +0.3 g.; also additional weights of 8V +0.3g.
Penetrometer With a scale calibrated in 0.1 mm., and.
fitted with a lens. The penetrometer made by Sommer and Runge, Berlin, is reco~mended, parti¢ularly for static use. The Hutchinson instrumènt c~n al~o be used; it requires ~o electricity SUpply3 bu$ mu~t be modi~ied for sati~factory operation. The accuracy o~
pene~rometer timing mechanisms must be checked regularly.
The use of a x3 magni~ication lens o~ about 6 - 8 cm.
diameter fitted to the penetrome$er facilitates the - setting of the cone tip on the sample surface9 and an unfocused light limited to the equivalent of a l-watt 2Q bulb at a distance of about 5 cm. (to aYoid heating t~e sample surface) is also advantageous.
Temperature probe Reading to within 0,lC. The temperature probe should have a ste~ about 1 ~m. in dia~eter and about 4 cm.
long. Its accuracy should be checked regularly in baths of know~ temperaturas.
::

.

~ c~.73l 1~5~7~:~

Temper~n~ f~cili-les (a) Room controlled at required temperature ~ 1C;
(b) Gonstant_temperature cabinets, tolerance + 0.2~C.
The forced-draught constant~temperature cabinets supplied by Zero N~Vo Rotterdam are satisfactory.
Process:
Samplin~
Samples should be convenient size and p~efera~ly with smooth surface~ to incre~se accuracy.
iO Tempering 2 Days at whatever temperature is required e.g.
- -20C. Measure te~perature aocurately before penetration.
Mea~urement Where possible, penetrations are made in the temperature-controlled room, and should be completed - within two minutes o~ removing -the sample ~rom the constant-temperature cabinet~

1. Insert ~the temperature probe as near horizoDtally a~
~20 possible ~at a few ~m. below the sa~ple surface; read and ~: note the sa~ple temperature a~ter 30 ~econds. (Reject any : ~amples di~fering by ~ore than 0.5C fro~ the nominal te~t - . ~
~: ~ temperature~) -~ , : ~:
~ :

.- .. . . .. , . . . .. ~ ... .

~-~ c~7~1 ~L63 5~71~

2. Place the samples on the levelled penetrometer table~
3.~ Set the cone tip accurately on the sample surface~
using a lens and, i~ necessary, oblique lighting.
4. Releaise the arresting device and allow the cone to penetra-te the sample for 15 seconds.

5 Read and note the penetration depth.

6. Should the penetration depth be less than 72 x 0.1 mm.
(equivalent to a C_value of more than 500 g./cm.2) the measurement should be repeated with the cone weight increased by 80 g.

Further 80 g. weights may be added as necessary to ensure adequate penetration of the sample and the C-value scale reading corrected--accordingly.

7. Penetration measurements should not be made within 2 cm. of the sample edge nor within 2.5 cm. o~ each othcr.
Determinations in which air bubbleis~ cracks, etc. inter~ere should be rejected. -Calculation of C_va~lues 20 ~ ~he C-value can be calcalated~frbm the penetration depth using the formu~a: -~ x F
C~ 6 where C = Yield value or C_value (g.~cm.2) P =~To~tal welght of cone and sliding stem (gO~ -; P = Penetration depth (0.1 mm.) `~ ~ = Factor dependlng on cone angle:

.: . . .

, . . , . , -. i -',~: . . ,, ' ~ ' ' ' ~5~7~ cQ731 Cone an~le K valwe 6~ 281 1~40 * Depending on the likely so~tne~s o~ the product, the cone weight should be adjusted, e.g.
at -10C use 80 gm at -15C use 1~0 gm at -20C ase 240 gm i.e. it depend~ on temper~ture o~ measure~ent.
C valaes will usually be taken a~ter hardening conventionally, a9 for inst,ance described on page 4, lines 18 to 209 and in the standard text-books~ -It should be noted that an ice cream according to ~he invention preierably has a mslt-down, determined as described above, o~ less thMn 25 ~lJhr and particularly pre~erably o~
between S and 20 ml/hr..
It should ~urther be noted that the log C at -20C oi an ice cream according to the invention should pre~erably not be less than 2,3.
Further details of suitable ~ircrocrystalline celluloses a~e available i~ pa~phlets ob$ainable iro~ FMC Corporation, A~icel Department~ Marcus Hook, Pe~nsyl~a~ia 19061 ~or i~stance 1n Bulletin ~C-16 and pamphlets RC-30 and ~C-34. RC-30 describes use o~ microcrystalline cellulose i~ ~ro~en desserts and states, inter alia9 that i'It is co~patible with all stabil-islng systems except:those containing Guary Loc~st Bean, and Na Alginate", ':
16 _ ~[35~7~ .,t', ~ 1--* ~ 1 0 ~
O O~ I~ 1~ ~ S IJ ~D
q ~ 3 ~
P'. E ~
~DX~ noooooooooooo O
g ~O ~ , g l_ ~ ~ O l_ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~n r~
~ (D ~ o\O D (D
--~ 3 1~ ~
- g ~ ~ *
O O O O o O O O I-- o O O
_ Q ~ ~1 3 ~D 3 Q ~ ~
n m j- i- ~ 3 It lq - n u~
O D~
~D ~ O O O O O O O O O O O O O U~
O Hl O ~ i- ~ i~ O ~ i' ~ i ~ i ~ O C
P' rt It ; i"

O ~ ~ . O
n ~ o o ~ O ~ O ~ ~ ; ~P
~ ~ ~O;
O ~ ~D
j_ X ~ V ~ V V V V V V V
O O O O O O O ~0 ~ O ~ O O O O ~rl O O' !~
' --3

8 ~
o ~ ~ w ~ w w ~ ~ O ~I ~ ~ ~ w w w O a~
~4 ~ n ~o o ~ ~ p~

H ~ ~ ~ ~ ~ r~ ~ O

n ~ rD n n n n ,~ ,q ~ ~n ,~ ~ n ~ n n ~, n ~r n n ~, ~a ~ ~a q Iq ~
n 3 ~ ~ ~ D n ~ n ~a (D i-- i-- i' (D (D lq

Claims (41)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY
OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. An ice cream containing a stabilising system characterised in that the stabiliser system comprises micro-crystalline cellu-lose and one or more of carboxymethyl cellulose and galactomannan gums, when no galactomannan gum is present, the weight ratio of micro-crystalline cellulose to carboxymethyl cellulose being not more than 3:1.

2. An ice cream as claimed in Claim 1 characterised in that the amount of micro-crystalline cellulose is at least 0.01% weight of the ice cream.

3. An ice cream as claimed in Claim 2 characterised in that the amount of micro crystalline cellulose is at least 0.1% by weight.

4. An ice cream as claimed in Claim 3 characterised in that the amount of micro-crystalline cellulose is not more than 0.8%
by weight.

5. An ice cream as claimed in Claim 4 characterised in that the amount of micro-crystalline cellulose is not more than 0.4%
by weight.

6. An ice cream as claimed in Claim 5 characterised in that the amount of micro-crystalline cellulose is in the range 0.15%
to 0.4% by weight.

7. An ice cream as claimed in Claim 6 characterised in that the range is 0.15% to 0.25%.

8. An ice cream as claimed in Claim 1 characterised in that the amount of total carboxymethyl cellulose (calculated as sodium carboxymethyl cellulose) and galactomannan gums is not more than 1% by weight of the ice cream.

9. An ice cream as claimed in Claim 8 characterised in that the amount of total carboxymethyl cellulose and galactomannan gums is not more than 0.5%.

10. An ice cream as claimed in Claim 9 characterised in that the amount of total carboxymethyl cellulose and galactomannan gums is in the range of 0.15% to 0.35%.

11. An ice cream as claimed in Claim 1 characterised in that the amount of galactomannan gums is not less than 0.01%.

12. An ice cream as claimed in Claim 1 characterised in that the amount of galactomannan gums is not less than 0.05%.

13. An ice cream as claimed in Claim 1 characterised in that the weight ratio of micro-crystalline cellulose to total carboxymethyl cellulose (calculated as sodium carboxymethyl cellulose) and galactomannan gums is in the range 4:1 to 1:4.

14. An ice cream as claimed in Claim 13 characterised in that the range is 2:1 to 1:3.

15. An ice cream as claimed in Claim 1 characterised in that the amount of the stabiliser system is in the range 0.15 to 1.0%.

16. An ice cream as claimed in Claim 15 characterised in that the range is 0.3 to 0.5%.

17. An ice cream as claimed in Claim 1 characterised in that the weight ratio of micro crystalline cellulose to carboxymethyl cellulose is not more than 3:1.

18. An ice cream as claimed in Claim 17 characterised in that the weight ratio is in the range 3:1 to 1:2.

19. An ice cream as claimed in Claim 1 characterised in that it has a log C as herein defined of less than 2.8 at -20° C.

20. An ice cream as claimed in Claim 19 characterised in that the log C is less than 2.5 at -20° C.

21. An ice cream as claimed in Claim 1 characterised in that it has a log C at -20°C as herein defined of not below 2.3.

22. An ice cream as claimed in Claim 1 characterised in that it is notionally obtainable by replacing water and sugar in the formulation of a conventional ice cream by a freezing-point depressant.

23. An ice cream as claimed in Claim 22 characterised in that it is notionally obtainable by use of sufficient freezing-point depressant to depress the log C at -20° C by between 0.25 and 1.

24. An ice cream as claimed in Claim 23 characterised in that the log C at -20° C is depressed by 0.4 and 0.75.

25. An ice cream as claimed in claim 22 characterised in that the notional conventional ice cream has a log C at -20° C in the range 2.9 to 3.7.

26. An ice cream as claimed in Claim 25 characterised in that the log C is in the range 3.3 to 3.7.

27. An ice cream as claimed in Claim 25 characterised in that the range is 2.9 to 3.3.

28. An ice cream as claimed in Claim 1 characterised in that it contains glycerol or sorbitol as the freezing-point depressant.

290 An ice cream as claimed in Claim 1 characterised in that it has a melt-down as herein defined of less than 25 ml/hr at 15° C.

30. An ice cream as claimed in Claim 29 characterised in that the melt-down is between 5 and 20 ml/hr at 15° C.

31. An ice cream as claimed in Claim 1 characterised in that the stabiliser system contains a galactomannan gum.

32. An ice cream as claimed in Claim 31 characterised in that the galactomannan gum is locust bean gum or tara gum.

33. An ice cream as claimed in any one of Claims 31-32 characterised in that the stabiliser system contains carboxy-methyl cellulose.

34. An ice cream claimed in Claim 1 characterised in that the ice cream has a PH in the range 3.0 to 5.2 and sufficiently below the isoelectric point of any acid precipitatable protein present in substantial amount that the protein is substantially uncoagulated.

35. An ice cream claimed in Claim 34 characterised in that the protein is whey protein.

36. An ice cream as claimed in Claim 35 characterised in that the whey protein has been purified by reverse osmosis.

37. A stabiliser system containing micro-crystalline cellulose and carboxymethyl cellulose characterised in that it contains a galactomannan gum.

38. A stabiliser system as claimed in Claim 37 characterised in that the galactomannan gum is locust bean gum.

39. A stabiliser system as claimed in Claim 37 characterised in that the galactomannan gum is tara gum.

40. A stabiliser system as claimed in Claim 37 characterised in that the weight ratio of micro-crystalline cellulose to total carboxymethyl cellulose and galactomannan gum is in the range 4:1 to 1:4.

41. A stabiliser system as claimed in Claim 40 characterised in that the range is 2:1 to 1:3.