Rapid freezing for storage of sheep milk

Authors: Prof Richard Archer, Prof Mohammed Farid, Jolin Morel, Dr Georg Ripberger, FIET

A pdf of the full article, with Figures, is here


It has been said that New Zealand’s prosperity was “built off the

sheep’s back”, but in recent years Bovine, ie cow Dairy has become

New Zealand’s largest agricultural earner.

Traditionally, the sheep industry in New Zealand has focused on the

production of meat and wool. In recent years however, a small but

rapidly growing sheep dairy industry has developed. The sheep dairy

industry produces a high value product, with a wide range of possible


In order to support the growth of this industry, a small team of

researchers from Massey University (MU), The University of Auckland

(UA) and GNS Science (GNS) are working to develop a rapid freezer

suitable for use on sheep farms, to allow raw sheep milk to be

stored for long periods without a loss of product quality. This has

the advantages that milk can be collected from remote locations

(previously deemed uneconomic), and allows small-scale farmers to

sell their milk in batches, as often required by cheese makers. This

gives farmers greater market power and allows them to establish a

profitable sheep-dairy business almost regardless of location.

This work is part of the larger MBIE-funded Food Industry Enabling

Technologies (FIET) research programme, which aims to develop

technologies that will grow export earnings by $250M pa by 2026.

Scope of work

This project is aimed at developing a rapid milk freezer, suitable for

on-farm applications, and taking it to the on-farm prototype stage. This

involves understanding the effects of freezing on milk quality (with a

focus on raw sheep milk in a New Zealand context) and using this

understanding when designing and optimising the on-farm freezer.

This is the focus of the MU and UA team while GNS is developing low

ice adhesion surfaces that will assist operation of the freezer unit and

reduce operational costs.

Sheep milk production

Globally, sheep milk accounts for about 1.3% of total milk production.

Sheep produce much less milk each day than cattle. In New Zealand,

dairy sheep produce approximately 1.5-2L of milk per day, with a total

yearly production between 150-300L of milk. This is significantly lower

than what is achieved in countries with more established industries

such as Israel (up to 750L per season) or Germany (450-550L per

season). The most popular dairy sheep breeds are the East Friesian,

Awassi, and Lacaune. These production systems are much more

intensive than those encountered in New Zealand..

The equipment used in milking, and the design of milking parlours, are

adapted from those used in goat dairy and are widely available.

Sheep milk nutrition

Sheep milk is higher in total solids than both cow and goat milks. This

difference is due mostly to higher levels of fat and proteins in sheep

milk. The lactose level is similar to those in cow and goat milks, and

the ash level is slightly higher. The higher solids content means that

cheese yields are higher with sheep milk.


Sheep milk’s fat is higher in valuable medium-chain triglycerides, and

may be more easily digested than the fat in cow milk due to the lower

diameter of sheep milk fat globules.

Sheep milk is claimed to have a superior amino acid profile when

compared with cow milk. Difference in the proteins between sheep

and cow milk may mean that sheep milk is less allergenic; however

this is still the subject of ongoing research.

Sheep milk products

Traditionally, sheep milk has been used to make cheeses, and this is

still its main use. Sheep milk cheeses include Feta, Pecorino, Manchego

and Roquefort. In New Zealand there are around 18 varieties of sheep

milk cheese being produced by producers ranging from artisan

cheesemakers to large commercial enterprises. Other food products

made from sheep milk in New Zealand include yogurts, ice-creams,

ready to drink products, fermented drinks such as kefir and, of course,

fresh milk .

Sheep milk is also used in the manufacture of whole milk powders

and infant formulas. The different allergenic profile of sheep milk is

helpful to some people who are sensitive to cow milk. Non-traditional

dairy products such as calcium chews, soaps and cosmetics have also

been developed.

Sheep milk products are generally aimed at the premium end of the

market, and towards high-value export markets. As a consequence,

the farm-gate price for sheep milk is significantly higher than cow milk,

on a milk-solids basis.

Sheep dairying in New Zealand

The New Zealand Sheep Dairy industry consists of approximately 16

producers as of 2017, with an annual growth of around 5 farms and

3000 ewes. These producers range in size from artisan cheesemakers

with 40 ewes, to a large, vertically integrated, operation that milks

20,000 ewes and produces milk powder, infant formula, and a range

of cheeses.

As can be seen in the figure on the facing page, the sheep dairy

industry is spread throughout the country. The geographic dispersion

of producers, the seasonal nature of milking, small volumes of daily

production, long distances to processors and intermittent demand

for large volumes of milk (to take advantage of economies of scale in

processing) suggest that a method for long term storage of fresh milk

would be beneficial to the industry.

Freezing— a method for long-term storage

Milk has been frozen on farms previously to overcome the issues of

seasonality, quantity, and market access. A common practice involves

freezing the milk in 2 L bladders, which are then stored and transported

for domestic use or exported to international markets (e.g. yoghurt

and cheese manufacturers). This method however has its drawbacks.

Frozen storage of milk can lead to a decrease in milk quality by several


• The proteins in milk that has been frozen and then thawed can

agglomerate and precipitate, leading to lower product yields and

undesirable textures

• The milk fat globules can be damaged during the freezing

process, leading to oiling-off, oxidation of fats, and off-flavours.


Consequently, the thawed milk becomes unsuitable for uses such as

liquid milk, UHT or drying as powder or formula. Previous research,

and current trials at MU demonstrate that the quality of freeze-thawed

milk is dependent on a number of factors. The most important are the

speed of freezing, the final storage temperature, and the duration of

storage. Broadly speaking, the best quality is achieved when freezing is

conducted as fast as possible, the frozen milk is kept as cold as possible

(below -20ÅãC), and the storage time is minimised.

Figure 4 illustrates the loss in protein stability that occurs during frozen

storage at higher temperatures. Milk samples were stored at -10ÅãC and

-30ÅãC for 9 weeks. After thawing at 20ÅãC, the samples were centrifuged

(3000 g x 60 min), and the amount of precipitate measured. It can

clearly be seen that the proteins in the -10ÅãC samples have precipitated,

whereas they are still in the liquid phase in the -30ÅãC samples.

The current practice of placing bladders or buckets of milk in blast

freezers or refrigerated shipping containers counts as slow freezing

(freezing times on the order of several hours), and is subject to

handling issues (e.g. lots of manhandling and labour required).

Freezing is regarded as processing from a regulatory point of view. Thus,

the unit operation of freezing has to take place in a food processing

environment, and the normal on-farm dairy Risk Management

Programme (RMP) must be extended to embrace processing and the

premises be appropriately registered.

In this project a rapid freezer has been developed that mitigates the

deleterious effects of traditional freezing methods, reduces labour

costs, complies with the regulatory requirements, and enables ease of

further processing.

Rapid freezing

Previous research suggests that the major causes of protein

destabilisation are the physical aggregation of casein micelles after

rejection from growing ice crystals, and the increased concentration of

salts in the unfrozen phase. The damage to fats during freezing occurs

due to physical aggregation of fat globules after similar rejection from

the solid phase.

Rapid freezing, followed by storage at temperatures below -20ÅãC should

counteract these mechanisms. Work reported on freeze concentrators

and falling film crystallisers shows that increasing ice growth rates

increases the partition coefficient of the system, reducing the amount

of solid rejected from the growing solid phase. The amount of a

food product that remains unfrozen is also reduced at lower storage

temperatures. This increases the stability of proteins during storage as

it reduces their aggregation, reduces the amount of unfrozen phase in

which destabilisation can occur, decreases the rates of any reactions

occurring, and increases the viscosity of the remaining unfrozen phase,

thereby further decreasing reaction rates. The damage to fats is also

reduced, as the fat globules are entrapped by the growing ice front

rather than being rejected.

Figure 5 demonstrates the differences between slow and fast freezing.

Sheep milk was sandwiched between 2 sheets of glass in a Hele-Shaw

cell, which was then placed in a Bridgman furnace, and the ice/milk

interface was observed using transmission light microscopy. At low ice

growth rates (0.5 μm/s) the interface is planar, and milk fat globules

are rejected from the growing ice front, leading to an increased

concentration in the unfrozen phase. At high freezing front velocities

(40 μm/s), the growth is columnar with secondary branches growing

at an angle to the main growth direction. At this speed fat globules are

trapped within the advancing ice, and there is no concentration in the

unfrozen phase.

The rapid freezer being developed operates by running a thin film of

liquid milk over a cooled curved surface, onto which the ice freezes.

Once the frozen milk layer has reached the desired thickness, the liquid

flow is stopped and the adhered ice removed. The sheet of ice is then

broken into flakes by an auger and transported to a storage vessel. This

geometry is commonly used for flake ice for fisheries but lends itself

to hygienic design.

In the system adopted the quality benefits of rapid freezing are

combined with ease of handling to reduce labour requirements

and allow for easier regulatory compliance. The thin flakes simplify

subsequent thawing and give a product suitable for freeze-drying,

should a processor desire this.

The technology may also enable processors to make raw milk cheeses,

even if they are not located near sheep milk producers, as milk can be

stored for extended times without thermal treatment.

Low-ice-adhesion surfaces

The key focus of the work being undertaken by GNS is the development

of icephobic surfaces, i.e. surfaces with a low-ice-adhesion. The

surface developed will be both food grade, and able to be applied

economically to large surfaces. Readily available 304 stainless steel is

the base substrate.

An icephobic surface will make it significantly easier to remove frozen

milk from the heat transfer surfaces, with a minimum of heat or force.

This will increase the thermal and mechanical efficiencies of the

system, decrease cycle time and lower ongoing freezing costs.

Other applications

The hygienic flake freezing method and equipment developed can

also be used to preserve in their native state without loss of functional

value, other valuable, perishable liquid raw materials like blood

plasma, whey and colostrum, that arise at times and places remote

from existing central processing facilities. This allows the launch of

new industries, which can secure and harvest their feedstock across

New Zealand. It will also enable the development of new products like

functional beverages (probiotics, ready-to-drink beverages) from niche

raw materials like donkey and red deer milk, which have been shown

to have bioactives.

Current progress and future plans

We have designed, built and instrumented a 50 L/hr pilot-scale unit that

is currently commissioned. Our initial trials are on sheep milk but we

would like to test other products as identified by businesses. Our next

step in the development involves constructing and trialling an on-farm

prototype of perhaps 250 L/hr capacity by mid-2018. We anticipate the

on-farm unit might look like a refrigerated container with freezer, plant,

controls and storage area inside, parked alongside the shed and fed by

a pipe and a power cord. The detailed design is currently being carried

out and we are working closely with MPI to comply with national and

international regulations. And we are currently considering partners

who might manufacture and market the units.

A pdf of the complete article, with figures, is here