Using Intermittent Solar Ammonia Absorption for Food Preservation

1
Development of the Idea of using the Intermittent Solar
Ammonia Absorption Cycle (ISAAC) Refrigerator,
Insulated Containers and Cold Storage for Preservation
of Food in Developing Countries.
The fisheries sector plays an important role in many developing countries, contributing towards
food security, generation of employment and procurement of foreign exchange. It is becoming
increasingly clear that many fish resources are being subjected to exploitation at or above their
capacity to remain sustainable. At the same time, there is a very large wastage of fishery
resources as a result of discarding unwanted catches at sea, and failure, particularly in smaller
vessels, to preserve effectively those fish that are marketable. Under these circumstances, the
need for improved handling, care and preservation of the catch is clear.
All species of fish, when properly chilled, will stay fresh for longer periods than those that are not
preserved in any way. The use of chilling techniques such as ice, therefore, effectively prolongs
the length of time available for fishing trips and makes it possible to increase the catch with
economic benefits for the vessel and crew. Products brought to market in a well-preserved
condition will generally command higher prices, both at wholesale and retail levels, and thus give
better returns to the fishing operation.
Given the above, it might be assumed that all types and sizes of fishing vessels would benefit
from the use of ice for catch preservation. However, in practice there are limitations. On the
smallest types of vessels, such as small rafts and the smallest dugout canoes, there is no space
to keep ice until it is needed. However, this may not be a problem as the fishery undertaken by
these very small craft usually only lasts a few hours and fish is consumed or sold on a daily basis.
In some of these very small fishing craft, owners are aware of the problems of catch deterioration
and often use wet sacking or palm leaves to cover the catch, lower the temperature and so
reduce spoilage.
1. Chilling versus freezing of fish
There are many factors to be taken into account when considering the differences between
chilling and freezing of fish products for various markets. Both chilling and freezing operations
can produce stable products and the choice of one or the other depends on many factors.
Table 1.1 lists some of the advantages and disadvantages of the two methods. It can be used to
help decide whether freezing or chilling is the option most appropriate to a particular situation
The preservative effects of chilling fish
The use of temperature reduction as a means of preserving fish and fishery products is very
important worldwide both for local and export markets.
Chilling is the process of cooling fish or fish products to a temperature approaching that of melting
ice.
The purpose of chilling is to prolong the shelf-life of fish, which it does by slowing the action of
enzymes and bacteria, and the chemical and physical processes that can affect quality. Fresh
fish is an extremely perishable food and deteriorates very rapidly at normal temperatures.

2
Reducing the temperature at which the fish is kept lowers the rate of deterioration. During chilling
the temperature is reduced to that of melting ice, 0 °C/32 °F.
TABLE 1.1
Advantages and disadvantages of chilling and freezing
Chilling Freezing
Short-term storage (up to one month
maximum for some species, only a few days
for others)
Long-term storage (a year or
more for some species)
Storage temperature 0 °C Storage temperature well
below zero, e.g. -30 °C
Relatively cheap Relatively costly
Product resembles fresh fish If poorly done can badly
affect quality
Relatively low-tech Relatively high tech
Low skills required High skills required
Portable refrigeration Generally static operations
The most common means of chilling is by the use of ice. Other means are chilled water, ice
slurries (of both seawater and freshwater), and refrigerated seawater (RSW). For the full benefits
of chilling to be realized, it is essential to maintain chill temperatures throughout the different fish-
handling operations.
Although ice can preserve fish for some time, it is still a relatively short-term means of
preservation when compared to freezing, canning, salting or drying, for instance. When used
properly it can keep fish fresh so that it is attractive in the market place.
The use of ice for preserving fish and fishery products has proved to be an effective handling
method on board fishing vessels for the following reasons:
 Ice is available in many fishing areas or ports.
 Purchasing patterns can be varied according to need (e.g. block ice of different sizes is
frequently manufactured, and crushed, small or fragmentary ice ready for use is sold by
weight).
 Ice has a very high cooling capacity.
 Ice is harmless, and in general relatively cheap.
 Ice can maintain a very definite temperature.
 Ice can keep fish moist and as it melts it can wash surface bacteria from the fish.
 Ice can be moved from place to place and its refrigeration effect can be taken to
wherever it is needed.
 Ice can be made on shore and used at sea.
However, packing fish in ice on board small fishing vessels, whether in boxes, shelves or pounds,
is a labour-intensive task and other methods have been introduced to reduce the time and labour
required. Among these, the most widely used are RSW and CSW. RSW is labour saving and an
acceptable chilling method, but requires onboard mechanical refrigeration, pumping and filtering
systems. It is also a relatively costly system. In CSW systems, sufficient ice is carried on a fishing
voyage and mixed first with seawater before fish are added to the ice and water slurry.

3
Both these systems offer the advantages of quick chilling, reduced physical damage to the fish
and quicker handling with less labour. However, they require more specialized installations on
board and have usually only been found suitable where large volumes of fish need to be handled
in a short time period, for instance when handling small pelagics on board purse seine vessels.
A typical comparison of temperature profiles for a medium-size round fish chilled in crushed ice,
RSW and ice slurry is shown in Figure 1.1. According to these data, the fastest and most efficient
chilling medium is ice slurry followed by RSW. The ice chilling rate is the lowest due to reduced
contact of ice with the fish (an air layer surrounding the fish was created during ice meltage). To
ensure maximum contact of ice with the fish, proper selection of the size of ice particles and good
stowage practices are needed. The rate of chilling is governed by:
 the size, shape and thickness of fish;
 the method of stowage;
 adequate mixing of ice, water and fish (in ice slurries);
 adequate contact of ice with the fish;
 the size of the ice particles.
Factors affecting the rate of spoilage in fish
The main factors that affect the rate of spoilage in chilled fish are:
 temperature
 physical damage
 intrinsic factors
It is well known that high temperatures increase the rate of fish spoilage and low temperatures
slow it down. Therefore, if the temperature of fresh fish is low, then quality is lost slowly. The
faster a lower temperature is attained during fish chilling, the more effectively the spoilage activity
is inhibited. Generally, the rate at which fish loses quality when stored in ice (0 °C) is used as the
baseline when comparisons are made regarding shelf-life at different storage temperatures. The
relationship between the shelf-life of fish at 0 °C and at t °C is known as the relative rate of
spoilage at t °C (RRS) and is defined below:
Relative rate of spoilage at t °C = keeping time at 0 °C
keeping time at t °C
Further information on spoilage rates can be found in FAO Fisheries Technical Paper No. 348,
Quality and quality changes in fresh fish (FAO, 1995a).
Physical damage
Fish is soft and easily damaged, therefore rough handling and bruising result in contamination of
fish flesh with bacteria and allow releases of enzymes, speeding up the rate of spoilage. In
addition, careless handling can burst the guts and spread the contents into the fish flesh.
Shelf-life of fish in ice
Chilling of fish can slow down the spoilage process, but it cannot stop it. Therefore, it is a race
against time and fish should be moved as quickly as possible.

4
The main question for fishermen, traders and consumers is how long fish will keep in ice. As
discussed previously, shelf-life will depend on several factors. However, the fish spoilage pattern
is similar for all species, with four phases of spoilage as outlined in Table 1.3.
There have been many research studies regarding the shelf-life of fish stored in ice. Based on
these studies, it is generally accepted that some tropical fish species can keep for longer periods
in comparison to fish from temperate or colder waters. This can be attributed to differences in the
bacterial growth rates, with a 1-2 week slow growth phase (or period of adaptation to chilled
temperatures) in tropical fish stored in ice. However, due to differences in the criteria used to
define the limit of shelf-life, and methodologies used, comparison between shelf-life of fish from
tropical and temperate waters is still difficult. Tables 1.4 and 1.5 show shelf-life of several fish
species stored in ice and RSW.
TABLE 1.2
Intrinsic factors affecting the spoilage rate of chilled fish
Intrinsic factors Relative spoilage rate of fish stored in ice
Slow rate Fast rate
Shape Flat fish Round fish
Size Large fish Small fish
Fat content in the flesh Lean species Fatty species
Skin characteristics Thick skin Thin skin
Source: FAO, 1995a.
TABLE 1.3
The four phases of fish spoilage
Phase I
(Autolytic changes,
caused mainly by
enzymes)
Fish just caught is very fresh and has a sweet,
seaweedy and delicate taste. There is very little
deterioration, with slight loss of the characteristic
odour and flavour. In some tropical species this period
can last for about 1 to 2 days or more after catching.
Phase II
(Autolytic changes,
caused mainly by
enzymes)
There is a significant loss of the natural flavour and
odour of fish. The flesh becomes neutral but has no
off-flavours, the texture is still pleasant.
Phase III
(Bacteriological
changes, caused
mainly by bacteria)
The fish begins to show signs of spoilage. There are
strong off-flavours and stale to unpleasant smells.
Texture changes are significant, flesh becoming either
soft and watery or tough and dry.
Phase IV
(Bacteriological
changes, caused
mainly by bacteria)
Fish is spoiled and putrid, becoming inedible.

5
TABLE 1.4 Shelf-life of some marine and freshwater fish species stored in ice
Species Shelf-life (days in ice) Remarks
Temperate
waters
Tropical
waters
Marine species 2-24 6-35 Shelf-life for tropical fish tends to be
longer.
Cod, Haddock 9-15 White-fleshed lean
Whiting 7-9 White-fleshed lean
Hake 7-15 White-fleshed lean
Bream 10-31 Lean/low fat
Croaker 8-22 Lean
Snapper 10-28 Lean
Grouper 6-28 Lean
Catfish 16-19 Lean
Pandora 8-21 Lean
Jobfish 16-35 Lean
Spadefish 21-26 Lean/low fat
Batfish 21-24 Lean
Sole, Plaice 7-21 21 Flat fish
Flounder 7-18 Flat fish
Halibut 21-24 Flat fish
Mackerel
1
4-19 14-18 Pelagic fish; high/low fat
Summer herring 2-6 Pelagic fish; high fat
Winter herring 7-12 Pelagic fish; low fat
Sardine 3-8 9-16 Pelagic fish; high fat
Freshwater
species
9-17 6-40 Shelf-life for tropical fish tends to be
longer.
Catfish 12-13 15-27 Lean
Trout 9-11 16-24 Low fat
Perch 8-17 13-32 Lean/low fat
Tilapia 10-27 Lean
Mullet 12-26 Lean
Carp 16-21 Lean/low fat
Lungfish 11-25 Lean/low fat
Shad 25 Medium fat
Corvina 30 Medium fat
Pacu 40 Fatty
Bagre (type of
catfish)
25 Medium fat
Chincuna 40 Fatty
1
Fat content and shelf-life are subject to seasonal variations.
Source: FAO, 1995a.

6
TABLE 1.5
Comparison of shelf-life of various fish species stored in ice, RSW and RSW with added
CO2
Species Shelf-life (days in the chilling medium) Storage
temperature in
RSW (°C)
Ice (0 °C) RSW RSW + CO2
Pacific cod 6-9 - 9-12 -1.1
Pink shrimp - 4-5 6 -1.1
Herring - 8-8.5 10 -1.0
Walleye pollock 6-8 4-6 6-8 -1.0
Rockfish - 7-10 >17 -0.6
Chum salmon - 7-11 >18 -0.6
Silver hake 4-5 4-5 >5 0 to 1
Capelin 6 2 2 +0.2 to -1.5
Source: FAO, 1995a.
2. Commercially manufactured insulated containers
There is a wide range of insulated containers available, offering a variety of features according to
different requirements of handling, size, insulation efficiency, modes of transportation, sturdiness,
durability and construction materials. However, these insulated containers are generally an
imported item in developing countries, which means that they are costly, especially when
compared with non-insulated boxes and locally made traditional fish containers.
Insulated container capacity or physical size has undoubtedly also been a limiting factor in
introducing these units to some markets in developing countries’ fisheries. Many such fisheries
lack the infrastructure and equipment, such as cranes and fork-lift trucks, for the physical
handling of large tubs when full of ice and fish. Most fish-landing sites at artisanal level still rely on
manual handling, which effectively places a limit on the size of items that can be used.
TABLE 2.1
Characteristics of some materials used in the manufacture of fish containers1
Type of material Density Thermal
conductivity
Material strength
Tensile
strength
Bending
strength
(kg/m3
) (W m-1
h-1
°C-1
) (kg/mm2
)
Wood (soft) 350-740 0.11-0.16 5-8 8-12
Wood (hard) 370-
1,100
0.11-0.255 8-14 10-15
Plywood 2
530 0.14 3.5-9.3 variable
Aluminium alloy 2 740 221 20-30 30-40
Mild steel 7 800 45.3 24-43 20-28
Fibreglass-reinforced plastic 64-144 0.036 20-50 30-100

7
High-density polyethylene 960 0.5 5-10 13-15
Eel grass between strong
paper (not compressed)
73.6 0.036 - -
Eel grass in burlap (not
compressed)
215 0.049 - -
Jute fibre (not compressed) 107 0.036 - -
Polyethylene sheet plus 2
layers of gunny (jute) fabric
plus polyethylene sheet (not
compressed)
500 0.054 - -
Polyethylene sheet plus 4
layers of gunny (jute) fabric
plus polyethylene sheet (not
compressed)
580 0.046 - -
Cane fibre insulation board
(not compressed)
216 0.048
1
Figures given must be taken only as indicative values due to the great variation within materials.
2
Tensile strength of three-ply wood parallel to grain of faces; under flexure plywood boards have
about 85 percent of the bending strength of a solid beam of the same wood and same size.
Common materials used for the manufacture of insulated containers are FRP and high-density
polyethylene (HDPE) often with plastic foams for insulation.
One of the most common types of insulated container used in fisheries consists of double-walled
HDPE with expanded polystyrene or polyurethane foam as insulation. These are usually
constructed in a single piece using a rotational moulding process. The HDPE walls vary in
thickness from 3 mm to 6 mm and the total thickness varies according to the size and capacity of
the insulated container. These types of containers are able to withstand relatively rough handling,
and are considered to be superior to those manufactured with other materials, such as FRP,
which tends to be more brittle and prone to impact damage and fractures.
TABLE 2.3
Examples of locally made insulated containers for small fishing vessels
Geographic
area
Characteristics of container Type of fishing
vessel
References
Asia (India) Capacity: 175-200 kg of ice/fish
mixture
Weight: 40 kg
Design: FRP and polyurethane foam
(70 mm thick). With 10 mm diameter
drainage
Technical and financial viability
studies were conducted, based on
field trials with eight vessels
Traditional 5-12 m
long wooden fishing
vessel known as
“navas”
FAO (1991)

8
Africa
(Senegal)
Capacity: 0.9-1.3 tonnes of fish
Weight: 200 kg (with two hatches)
and 270 kg (with 3 hatches)
Design: FRP and polyurethane foam
(100 mm thickness). With PVC pipe
drainage. Tailor-made to fit local
pirogues
Technical and financial viability
studies were conducted, based on
field trials with several vessels
(Evaluation Report, Phase I, Project:
“Amélioration de la conservation de
poisson à bord des pirogues”,
CAPAS, Senegal, 1983 and FAO
project GCP/INT/398/NOR)
Later on, based on this design,
several experiences on the
introduction of insulated containers
for small-scale fishing vessels were
undertaken in the United Republic of
Tanzania, Kenya, the Gambia,
Guinea-Bissau and Guinea.
Traditional 14-18 m
long wooden
pirogues with
outboard engines
FAO (1985)
Asia
(Indonesia)
Nominal volume: 5.15 m3
.
Approximate external dimensions: L
= 4.96 m × W = 1.33 m × H = 0.78 m
Design: CSW insulated container
constructed of wood board,
expanded polystyrene foam, FRP
lining on the inner surface. Tailor-
made to fit local purse seiner.
Technical viability studies were
conducted by the Research Institute
for Fish Technology in Jakarta, based
on field trials with one prototype
vessel. In 1986, three years after the
introduction of the CSW container in
the Bali Strait area, about 78 purse
seiners were equipped with CSW
containers.
Traditional 12.2 m
long purse seiner
(main species
caught: Sardinella
longiceps).
Putro (1986)
Asia
(Indonesia)
Capacity: 48 kg ice/fish mixture
Weight: 24 kg
External dimensions: L= 1 025 mm ×
W = 295 mm (top) and 260 mm
(bottom) × H = 400 mm
Non-motorized
traditional fishing
dugout canoes (4-6
m Loa). In practice
these canoes were
catching on average
10-20 kg/day of high-
value red snappers
FAO (1992a)

9
Design: FRP laminated onto an
insulated container consisting of
expanded polystyrene foam (25.4
mm thickness) lined with 6.5 mm
plywood with an aluminium angle bar
framework. Access to the insulated
container is through two lids of 505
mm × 280 mm, located at both ends
of the top surface. A rubber gasket
20 mm × 10 mm is fitted around both
insulated lids. Heavy duty ropes were
fitted on both ends as handles to
assist in lifting the container. A single
10 mm diameter PVC pipe and plug
was fitted through the insulated
container at one side of the end of
the container. Tailor-made to fit local
canoes. Two versions were
constructed. The second version was
without FRP lining; it was made of a
wooden framework with plywood as
the outer shell, with 25.4 mm
expanded polystyrene foam as
insulation and an aluminium sheet
lining. Both versions had similar
external dimensions. The second
prototype weighed 17 kg. The cost of
the wooden version was about 50
percent of the FRP container’s cost.
These containers were designed
specifically for chill storage of red
snappers, which were the most
important species in the deep water
operations and export marketing of
the project.
Initial field trials were carried out on
two canoes; later, the project
“Cenderawasih Bay Coastal Area
Development” (UNDP/FAO INS/88/
0911) continued with the field tests
with deep water
handlines. Only
occasionally could
they catch large
quantities of, say, 40
kg/day. These
canoes were
generally working
within a motherboat
system, transferring
the catch at the end
of the day to the
motherboat’s fish
hold
South
America
(Ecuador)
Capacity: several types of containers
were made with nominal volumes
from 1.166 m3
to 1.224 m3
Weight: complete with tops, from 72
to 89 kg
Design: Three types of insulated
containers were made, namely: a)
wooden sides (consisting of a
wooden framework and plywood) and
wood as lining with 25 mm thick
expanded polystyrene insulation; b)
Traditional small-
scale fishing vessels
and canoes, but
these containers
(type (a), (b) and (c))
were specifically
designed for FRP
launches (7.2-7.5 m
Loa) engaged in
fishing for dolphinfish
(Coryphaena sp.)
Wood and
Grijalva
(1988)
Acero,
(1997)1
;
Tilman
(1999)2

10
same as container (a) but with
galvanized iron sheet as lining; c)
same as container (a) but with
polyurethane foam and FRP lining.
The FRP laminated container was
considered the most promising in
terms of lightness and durability
Four FRP-laminated insulated
containers were constructed and
long-term durability tests were carried
out by the project ODNRI/ODA
(Overseas Development Natural
Resources Institute/Overseas
Development Agency) with the
Instituto Nacional de Pesca. In recent
years, another tailor-made FRP-
laminated container (with 50 mm
polyurethane foam as insulation) was
field-tested and introduced in several
small-scale fishing communities by
the project “Technical Cooperation
Programme for Fishing” UE-VECEP
ALA 92/43, with promising results.
Similar FRP-laminated containers
were also field-tested by the UE-
VECEP ALA 92/43 project in
Colombia
1
Acero, 1997. Personal communication regarding field-testing of insulated containers in small-
scale fishing vessels in Colombia and references to a pilot activity in Ecuador for the introduction
of insulated containers by Project EU-VECEP ALA 92/43.
2
Tilman, 1999. Personal communication and references regarding the introduction of insulated
containers in small-scale fishing vessels in Ecuador by Project EU-VECEP ALA 92/43.

11
FIGURE 2.1 Example of a customized insulated fish box (West Africa)
Source: FAO, 1985.

12
3. Calculations and examples for insulated containers
and fish holds
In this chapter some examples of basic calculations for insulated containers and fish holds are
given. There is also a section on calculating ice requirements for cooling fresh fish and a section
on methods for the determination of fish hold volumes.
Calculating the specific ice melting rate for an insulated
container or fish hold
There are a number of methods for calculating the ice melting rate for an insulated container,
such as:
a) theoretical mathematical and numerical methods;
b) practical methods based on ice meltage tests.
Theoretical mathematical and numerical methods are available for the calculation of ice
meltage rates for containers, based on the coefficient of heat transmission (U), area through
which heat is transferred (A) and the latent heat of fusion of ice (L), which is 80 kcal/kg for pure
freshwater and 77.8 kcal/kg for seawater. The specific ice melting rate of a container (K1
),
expressed in kg of ice/ h °C, can be calculated from the following equation:
(equation 1)
The coefficient of heat transmission (U) (kcal m-2
h-1
°C-1
) is the rate of heat penetration through
the container walls per m2
of surface area per degree centigrade of temperature difference
between inside and outside. This value depends on the thermal conductance coefficient of the
materials used in the container wall (l), the thickness of these materials and the speed at which
heat can be transferred from the outside environment to the outside wall of the container, as well
as from the inside wall to the contents (e.g. fish and ice mixture).
For an insulated container made up of different layers of different materials, the coefficient of heat
transmission can be calculated from the following equation:
(equation 2)
where:
a1 = coefficient of heat transmission on the outside of the wall

13
a2 = coefficient of heat transmission on the inside of the wall
di = thickness of material layers used in the wall
l = thermal conductance coefficient of the materials used in the wall
f = coefficient taking into consideration the influence of stiffening ribs used in the ship’s structure
as well as supporting construction for the insulating wall (frames, deck beams, various elements,
etc., which can create heat leakage bridges)
For simplicity, the coefficients a1 and a2 are sometimes disregarded, as these factors may have
a relatively small influence on the result. However, all these methods require adequate heat
transmission knowledge, laboratory facilities and computer hardware and software, which are
generally not available in small-scale fisheries.
The heat leak through an element can be calculated using the following equation:
Q = A × U × (to - ti)
(equation 3)
where:
Q = total heat transfer rate through the element (kcal/h)
A = area of the element (m2
)
U = coefficient of heat transfer for the element (kcal m-2
h-1
°C-1
)
to = outside temperature of the element (°C)
ti = inside temperature of the element (°C)
Example: The following calculations are made for a steel fishing vessel with an insulated CSW
hold or tanks. The calculation of the rate of heat transmission (U value) for the vessel’s fish hold
requires knowledge of the heat transfer coefficients for all elements involved. The following
relationship is derived from equation 1.
where:
H1 = outside heat transfer coefficient (kcal h-1
m-2
°C-1
)
H2 = inside heat transfer coefficient (kcal h-1
m-2
°C-1
)
K1 = thermal conductivity of steel plate, ship’s side (kcal h-1
m-1
°C-1
)
K2 = thermal conductivity of polyurethane insulation (kcal h-1
m-1
°C-1
)
K3 = thermal conductivity of steel plate, tank lining (kcal h-1
m-1
°C-1
)
X1 = thickness of steel plate ship’s side (m)
X2 = thickness of polyurethane insulation (m)
X3 = thickness of tank lining (m)
Once the heat transfer coefficients have been calculated for each element of the fish hold
(deckhead, fish hold flooring, engine room bulkhead, forward bulkhead, ship’s sides above water
and ship’s sides below water), the heat leak through each surface can be calculated using

14
equation 3. The area of each element has to be determined and design temperatures for inside
and outside chosen.
For the fish hold or fish container as a whole, the total heat exchange rate will result from the sum
of individual calculations of the Q values. Table 7.1 shows the calculated heat leaks (kcal h-1
m-2
)
in a steel hull fishing vessel installed with a CSW steel tank with 100 mm thick polyurethane
insulation and one without insulation. This calculation is based on ideal conditions (without frames
or hangers penetrating the insulation, thus no heat leakage bridges). The main areas of the
fishing vessel through which heat enters into the uninsulated CSW tank are: the deckhead, the
ship’s sides above the water line and the ship’s sides below the water line, with overall heat
transfer coefficients of 27.6, 27.6 and 374 kcal m-2
h-1
°C-1
. Other areas of the vessel, such as the
engine bulkhead and tank floor, have overall heat transfer coefficients of 7.03 and 7.73 kcal m-2
h-1
°C-1
. However, the average overall heat transfer coefficient for a fully insulated CSW tank (under
ideal conditions) was calculated to be only 0.21 kcal m-2
h-1
°C-1
. Temperature differences between
the internal surface of the CSW tank (0 °C) and other areas of the vessel and heat leakages were
as shown in Table 7.1.
Since 1 kg of ice will absorb 80 kcal when it melts, the total heat leakage load in our example for
a fully insulated CSW tank will require about 7.7 kg/h of ice (185 kg/day). Therefore, for a four-
day trip the ice required to absorb the heat infiltration load will be 740 kg. However, for the above-
mentioned example, in practical circumstances, the heat leakage bridges in similar CSW tanks
insulated to a commercial standard (due to partial insulation of the CSW tanks) can be estimated
to be about 7 percent of the total heat leakage in the insulated CSW tank.
A practical and simple calculation method of the experimental specific ice melting rate of a
container (Kexp
), based on data of the mass of ice melted during a given time period, has been
developed and found suitable for small-scale fisheries. The proposed model is based on the
assumption that there is a linear relationship between the ice meltage rate in a container and time
when the ambient temperature remains constant. However, in practice, there are differences of
ice meltage rates for containers stored in the shade and in the sun.
TABLE 2.1
Heat leaks in a steel hull fishing vessel installed with a CSW steel tank
Surface D
Temperature
Heat leakage
Uninsulated CSW
tank
Fully insulated
CSW tank
(°C) (kcal/h) (%) (kcal/h) (%)
Deckhead 30 24 543 18.8 186 30.3
Tank floor 25 5 744 4.4 152 24.8
Engine room
bulkhead
35 4 700 3.6 137 22.4
Forward bulkhead 8 1 044 0.8 31 5.1
Ship’s sides above
water line
30 7 702 5.9 58 9.5
Ship’s sides below
water line
25 86 814 66.5 49 8.0
Total 130 548 100 613 100.0
Note: mild steel plate thickness: ship’s side: 6 mm; CSW tank lining: 5 mm.

15
Source: FAO, 1992b.
The specific ice meltage rate (K1
) can be calculated with the following equation:
Mi (t) = Mi (0) - K1
× Te (a) × t
(equation 4)
where:
Mi (t) = mass of ice inside the container at time t (kg)
Mi (0) = initial mass of ice inside the container at time t = 0 (kg)
K1
= specific ice melting rate of the container (kg of ice/h °C)
Te (a) = average temperature outside the container (°C)
t = time elapsed since icing operations in the container (h)
This methodology, described by Lupin (FAO, 1986), consists of the following steps:
1. Determine the technical characteristics of the insulated container to be tested.
2. Weigh the insulated container accurately (empty and dry).
3. Completely fill the insulated container with ice and weigh container again.
4. Record the time when the insulated container was filled with ice as well as the weight of ice
placed inside (calculated by weight difference).
5. Store and handle the containers in the shade, accurately record the prevailing working
conditions.
6. Monitor the ambient temperature at regular intervals so that an average temperature can be
estimated. It is recommended to monitor air temperature each hour during daytime (for short trials
of six to eight hours long) and use maximum-minimum thermometers for monitoring overnight
experimental works. However, time-temperature recorders can be also utilized for better results if
they are readily available.
7. For containers that allow melt water to drain away, the weight losses can be measured at
regular intervals (say every two hours) to monitor accurately the rate at which the mass of ice
placed inside the container melts.
8. This type of ice meltage test should be done using ice only, though the method can be equally
valid for fish and ice mixtures (provided that ice for chilling the fish load is accounted for). In
addition, some of the initial ice meltage will be the result of the heat removed for cooling the
container and, in some cases, some melted water may be absorbed by the container (depending
on the type of material).
9. The data of mass of ice melted (weight loss) should be plotted against time on a graph. These
data should give a more or less straight line graph (however, this will depend on the variability of
the external air temperature).
Typical experimental data of ice meltage plotted against time for a 90 litre capacity insulated
container stored in the shade. The experimental specific ice melting rate (Kexp
) value represents
the value of the slope of the plotted line. In the example in Figure 7.1 the slope of the line is

16
0.1498, therefore Kexp
= 0.1498 kg of ice/h (3.6 kg of ice/day) at an average ambient temperature
of 28 °C.
To determine the specific ice meltage rate (Kexp
) for the same container at different ambient
temperatures, it will be necessary to conduct several trials at the desired temperatures. Table 7.2
shows the experimental values of specific ice meltage rates (Kexp
) obtained during tests carried
out at different ambient temperatures.
Kexp
(kg of ice /day) = 0.1233 Te (a)
(equation 5)
Example: Determine how much ice will be consumed in the insulated container if it is stored in
the shade in a fishing boat over a five-day period at an average ambient temperature of 40 °C
(without considering the quantity of ice needed for chilling fish).
TABLE 3.2
Experimental values of specific ice meltage rates (Kexp) at different temperatures
Kexp
Average ambient temperature
(kg of ice/day) (°C)
4.8 40
4.2 35
3.6 28
3.0 23
2.4 20
2.0 16
1.6 12
1.2 10
0.9 5
From equation 5 the specific ice melting rate can be calculated:
Kexp
(kg of ice /day) = 0.1233 Te (a)
Kexp
= 0.1233 × 40 °C = 4.932 kg/day
Therefore, the total ice consumed in the insulated container to compensate for heat losses will
be: 4.932 kg/day × 5 days = 24.660 kg @ 25 kg of ice.
Note: Data resulting from the above ice meltage tests for insulated containers should be used
with caution. If the containers cannot be protected from direct sunlight or other radiated heat
source during field-working conditions, the above calculated values of specific ice melting rates
should be upgraded. In practice, it is best to store and handle insulated containers in the shade
and if possible complemented by covering the container in some way (e.g. wet insulating blanket
or tarpaulin laid over the container) to minimize the effects of radiated heat.

17
3.2 Methodology for the calculation of ice requirements for
cooling fresh fish
In general, the total amount of ice required for cooling fresh fish from any initial temperature to a
final temperature (ideally to 0 °C) using ice can be calculated from the following equation:
Mi × L = Mf × Cpf × (Tfi - Tfo)
(equation 6)
where:
Mi = mass of ice which melts (kg)
L = latent heat of fusion of ice (80 kcal/kg)
Mf = mass of fish to be cooled (kg)
Cpf = specific heat of fresh fish (kcal/kg °C)
Tfi = initial temperature of fresh fish (°C)
Tfo = final temperature of fresh fish (°C), normally 0 °C
From equation 6 the requirement of ice for cooling fresh fish to 0 °C will be:
The specific heat of fish varies according to its chemical composition; for example for lean fish
this value is about 0.8 (kcal/kg °C) and for fatty fish about 0.75 (kcal/ kg °C). For practical
purposes, however, it is acceptable to use the value of 0.8 (kcal/ kg °C) in all calculations for
fresh fish. This will give the simplified equation:
Example: Determine the ice requirement for chilling 40 kg of fresh fish at an initial temperature of
40 °C.
kg of ice
In practice, in tropical conditions much more ice is needed to compensate for losses of ice cooling
capacity due to meltage during ice storage at room temperature. There is evidence that when ice
is stored at 27 °C, there is a certain amount of water on the surface of flake ice particles at steady
conditions, which represents about 12-16 percent of the total weight. For crushed block ice, this
water on the surface can be about 10-14 percent of the total weight. The amount of water in
equilibrium on ice particles will depend on the type of ice and storage temperature.
Additional ice losses during chilling and storage of fish are due to bad handling practices, such as
ice wasted during icing operations. These losses are estimated to be about 3-5 percent of the
total amount of ice used.

18
The total requirements of ice to chill 40 kg of fish from 40 °C to 0 °C and maintain it chilled for five
days in a 90 litre insulated container are presented in Table 7.3. As can be seen, the required
amount, 50 kg, is slightly above the “rule of thumb minimum” of 1 kg of ice for 1 kg of fish.
TABLE 3.3
Summary of ice requirements for chilling fresh fish in a 90 litre insulated container
Consumption/loss factor Ice requirements
(kg)
To compensate for heat losses in the
insulated container
24.7 (4.932 kg/day × 5 days)
To cool down 40 kg of fish from 40 °C to 0 °C 16
To compensate for bad ice-handling
practices
2.5 (estimated as 5% of total
ice used)
To compensate for water in equilibrium in ice 7 (estimated as 14% of total
ice used)
Total consumption 50.2
Note: all values are based on previous worked examples.
Isaac Solar Ice Maker
Up to 1000 pounds of ice per day! Low cost and reliable
source of ice for situations requiring 25 to 1000 pounds
of ice per day (12-450kg). 'Isaac' is the acronym for
Intermittent Solar Ammonia Absorption Cycle. Ammonia
absorption refrigeration technology was developed in
the 19th century and is still used in industrial
applications. Energy Concepts has adopted this
technology to a machine which uses the sun as the only
energy output. The particular advances in the design
and configuration have resulted in a low cost and
reliable method of making significant quantities of ice in
areas without electricity. How Isaac Works... During the
day the solar collector focuses the energy of the sun
onto the ammonia generator in the collector trough.
Solar heat distills pure ammonia vapor from the water-
ammonia solution in the generator. The vapor
condenses in the cooling coils and collects as liquid
ammonia in the receiving tank in the evaporator.
At the end of the day, the user switches three valves from the Day to Night position to allow the
ammonia to evaporate in the ice compartment, providing the refrigeration to freeze the water. The
resulting vapor is absorbed back in the generator. Critical to the operation of Isaac is a passive
thermosyphon that operates in the Night mode to remove the heat from the generator and allow the
ammonia vapor to absorb into the solution at lower pressure and temperature. At the beginning of the
day, the operator harvests the ice from the ice trays, operates a drain sequence to remove traces of
absorbent from the evaporator, and places the unit back into Day mode to begin the next cycle. The solar
collector is re-aimed weekly to follow the seasonal movement of the sun. Occasionally the solar collector
should be rinsed with water to remove dust. The primary effort involved in operating the Isaac is filling
the ice trays with water in the evening and removing the ice each morning. The valve operation only adds
a few seconds to the tasks.

19
Isaac Solar Ice Maker
Solar Icemaker Up to 1000 pounds of ice per day! Low cost
and reliable source of ice for situations requiring 25 to 1000
pounds of ice per day (12-450kg).
"Isaac" is the acronym for Intermittent Solar Ammonia
Absorption Cycle. Ammonia absorption refrigeration
technology was developed in the 19th century and is still
used in industrial applications. Energy Concepts has
adopted this technology to a machine which uses the sun as
the only energy output. The particular advances in the
design and configuration have resulted in a low cost and
reliable method of making significant quantities of ice in
areas without electricity.
How Isaac Works...
During the day the solar collector focuses the energy of the
sun onto the ammonia generator in the collector trough. Solar heat distills pure ammonia vapor from the water-
ammonia solution in the generator. The vapor condenses in the cooling coils and collects as liquid ammonia in the
receiving tank in the evaporator.
At the end of the day, the user switches three valves from the Day to Night position to allow the ammonia to evaporate
in the ice compartment, providing the refrigeration to freeze the water. The resulting vapor is absorbed back in the
generator. Critical to the operation of Isaac is a passive thermosyphon that operates in the Night mode to remove the
heat from the generator and allow the ammonia vapor to absorb into the solution at lower pressure and temperature.
At the beginning of the day, the operator harvests the ice from the ice trays, operates a drain sequence to remove
traces of absorbent from the evaporator, and places the unit back into Day mode to begin the next cycle.
The solar collector is re-aimed weekly to follow the seasonal movement of the sun. Occasionally the solar collector
should be rinsed with water to remove dust.
The primary effort involved in operating the Isaac is filling the ice trays with water in the evening and removing the ice
each morning. The valve operation only adds a few seconds to the tasks.
Isaac Availability
Energy Concepts Company offers Isaacs in three standard sizes:
1. The Midi Isaac (37 ft collector) is designed to supply vaccine storage capacity for remote health clinics: daily
capacity is 35 lbs/15kg of ice per day.
2. The Standard Isaac (63 ft collector) will supply a small farm or fishing operation, yielding 70 lbs/32 kg of ice per
day.
3. The Double Isaac (125 ft collector) is suitable for a small village, with a yield of 150 lbs/68 kg of ice per day.
Typical Application Village Farm Health Clinic Collector Size 11.9 5.9 3.1 Daily Ice Production 68 30 16 Unit Weight 660
380 220 Absorbent Weight (kg) 164 82 45 Gross Weight Export Packing 940 520 340 Container Size (M3) 12 7 5
ordering code numbers RF3021 RF3020 RF3019 Unit Cost (US) Annapolis, MD $11,750 $7,950 $5,150 Delivery 90 days
90 days 60 days
Terms and Conditions
Terms: 25% of unit cost in US funds upon receipt of order, balance including freight and insurance prior to shipment.
Freight: Customer to advise preferred method.
Warranty: 1 year from date of shipment, not including freight
Installation assistance at customer location available at $300/day plus travel.
Item # Description Price
Developing
Countries Only

20
We may only be
able to ship this
product if you
have an address
in a developing
country.
RF3019 Isaac Solar Ice Maker - Health Clinic size $11,000.00
RF3020 Isaac Solar Ice Maker - Farm size $13,000.00
RF3021 Isaac Solar Ice Maker - Village size $17,000.00
Isaac Solar Ice
Maker - Farm size

21
.
A solar-powered absorption refrigeration apparatus comprised of:
(a) a pressure vessel of approximately circular cross-section for containment of a liquid
absorbent solution containing absorbed refrigerant;
(b) a solar-radiation reflector for concentrating solar radiation onto said pressure vessel;
(c) a condenser for condensing refrigerant vapor desorbed from said liquid absorbent
solution, and a conduit connecting the upper portion of said pressure vessel to said
condenser;
(d) a refrigeration evaporator which is supplied refrigerant which has been condensed in
said condenser via a means for pressure reduction;
(e) a conduit for returning evaporated refrigerant from said refrigeration evaporator to
said liquid absorbent solution in said pressure vessel; and
(f) a means for removal of heat from said liquid absorbent solution comprised of:
(i) a thermosyphon evaporator in heat exchange contact with said liquid absorbent
solution;
(ii) Conduit for transporting refrigerant condensed in said condenser to said
thermosyphon evaporator; and
(iii) conduit for transporting evaporated refrigerant from said thermosyphon evaporator to
said condenser.
2. Apparatus according to claim 1 additionally comprised of
(a) a cutout valve in said conduit for transporting evaporated refrigerant; and
(b) a one-way valve in said conduit between the upper portion of said pressure vessel and
said condenser, whereby refrigerant vapor can only flow out of said pressure vessel via
said conduit.
3. Apparatus according to claim 2 additionally comprised of:
(a) a receiver vessel for collecting condensed refrigerant from said condenser and
supplying it to said means for pressure reduction and said conduit for transporting
refrigerant; and
(b) a means for cutting out flow of condensed refrigerant to the evaporator when the
pressure vessel is received solar radiation.

22
4. Apparatus according to claim 3 further characterized by said receiver vessel being
located at an elevation no lower than the elevation of said thermosyphon evaporator, and
also characterized by the presence of thermal insulation material on said conduit
connecting the upper portion of said pressure vessel to said condenser, whereby
essentially none of the refrigerant vapor in that conduit is condensed so as to flow back to
said pressure vessel.
5. Apparatus according to claim 4 further characterized by a means for sensible heat
exchange between refrigerant to and from said refrigeration evaporator; a means for
maintaining the exterior of said receiver vessel wet with water; and at least one
transparent glazing between the sun and said pressure vessel.
6. Apparatus according to claim 4 further characterized by a flat glazing covering at least
the major portion of said reflector including said pressure vessel; a plurality of external
longitudinal fins on said pressure vessel; and a second transparent glazing draped over
said fins; and wherein the refrigerant is ammonia and the liquid absorbent is aqua
ammonia.
7. A solar-radiation energized absorbent apparatus comprised of:
(a) a pressure vessel for containment of an absorbent composition containing absorbed
refrigerant;
(b) a means for directing said solar radiation onto said pressure vessel;
(c) a condenser for condensing refrigerant vapor desorbed from said absorbent
composition, said condenser being connected to the upper portion of said pressure vessel
by a conduit;
(d) a receiver vessel which collects condensed refrigerant from said condenser;
(e) a check valve in said conduit;
(f) a means for pressure reduction which supplies liquid refrigerant from said receiver
vessel to a refrigeration evaporator;
(g) a conduit which conveys vapor from said refrigeration evaporator to said absorbent in
said pressure vessel;
(h) a thermosyphon evaporator in thermal contact with said absorbent in said pressure
vessel;
(i) a liquid refrigerant conduit connecting said receiver vessel to the bottom portion of
said thermosyphon evaporator;
(j) a conduit connecting the upper portion of said thermosyphon evaporator to said

23
condenser; and
(k) a cutout valve in one of the said fluid paths connected to said thermosyphon
evaporator.
Description:
TECHNICAL FIELD
This invention relates to apparatus for producing refrigeration at temperatures below
0.degree. C. and/or producing ice using heat from the sun. More particularly it relates to
intermittent solar-absorption cycles (heated by day, make ice at night) which use
NH.sub.3 as refrigerant and preferably liquid-phase absorbent solution for the NH.sub.3.
BACKGROUND
Solar-powered refrigeration holds great promise for extending the benefits of
refrigeration to areas not served by reliable central station generated electricity. However,
in spite of a keen interest in this prospect for over thirty years, no solar-powered
refrigerator has yet achieved all the desirable objectives of simplicity, reliability, and low
cost. Simplicity implies both ease of operation and also ease of manufacture, particularly
in lesser developed countries. Cost is a function of cycle efficiency as well as design
techniques. Since lower efficiency requires greater collector area, extreme simplicity at
the expense of very low efficiency imposes a costly tradeoff.
Three fundamental initial choices are made when categorizing solar thermally-activated
refrigerators. They are: whether the cycle is continuous (absorption and generation occur
simultaneously or intermittent): what working fluid (i.e., refrigerant) is used (e.g.,
NH.sub.3, MMA, halogenated hydrocarbons, methanol, sulfur dioxide, or H.sub.2 O);
and whether a solid-phase or liquid-phase absorbent is used.
This disclosure is directed toward intermittent cycle (also known as periodical)
refrigerators employing preferably NH.sub.3 as working fluid and preferably liquid-phase
absorbents.
Within this limited category, numerous problems are encountered in the various prior art
solar-powered refrigerators. Various continuous-cycle solar refrigerators are known in the
prior art, e.g., as described in U.S. Pat. Nos. 4,362,025 and 4,178,989. Those cycles have
the disadvantages that a solution pump is required (e.g., percolator type, with very
imprecise flow rate), and also that the evaporation and absorption steps must occur only
during the very limited time (about 5 hours per day) when generation and condensation
are also occurring.
Various intermittent-cycle solar refrigerators are known which employ solid-phase
absorbents, e.g., as described in U.S. Pat. No. 4,586,345. When the solid absorbent is
anhydrous (CaCl.sub.2, SrCl.sub.2, KSCN, or others) and the refrigerant is NH.sub.3,
those refrigerators have the advantage that only dry NH.sub.3 is present, and hence

24
aluminum construction is possible. However, they suffer the disadvantages that heat
transfer with the solid particles is very poor; that shrinkage and swelling occurs leading to
voids and possible ruptures; and that the heat necessary to desorb NH.sub.3 from a solid
is characteristically substantially higher than that necessary to desorb NH.sub.3 from a
liquid, thus requiring more solar energy for production of a given amount of NH.sub.3
refrigerant.
Various intermittent-cycle refrigerators employing either liquid or solid absorbents, but
which use flame combustion as a source of heat, are known in the prior art. Included
among them are the disclosures of U.S. Pat. Nos. 1,711,804, 2,446,636, 2,452,635, and
2,587,996 (all solid-phase absorbents), and also U.S. Pat. No. 2,185,330 (liquid-phase
absorbent).
Numerous solar-powered liquid absorbent intermittent cycle refrigerators employing
NH.sub.3 as refrigerant have been described in the technical literature over the past thirty
years. Representative examples include: F. Trombe and M. Foex, "Production of Cold by
Means of Solar Radiation", Solar Energy, 1, 1957, p. 51-52; R. K. Swartman, et al.,
"Comparison of Ammonia-Water and Ammonia-Sodium Thiocyanate as the Refrigerant-
Absorbent in a Solar Refrigeration System", Solar Energy, 17, 1974, p. 123-127; A.
Venkatesh and M. C. Gupta, "Experimental Investigations of an Intermittent Ammonia-
Water Solar Refrigerator", National Solar Energy Convention, Report CONF-781261,
December 1978, p. 675-784; and R. H. B. Exell, et al., "Design and Testing of a Solar
Powered Refrigerator", Asian Institute of Technology Research Report 126, 1981,
Bangkok.
One generic problem which all solar-powered, inermittent-absorption-cycle refrigerators
share is how to efficiently and economically collect and retain a maximum amount of
solar heat into the absorbent solution by day when it is generating, yet equally effectively
cool it by night when it is absorbing. Various solutions to this problem have been tried.
Due to the elevated temperature above ambient while generating, at least one layer of
glazing is normally employed to admit the solar radiation to the generator yet minimize
the escaping thermal radiation (heat leak). Some designs have removable glazing for
nighttime cooling. Others have dampers which can be opened to admit convective air
flow under the glazing at night. Designs with flat plate collectors will normally have
insulation on the side away from the sun, and that can be removed at night.
Clearly none of the above techniques presents much air-cooled surface for cooling the
absorption mode, typically no more surface than the solar aperture dimensions. Since air
contact cooling has a very low heat transfer coefficient, this cooling technique is not
efficient. Removable glazing is unwieldy in larger sizes, and mitigates against retaining a
good seal against rain and dust.
Another solution preferred is to have a separate heat removal circuit built into the
generator which is only activated at night when it is in the absorption mode. The
evaporator end of a thermosyphon can be incorporated in the generator, and an efficient
air-cooled condenser end of the thermosyphon is located at a higher elevation such that

25
liquid returns from the condenser to the evaporator by gravity. The thermosyphon
technique for removal of absorption heat is illustrated in U.S. Pat. No. 4,586,345. Note
that a cutout valve is necessary in order to block liquid flow to the thermosyphon while
the sun is shining, and the valve mechanism is solar-actuated. This system has several
disadvantages: a completely separate air-cooled condenser plus associated refrigerant is
required, which is only used at night; and when the liquid flow to the thermosyphon is
cutout in order to stop the thermosyphoning action, all the liquid inventory in the
thermosyphon evaporator must be boiled away before the heat removal ceases.
The five flame-actuated intermittent absorption refrigerators referenced above
incorporate thermosyphons for removal of absorption heat. The U.S. Pat. Nos. 1,711,804
and 2,185,330 incoporate a separate closed-cycle thermosyphon with its own internal
refrigerant, similar to the U.S. Pat. No. 4,586,345 solar patent. The other three patents,
U.S. Pat. Nos. 2,446,636, 2,452,635, and 2,587,996, all directed to solid absorbents,
incorporate open cycle thermosyphons which utilize the same condenser(s) and the same
refrigerant as the refrigerator itself. Note they all incorporate two generator/absorber
vessels, and they all locate the thermosyphon cutout valve in the liquid supply line.
Other problems found in the prior art practice of intermittent solar-powered refrigerators
using high pressure refrigerants such as NH.sub.3 or monomethylamine include:
1. The choice of solar collector geometry. With flat plate collectors, providing both the
necessary storage volume of refrigerant absorbent and also the good thermal contact
between solar radiation and the absorbent requires both large storage vessels and many
small pressure tubes welded to the storage vessel. Also, flat plate geometry presents
much surface area for thermal leakage. However, the alternative to flat plate--
concentrating collectors--may require tracking or frequent repositioning, which greatly
increases the cost and complexity of the solar collector. This is especially true for high
concentration ratios, e.g., 2.5 or more.
2. Many prior art designs incorporate the receiver directly in the evaporator, or at the
same pressure as the evaporator. This requires that all of the refrigerant liquid in the
receiver cools down by adiabatic flashing as the evaporator cools down. Thus much of
the refrigerant is wastefully consumed, and a larger cold thermal boundary is present.
When the receiver is integral with the evaporator, warm refrigerant liquid is collected in
the cold box by day, contrary to the objective of keeping the cold spaces cold. Even
worse is when the condenser and evaporator are physically the same component, which
introduces latent as well as sensible heat to the evaporator (condenser) coil.
3. In prior art ammonia-water solar-powered refrigerators, frequently a rectifier is
incorporated to reduce the H.sub.2 O content of the desorbed NH.sub.3 vapor. However,
without a rectifier only about 2% H.sub.2 O accumulates in the liquid NH.sub.3 each day
in a well-designed system. Furthermore, by proper design of the evaporator, and by
incorporating a sensible heat exchanger between liquid NH.sub.3 to the evaporator and
the fluid effluent the evaporator, it is possible to recapture refrigeration from that H.sub.2
O by subcooling the NH.sub.3.

26
4. In less humid climates evaporative cooling can be much better than dry air cooling,
permitting wet bulb cooling temperatures on the order of 5.degree. C. below ambient
temperature. Similarly, water cooling provides much better heat transfer than air cooling.
Unfortunately the prior art attempts to obtain these benefits have involved very large and
costly water tanks, constructed, for example, out of porous cement, thus requiring
extensive field construction. Also the porosity decreases with time, providing excessive
water loss early on and insufficient wicking later on. Water availability can also be a
problem.
What is needed, and included in the objectives of this invention, is a solar thermally-
actuated intermittent-absorption-cycle refrigerator using NH.sub.3 as working fluid and
liquid-phase material as absorbent, which:
requires only a single simple vessel for containment of the absorbent solution, and the
solar radiation is directed on that vessel;
achieves a concentration ratio between about 1.5 and 2.5 but does not require tracking;
has a liquid NH.sub.3 receiver vessel separate from the NH.sub.3 evaporator and outside
the cold boundary; and has an efficient and economical means of cooling the absorbent
container which does not require removing insulation or glazing, or a water tank of
capacity greater than the absorbent vessel.
DISCLOSURE OF INVENTION
A solar-powered intermittent-absorption-refrigeration apparatus is disclosed which
overcomes the prior art limitations by: providing a single cylindrical pressure vessel at
the focus of an arcuate solar reflector having an aperture between about 1.5 and 2.5 times
the diameter of the cylinder providing a charge of liquid absorbent for ammonia
(preferably aqua ammonia) in the pressure vessel providing efficient cooling of the
nighttime absorption step by locating a thermosyphon evaporator in the genrator which is
supplied liquid NH.sub.3 from the receiver vessel (located at or above the thermosyphon
evaporator height) and using the same condenser for the thermosyphon as is used by day
for condensing desorbed NH.sub.3 vapor. The condenser is preferably finned for efficient
air cooling, at least in part. The thermosyphon incorporates a cutout valve which is used
to halt its operation during the solar radiation period; the cutout valve may be located in
the liquid leg of the thermosyphon, but is preferably located in the vapor leg. The valve
also may be solar-actuated, but is preferably simply a manually operated valve. Also, in
order to preclude backflow of NH.sub.3 vapor from the condenser into the (absorber)
pressure vessel by night, a check valve (one-way valve) is positioned in the conduit
between the vapor space (upper portion) of the pressure vessel and the condenser.
By using a horizontal cylindrical pressure vessel, a low cost containment is obtained for
the absorbent. By locating the cylinder at the force of the solar reflector, a large target is
presented which gives rise to a solar acceptance angle on the order of 60.degree., while

27
also having a concentration ratio in excess of 1.5. The preferred reflector geometry is the
"truncated compound parabolic collector (CPC)", which is known in the prior art. Thus
heat leakage is minimized, and yet no tracking is necessary--the collector is stationary,
except for seasonally adjusted extensions.
With the "open cycle" thermosyphon, only a single efficient ambient-cooled condenser is
required. With ammonia-water as the working pair, the generator can be constructed of
low cost mild steel. All other components which are exposed to ambient weather
conditions may be constructed of a more corrosion-resistant material such as stainless
steel.
The liquid NH.sub.3 is collected in a reservoir or receiver vessel by day, which provides
the inventory of NH.sub.3 necessary to support both evaporation and thermosyphoning
throughout the night. By locating the thermosyphon evaporator below the receiver, it will
operate until the receiver is empty, at which time it is no longer necessary to
Thermosyphon. The pressure reduction device (orifice, valve, thermostatic expansion
valve, or the like) supplying liquid NH.sub.3 from the receiver to the evaporator must be
located downstream of the supply to the thermosyphon, since the thermosyphon must
operate at the pressure of the condenser, not of the evaporator.
The receiver vessel is not insulated from ambient temperature. Indeed, it may
advantageously be designed for maximum heat exchange with ambient, thereby further
increasing the effectiveness of the condenser. One method of doing this is to wrap the
receiver in a water-wicking material such as fiberglass cloth, and then position a tray of
water in contact with the lower portion of the wick. This will give rise to efficient
evaporative cooling.
As described, the apparatus has no moving parts other than approximately three valves,
and requires no electrical supply and minimal operator action.
The disclosed open-cycle thermosyphon absorber heat removal construction will apply to
solar-intermittent-absorption refrigerators employing solid absorbents also. It also applies
to systems using other refrigerants than NH.sub.3. It also applies to flat-plate collector
geometrics as well as to various concentrating configurations.
BEST MODE FOR CARRYING OUT THE INVENTION
A generally cylindrical pressure vessel 1 containing refrigerant and absorbent (e.g., aqua-
ammonia) is mounted horizontally in an arcuate solar collector 2 having internal
reflective surface which reflects solar radiation onto cylinder 1. The reflector geometry
may be of generally parabolic shape, and preferably has a cusp to prevent escape of some
reflections. The truncated compound parabolic shape is particularly preferred, where the
ratio of solar aperture to effective cylinder diameter is in the range of about 1.5 to 2.5. A
vapor exit conduit 3 connects the upper portion of cylnder 1 to condenser 5 via one-way

28
valve 5, which prevents backflow of vapor into the cylinder. Alternatively a manual valve
could be used. Condenser 5 is preferably air-cooled finned tubing. The condensate from
condenser 5 collects in receiver 6 by day, and is discharged to evaporator coil 8 via
means for pressure reduction 7 by night. The evaporator coil is located in the space to be
refrigerated, e.g., cold box 9. A conduit 10 conveys the evaporated refrigerant to cylinder
1 for absorption into the absorbent by night. Bubbler tube 11 ensures good mixing of the
vapor into the absorbent, and the horizontal orientation of the cylinder minimizes the
liquid hydrostatic backpressure. Optional check valve 12 will prevent backflow of a small
amount of absorbent into the evaporator coil during the day. Reservoir 6 also supplies
liquid NH.sub.3 to thermosyphon evaporator 13, which is a continuously uphill-sloped
length of pipe or tubing extending inside most of the length of cylinder 1. The exit vapor
passes through conduit 14 containing cutout valve 15 to condenser 5. Optional water tray
16 is fitted to receiver 6 to keep its optional wick jacket wet, should evaporative cooling
be desired. Conduit 3 is well-insulated to prevent any significant amount of condensate
forming therein and draining back to cylinder 1. Condenser 5 is necessarily located above
receiver 6, and thermosyphon evaporator 13 is located slightly below the height of
receiver 6. Sensible heat is preferably exchanged between the pressurized liquid
NH.sub.3 en route to pressure reduction device 7 and the almost entirely evaporated fluid
exiting from evaporator 8, in sensible heat exchange 17.
The general cross-sectional shape of a truncated CPC 20 is shown. For cylinder 21
diameters of larger than about 0.2 meter, the decreasing surface to volume ratio requires
that either more solar radiation be intercepted or less absorbent volume be contained than
that characteristic of simple cylindrical geometry. Thus for larger cylinders it is desirable
to incorporate fins 22, which will increase the effective target area. The outer flat glazing
23 both reduces heat leak and protects the reflective surface from rain and dust. It is
preferably polycarbonate or similar non-shattering material. The optional inner glazing
24 is primarily for further reduction of heat leak. It may be simply draped over cylinder 1
as illustrated, and is preferably of very thin TFE film, to maximize transmittance and to
withstand high local temperatures. The CPC geometry is preferably extended by a piece
of repositionable reflector 25, which is relocated each March and September to the side
away from the sun for the ensuing six months.

29
NEW SOLAR POWERED ADSORPTION REFRIGERATOR WITH
HIGH PERFORMANCE
An adsorptive solar refrigerator was built in September 2000 in Yverdon-les-Bains, Switzerland.
The adsorption pair is silicagel + water. The machine does not contain any moving parts, does not
consume any mechanical energy except for experimental purposes and is relatively easy to
manufacture. Cylindrical tubes function as both the adsorber system and the solar collector (flat-
plate, 2 m2, double glazed); the condenser is air-cooled (natural convection) and the evaporator
contains 40 litres of water that can freeze. This ice functions as a cold storage for the cabinet (320
litres).
The first tests (September 2000) showed a very promising performance, with a gross solar
cooling COPSR of 0.19. After minor modifications, a second test series was carried out during
summer 2001. This test series shows how the external parameters influence the machine with
respect to the COPSR (irradiation and external temperature). The latter varies between 0.10 and
0.25 with a mean value of 0.16. These values are higher than those obtained by earlier solar
powered refrigerators (0.10-0.12).
This paper describes the principle of the cycle, the different components of the machine, and the
test procedure. The test procedure includes a constant daily cooling requirement. The
experimental results presented were taken over a period of two months.
1. INTRODUCTION
The concept of using solar energy for powering a refrigerator appeared forty years ago
(Chinnappa, 1962) with a prototype using a liquid sorption cycle. Solar-powered
refrigeration can also use solid sorption, with either a chemical reaction (Worsøe-Schmidt
P., 1983; Erhard et al, 1998; Spinner et al., 1999), or adsorption. Meunier published a
comparison of those three sorption systems for solar cooling (Meunier, 1994). The solid-
gas system used in the present study is adsorption.
The solar adsorption refrigerators have been developed mainly to be used in hot regions
with no electricity supply. There is an urgent need in the health sector (for the
conservation of medicines and vaccines). These systems have the advantage of not
requiring any energy other than solar
energy.
All the machines reported in the articles (Headley O.StC. et al., 1994; Liu Z. et al., 1998;
Niemann et al., 1997; Boubakri A. et al. 1992; Critoph R.E., 1994; Critoph R.E. et al.,
1997; Grenier et al., 1988; Marmottant B. et al., 1992; Mhiri F. and El Golli S., 1996;
Pons M. and Grenier Ph., 1987; Pons M. and Guilleminot J.J., 1986; Pralon Ferreira-Leite
A. and Daguenet M., 2000) using either achemical reaction or adsorption, follow an
alternative cycle of heating/cooling, also known as 'intermittent', the period of which
corresponds to the alternation of day and night.
Regarding performance, the highest values of COPSR (0.10-0.12) were obtained with the
adsorption systems zeolite + water (Grenier et al., 1988) and activated carbon + methanol
(Boubakri A. et al. 1992; Pons M. and Grenier Ph., 1987). As methanol can easily
evaporate at temperatures below 0oC, thus favouring the production of ice, the most
environmentally friendly refrigerant must be water. Using water, ice can be produced
within the evaporator, acting as a ‘cold storage’. Both refrigerants, water or methanol,
operate at below atmospheric pressure and therefore require vacuum technology.
The main purpose of the present study is to obtain better performances than those
reported above, with what is, technically speaking, a simple machine. This aim seems

30
reasonably achievable with an adsorptive machine, operated in a 100% solar-powered 24
hour cycle with a flat-plate solar collector containing the adsorbent. However, when
referring to the work reported above, both the efficiency of the solar collector and that of
the adsorption thermodynamic cycle could be improved.
These requirements were crucial to the design of the 'advanced' machine.
The laboratory of solar energy of the Engineering school of the Canton de Vaud (EIVD,
Yverdonles-Bains, Switzerland) has been developing adsorptive solar refrigerators since
1999. The first systems built used the adsorption pair of activated carbon + methanol. For
reasons of reliability and respect for the environment, this pair has been abandoned in
favour of a silicagel + water pair.
The prototype described and analyzed in this article has been functioning since the
summer of 2000 on the site of the EIVD. A thorough measurement system allows us to
characterise it in a complete way. During the summer of 2001, a constant procedure of
thermal load in the cold cabinet allowed us to observe the behaviour of the adsorption
system over a continuous period of 68 days. We have highlighted the great influence of
both external temperature and daily irradiation upon the daily coefficient of performance
(COPSR). Previously, few articles were interested in the analysis of the storage.
2. DESCRIPTION OF ADSORPTION AND OF THE ADSORPTION COOLING CYCLE
Adsorption, also known as physisorption, is the process by which molecules of a fluid are
fixed on the walls of a solid material. The adsorbed molecules undergo no chemical
reaction but simply lose energy when being fixed : adsorption, the phase change from
fluid to adsorbate (adsorbed phase) is exothermic. Moreover this process is reversible. In
the following, we will focus on adsorption systems mainly used in cooling (or heat-
pumping) machines : a pure refrigerant vapour that can easily be condensed at ambient
temperature and a microporous adsorbent with a large adsorption capacity.
The main components of an adsorptive cooling machine are the adsorber (in the present
case, the solar collector itself), the condenser, the evaporator and a throttling valve
between the last two devices, see figure 2. An ideal cycle is presented in the Dühring
diagram (LnP vs. –1/T), figure 1.

31
Figure 1 : An ideal adsorption cooling cycle in the Dühring diagram. Saturation liquid-vapour curve
for the refrigerant (EC dashed line), isoster curves (thin lines), adsorption cycle (thick lines). Heating
period : step AB (7 a.m.→10 a.m.) and step BD (10 a.m.→4 p.m.) ; cooling period : step DF (4 .m.→7
p.m.) and step FA (7 p.m. →7 a.m.).
The cycle is explained in detail in (Buchter F., 2001). We can summarize it in four
stages:
Step 1 : Isosteric heating (A→B) :
The system temperature and pressure increase due to solar irradiance.
Step 2 : Desorption + condensation (B→D) :
Desorption of the water steam contained in the silicagel ; condensation of the water steam
in the condenser ; the water in the evaporator is drained through the valve.
Step 3 : Isosteric cooling (D→F) :
Decrease of the period of sunshine ; cooling of the adsorber ; decrease of the pressure and
the temperature in the system.
Step 4 : Adsorption + evaporation (F→A) :
Evaporation of water contained in the evaporator ; cooling of the cold cabinet ;
production of ice in the evaporator ; readsorption of water steam by the silicagel.
3. DESCRIPTION OF THE MACHINE TESTED IN YVERDON-LES-BAINS,
SWITZERLAND
Adsorptive pair : The refrigerant is water, and the adsorbent is a microporous silicagel
(Actige SG®, Silgelac).
Collector-adsorber : The solar collector (2 m2, tilt angle of 30°) is double-glazed : a
Teflon® film is installed between the glass and the adsorber itself. The adsorber consists
of 12 parallel tubes (72.5 mm in diameter) that contain the silicagel (78.8 kg). The tubes
are covered with an electrolytic selective layer (Chrome-black, Energie Solaire SA),
which absorbs 95% of the visible solar radiation while presenting an emissivity of 0.07 in
the infrared wave-lengths. The tubes are layered with a material which presents high

32
conductivity but low specific heat capacity (sheets of graphite : Papyex®, Le Carbone
Lorraine).
A central tube is made out of a grid (diameter 15 mm, mesh 1 mm, wire 0.45 mm
diameter). The ventilation dampers mentioned in the previous sections consist of a
mechanism that allows the thermal insulation to be opened on the rear side of the
collector (50 mm glass fibre), to provide efficient cooling by natural convection during
the night.
Condenser : Eight parallel finned tubes make a condenser, and are cooled by natural
convection of air. The total fin area is 6.9 m2.
Evaporator, ice storage and cold cabinet : The evaporator consists of three rings made
of square tubes. The total heat exchange area is 3.4 m2. The evaporator contains 40 litres
of water which can be transformed into ice during the evaporation stage. The cold cabinet
is chest-type and is well insulated (170 mm of expanded polystyrene) with an internal
volume of 320 litres.
Valve : A valve located between the graduated tank and the evaporator is needed on this
machine.
For control strategy reasons, this valve is electrically powered.
Figure 2 : Photograph and plan of an adsorptive solar refrigerator : solar collector-adsorber (1) with
detail :
glass cover (A), teflon® film (B), tube covered with selective surface (C) and internally layered with
Papyex®, central tube for vapour transport (D), silicagel bed (E), thermal insulation around the
collector (F) ; ventilation dampers (2) closed (2a) and open (2b), condenser (3), cold cabinet (4),
graduated tank (5), valve (6), evaporator and ice storage (7).
4. MEASUREMENTS AND OPERATIONS
The objective of the 2001 series of measurements was to obtain a high number of
measurements continuously, in order to characterise the working of our adsorption
machine. To do this, a system of measurement and a constant procedure of load has been
established.
4.1 Measurements
The temperature is measured (probes Pt 100) in the silicagel of a central tube of the
collectoradsorber (7 sensors), on two condenser tubes and three evaporator tubes ; and the
ambient air temperature is also measured. The vapour pressure is measured by a piezo-

33
gauge in the collectoradsorber, in the condenser and in the evaporator. The global
irradiance in the plane of the collector is recorded by a pyranometer. A graduated tank
(6.5 litres) collects the condensed water. The level of liquid water is automatically
measured by a level detector.
4.2 Acquisition system and command
A Labview® program takes measurements and administers various commands (valve,
dampers and load). A measurement is made every 30 seconds.
4.3 Automatic load
In order to simulate a thermal load in the refrigerator, every day at 1 a.m. a quantity of
water is automatically introduced, allowing to the load to be renewed. This is obtained by
a 32 litres water bottles system.
The input temperature is about 35°C. The load introduced daily corresponds to 4.1 MJ.
The water flow is stopped when the difference between the input and the output
temperatures is lower than 0.5 K, which guarantees a good homogeneity of the
temperature in the bottles.
4.4 Valve management
Closing : when solar irradiance is above 100 W/m2.
Opening : when, at the end of afternoon, the difference of pressure between evaporator
and condenser is lower than 100 Pa.
4.5 Ventilation damper management
Closing : when the irradiance goes above 100 W/m2.
Opening : at the end of afternoon when the angle for the solar beam radiation incident
upon the aperture plane of collector (angle of incidence) is above 50°.
5. METEOROLOGICAL CONDITIONS
The series of measurements took place from July 25th to September 30th 2001 (68 days) in
Yverdonles-Bains (altitude : 433 m, longitude : –6.38°, latitude : 46.47°).
1) From July 25th to the beginning of September : during this summer period, the mean
external
temperature is above 20°C and the mean daily irradiation reaches 22 MJ/m2. This fine
weather
period is interrupted between the 3rd and 9th August by less favourable weather.
2) From the beginning of September to the end of the measurement : the mean external
temperature
and the daily irradiation are distinctly lower (13°C and 13 MJ/m2). Furthermore, the
conditions are
very variable from one day to the next.
6. PERFORMANCE OF THE TESTED UNIT
For each day, a gross solar COPSR can be defined as the ratio of the heat extracted by
evaporation of water to the solar heat supply, see equation 1. The first one, Qe, is
obtained by multiplying the mass of processed water, mL, by the enthalpy difference
between the saturated vapour at Te and the saturated liquid at Tc. The second one, Qh, is
the product of the surface A of collector and the solar

34
• At nightime : the lower the night-time temperature is, the better the readsorption of
water in the silicagel will be. When the readsorption is high, so is evaporation, and
consequently a large amount of ice is produced in the evaporator. Furthermore, the next
morning, the adsorber is in a favourable state for a good desorption during the day which
follows.
• In the daytime : The temperature of the ambient air defines the level of pressure of
condensation (Pc in figure 1). The lower the day temperature is, the lower this pressure is
and the earlier in the day condensation will begin. To sum up, for two days with equal
irradiation, the performance will be better for the day characterised by the lower external
temperature.
Furthemore, the COPSR is influenced by the heat losses during night and day, which are
themselves directly linked to the external temperature.
This variation of COPSR . For these few days, a decreasing COPSR (0.22 to 0.10)
corresponding to an increasing external mean temperature (17°C to 24°C) can be
observed while the daily irradiation is almost constant. The COPSR falls when the outside
temperature increases.
In addition to the temperature and the daily irradiation, we could note that the state of
what is being stored (completely, partially or not at all frozen) influences the value of the
COPSR of the next day.
We see that these increases in temperature do not necessarily correspond to a similar
sequence of lack of irradiation. On the one hand, these variations can be explained by the
evolution of the COPSR as described previously. On the other hand, we note that the load
imposed daily on the refrigerator corresponds to the mean of daily net production of cold
in the evaporator (equations 4 and 5). It explains the sensitivity of the system to
meteorological conditions.
There follow some values permitting to draw up a balance sheet of energy over a period
where the temperature of the evaporator remained virtually constant from August 24th to
September 22nd 2001
(30 days) :
Irradiation Qh : 923 MJ
Cold energy produced Qe : 146 MJ
Cabinet thermal losses Qtl : 26 MJ
Load energy Ql : 123 MJ
The difference observed between the sum of the cold cabinet thermal losses and the load
energy (Qtl + Ql) and the cold energy produced (Qe) corresponds to the variation of the ice
storage in the evaporator between August 24th and September 22nd.
The rough and net COPS can be defined :
The mean net production on this period is :

35
and the mean load :
It should be remembered that the load was maintained constant during the whole
measurement period. This procedure did not favour the system. During a real-life use of a
sun refrigerator (adsorption or photovoltaic), it is necessary to adapt the load to the
meteorological conditions. In fact, when the weather is not good, the user has the choice
between :
- not loading the refrigerator, if possible
- loading the refrigerator and letting the temperature of the whole stock rise above the
temperature desired (5-10°C), thus risking damaging the contents if the unfavourable
meteorological conditions persist.
7. CONCLUSIONS
This series of measurements carried out over a long period reveals the influences of the
meteorological conditions (external temperature and irradiation) on the performance of
the system.
Our adsorption solar refrigeration system presents very interesting performance
coefficients (COPSR : 0.16 ; COPSN : 0.13). These values are better than other adsorptive
system (the highest values of COPSR (0.10-0.12) were obtained with the adsorption
systems of zeolite + water (Grenier Ph. et al., 1988) and activated carbon + methanol
(Boubakri A. et al., 1992; Pons M. and Grenier Ph., 1987).
We choose the dimensions of the solar installation with the aim of storing cold by
freezing water during sunny days so as to maintain an appropriate temperature 3 days of
bad weather. The data presented in this paper show that we are near to obtaining this.
The meteorological conditions of the climate in which the tests took place (Switzerland)
are relatively favourable. Since such systems are intended to be used in a Sahel climate
type, it is necessary to take into account the influence of the external temperature by
selecting the size of the condenser and insulating the cold cabinet well. On the other
hand, in these regions, the high level of daily irradiation encourages the use of the solar
refrigerator based on adsorption.
In addition to the developments made in Switzerland, a technology transfer to Sahel
country (Burkina Faso) is to be carried out. An activated carbon + methanol refrigerator
was built in 1999 in Ouagadougou for a NGO (Centre Ecologique Albert Schweitzer). It
has worked well so far (Buchter F., 2003). A new compact silicagel + water refrigerator
was tested in summer 2002 in Ouagadougou.
We should emphasize that the behaviour of the users of a solar refrigeration system
without auxiliary energy resource must be adapted to the variable solar resource.

36
NOMENCLATURE
COP Coefficient of Performance [-]
Cp Specific heat [J·kg-1·K-1]
G Global Irradiance [W·m-2]
H Global Irradiation [J]
L Evaporation latent heat for water[J·kg-1]
m Mass [kg]
n Number [-]
P Pressure [Pa]
Q Heat quantity [J]
A Area [m2]
T Temperature [K]
t Time [s]
Indexes
c condenser
d day
e evaporator
fs collector front side
h solar heat supply
l load
L liquid
m maximum
N net
R rough
S solar
tl thermal losses (of the cabinet)
w water

37
Design And Construction Of A Solar Energy Refrigeration
System Using Vacuum Concentric Tubes With Adsorption.
The R&D project that is envisioned here, will consist in the physical construction of a Solar
Refrigeration System. The theory, on which our design will be based, will be that of using an
adsorption thermodynamic system based on the use of an Activated Charcoal and Methanol as
working fluid. The Solar collector will consist of a Vacuum Concentric Tubes (VCT)(Tubes that
are actually in our position
in the Dom. Rep.) The refrigeration cycle will consist of a de-adsorption of the methanol during
the heating period of the solar radiation of the day. The vapor will condense and accumulate as a
liquid during this period. During the night the process will invert and methanol will be absorbed
by the Activated Charcoal and evaporate, lowing its pressure and temperature, producing ice in a
refrigerated isolated box. This box can be used as a refrigerated in remote places. It is our
intention to establish an R&D process in order to fabricated a working commercial model that can
produce refrigeration plus hot water simultaneously.
Introduction
The Dominican Republic is at the present stage immersed in a critical situation due to the
inability to satisfy the energy needs of its people. It struggles day to day in order to make the
needed equilibrium between supply and demand of its electric power in the utility grid. Costs of
producing electric power from fossil fuels are to big a burden to the nation. Therefore it is our
intention with this R&D project to try within ourselves to make the difference, although we are
aware and recognize our intrinsic limitations of an underdeveloped country. Substitution of fossil
fuels by solar energy is our goal in this segment of energy requirements, which represent more
then 60 % of the consumption. (Air Conditioning, refrigeration and water heating.)
RESEARCH AND DEVELOPMENT PROCEDURE.
The process that will be executed in this project is divided in three main sections. First, will be
involved in the establishment of the thermodynamic theory used for the proper formulations and
modeling, plus bibliographic search. Second, design of our prototype and selection and
procurement of materials. Third, Construction and experimentation of model, with all possible
scenarios taken into consideration. In all sections of this project, students of last year of
Mechanical Engineering will be involved as part of they thesis requirements. The University
possesses mechanical engineering laboratories that can do all the necessary works indoors. We
have all machine shops and tolls needed in our Campus Oriental in Santo Domingo. At the
present time, we have all the needed Solar Vacuum Tubes in our possession. These tubes were
bought in China, and specifications are given further in this paper. The central part of this R&D
project is the use of the Solar Vacuum Tubes. Supplier was: Xu Guangwen, COAST CORP LTD,
Tel: 86 4512735906 Fax:864512735876864512308891 Email:guangwen@public.hr.hl.cn
Section 1:
THERMODYNAMIC CYCLE OF REFRIGERATION PROPOSED.
1. ADSORPTION CYCLES FOR SOLAR COOLING
Adsorption cycles for solar cooling are described and past work reviewed, but is our intention to
built a refrigeration system that will work with solar energy. Because of the fact that Vacuum
Solar Tubes are available at economic values, it is our belief that due to their high efficiencies we
can make a difference from previous research work done in this subject. Zeolites have been used
as adsorbents in many systems, but this work concentrates on activated charcoal adsorption.
A general study of the cycle thermodynamics shows that provided the latent heat of the
refrigerant exceeds about 10(K) kJ/kg and the concentration change exceeds about l0% then the
coefficient of performance (COP) can only be slightly improved by further increase. Looking at
activated charcoal in particular, the Dubinin- Astakhov (D--A) equation is used to predict cycle

38
COPS based on limited data available for chosen refrigerants and carbons. Of the refrigerants that
are sub atmospheric at — 10°C. methanol, acetonitrile, methylamine, and NO2 are suitable, with
methanol giving the best COP. Of the refrigerants above atmospheric pressure at – l0oC,
ammonia, formaldehyde, and SO2 are suitable. Overall, methanol gives the best COP, with 0.5
being achievable in a single-stage cycle.
Solar cooling could be a useful technology in areas of the world where there is a demand for
cooling, high insolation levels, and no firm electricity supply to power conventional systems. The
most promising applications are vaccine storage and food storage (particularly fish). Solar
thermal systems for refrigeration have been studied for some years, and many refrigerators built
and tested. Earlier work concentrated on intermittent absorption cycles such as the ammonia—
water machines built Exell [1. 2] and Van Paasen[3]. More recently, solid adsorption cycles have
been examined. These have the advantage of requiring no rectifiers, valves, or liquid seals such as
are needed in the ammonia water cycle.
The adsorption cycle is best understood with reference to the p-T-x (pressure-temperature
concentration) diagram of Fig. 1 and the schematic diagram of Fig. 2. The processes involved are
as follows:
Figure 1.- p-T-x diagram of a simple adsorption cycle.
1. Starting in the morning with the valve open and at ambient temperature of about 30 C (Ta2 the
rich concentration adsorbent in the generator/absorber is heated by solar energy until the pressure
reaches a level that enables refrigerant to desorbed and be condensed in an air or water cooled
condenser.
2. Refrigerant is driven off at constant pressure, the adsorbent becoming more and more dilute
until the maximum cycle temperature of about 100°C (Tg2) is reached. The condensed liquid is
collected in a receiver.
3. The valve is shut, and the adsorbent cools and reduces its pressure. At some stage of the
evening or night its pressure will be the saturated vapor pressure of refrigerant at -10°C, (Te)
which is sufficiently low for ice production.
4. The valve is now opened, and the liquid refrigerant starts to boil in the evaporator. Initially the
refrigerant within the evaporator and receiver simply cools itself, but having dropped below 0°C
it can start to freeze water. Adsorption is completed by the following morning, completing the
cycle.
During this process heat is released in the absorber and so the generator/absorber must be cooled
be ambient temperature air or water.
This ideal cycle can be used to calculate COPs with reasonable accuracy, but in practice an
isolating valve is not necessary and processes c and d merge into a smooth curve and there are no
sharp corners on the experimental p-T-.v diagram. The diagram used is essentially the same for
all adsorption cycles.

39
Although silica gel has been used in at least one closed-cycle refrigerating system, the vast
majority of experience is with zeolites or activated carbons. Critoph and Vogel [8] compared
zeolites and active charcoal with refrigerants Rll. Rl2. R22, and R14 and in all cases found
charcoal a preferable adsorbent for solar cooling. Meunier et al. [7..-...l0] have compared
synthetic zeolite —water, synthetic zeolite —methanol and charcoal—methanol combinations.
They find that activated charcoal-methanol gives a better COP generally but that when the
nighttime ambient temperature to evaporating temperature difference is particularly high then a
zeolite—water combination is better. However, this will also require a higher generating
temperature from the solar collector during the day. Meunier, comparing AC35-methanol with
zeolite 13X-water, suggests that the zeolite combination will only be superior when the
temperature lift (adsorption-evaporating temperature) exceeds 45°C. The COP is based on heat
input to the adsorbent rather than to the solar collector and so the reduced solar collector
efficiency at higher temperatures may actually make charcoal—methanol combinations superior
at even higher temperature lifts.
There are other reasons for preferring charcoals:
1. Activated charcoals are cheaper than zeolites.
Figure 2.- Schematic diagram of the simple adsorption cycle.
2. Activated charcoals can be made with properties to suit particular applications by
varying the activation time and temperature etc.
3. Activated charcoals (particular coconut shell charcoal can be manufactured in the
Dominican Republic.)
2. THERMODYNAMICS OF ADSORPTION CYCLES
Consider the p-T-x diagram of Fig. 1. The various temperatures are:Ta1 = temperature at start of
adsorption, Ta2 = temperature at end of adsorption, Te = evaporating temperature, Tc =
condensing temperature, Tg1 = temperature at start of generation, Tg2 = temperature at end of
generation. Te depends on the application; air conditioning, ice making, deep freeze, etc. Tc is a
heat rejection temperature and should be as near to ambient as heat transfer and economics will
allow. Ta2 should be as low as possible so that the strong concentration is as high as possible—
this maximizes the concentration change in desorption, thus minimizing the quantity of charcoal
that must be wastefully heated and cooled with the adsorbed refrigerant. Ta2 is also limited by
ambient temperature and heat transfer considerations.
Although Ta2 is not necessarily equal to Tc (different heat exchangers may be used and the
ambient temperature may change between adsorption and desorption) it is useful to consider the
simple case where they are equal.

40
Both the pure refrigerant and the isosteres approximately obey Trouton’s rule, and assuming that
the isosteres on the In pv vs. 1/T diagram are indeed linear, it can shown that .
In the case of solar cooling, Ta2 might vary between 25oC and 45°C depending on ambient
conditions, and Te could vary between —20oC for cold storage to +5oC for air conditioning. A
few possible combinations are listed with calculated value of Tg1 in table 1.
The values of Tg2 are normally within 2 to 5oC of the results for detailed calculations on specific
pairs.
Given that Tg1 is the temperature at the start of generation and that in practice Tg2 will be at least
10°C higher (probably more) the importance of keeping Ta2 low can be seen. Heat rejection from
the condenser and adsorber should be as close to ambient as possible to avoid excessively high
solar collection temperatures.
The assumption that Ta2 = Tc, is reasonable in many circumstances but may not be true for a
diurnal cycle in a climate with large diurnal temperature variations. The maximum cycle
temperature Tg2 is not fixed by the above relationships but is a design variable that must be
optimized. The higher Tg2., the more refrigerant is driven off and so the greater the cooling effect
per Unit mass of adsorbent. However, progressively less refrigerant is desorbed as the
temperature rises, and the extra heat input is mainly used in sensible heating of adsorbent and
refrigerant, rather than in desorption.
Considering the ideal case of an adsorbent with negligible thermal mass and a refrigerant of negligible
specific heat compared to enthalpy of vaporization, then the COP = Enthalpy of vaporization (at Te) / Mean
heat of adsorption.
Taking a base of Te = -10°C, Tc = 30°C, and Tg = 90°C, then the ideal COP is 0.83. This may be
compared with the Carnot COP of a similar cycle Te = - 10°C, TA = 30°C. Tg2 = 9OoC which is 1.09. The
difference is, of course, due to the constant heat source and sink temperatures assumed in the Carnot cycle.
This is a new and useful
Application Ambiente
Temperature
Te oC Ta2oC Tg1 oC
Freezing Moderate -20 25 79
Hot -20 45 126
Ice Making Moderate -10 25 65
Hot -10 45 112
Air Conditioning Moderate 5 25 46
Hot 5 45 91
Table 1.- Typical Cycle Temperatures
When taking account of the sensible heating and cooling of adsorbent and refrigerant, the COP is
further reduced. For any particular application. Te,, Tc,. and Ta2 are fixed, and Tg1 and DH/L can be
calculated. It can be seen that the refrigerant needs a high, and if possible a low cpR.
Unfortunately. high latent heats tend to be associated with high specific heats and so a high
specific heat must be tolerated. The adsorbent should have as low a specific heat as possible, but
its main requirement is that it gives a high concentration change between Tg1 and Tg2 The
reasoning presented as a theory and postulate, will provide a practical cycle COP with limitation
lower than that of pure calculations, but more specific information must be provided to calculate
the COP’S. The objective of our research project will be to construct with the Vacuum Tubes in
our possession a working model and find the parameters that will fit these equations.
Section 2:
TUBULAR SOLAR ENERGY COLLECTORS
1.Solar Energy With Vacuum Tubes. Our Method Of Choice.
Two general methods exist for significantly improving the performance of solar collectors above
the minimum flat-plate collector level. The first method increases solar flux incident on the
receiver; the second method involves the reduction of parasitic heat loss from the receiver

41
surface, such as the Tubular Vacuum Solar Collectors. Tubular collectors, with their inherently
high compressive strength and resistance to implosion, afford the only practical means for
completely eliminating convection losses by surrounding the receiver with a vacuum on the order
of 1O-4 mmHg. The analysis of evacuated tubular collectors is the principal topic of this section.
2. Evacuated-Tube Collectors.
Evacuated-tube devices have been proposed as efficient solar energy collectors since the early
20th
Century. In 1909, Emmett6 proposed several evacuated-tube concepts for solar energy
collection, two of which are being sold commercially today. Speyer2° also proposed a tubular
evacuated flat-plate design for high-temperature operation. With the recent advances in vacuum
technology, evacuated-tube collectors can be reliably mass-produced. Their high-temperature
effectiveness is essential for the efficient operation of solar air-conditioning systems and process
heat systems.
The level of evacuation required for suppression of convection and conduction can he calculated
from basic heat-transfer theory. As the tubular collector is evacuated, reduction of heat loss first
occurs because of the reduction of the Rayleigh number. The effect is proportional to the square
root of density.
When, the Rayleigh number is further reduced below the lower threshold for convection, the heat-
transfer mechanism is by conduction only. For most gases the thermal conductivity is
independent of pressure if the mean free path is less than the heat— transfer path length.
3. Optical Analysis of the CTC. (Concentric Tube Collector)
Since close packing of CTC tubes in an array can result in shading losses at any angle other than
normal incidence, it is cost-effective to space the tubes apart and to use a back reflector in order
to capture any radiation passing between the tubes. Evacuated—tube designs and concentrating
designs should be closely packed to optimize solar energy collection. Beeklcy and Mather (ref)
base analyzed the CTC in detail and their analysis is recommended for further studies. CTC
arrays can collect both direct and diffuse radiation. Each radiation component must be analyzed
in turn.
4. Thermal Analysis of the CTC
The heat loss from a CTC occurs primarily through the mechanism of radiation from the absorber
surface.
The rate of heat loss per unit absorber area qL then can he expressed as Total thermal resistance
1/Uc. is the sum of three resistances:R1 — radiative exchange from absorber tube
to cover tube R2 —conduction through glass tube R3___ convection and radiation to environment.
The CTC energy delivery rate qu on an aperture area basis can be written as where At, is the
projected area of a tube (its diameter) and Ar is the receiver or absorber area. The receiver-to-
collector aperture area ratio
5. Physical and Thermal Characteristics of the Evacuated-Tube Solar Collector
The tubes that are actually in our possession have the following characteristics:
• Tube is a new generation of sun-heat collect device with low cost but high efficiency of solar
thermal conversion.
• Tube is made of all-glass body and has the configuration of two concentric Borosilicate glass
tubes, Inner glass tube (Absorber glass tube) and Outer glass tube (Cover glass tube).

42
• There is a selective absorbing surface on the outside of inner glass tube (Absorber glass tube),
this selective absorbing surface is based as a layer of aluminum-nitrogen, this aluminum-nitrogen
layer was manufactured using a special coated surface by using the process of magnetron
sputtering.
• The jacket between cover (outer) and inner glass tubes is evacuated and permanently sealed off.
• These Tubes are widely utilized due to their high efficiency, low heat losses, long lifetime and
low costs. Outside Diameter OD47 mm.
Specifications of Glass Vacuum Tubes:
Tube material: Borosilicate glass, Configuration: Two concentric tubes Length: 1.5 meters, Cover
tube diameter: 47mm Absorber tube diameter: 37mm Net weight of tube: 1.4kgs of 1.5 meter.
Tube wall thickness: 1.6mm Transmittance of cover tube: 91% Solar absorptance (AM1.5):
93%Emittance (80 c): 6% Pressure of vacuum space: < 0.005Pa Stagnation temperature (typical):
200 c degree Heat loss coefficient of tube: < 0.8W/m2.c Impact resistance: withstand 25mm
diameter hailstone without breaking glass strength (pressure tested): 1 Mpa Lifetime: 15 -
20Years

Using Intermittent Solar Ammonia Absorption for Food Preservation

Using Intermittent Solar Ammonia Absorption for Food Preservation

Recommended

Recommended

More Related Content

What's hot

What's hot (13)

Similar to Using Intermittent Solar Ammonia Absorption for Food Preservation

Similar to Using Intermittent Solar Ammonia Absorption for Food Preservation (20)

More from Bruno B Garcia

More from Bruno B Garcia (7)

Recently uploaded

Recently uploaded (20)

Using Intermittent Solar Ammonia Absorption for Food Preservation