Thermophysical Properties of

Aqueous Solutions Used as

Secondary Working Fluids

Doctoral Thesis
By
Åke Melinder

Division of Applied Thermodynamics and Refrigeration

Dept. of Energy Technology
School of Industrial Engineering and Management
Royal Institute of Technology, KTH
Stockholm, Sweden 2007

TRITA REFR Report No 07/60

ISSN 1102-0245
ISRN KTH/REFR/07/60-SE
ISBN 978-91-7178-707-1
1 S e c o n d a r y wo r ki n g f l u i d s
used in indirect systems

1.1 Indirect systems with secondary

working fluids
Secondary working fluids (secondary refrigerants, heat transfer
fluids, antifreezes, brines) have long been used in various indirect
refrigeration and heat pump systems. One secondary working fluid
circuit transports heat from the cooling object or heat source to
the evaporator and another may cool the condenser. Many terms
are used for the fluids we here consider.
Many terms are used for the fluids we here consider. Why has sec-
ondary working fluid been chosen for this publication? The term is
not restricted to just one function of the fluid, such as heat carrier,
heat transfer fluid or antifreeze. It draws attention to an indirect or
secondary system without using the often misunderstood term
secondary refrigerant. The commonly used term brine should
strictly only be used about aqueous salt solutions. The term secon-
dary working fluid is in agreement with the Call for papers to the
22nd Int. IIR Congress of Refrigeration in Beijing, August 2007.
Aqueous solutions (water solutions) have long been used as sin-
gle phase (liquid only) secondary working fluids for cooling in su-
permarkets, for ice rinks, heat recovery systems, heat pumps and
other applications. However, aqueous solutions are increasingly
used also for freezers in supermarkets and other applications in
low temperature refrigeration. Aqueous solutions are among else
also used in sorption based air conditioning equipment for dehu-
midification of air, usually operating at rather high temperatures
(Arias 1999, Conde 2004, Kruse 2005, Lundqvist 2000, Mao and
Hrnjak 1999, Pearson 1993).
Figure 1.1 shows the principles of a completely indirect cooling
system, i.e. for a supermarket. In such an indirect system, the pri-
mary refrigerant can be confined to a compact chiller unit in the
machine room. One secondary working fluid (1) cools the con-
denser and the dry cooler is usually placed outdoors. Another sec-
ondary working fluid (2) is cooled in the evaporator and used for
cooling cabinets and cold storage.

1
 Outdoor temperature

Dry cooler
Secondary working fluid (1)

Pump Condenser

Compressor Chiller unit

Pump

Evaporator

Secondary working fluid (2)

Cooling Cold
cabinet storage

Figure 1.1 Completely indirect system

Research has the last decade been made with aqueous solutions as
two-phase ice slurry for cooling in supermarkets, cooling of fish,
shrimps and vegetables, transport cooling, energy storage as well as
other applications. (Aittomäki 1999, Ben Lakhdar 2002, Egolf and
Kauffeld 2005, Field and Kauffeld 2003, Fukusako 1999, Granryd
2002, Kauffeld, Egolf and Kawaji 2005, Malter 1996, Paul 1997,
Snoek 1993, Wang 1999).

1.1.1 Reasons for the move toward more

indirect systems
Indirect refrigeration plants and heat pumps with secondary work-
ing fluids are more and more used in in many countries. What are
the reasons? The discovery that some refrigerants contribute to
global environmental effects has led to the phasing out of CFC re-
frigerants and also an increased awareness that the use of any pri-
mary refrigerant should preferably be kept to a minimum. Large
primary refrigerant leaks in supermarketsand large heat pumps
have also focused attention on the use of indirect systems with
secondary fluids.

2
With an indirect system it is possible to manufacture large quanti-
ties of factory built cooling units that can be used for a variety of
applications. It is also possible to design the primary refrigeration
unit as a compact chiller in a machine room with an extremely
small refrigerant charge. All refrigerant piping work can be per-
formed at the factory with minimal risk of leakage. Local construc-
tion of refrigerant piping can be avoided and installation work can
be made in a more simple way (Lindborg 2003, Lundqvist 2000).
See also PAPER 3 (Melinder 2003b).

1.1.2 Temperature profiles of an indirect

system
Before mentioning pro’s and con’s of an indirect refrigerating sys-
tem, let’s consider a schematic figure with temperature profiles of
an indirect refrigerating system with temperature profiles of the
evaporator side as given in Figure 1.2.

Condenser

Compressor

Evaporator Pump Heat source

Secondary working fluid circuit (Cooling coil)

Temp.
' t HS

Aver. heat source temp., tHS

T HS-SF
SF

Aver. sec. fluid temp., tSF

't
T SF-EV

Evaporation temp., t2

Figure 1.2. Scheme of indirect refrigerating system with temperature

profiles of evaporator side

The following temperatures are indicated: evaporating temperature

of refrigerant, t2, average temperature of heat source, tHS, and av-
erage temperature of secondary working fluid, tSF. The following
temperature differences are indicated: average secondary working

3
fluid temperature change, 'tSF, the difference between heat source
temperature and average secondary fluid temperature, T2,HS = tHS –
tSF and the difference between average secondary fluid temperature
and evaporating temperature, T2,EV = tSF - t2. (Compare Arias,
Lundqvist 1999).

The additive concentration of the secondary working fluid is usu-

ally chosen to give a freezing point below the lowest temperature
in the secondary fluid loop. This will be considered more in Chap-
ter 5.3. (Compare Granryd and Melinder 2005, Melinder 1997).

Indirect systems with a secondary working fluid circuit may have a

number of advantages but also some drawbacks in comparison
with direct expansion (DX) systems depending on how the sys-
tems are built.

1.2 Advantages and drawbacks of

indirect systems
1.2.1 Advantages of indirect systems
compared to direct ones
• The primary refrigeration unit is usually designed as a factory
built compact chiller in a machine room with an extremely small
refrigerant charge. Such cooling units can be built in larger quan-
tities as they can be used for a variety of purposes.
• Local construction of primary refrigerant piping is avoided and
installation work is made in a more simple way. Such prefabri-
cated chillers of high quality and with correct charge minimize re-
frigerant leakage. Equipment in contact with the refrigerant is
confined to the machine room.
• High performing refrigerants, such as ammonia and hydrocar-
bons, may be used.
• Flooded evaporators in chillers for indirect systems can be used
to 100 % while thermostatic expansion valves used in direct sys-
tems may limit the effective evaporator surface down to some 70
%.
• There are less technical limitations for the distance between
cooled spaces and the machine room. Plastic piping may be used
for secondary working fluid circuits.

4
• A more even temperature distribution gives more uniform for-
mation of frost. Warm secondary fluid can be used for defrosting
which shortens the time needed for each defrosting, gives longer
intervals between defrosting periods and more even and favor-
able product temperatures.
• Indirect systems are usually designed and built as large systems
and variations in load can be evened out and shared within dif-
ferent parts of a large system.
• The energy consumption of a well designed indirect system may
over a year even be less than of a traditionally built direct system,
partly also due to more even temperature distribution at the cool-
ing object and fewer defrosts needed!

1.2.2 Drawbacks of indirect systems in

comparison with direct ones
• The investment cost for an indirect system is usually higher than
a direct (DX) system for the same application, as it introduces
the added cost for pump, heat exchanger and secondary fluid, al-
though the cost of primary refrigerant charge is much less.
• Added temperature difference is introduced, lowering the evapo-
ration temperature as well as the coefficient of performance
somewhat.
• Added pumping power, though it can be quite small compared
to the cooling capacity.
In spite of these drawbacks of indirect systems in comparison with
direct ones, the earlier mentioned advantages have been main rea-
sons for the move toward more indirect systems. (Aittomäki 2003,
Lindborg 2003, Lundqvist 2000, Mao and Hrnjak 1999, Melinder
2003b, Ure 2003, Yang 2001).

1.3 Choice of secondary working fluid

1.3.1 What is involved in choosing a
suitable secondary working fluid?
On the warm side, for the condenser, water can usually be used as
a secondary working fluid. Water is an excellent secondary working
fluid for air conditioning and other applications where tempera-
tures down to about +3°C are enough. The problem is to find flu-

5
ids for applications below 0°C, and the challenge is especially great
for freezer applications.
There is a need for a good knowledge of thermophysical proper-
ties and other aspects of available secondary working fluids in or-
der to make a good choice of fluid and to make reliable technical
calculations of indirect refrigeration and heat pump systems. This
thesis is meant to fill this need and to point out the requirements
of such fluids.

1.3.2 What requirements are to be

fulfilled by an ideal secondary
working fluid?
The fluid should possess good thermophysical properties, making
it possible to:
• Give sufficient freezing protection, yet not have higher concen-
tration than necessary, in order to benefit more from the good
properties of water.
• Transport large refrigeration capacity through pipes or tubes
with a certain volume flow and temperature change of the fluid.
• Give good heat transfer leading to small temperature differences
at heat transfer in evaporator and cooling object or heat source.
• Give small pressure drop for the system fluid flow in order to be
able to use a simple pump with small pumping power.
The fluid should also be:
• Environmentally acceptable, non-toxic, biodegradable. Not
flammable or dangerous to handle and be available at a reason-
able price.
• Compatible with materials used in the system so that it does not
give cause to problems in the plant due to corrosion.
No secondary working fluid fulfills all the mentioned require-
ments fully. We should find out which of the parameters that
are especially important in the application we are considering
and then try to choose the fluid that is best suited for that par-
ticular case (Granryd and Melinder 2005, Melinder 1997).

6
2 Basic thermophysical properties of
aqueous solutions – description
and comparison of fluids
2.1 Introduction
Of importance when comparing different secondary working flu-
ids (secondary refrigerants, heat transfer fluids, brines antifreezes)
for indirect systems are the following basic thermophysical proper-
ties: freezing point, density, viscosity, specific heat, and thermal
conductivity. Reliable data of these properties are also needed to
choose suitable fluid and to make technical calculations of an indi-
rect refrigeration or heat pump system. Many handbooks over the
decades have contained tables and/or charts with thermophysical
properties of various secondary fluids (ASHRAE Fundamentals
2001, Bäckstöm 1965, Carrier’s Handbook of Air Conditioning
System Design 1965, CRC Handbook of Chemistry and Physics
1988-89, DOW The Glycols 1974, Hofmann1959, Int. Critical Ta-
bles 1928-29, Plank 1988, Raznevic 1976, VDI Wärmeatlas 2002).
In order to mix the aqueous solutions to the right proportion this
chapter also presents a conversion chart and table of additive con-
centrations in weight-% and volume-% (Figure 2.2 and Table 2.2).
The chapter also gives some chemical information to determine
the density, such as, molar concentration, mixing masses and re-
fractive index of aqueous solutions. Thermal volume expansion is
also given as an aid to determine the size of expansion vessels.
The types of aqueous solutions included and symbols used are
given in Figure 2.1. A comparison of the fluid properties is here as
an example given for additive concentrations of the solutions giv-
ing the freezing point tF = -15ºC. This freezing point is often used
in heat pumps with ground or rock as heat source and they may be
used for many cooling applications.
A brief definition and description of each of the following basic
thermophysical properties and some other properties is also given
(CRC Handbook 1986-87, Geiringer 1962, Melinder 1997).

7
2.2 Freezing point, eutectic point and
additive concentration
The freezing point temperature, tF [°C], here means the tempera-
ture at which ice crystals begin to form in equilibrium if there is no
sub-cooling. The freezing point curve is then the solubility curve
of ice in the solution. It represents the temperatures at which ice
and solution are in equilibrium for various additive concentrations.
The freezing point should be below the lowest expected tempera-
ture and sufficiently (| 10K) below the normal operating tempera-
ture of the secondary working fluid so that the fluid without diffi-
culty can be pumped through the system.
0

-5

-10
Freezing point temp. [°C] .

-15
NaCl
GL
-20
PG
-25
EG
MgCl2 MA
-30 EA
-35
K2CO3
CaCl2
-40
NH3 KAC KFO
-45
LiCl MA
-50

-55
0 5 10 15 20 25 30 35 40 45 50 55 60
Additive concentration [weight-%]

EG: Ethylene glycol PG: Propylene glycol

EA: Ethyl alcohol MA: Methyl alcohol
GL: Glycerol NH3: Ammonia
K2CO3: Potassium carbonate CaCl2: Calcium chloride
MgCl2: Magnesium chloride NaCl: Sodium chloride
KAC: Potassium acetate KFO: Potassium formate
LiCl: Lithium chloride

Figure 2.1. Freezing point temperature, tF, of aqueous solutions

as function of additive concentration by weight, cA

8
Figure 2.1 gives freezing point curves of aqueous solutions as
function of the concentration. Note the abbreviations or symbols
used for the fluids in Figure 2.1. They are also used in several of
the following figures and tables in chapters 2 and 5. Some fluids
(NaCl, MgCl2, CaCl2, K2CO3) have vell defined eutectic tempera-
ture, tE [°C], as can be seen in the figure. The eutectic temperature
is the highest temperature at which full solidification can be
achieved (Fennema 1973).
The aqueous solutions do not usually freeze to solid ice when ice
crystals begin to form and mix with the liquid. The risk for serious
damage due to freezing up of a tube or heat exchanger is small
with aqueous solutions. Total freezing takes place first when the
temperature goes down to the eutectic point for the mixture.
(Granryd and Melinder 2005). ASHRAE Fundamentals (2005)
recommends at least 30% ethylene glycol to avoid damage of the
system. However, if we know the lowest fluid temperature in the
system, we may conclude that less additive concentration can be
chosen. Compare discussion in chapter 5.3.
Note also in Figure 2.1 that the temperature often rises sharply at
concentrations above the eutectic, as is the case with the chlorides
where the over-eutectic curve is the solubility curve of additive in
the solution.
Table 2.1. Chemical data, additive concentrations for tF = -15°C
and approximate eutectic points of aqueous solutions
Type of solution Chem. Mol. wt-% Appr. eutectic
symbol weight tF=-15°C temp./ conc.
Ethylene glycol (EG) C2H602 62.07 30.5 | -51°C/57%
Propylene glycol (PG) C 3H 80 2 76.09 32.9 | -50°C/60%
Ethyl alcohol (EA) C2H60 46.07 24.5 | -118°C/93.5%
Methyl alcohol (MA) CH40 32.03 20.0 | -116°C/87.7%
Glycerol (GL) C 3H 80 3 92.09 39.2 | -47°C/67%
Ammonia NH3 17.03 10.8 | -100°C/33%
Potassium carbonate K2CO3 138.20 27.0 | -37.5°C/40%
Calcium chloride CaCl2 110.99 17.9 | -50°C/30.5%
Magnesium chloride MgCl2 95.23 14.7 | -33°C/21.8%
Sodium chloride NaCl 58.44 18.8 | -21°C/23.4%
Potassium acetate (KAC) KC2O2H3 98.15 23.9 | -52°C/45%
Potassium formate (KFO) KCO2H 84.12 24.0 | -51°C/48%
Lithium chloride LiCl 42.40 11.2 | -75°C/25%

9
Some of the organic aqueous solutions such as propylene glycol do
not have a clear eutectic point but rather seem to enter a glassy
state at very low temperatures. This seems to be the case also with
some organic potassium salt mixtures.
Table 2.1 lists the aqueous solutions and gives the additive’s
chemical symbol, molecule weight and the approximate tempera-
ture and additive mass fraction or concentration of the eutectic so-
lutions.
Note that Table 2.1 lists additive concentrations for the freezing
point tF = -15°C as the rest of this chapter makes a comparison of
thermal properties for aqueous solutions with tF = -15°C.
In order to mix an aqueous solution to the right proportions it is
often the concentration by volume rather than by weight, volume
fraction rather than mass fraction of the additive, that is asked for
or needed. This is especially true when the additive is in liquid
form, as is the case with glycols and alcohols.
Figure2.2 is based on ASHRAE Handbook - Fundamentals (2005)
and Mellan (1977) and converts weight-% to volume-% for aque-
ous solutions of ethylene glycol, propylene glycol, ethyl alcohol
and glycerol. For ethylene glycol and glycerol the concentration in
volume-% is lower than in weight-%. For propylene glycol the val-
ues are about the same while for ethyl alcohol the concentration in
volume-% is up to 8 % higher than in weight-%. For detailed val-
ues, see Appendix 1, Table 2.2.

2.3 Density, ways of measuring density,

mixing proportions and thermal
volume expansion
2.3.1 Density
Density of a homogenous material is the concentration of matter,
measured by mass per unit volume. Correct density, U [kg/m3],
values of a secondary working fluid are important for several rea-
sons. The concentration of a known solution is often measured
with gas chromatography or refractive index. It may also be de-
termineded by checking the density with an aerometer (densime-
ter) or by weighing a known volume of the fluid or some other
method. Most methods are today usually performed at 20°C.

10
 100

90

80
Additive conc. [volume-%] .

70

60

50

40

30 Ethylene glycol, EG
Propylene glycol, PG
20 Ethyl alcohol, EA
Glycerol, GL
10
wt-% = vol-%
0
0 10 20 30 40 50 60 70 80 90 100
Additive conc. [weight-%]

Figure 2.2. Conversion chart of additive concentration

Figure 2.3 gives density curves of aqueous solutions at 20°C as

function of the additive concentration. The increase of density
with additive concentration is rapid for the salts while the increase
for the glycols is quite slow. Only the density of alcohols and am-
monia is lower than that of water. The aerometer method to
measure density is not very accurate with glycols as their density is
near that of water and as inhibitors and other substances added to
a commercial product usually changes the density. Glycol concen-
trations may be determined more accurately by index of refraction
or gas chromatography. See also Table 2.3 in Appendix.

2.3.2 Ways of measuring density, index of

refraction and mixing proportions
The refractive index, also called index of refraction, n, is defined
as the ratio of the velocity of light in a vacuum to the velocity of
light of a certain wave length in a given substance. The index of
refraction of a fluid is then the measure of refraction experienced
by a ray as it passes through a refracting interface, a surface sepa-
rating two media of different densities.

11
 1400
EG
PG
1350 EA
MA KFO
GL
1300 NH3 CaCl2 KFO
K2CO3 K2CO3
CaCl2
Density at 20°C [kg/m3] . 1250 MgCl2 KAC
NaCl
KAC
1200 KFO
LiCl MgCl2
NaCl
1150 LiCl
GL

1100
EG
1050 PG

1000
EA
950 MA
NH3
900
0 5 10 15 20 25 30 35 40 45 50 55 60

Additive concentration [weight-%]

Figure 2.3. Density at 20°C, U20°C, as function of additive con-

centration by weight, cA
1,42
EG MgCl2
PG CaCl2
1,41 EA
MA
GL K2CO3
1,40 LiCl
Index of refraction at 20°C [-] .

NH3
K2CO3
CaCl2
1,39 MgCl2 GL
NaCl
LiCl PG
1,38 NaCl
EG
1,37

1,36 EA

1,35 NH3
MA
1,34

1,33
0 5 10 15 20 25 30 35 40 45 50 55 60
Additive concentration [weight-%]

Figure 2.4. Index of refraction at 20°C, n20°C, as function of ad-

ditive concentration by weight, cA

12
Figure 2.4 gives values of the index of refraction of aqueous solu-
tions at 20°C, n20°C, as function of the additive concentration. The
increase with additive concentration is largest with salts and small-
est with methyl alcohol and ammonia. As mentioned, refractive
index is a common way to measure concentrations of glycols. For
detailed values, see Table 2.4 in Appendix.
However, it should be possible to measure the additive concentra-
tion of any of these fluids with a “glycol meter” that has a scale for
refractive index.
Table 2.2 is given as an aid to mix additive and water in the right
proportions to get a certain freezing point. The table gives mixing
proportions and values of refractive index, n20°C, of each of the
aqueous solutions at 20°C, all as a function of the additive concen-
tration that corresponds to the freezing point tF = -15°C. Table
2.2 also lists molar concentration and is mainly constructed by in-
terpolation of values in CRC Handbook of Chemistry and Physics
(1986-87). Values are not available for the organic potassium salts.
For explanation of fluid symbols or abbreviations see Figure 2.1.

Table 2.2. Density, molar conc., mixing masses and refractive

index of aqueous solutions with conc. giving tF = -15°C.
Type of Additive Density Molar Additive Water Index of
additive Conc. at 20°C Conc. Mass Mass Refraction
CA U20°C M Cs Cw n20°C
wt-% kg/m³ kg-mol/l kg/m³ kg/m³ -
EG 30.46 1038.4 5.097 316.4 722.1 1.363
PG 32.90 1025.5 4.434 337.4 688.1 1.370
EA 24.55 962.4 5.128 236.3 726.2 1.350
MA 19.98 966.6 6.029 193.1 773.4 1.338
GL 39.17 1096.2 4.663 429.5 666.7 1.383
NH3 10.83 954.4 6.071 103.4 851.1 1.339
K2CO3 27.00 1264.7 2.473 341.8 922.9 1.379
CaCl2 17.92 1157.2 1.869 207.4 949.8 1.378
MgCl2 14.68 1125.2 1.736 165.3 959.9 1.371
NaCl 18.82 1138.3 3.665 214.2 924.10 1.366
LiCl 11.20 1062.9 2.809 119.1 943.8 1.357

2.3.3 Thermal volume expansion

The needed size of an expansion vessel can be determined by cal-
culating the coefficient of thermal volume expansion. The expan-
sion of a volume from V0 to VT over the temperature change 'T

13
can with the coefficient of thermal volume expansion E>.@be
expressed as:
VT = V0(1+E'T) (2.1)
The value of E is a function of temperature. It is convenient for
practical reasons to use the concept of thermal volume expan-
sion of a fluid, TVE [%], estimated from the freezing point, tF, to a
chosen temperature, t. It can be calculated with the relation
TVE = 100·(UF/U-1) (2.2)
where UF is the density at the freezing point, tF, and U is the den-
sity at a chosen temperature, t.
The largest expected volume change, 'V [m3], from the freezing
point, tF, to a chosen temperature, t, is then for a volume V [m3]:
'V = V·(TVE/100) (2.3)

Figure 2.5 gives as an example TVE curves for concentrations of

the aqueous solutions giving tF = -15ºC and water (Compare Table
2.1). The figure shows that TVE from the freezing point tF = -15ºC
to t = 30°C for all the aqueous solutions are between 1% and 2%.
The TVE from the freezing point tF = -15ºC to t = 100°C is for
glycols about 6%. The volume change between any two tempera-
tures in the chart can be estimated by noting the values between
these temperatures. For detailed values, see Table 2.5 in Appendix.
Thermal volume expansion is of interest for designers of heat
pump and refrigeration systems as an aid to determine the size of
expansion vessels.

2.4 Specific heat and volumetric heat

capacity
The specific heat of a substance is simply the quantity of thermal
energy (in J) required to raise the temperature of one unit mass (1
kg) of a substance one temperature unit (1 K). A high value of a
fluid’s specific heat, cp [J/kgK], its thermal capacity, is favorable as
it affects mass flow rate as well as the heat transfer coefficient.

14
 2,2
EG
2,0 PG
EA
MA EA
1,8
GL PG
Thermal volume exp. [%]

NH3 NaCl
1,6 K2CO3 GL
CaCl2 K2CO3 EG
1,4 NaCl
KAC KAC
CaCl2
1,2 KFO MA
LiCl KFO
H2O NH3
1,0 tF
LiCl
0,8
H2O
0,6

0,4

0,2

0,0 LiCl LiCl

-15 -10 -5 0 5 10 15 20 25 30 35 40
Temperature [°C]

Figure 2.5. Thermal volume expansion, TVE, (tF = -15ºC) as

function of temperature

4500

4250

4000

3750
Specific heat [J/kgK] .

3500
EG
3250 PG
EA
MA
3000 GL
NH3
2750 K2CO3
CaCl2
NaCl
2500 KAC
KFO
2250 LiCl
H2O
tF
2000
-20 -10 0 10 20 30
Temperature [°C]

Figure 2.6. Specific heat of aqueous solutions (tF = -15ºC) as

function of temperature

15
Figure 2.6 gives specific heat for concentrations giving tF = -15ºC.
Note that the cp-values for alcohols and ammonia are almost the
same as for water while the cp-values for salts are 20-30% lower.
An added reason for the importance of a high specific heat is that
the volumetric heat capacity, U cp [kJ/m3K], has a direct influ-
ence on the volume flow rate, V , needed for a given heating
power, Q , and temperature change, 't. The simple relation can be
written:
Q = V U cp't (2.4)
The volumetric heat capacity is then an index of the ability of a
fluid to convey heat directly, by its motion, to a given destination.
Note in Figure 2.7 that the volumetric heat capacity, U cp of water
and all aqueous solutions are within 15%, ethyl alcohol having
highest U cp-value besides water. Glycerol, calcium chloride and the
other salts have the lowest values.

2.5 Thermal conductivity

The thermal conductivity, k, is defined by Fourier’s law for heat
conduction (eqn. 2.5) as the negative of the rate of heat flux and
the temperature gradient dT/dx over the thickness x and, the cross
section area A. The relation is written
dQ/dt = -kAdT/dx (2.5)
where the negative sign indicates that heat flows in the direction of
decreasing temperatures.
A high value of thermal conductivity, k [W/mK], of a fluid is de-
sirable as it contributes to good heat transfer and thereby decreases
the temperature difference between fluid and tube wall in the
evaporator and heat exchangers.
Figure 2.8 gives thermal conductivity curves for concentrations of
the aqueous solutions giving tF = -15ºC. The chloride salts and po-
tassium formate have quite high values, close to water and about
30% higher values than propylene glycol, ethyl alcohol and glyc-
erol.

16
 4250

4000
Vol. heat capacity [kJ/m3K] .

3750
EG
PG
EA
MA
3500 GL
NH3
K2CO3
CaCl2
NaCl
3250 KAC
KFO
LiCl
H2O
tF
3000
-20 -10 0 10 20 30
Temperature [°C]

Figure 2.7. Volumetric heat capacity of aqueous solutions (tF = -

15ºC) as function of temperature

0,60

0,50
Thermal conductivity [W/mK]

0,40
EG
PG
0,30 EA
MA
GL
.

NH3
0,20 K2CO3
CaCl2
NaCl
KAC
0,10 KFO
LiCl
H2O
tF
0,00
-20 -10 0 10 20 30
Temperature [°C]

Figure 2.8. Thermal conductivity (tF = -15ºC) as function of

temperature

17
2.6 Dynamic and kinematic viscosity
It is important to determine the absolute or dynamic viscosity, P
[mNs/m2], and/or the kinematic viscosity, Q = P/U [mm2/s], at
the operating temperature of a secondary fluid. The absolute vis-
cosity P or internal friction of a fluid is defined by the following
relation
W = P˜du/dy (2.6)
where W is the shearing stress at the surface of contact and du/dy
is the velocity gradient. The viscosity should be as low as possible
as it directly influences the needed pumping power and the heat
transfer.
which determines the type of flow that will prevail. Here w [m/s]
is the fluid velocity and d [m] is the inner tube diameter.
The viscosity value is used when calculating the Reynolds number,
Re,
Re = wd/Q = wdU/P (2.7)
Figure 2.9 gives as an example kinematic viscosity for concentra-
tions giving tF = -15ºC. The values differ very much between the
aqueous solutions and the viscosity increases sharply at low tem-
peratures, especially with propylene glycol, ethyl alcohol and glyc-
erol, even more so at higher additive concentrations and lower
temperatures.

2.7 Boiling point

A low boiling point temperature, tB [°C] of the secondary fluid
can create problems in a system, especially at higher temperatures.
Evaporation of fluid can also take place at lower temperatures than
the boiling point if the fluid is exposed to atmospheric air.
High concentrations of ammonia and alcohols give low boiling
point temperature, tB, and flash point, tFL, which may make the
fluid flammable. Ethyl alcohol solutions should in many applica-
tions not exceed 30 wt%, mainly because of low flash point. Figure
2.10 gives boiling point, tB [°C], as function of the additive concen-
tration. Values mainly from Timmermans (1960), ASHRAE Fun-
damentals (2001) and Plank (1988). See further Melinder (1998a).

18
 16
EG
GL EA PG
PG
14 EA
MA
Kinem. visc. [mm2/s = cSt] .

GL
12 NH3
K2CO3
10 CaCl2
NaCl
EG KAC
8 KFO
LiCl
MA H2O
6 tF
KAC
K2CO3
4 xxCl
KFO

2
H2O

0
-20 -10 0 10 20 30
Temperature [°C]

Figure 2.9. Kinematic viscosity (tF = -15ºC) as function of tem-

perature
120
MgCl2 CaCl2 KAC
115
Boiling point temp. at 1 atm [°C] .

110 PG
K2CO3 EG
NaCl GL
105
EG
100 PG
EA
MA
95 GL
NH3
K2CO3
90 CaCl2
MgCl2
NaCl
85 KAC
EA KFO
NH3
80 MA

75
0 5 10 15 20 25 30 35 40 45 50 55 60
Additive concentration, [weight-%]

Figure 2.10. Boiling point temperature, tB, of aqueous solutions

as function of additive concentration by weight, cA

19
In view of the flammability risk when using ethyl alcohol in indi-
rect systems, safety regulations in Sweden rewquire that the highest
temperature in cooling systems has to be at least 5ºC below the
flash point for the current ethyl alcohol (ethanol) concentration. If
the mass concentration is 30% or more the safety regulations are
quite strict. Figure 2.11 gives the flash point, tFL [°C], as function of
the additive concentration for ethyl alcohol solutions (Svensk Kyl-
norm, 1992).

2.8 Surface tension

The surface tension, J [mN/m], of a fluid is the force per unit
length that strives to keep the surface as small as possible. A low
surface tension increases the tendency of the solution to leak out
and increases the risk of foaming in the system and of cavitation in
the pump. Figure 2.12 gives surface tension as function of the ad-
ditive concentration. Alcohols and glycols have rather low surface
tension. Surface tension decreases with higher temperature as can
be seen from values of water (at 0%). Values of surface tension are
obtained among other sources from American Inst. of Physics
(1927), CRC Handbook of Chemistry and Physics (1988-89),
DOW - The Glycols (1974) and Int. Critical Tables (1928-29). See
further Melinder (1998a).

2.9 Conclusions
A brief definition and description has in this chapter been given of
each of the basic thermophysical properties and some other prop-
erties. A comparison of various fluids has been made for concen-
trations giving the freezing point, tF = -15ºC.
It is evident that no secondary working fluid is ideal for all applica-
tions and we should look for the best one for each type of system
and application. It is also essential to consider company informa-
tion sheets of the commercial products.
In chapter 5 a comparison is made of different freezing points or
concentrations of two of these aqueous solutions to see what im-
pact the concentration has on the thermal behavior of the fluid.

20
 50
48 15 20 25 30 35
Flash point temperature [°C] .

[volume-%]
46
44
42
40
38
36
34
32
30
10 12 14 16 18 20 22 24 26 28 30
Additive concentration [weight-%]

Figure 2.11. Flash point temperature, tFL, of aqueous solutions of ethyl

alcohol as function of additive concentration

100
LiCl EG
90 PG
EA
MgCl2 CaCl2 MA
Surface tension at 25°C [mN/m] .

GL
80 NaCl K2CO3 NH3
0°C CaCl2
GL MgCl2
70 NaCl
50°C K2CO3
KAC
PG 100°C
60 100°C 50°C
NH3 0°C
50
EG
MA
40
EA
30

20
0 5 10 15 20 25 30 35 40 45 50 55 60
Additive concentration, [weight-%]

Figure 2.12. Surface tension, J, of aqueous solutions as function

of additive concentration by weight, cA

21
2.10 Nomenclature
The following nomenclature is used in chapter 2.
A cross section area [m2]
cA additive concentration by weight [weight-%]
cp specific heat [J/kg K]
d tube diameter [m]
dT/dx temperature gradient [ - ]
du/dy velocity gradient [ - ]
k thermal conductivity [W/m K]
Q cooling capacity [W]
Re Reynolds number; Re = (w˜d)/Q [ - ]
t, T temperature [°C]
tB boiling point temperature [°C]
tE eutectic temperature [°C]
tF freezing point temperature [°C]
tFL flash point temperature [°C]
TVE thermal volume expansion [%]
V volume [m3]
V volume flow [m3/s]
w fluid flow velocity [m/s]
E coefficient of thermal volume expansion >.@
't, 'T temperature change [K]
'V volume change [m3]
J surface tension [mN/m]
P dynamic viscosity; P = Q˜U [Pa˜s]
Q kinematic viscosity [m2/s]
U density [kg/m3]
U cp volumetric heat capacity [kJ/m3K]
W shearing stress at surface of contact [Pa]

22
3 S e l e c t i o n o f p r o p e r t y va l u e s
of aqueous solutions

3.1 Summary
One main purpose of this chapter is to summarize the review work
reported on in the Engineering Licentiate Thesis: Thermophysical
properties of liquid secondary refrigerants, A Critical Review on Lit-
erature References and Laboratory Measurements (Melinder 1998a).
That material forms the basis for the charts and tables used in the
IIR-publication Thermophysical properties of liquid secondary re-
frigerants (Melinder, 1997). An additional purpose of this chapter
is to report on more recent review work done, including two addi-
tional fluids not covered in Melinder (1998a).
The main intention of this work has been to select thermophysical
property data with good or acceptable technical accuracy of a
number of aqueous solutions that can be used by the refrigeration
and heat pump industry, rather than focusing on a limited number
of property values or scientifically accurate measuring techniques.
Hence, a thorough literature search was made to find the most re-
liable property values of aqueous solutions for choice of secondary
working fluid and technical calculations of refrigeration and heat
pump systems. Detailed literature references are given for thermo-
physical properties of each aqueous solution. Some laboratory
measurements were made of most of the fluids types when litera-
ture values were incomplete or deemed unreliable. Methods used
are briefly described and results are given.
The aqueous solutions, without other additives, presented in
Melinder (1998a) and Melinder (1997) are ethylene and propylene
glycol, ethyl and methyl alcohol, glycerol, ammonia, potassium
carbonate, calcium, magnesium and sodium chlorides as well as
potassium acetate. The types of aqueous solutions that have been
added in this thesis */ are potassium formate and lithium chloride.
*/ A commercial product based on betaine has been introduced
on the market and in conference papers but property values from
other sources are lacking (Jokinen 2003, Jokinen, Willems 2004).
Thermophysical property data given in Melinder (1997) are in the
Melinder (1998a) compared with measured values and other litera-

23
ture reference data. A detailed list with reference data is given as
well as charts with freezing point, density, viscosity, specific heat
and thermal conductivity. For boiling point, and surface tension
the information is limited to the data list. A brief discussion fol-
lowing each chart give main sources for the tables and charts in
Melinder (1997) deviating literature values and other comments
that may be of interest. A few of the charts show that some values
in handbooks for the refrigeration industry deviate much from
other basic literature sources.
Examples of this are viscosity of aqueous solutions of ethyl alco-
hol, specific heat of ammonia, thermal conductivity of sodium
chloride, the glycols and alcohols. Viscosity measurements by au-
thor of potassium carbonate and calcium chloride (down to about
-20°C) differ much from values in refrigeration handbooks at
lower temperatures but are supported by laboratory loop tests at
KTH and values from basic sources. See further Melinder (1998a)
and Melinder (1998b).
The agreement in many of the charts is satisfactory from a techni-
cal standpoint. However, some of the charts indicate that more re-
search and measurements would be of value to ensure that prop-
erty values are sufficiently correct, especially at low temperatures
wher little data is found at temperatures below 0°C.
It is good to keep in mind that the commercial products on the
market have corrosion inhibitors and or other added substances
that will change property values to some extent.

3.2 Summary of critical evaluation of

thermo-physical property values
given in Tech. Lic. Thesis
A main purpose of this report is as mentioned to provide a critical
review on the literature references and laboratory measurements
that form the basis for charts and tables in Melinder (1997).
A detailed list of the sources of literature references and laboratory
measurements for each thermophysical property is given for each
aqueous solution. Many of these sources are introduced in charts
from Melinder (1997) and observations and an evaluation is made.

24
3.2.1 List of literature references for the
evaluation of property values
Literature sources and other references for thermophysical prop-
erty values are in a – b listed with numbers and given in the same
order as in Melinder 1997 and Melinder 1998a, references in c.
only in Melinder 1998a. Literature sources and other references
added in this Thesis are listed with numbers in d. In this chapter,
these references are usually only marked with the number xx or
[xx].

a. Main literature sources of aqueous solutions (Ref. 1 – 6)

1. CRC Handbook of Chemistry and Physics, 67th edition (1986-87)
2. Dow, Organic Chemicals / The Glycols (1974)
3. International Critical Tables of Numerical Data, Physics, Chemistry
and Technology (1928-29)
4. Landolt-Börnstein; Eigenschaften der Materie in Ihren
Aggregatzuständen (1960-1971)
5. Landolt-Börnstein; Numerical Data and Functional Relationships in
Science and Technology; Vol. 1b (1977)
6. The Physico-chemical Constants of Binary Systems in Concen-
trated Solutions, Vol. 3 - 4; J. Timmermans (1960)

b. Additional literature sources and references (Ref. 7 – 42)

7. Akzo Chemie, ”Hoesch Pa 9 rot” company data (1978)
8. American Institute of Physics Handbook, 3rd edition (1972)
9. ASHRAE Handbook, Fundamentals (1993)
10. Bayer, ”Baysilone-Fluids” company data (1993)
11. Carrier, Handbook of Air Conditioning System Design, (1965)
12. Chemical Engineering, Vol. 1, J. M. Coulson, J F Richardson
(1960)
13. Chevron Chemical Co. (Cryotech); ”Chevron Heat Transfer Fluid”
company data (1991)
14. Development of Improved Secondary Refrigerants; Int. Institute of
Refrigeration (U K), S. F. Pearson (1993)
15. Die Kältemaschine, M. Hirsch (1932)
16. Dow Europe S.A., ”Syltherm” company data
17. Handbook of Heat Transfer; W. M. Rohsenow, J. P. Hartnett
(1973)

25
18. Handbook of Heat Transfer Media; P. L. Geiringer (1962)
19. Handbook of Refrigerating Engineering, Vol. 1, Fundamentals; W.
R. Woolrich (1965)
20. Handbook of Thermodynamic Tables and Charts; K. Raznjevic
(1976)
21. Handbuch der Kältetechnik, VI/B; R. Plank, (1988)
22. Heat transmission, W. H. Mc Adams (1954)
23. Hoechst, ”Antifrogen KA” company data (1995)
24. Kältemaschinen Regeln (1958)
25. Kältetechnik (”Kylteknikern”), M. Bäckström, E. Emblik (1965)
26. Kältetechnik, Zeitschrift für das gesamte Gebiet der
Kälteerzeugung und Kälteanwendung, Heft 4, 1950
27. Köldbärare för värmepumptillämpningar. Studier av termodyna-
miska egenskaper och korrosion; Å. Melinder; Report R114:1985
to the Swedish Council for Building Research (1985)
28. Köldbärare för värmepumptillämpningar. Termodynamiska egen-
skaper och korrosion för kaliumkarbonat samt jämförelse med
andra köldbärare; Å. Melinder, J. Berendson, E. Granryd, B.
Kyrk; Report R18:1989 to the Swedish Council for Building Re-
search (1989)
29. Lange’s Handbook of Chemistry, 13th edition (1985)
30. Macmillan’s Chemical and Physical Data (1992)
31. Mellor’s Comprehensive Treatise on Inorganic and Theoretical
Chemistry, Vol. 2, Suppl. 3, Part 2 (1963)
32. Secondary Heat Transfer Systems and the Application of a New
Hydrofluoroether; G. Sherwood, The 1995 International CFC and
Halon Alternatives Conference (1995)
33. Secondary Refrigerants for Heat Pumps and Low Temperature Re-
frigeration. A comparison of thermodynamic properties of aque-
ous solutions and non-aqueous liquids; Å. Melinder, E. Granryd;
Trita-REFR Report No 92/6, ISSN 1102-0245 (1992)
34. Secondary Refrigerants for Indirect Refrigeration and Heat Pump
Systems; E. Granryd, Å. Melinder; Scanref No. 4, 1994
35. Thermodynamic Properties for Secondary Refrigerants - Tables
and Charts; Å. Melinder, Handbook No. 12 of the Swedish Soci-
ety of Refrigeration (1994)
36. Thermo-physical properties of liquid secondary refrigerants. Ta-
bles and Diagrams for the Refrigeration Industry, Melinder, Å.,
IIF/IIR, Paris (1997)
37. Tyforop; ”Tyfoxit” company data (1995)

26
38. Verfahren der Kälteerzeugung und Grundlagen der
Wärmeübertragung, E. Hofmann (1959)
39. VDI Heat Atlas (1993) (translation of VDI Wärmeatlas 6-th edi-
tion, 1991)
40. Wärmelehre und Wärmewirtschaft in Einzeldarstellungen; F.
Bosnjakovic (1961)
41. Zeitschrift für Die Gesamte Kälte-Industrie (certain issues from
1918 - 1943)
42. 3M, ”HFE-7100” company data (1996)

c. References of phase changing secondary fluids (Ref. 43 –

49)
43. Ammonia, carbon dioxide and water - the commercial refrigerants
of the future?, M. Kauffeld, DTI Energy, Refrigeration and Heat
Pump Technology (1995)
44. Binary Ice as a secondary refrigerant; J. Paul; 19th Int Congress of
Refrigeration, Vol.4b (1995)
45. Ice slurry based district cooling systems; C. V. Snoek; D. Hamp-
ton; Proceedings, Int District Heating and Cooling Ass. 8th Ann.
Cooling Conference, Tulsa, Oklahoma (1993)
46. Pressure drop characteristics of ice - water flows; C. V. Snoek, S.
U. Joanis,R. V. Gupta; Int. Symp. on Two-Phase Flow Modelling
and Experimentation, Rome (1995)
47. Properties of ice slurry, P. W. Egolf, et al; Applications for Natu-
ral Refrigerants, Aarhus (1996)
48. Recent advances in CO2 refrigeration; J. Pettersen et al; 19th Int.
Congress of Refrigeration; Vol.4b (1995)
49. Thermal study of an ice slurry used as refrigerant in a cooling
loop; O. Bel et al; Applications for Natural Refrigerants, Aarhus
(1996)

d. References added in Melinder 1998a (Ref. 50 – 63)

50. Chemical Engineers Handbook, J.H. Perry (1950)
51. Data Book of the Viscosity of Liquid, D. S. Viswanath; G. Natara-
jan (1989)
52. Handbuch der Kältetechnik, VII; R. Plank, (1959)
53. Heat Transfer and Pressure drop of Liquid Secondary refrigerants
- Experiments and theory; Engineering Thesis work, Olivia
Florian (1997)

27
54. Int. Journal of Thermophysics, Vol. 8, p.351 (1987) and Vol. 15,
p.165 (1994)
55. Internal report of measurements at KTH for ”Kylteknikern” [25];
(1957)
56. Internal report with measurements at Umeå university primarily of
specific heat and thermal conductivity (1996)
57. MIDAS Database for Thermophysical Properties, Excerpt from R.
Krauss, ITT, Stuttgart University, Germany (1995)
58. Refrigeration, Air Conditioning, and Cold Storage, R. C. Gunther
(1957)
59. Tables of Physical and Chemical Constants, G. C. Kaye, T. H.
Laby (1986)
60. Thermal Conductivity, Vol. 2, R. P. Tye (1969)
61. Thermophysical Properties of Matter, Volumes 3; Thermal Con-
ductivity. Nonmetallic liquids and gases; Y. S. Touloukian et al
(1970)
62. Thermophysical Properties of Matter, Volumes 6; Specific heat.
Nonmetallic liquids and gases; Y. S. Touloukian et al (1970)
63. Wärme und Kältetechnik, Zeitschrift für Klimatechnik,
Trockentechnik, Wärme- und Schallschutztechnik, February 1941

e. References added in this thesis (Ref. 64 – 88)

64. Aittiomäki, A., Lahti, A.., 1997, Potassium formate as a second re-
frigerant, Int. Journal of Refrigeration, vol. 20, No 4, p.276-282.
65. Andersson, O., 1996, 2002, Internal reports of measurements of
thermal conductivity and volumetric heat capacity with transcient
hot wire technique at Umeå univ.
66. Clariant, 1998, ”Antifrogen KF”, company data
67. Conde,M.R., 2004, Properties of aqueous solutions of lithium and
calcium chlorides: formulations for use in air conditioning
equipment design, Int. Journal of Thermal Sciences, Volume 43,
p.367-383.
68. -
69. Evenmo, K., 1998, A New Secondary Fluid, Natural Working Flu-
ids’98, IIR – Gustav Lorentzen Conf., Oslo.
70. Evenmo, K., 1998, Personal communication
71. Fransson, Å., 1996, 2002, Internal reports of measurements of
specific heat with differential scanning calometry at Umeå univ.
72. Gmelins Handbuch der Anorganischer Chemie, Kalium, 22, 1937.
73. Gmelins Handbuch der Anorganischer Chemie, Lithium, 20, 1960.

28
74. Int. Journal of Thermophysics, Vol. 8, p.351, 1987
75. Int. Journal of Thermophysics, Vol. 15, p.165, 1994
76. Jokinen, J., Betaine Based Heat Transfer Fluid as a Solution for
Toxicity and Corrosion Problems in Heating and Cooling Sys-
tems, 21st Int. Congress of Refrigeration, IIF/IIR, Washington
D.C.
77. Kemira, 1998-99, ”Freezium” company data
78. Melinder, Å., 1983-1999, Internal reports on measurements of vis-
cosity at KTH with falling ball viscosimeter. (See Appendix).
79. Melinder, Å. 1998, Thermophysical properties of liquid secondary
refrigerants, A Critical Review on Literature References and
Laboratory Measurements (Tekn. Lic. Thesis), KTH, Stockholm.
80. Melinder, Å, 1998, A Critical Review on Thermophysical Proper-
ties of Liquid Secondary Refrigerants, Natural Working Fluids’98,
IIR – Gustav Lorentzen Conf., Oslo.
81. Mellan, I., 1977, Industrial Solvents Handbook, New Jersey:
Noyes Data Corporation, 133 p
82. Mellor’s Comprehensive Treatise on Inorganic and Theoretical
Chemistry, Vol. 2, Suppl. 3, Part 2 (1963)
83. Stakelbeck, H., Plank, R., 1929, Ueber die Zähigkeit von
Chlornatrium-, Chlorkalzium- und Chlormagnesium-lösungen in
Abhägigkeit von Temperatur and Konzentration, Zeitschrift für
Die Gesamte Kälte-Industrie, p.105-112.
84. Söhnel, O., Novotny, P., 1985, Densitites of solutions of inorganic
substances, Physical Science Data 22.
85. Tyforop, 1997, ”Tyfoxit F”, company data
86. Vervoerd, M., Liem, N., van Gerwen, R.J.M., 1999, Performance
of a frozen food display cabinet with potassium formate as secon-
dary refrigerant, 20th Int. Congress of Refrigeration, IIF/IIR, Syd-
ney.
87. Wimby, J.M., Berntsson, T.S., 1994, Viscosity and density of
aqueous solutions of LiBr, LiCl, ZnBr2, CaCl2 and LiNO3”, 1. Sin-
gle salt solutions, J. Chem. Engineering Data 39, p.68-72.
88.Yara formates - Coolants – “Hycool”, 1997-2005, company data
(from Hydro)

29
3.2.2 Reference details and symbols used
The IIR-publication (Melinder 1997) contains numerous charts
and tables with basic thermophysical properties of secondary
working fluids. A detailed list of references for each fluid and
property is given in this chapter followed by a critical review with
charts. Reference details and symbols used in the lists are here
given.
The reference numbers from Melinder (1998a) are indicated with
bold letters and refer to the references listed in 3.5. Some of these
references indicate sources of measurements. These sources, or the
number of sources are then given in brackets [ ... ]. The numbers
of main references and sources of measurements used to make up
the tables and charts are underlined. The estimated significance of
each source of the reference material is indicated by the following
symbols given after the reference details:
(e,..) experimental values from one source (sometimes smoothed)
(x s,..) selected values from a number (x..) of sources
(t,..) table giving property values (usually in handbook)
(c,..) chart giving property values (usually in handbook)

(..,1r) good accuracy, rather complete, incl. low temperatures

(..,1) good accuracy, rather complete, not incl. low temp.
(..,1y) good accuracy, not complete (scattered data or few temp.)
(..,2r) less good accuracy, rather complete, incl. low temperatures
(..,2) less good accuracy, rather complete, not incl. low temp.
(..,2y) less good accuracy, not complete (scattered data or few
temp.)
(..,3r) poor accuracy, rather complete, incl. low temperatures
(..,3) poor accuracy, rather complete, not incl. low temp.
(..,3y) poor accuracy, not complete (scattered data or few temp.)
(..,4) whole or vital part clearly inaccurate
(..,..°) indicate that values are not given in same unit, conversion
needed
(..,..*) indicate product with corrosion inhibitor or other substance
added

30
3.3 Critical evaluation of some
thermo-physical property values
Some of the charts and discussions in Melinder (1998a) are cov-
ered in Melinder (1998b). These include density of ethylene glycol,
thermal conductivity of sodium chloride, freezing point and vis-
cosity of calcium chloride.
Certain property values of some aqueous soutions have been
changed especially at low temperatures since Melinder (1998a) was
written. A discussion of some of these changed property values
will be presented here, including specific heat of ethyl alcohol. The
viscosity values of calcium chloride at low temperatures are re-
viewed. The reference lists of these aqueous solutions are also
given.
Property values of aqueous soutions of lithium chloride and potas-
sium formate have been added since the publication of Melinder
(1998a) and are considered with a detailed reference list and charts
with discussion. Charts give for these two aqueous solutions the
basic thermophysical properties: freezing point; density; dynamic
viscosity; specific heat and thermal conductivity.

3.3.1 Critical evaluation of ethyl alcohol

A detailed list of references for each property of ethyl alcohol is
given while similar lists of glycols and other aqueous solutions can
be seen in Melinder (1998a) that in chapter 3 is marked [79].
Freezing point: 1. p.D-228 (t,1); 3. Vol. IV, p.262 (4 s,2°); 4. Vol.
2a, p.885 (3 s,1°); 6. Vol 4, p.181-2 [Ross, 1954 (e,2r); Lalande,
1934 (e,2r); Benjamin, 1932 (e,2r°); Tarasenkov, 1928 (e,2r); Pushin
and Glagoleva, 1923 (e,2°r); Pictet and Atschul, 1895 (e,2r);
Pickering, 1893 (e,2r)]; 11. p.4.26 (t,2y); 18. p.117 (c,2*y); 19.
p.371 (t,2°r); 21. p.670 (t,2y); 25. p.52 (c,3r); 29. p.10-75,76
(t,2°r); 39. p.Dd15-16 (t,2r).
Density: 1. p.D-228 (t,2); 3. Vol. III, p.116,117 (19 s,1); 5. p.129
(25 s,1); 6. Vol 4, p.183-7 [Harms, 1938 (e,1°y); Spells, 1936
(e,1y); Wiley and Harder, 1935(e,1°y); Brun, 1931 (e,1y); Kreitling,
1892 (e,1); Mendelejeff, 1869 (e,1)]; 11. p.4.41 (c,2°r); 18. p.115
(t,2°y); 20. p.72 (t,2y); 21. p.670 (t,2y); 25. p.52 (c,3r); 39.
p.Dd15-16 (t,2r).

31
Dynamic viscosity: 1. p.D-228 (t,2); 3. Vol. V, p.22 (2 s,1); 4.
Vol. 5a, p.367 (4 s,1); 6. Vol 4, p.190-3 [Spells, 1936 (e,2y);
Whightman et al, 1914 (e,2); Bingham et al, 1913 (e2); Dunstan et
al, 1909 (e,2); Noack, 1886 (e,2)]; 11. p.4.40 (c,4°r); 18. p.115
(t,3°y); 20. p.85 (t,2y); 21. p.670 (t,2y); 25. p.52 (c,3r); 39. p.Dd15-
16 (t,2r); 79 and App.1 [Melinder, 1992 (e,2r)].
Specific heat: 3. Vol. V, p.116 (7 s,2°); 5. p.310-11 (3 s,c,2); 6.
Vol 4, p.205 [Blacet et., 1931 (e,2y); Bose, 1907 (e,2)]; 11. p.4.42
(c,2°r); 20. p.65 (t,2y); 21. p.670? (t,2y); 25. p.52 (c,3r); 39.
p.Dd15-16 (t,2r); 71. [Fransson, 1996 (e,2y)].
Thermal conductivity: 4. Vol V/b, p.94 and Vol. T. 4b, p.567 (t,1r);
11. p.4.43 (c,2°r); 18. p.115 (c,2°); 20. p.96 (t,2y); 21. p.670 (t,2y);
25. p.52 (c,3r); 39. p.Dd15-16 (t,2r).
Boiling point: 6. Vol 4 p. 171-80 [Bosnakovic and Grumbt, 1931
(e,2), Brun, 1931 (e,2); Öman and Gunnelius, 1925 (e,2); 21.
p.668-70 (t,2y).
Surface tension: 1. p.F-32 (t,1°); 3. Vol. IV. p.467 (4 s,1); 6. Vol
4, p.194-7 [Teitelbaum et al, 1951 (t,1°); Valentine and Holes, 1938
(t,2°); Ernst et al, 1936 (t,2y)]; 8.p.2-209 (t,2y); 17. p.2.61 (t,1y);
30. p.30 (t,1°).
Specific heat of ethyl alcohol – water (See Fig 3.1)
A discussion of changed specific heat values of ethyl alcohol are
presented here followed by some comments about specific heat
values of ethylene and propylene glycols. Similar discussions are
made of the other basic properties of ethyl alcohol in Melinder
(1998a) on p. 27 - 31.
Discussion: Curves for high concentrations and low temperatures
in Fig 3.1 have been adjusted to fit measured values by Bose [3]
and [6], measured values of 47% and 50% by Fransson [71] (with
DSC, see also 3.3.2) that all show a steep slope below room tem-
perature. Values in [11] and [21] are in rather good agreement,
though curves for high concentrations and low temperatures in
[11] are less steep. Note also measured values of 10% from Umeå
univ. are in good agreement with measured values by Bose. Note
the curve for water, 0%, that is of right level, though a little steep.

32
 4600
% by w t

[3], [6]
4400 20 10
Bose 5 [69] 10%
0, 10, 20, 25
30, 40, 0 .
4200 30
50, 60% [69] 0%
o---o
4000
Specific heat [J/(kg·K)] .

40
3800
[21]
20, 40%
3600
50

3400
60 [71]
0, 10, 47%
3200 ++++++++

[71] 50%
3000 ++++++

2800 Ethyl alcohol

Specific heat
2600
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 3.1. Specific heat of ethyl alcohol - water

Specific heat of ethylene and propylene glycols

Values of specific heat for ethylene glycol were already adjusted for
high concentrations and low temperatures. Values for propylene
glycol have now been adjusted in about the same way as with eth-
ylene glycol and ethyl alcohol. See Melinder (1998a).

3.3.2 Critical evaluation of viscosity of

calcium chloride
A discussion of viscosity values of calcium chloride are presented
here to draw attention to the fact that less accurate values for low

33
temperatures still are used in main sources of literature values such
as ASHRAE Fundamentals, 2005 (Compare Melinder, 1998a).

Calcium chloride - water:

Dynamic viscosity: 1. p.D-225 (t,2); 3. Vol. V, p.14 (5 s,2°y); 4.
Vol. 5a, p.342 (3 s,1°y); 6. Vol 3, p.762-63 [Lyons and Riley, 1954
(e,2y); Stackelbeck and Plank, 1929 (e,2-4r); Hechler, 1904 (e,2°r);
9. p.18.2 (c2-4r); 11. p.4.32 (c,2-4°r); 15. p.531 (c,2-4°r); 21.
p.618 (t,2-3y); 24. p.86 (t,2-4r); 25. p.47 (c,3r); 38. p.454-5 (t,2-
4°r); 39. p.Dd10 (t,2-4r); 41. 1927, p.144 (t,2-3°); 1929, p.110
(t,2-4°r); 79 and App.1 [Melinder, 1983-1996 (e,2r)].

100

Calcium chloride
Dynamic viscosity

[6]
Stakelb.
Dynamic viscosity [mPa·s] .

& Plank,
21,8%
28,7%
---- [3]
10 Sprung
[1]
18,2%,
20°C
30,8%
o---o
[79], [App 1] % by w t
Melinder [21]
30
18, 22, 28% 20, 30%
28
x -- -- x 25

20
[4]
Buche 15
1927 10
5
- - 0
1
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2006

Figure 3.2. Dynamic viscosity of calcium chloride – water

34
Discussion of dynamic viscosity of calcium chloride - water

Curves from Chart C8b in [36] are based on [1] and measurements
made by Sprung (1876) [3], Buche (1927) [41] and author (see Ap-
pendix). Values from Stakelbeck & Plank [41] (1929, p.105-112)
used by ASHRAE Fundamentals, 2005 and most other publica-
tions are about 30% higher at -20°C than values from measure-
ments made by author [App.1].
The article in [41] explains that a constant used for the calculations
had to be extrapolated between -5.5°C and -25°C. Note that the
curves from [41] are very straight for temperatures below 0°C in a
linear-logarithmic representation. The low temperature values by
author are supported by tests made by Björn Kyrk in a laboratory
loop at KTH as described in [28] Melinder (1989), on p.61-63 in
[79] (Melinder 1998a), in Melinder (1998b) and on the following
pages.
The laboratory test rig by Kyrk was designed to measure heat
transfer and pressure drop of liquids flowing in an tube and the
liquid was circulated with a pump in a closed tube loop. The pres-
sure drop, 'p, over this section, that had a sharp inlet devise, to
ensure transition between turbulent and laminar flow at Reynolds
number, Re | 2300, was measured with a mercury manometer and
the velocity w with a flow meter. Density, U, and dynamic viscos-
ity, P, for the liquid were determined at the mean temperature of
the liquid [28].

Pressure drop results with calcium chloride - water

Pressure drop results from tests with a concentration of calcium
chloride - water, giving the freezing point -25°C, were represented
in charts where the friction coefficient f1 (= f/2) was plotted as a
function of the Reynolds number, Re. (Compare equations 5.4 and
5.5 in chapter 5.2). See Figures 3.3 and 3.4 and discussion below.
Note that in Figure 3.3 with values from [41] that transisition be-
tween laminar and turbulent flow takes place at lower Re-values
for lower temperatures (t = -10°C and -20°C). In Figure 3.4 with
values from author [79], transition takes place at approximately the
same Re-values for all temperatures.
Similar tests and observations were made with heat transfer meas-
urements and these are recorded on p.62-63 in Melinder (1998a).

35
Figure 3.3. Results of pressure drop tests of calcium chloride -
water with property values from Stakelbeck [41]

Figure 3.4. Results of pressure drop tests of calcium chloride -

water with property values from author [79]

36
These results show that the viscosity values presented by author
are basically correct and that values from [41] are too high at these
lower temperatures. See also Melinder (1998a), p. 59–65.

3.3.3 Critical evaluation of property

values of lithium chloride
Recently aqueous soutions of lithium chloride have been consid-
ered for use in refrigerating and airconditioning systems (Conde
2004, Melinder 2004, Wimby et al 1994). A list and detailed discus-
sion of property values of aqueous soutions of lithium chloride
that may be used at very low temperatures is here for the first time
presented in detail by author.

Lithium chloride - water

Freezing point: 1. p.D-235 (t,1y); 6. Vol 3, p.614-615 [Maran, 1956
(e,3r); Voskresenskaja & Yanatieva, 1937 (e,2r); Rodebush, 1918
(e,2y)]; 65. Andersson, 1996 (e,2y); 73. p.337 (t,2r) p.354-55 (4 s
,2°r).
Density: 1. p. D-235 (t,1y); 3. Vol. III, p.77 (51 s,1); 5. p.59 (29
s,1); 6. Vol 3, p.619 [Fontell, 1927 (e,1y); 73. p.344 (31 s,1); 84.
p.135 (t,2).
Dynamic viscosity: 1. p. D-235 (t,1y); 4. Vol. 5a, p.314-315 [Lan-
gyel & Tamaz, 1964 (s,1°y); 6. Vol 3, p.621-623 [Jacopetti, 1942
(e,2y); Alfimoff, 1917 (e,2°y); Green, 1908 (e,2°y)]; 73. p.349-51
(t,2°y); App.1. [Melinder, 1996 (e,2r)].
Specific heat: 5. p.263-264 (6 s,2°y); 6. Vol 3, p.627 [Jauch, 1921
(e,2°y); Tucker, 1920 (e,2°y); Green, 1908 (e,2°y)]; 71. [Fransson,
1996 (e,1y)];
Thermal conductivity: 4. Vol V/b, p.92 [Riedel, 1948 (t,2y)]; 6.
Vol 3, p.628 [Kapustinski & Ruzavin, 1955 (t,2y)]; 65. [Andersson,
1996 (e,1y)]; 67. Conde, 1967 [Uemura, (c,2y)]; 73. p.355 (e,2°y).
Boiling point: 6. Vol 3, p.614 [Johnston, 1906 (e,2); Gerlach, 1886
(e,2)].
Surface tension: 6. Vol 3, p.622 [De Block, 1925 (e,2); Lineberger,
1899 (e,2)]; 73. p.349 (t,2°y).

37
 Lithium chloride - water - Freezing point temp.
0

-5

-10

-15
Freezing point temperature, [°C] .

-20

-25

-30

-35

-40

-45

-50
Chosen values
-55 [1] CRC Handbook of Chem. & Physics
[6] Maran, 1956
-60 [6] Voskreneskaya, Yanatieva, 1937
[6] Rodebush, 1918
-65 [65] Andersson, 1996, 20%
[73] Gmelins Handbook, Lithium, 1960
-70
0 2 4 6 8 10 12 14 16 18 20 22 24 26
Additive concentration [weight-%]

Figure 3.5. Freezing point of lithium chloride – water

Discussion: The curve for the chosen values is in good agreement

with [1] (values given to 14%), Rodebush [6] (values given to 18%)
and fit measured values by Voskreneskaya and Yanatieva [6] rather
well. Note also the good agreement of the value for 20% obtained
from measurements of thermal conductivity at Umeå univ. [65]
(Compare Fig 3.9). Values by Maran [6] and in [73], p.336-37 are
for the “ice curve” and indicate a lower eutecticum (tE = -75qC)
than Voskreneskaya and Yanatieva [6] (tE = -66qC).

38
 1160
26
1150 w t-%
25
1140 24

1130
22
1120
[1] 20
1110 20°C
and 0% 18
1100
Ƒ Ƒ
Density [kg/m3] .

1090
[3], [73] 15
1080 2, 6, 10,
14, 18, 22, 14
1070
24, 26%
1060 x---x
10
1050 [5]
2, 5, 10, 15
1040 20, 25%
o o 6
1030
[6] 5
1020 0, 5.1,
[84] 15.2,
1010 5, 10, 15, 25.3% 2
20, 25% 6, 25, 50°C
1000 6
- - ¨ ¨ 0
990
-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60
Temperature [°C] Melinder 2007

Figure 3.6. Density of lithium chloride – water

Discussion: The curves for the chosen values for density are in
good agreement with values from many sources, some with values
down to 0qC in [1], [3], [5], [6], [73] and [84]. No values are given
for temperatures below 0qC. The curves follow the pattern of
other chlorides, such as calcium chloride, below 0qC (Melinder,
1998a).

39
 100

Dynamic viscosity [mPa·s] .

[6]
[6] Jacopetti
Alfimoff 18, 40°C
0°C - -
¨
10
[4]
15, 35°C
[App 1]
o o
Melinder
15, 20 % [1]
w t-%
+, x 24 20°C
20 Ƒ

15
10
5
0

1
-60 -50 -40 -30 -20 -10 0 10 20 30 40

Temperature [°C] Melinder 2007

Figure 3.7. Dynamic viscosity of lithium chloride – water

Discussion: The curves for the chosen values for viscosity are in
good agreement with values from [1] as well as measured values by
Langyel & Tamaz [4], Alfimoff and Jacopetti [6] down to 0qC.
However, only measured values by author [App.1] are given below
0qC, down to -15 to -20qC. The curves below -20qC are extended
much based on the freezing point values.

40
 4,4

w t-%
4,2 0

4,0
5

3,8
Specific heat [kJ/kgK] .

10
3,6

15
3,4
18

3,2 20
[71] 22
24
20, 23%
3,0 x,+ [5]
[6] 26,5°C
[6] Tucker Ƒ
2,8 Jauch 20°C
18°C o
¨
2,6
-60 -50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 3.8. Specific heat of lithium chloride – water

Discussion: The curves for the chosen values for specific heat are
in good agreement with values from 5 sources in [5] as well as val-
ues by Jauch and Tucker [6] all for room temperatures. However,
measured values by Fransson [71] for 20% and 23% are given
down to -45 and -50qC. The slope of the freezing point is esti-
mated and curves for lower additive concentrations can be drawn
to their freezing point values that are established in Figure 3.5.

41
 0,62

[67]
0,60 Uemura
w t-%
10, 30°C 0
10, 20%
0,58 ¨ ¨ 5
Thermal conductivity [W/mK] .

10
0,56 15
18
[65] 20
Andersson 22
0,54 24
20% [6]
x--x x Kapus-
0,52 tinskii
& Ruzavin
[4]
25°C
Riedel
0,50 o
20°C
0, 5, 10,
0,48 15, 18, 20,
22, 24%
Ƒ
0,46

0,44
-60 -50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 3.9. Thermal conductivity of lithium chloride – water

Discussion: The curves for the chosen values for specific heat are
in good agreement with values from Riedel [4] as well as values by
Kapustinski & Ruzavin [6] and Uemura [67]. However, measured
values by Andersson [65] for 20% are given down and passed the
freezing point at about -45qC. The slope of the freezing point is
estimated and curves for lower additive concentrations can as in
Figure 3.8 be drawn to their freezing point values that are estab-
lished in Figure 3.5.

42
3.3.4 Critical evaluation of property
values of potassium formate
Over a decade ago aqueous soutions of potassium formate were
considered for use in refrigerating systems as a secondary fluid
with very low viscosity at low temperatures (Aittiomäki, Lahti
1997, Evenmo 1998, Vervoerd et al 1999). A list and detailed dis-
cussion of property values of potassium formate is here presented
in detaile by author.
Very little data material was fond in the general literature search.
However, when doing this research there were four commercial
products providing data that seemed rather independent from each
other. Data for two of these products, Antifrofen KF and
Freezium, were presented as function of additive concentration,
the other two, Hycool and Tyfoxit F, were given only as function
of cooling limit. How could these data properly be compared and
combined? My approach was first to establish a freezing point
curve as there were some literature values to compare with. The
next step was to determine the additive concentration of the prod-
ucts from their density value at 20qC as that is the property value
that a company would find of high importance and could easily es-
tablish.

Potassium formate - water (not yet done)

Freezing point: 6. Vol 3, p.588 [Sidgwich and Gentle, 1922 (e,2r)];
66. (c,3*r); 69. (e,2r); 70. (e,3y); 77. (t,2r).
Density (and TVE): 6. Vol 3, p.588 [de Garcia 1920, (t,2y); 84.
p.199 (t,2y).
Dynamic viscosity: App.1 [Melinder, 1996 (e,2*r).
Specific heat: Only company data.
Thermal conductivity: 65. (c,2r).
Boiling point and Surface tension: Only company data.
In view of a strong interest from the industry for property values
of potassium formate I tried to find an acceptable way to make use
of information from the mentioned four commercial products. See
Figures 3.10-15 and the discussions.

43
 Potassium formate - Freezing point temp.
0
Chosen values
[70] Evenmo.
-5 [6] Sidgw ick
.
[53] Florian
-10 [77] Freezium
[66] Antifrogen KF
Freezing point temperature [°C] .

-15

-20

-25

-30

-35

-40

-45

-50

-55
0 5 10 15 20 25 30 35 40 45 50 55
Additive concentration [weight-%] Melinder 2007

Figure 3.10. Freezing point of potassium formate – water

Discussion: The curve for the chosen values is in good agreement

with measured values by Sidgwich and Gentle [6] as well as values
given by Evenmo [70] and Florian [53]. Note also the values by
two commercial products, Antifrogen KF [66] and Freezium [77].
The agreement with the other sources is not as good but accept-
able. The low freezing point indicated near 50% by [77] is not
supported by [66] nor by measurement results given by Evenmo
[69]. Below -52qC we may get a metastable phase that could turn
glassy state as may also be the case with propylene glycol at high
concentrations (See Melinder, 1998a, p.21).

44
 Potassium form ate - Freezing point tem p.
Com m ercial product concentrations
0
Chosen values
.
-5 [77] Freezium
[88] Hycool
-10 [85] Tyfoxit F
Freezing point temperature [°C] .

[66] Antifrogen KF
-15

-20

-25

-30

-35

-40

-45

-50 EA

-55
0 5 10 15 20 25 30 35 40 45 50
Additive concentration [w eight-%] Melinder 2007

Figure 3.11. Freezing point of some commercial products based

on potassium formate – water

Discussion: The curve for the chosen values (from Figure 3.10) is
in Figure 3.11 plotted with four commercial products, besides An-
tifrogen KF [66] and Freezium [77], also Hycol [88] and Tyfoxit F
[85] that both give cooling limit rather than freezing point. From
the figure it is obvoious that the additive concentrations chosen
for Tyfoxit F are unnecessarily high leading to poorer performance
of the fluid.
In each of Figures 3.12-15 the following sources appear and are
commented on in the discussions.
[66] Antifrogen KF -40; 43% [77] Freezium -40; 43%
[88] Hycool -40; 43,5% [85] Tyfoxit F -40; 49%
[66] Antifrogen KF -20; 30% [77] Freezium -20; 29%
[88] Hycool -20; 30% [85] Tyfoxit F -20; 40%
8 0 0

[77] Freezium -10; 19%

6 0 0

4 0 0
2 0 0

0 1 0 2 0 3 0 4 0

45
 1360
% by w t
1340 48
1320
45
1300
1280
40
1260
1240 35
1220
Density [kg/m3] .

1200 30

1180
25
1160
1140
20
1120 [84] from
1100 de Garcia,
1920, 21°C
1080 also in [6]
10
1060 Ƒ

1040 5
1020 .

0
1000
980
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

[66] Antifrogen KF -40; 43% [77] Freezium -40; 43%

[88] Hycool -40; 44% [85] Tyfoxit F -40; 49%
[66] Antifrogen KF -20; 30% [77] Freezium -20; 29%
[88] Hycool -20; 30% [85] Tyfoxit F -20; 40%
8 0 0

[77] Freezium -10; 19%

6 0 0

4 0 0
2 0 0

0 1 0 2 0 3 0 4 0

Figure 3.12. Density of potassium formate – water

Discussion: The only literature data found beside the commercial
product data are from de Garcia [6], [84]. Density values for the
commercial products put in the same figure and from this their
additive concentrations are estimated and used also in Figures 3.13
to 3.15.

46
 100
Dynamic viscosity [mPa·s] .

10

% by w t

[[App. 1] 48
Melinder, 45
Hycool 40
38, 44, 35
47% 30
20
+--+--+
10
0
1
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

[66] Antifrogen KF -40; 43% [77] Freezium -40; 43%

[88] Hycool -40; 44% [85] Tyfoxit F -40; 49%
[66] Antifrogen KF -20; 30% [77] Freezium -20; 29%
[88] Hycool -20; 30% [85] Tyfoxit F -20; 40%
8 0 0

[77] Freezium -10; 19%

6 0 0

4 0 0
2 0 0

0 1 0 2 0 3 0 4 0

Figure 3.13. Dynamic viscosity of potassium formate – water

Discussion: The only literature data found beside the commercial

product data are from measurements of three additive concentra-
tions of Hycool by author [Appendix 1]. The agreement is rather
good though Freezium -10 give too low values.

47
 4300
% by w t
4200 0

4100
5
4000 .

3900
10
3800

3700
Specific heat [J/kgK] .

3600
3500 20
3400

3300 25

3200
30
3100

3000 35

2900
40
2800
45
2700
2600 30

2500
-50 -40 -30 -20 -10 35 0 25 10 20 30 40
Temperature [°C] Melinder 2007

[66] Antifrogen KF -40; 43% [77] Freezium -40; 43%

[88] Hycool -40; 44% [85] Tyfoxit F -40; 49%
[66] Antifrogen KF -20; 30% [77] Freezium -20; 29%
[88] Hycool -20; 30% [85] Tyfoxit F -20; 40%
8 0 0

[77] Freezium -10; 19%

6 0 0

4 0 0
2 0 0

0 1 0 2 0 3 0 4 0

Figure 3.14. Specific heat of potassium formate – water

Discussion: The only data found are commercial product data. The
values of Tyfoxit F have a different slope and are higher that what
can be expected of the high additive concentrations. One reason
for this devition could be that the values are based on volumetric
heat capacity measurements with the transient hot-wire technique
(See 3.4.2 and also Melinder, 1998a, p.18 and 24).

48
 0,620
% by w t
0,610
0,600
0
0,590
0,580 5
10
0,570
Thermal conductivity [W/mK] .

0,560 20
[65]
0,550 Andersson 25
0,540 Hycool -25
30
+--+--+
0,530 35
0,520
40
0,510
0,500 45
48
0,490
0,480
0,470
0,460
0,450
0,440
0,430
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

[66] Antifrogen KF -40; 43% [77] Freezium -40; 43%

[88] Hycool -40; 44% [85] Tyfoxit F -40; 49%
[66] Antifrogen KF -20; 30% [77] Freezium -20; 29%
[88] Hycool -20; 30% [85] Tyfoxit F -20; 40%
8 0 0

[77] Freezium -10; 19%

6 0 0

4 0 0
2 0 0

0 1 0 2 0 3 0 4 0

Figure 3.15. Thermal conductivity of potassium formate – water

Discussion: The only data found beside the commercial product

data are measured values by Andersson [65] for Hycool -25 given
down and passed the freezing point at about -25qC. The higher
values below the freezing point are no doubt due to ice crystals.
The values of the commercial products have different slopes and
the chosen values take this into consideration.

49
3.4 Measurements of some properties
Some laboratory measurements of basic thermophysical properties
have as a background to this work been made at KTH since 1983
and also at Umeå university since 1996. These measurements were
mainly made when literature values were incomplete or seemed
unreliable. The different thermophysical properties are already
briefly described (Chapter 2). A brief description of the methods
and equipment used for these measurements is given here.

3.4.1 Measurements of viscosity at KTH

Measurements of viscosity were carried out with a Höppler Falling
ball Viscosimeter (Figure 3.16) attached to a refrigerating unit (See
Figure 3.17). The falling tube of glass of the viscosimeter was filled
with the sample of the liquid to be tested and a falling ball of glass.
The falling time of the ball between the two marking lines was
measured manually with a chronometer. A falling ball was chosen
that gives a falling time of 30 - 300 seconds. The time was meas-
ured a manually with a chronometer.

The calculation of the dynamic viscosity is done with the following

formula:
P = W˜(U1 - U2)˜K (3.1)
where P = dynamic viscosity (Pa˜s); W = average falling time of ball
(sec.); U1 = density of ball (kg/m3); U2 = density of liquid (kg/m3);
K = ball constant.
The set up for measurements before 2002 is described in Melinder
(1998a). The set up for measurements since 2002 is as follows.
Stable low temperatures were obtained in the following way: The
glass enclosed container around the falling tube was connected
with a circuit consisting of the heat exchanger of a refrigerating
unit, a small pump, an electric heater and a thermostat that could
be set to desired temperature. This whole circuit was filled with an
alcohol - water solution that was pumped around and deaired. Sta-
ble temperatures were obtained with a thermostat. The equipment
put a limit to about -20°C for the lowest temperature. The tem-
perature was measured with thermocouples connected to a re-
corder. In Figure 3.17 the viscosimeter is upside down ready for
start of a measurement.

50
Measurements and calculations
were carried out with almost all
aqueous solutions. The number
of measurements, mean tem-
perature, maximum tempera-
ture variation, density of liquid,
ball constant, ball density and
average falling time are also
given and the viscosity result is
recorded both as dynamic (ab-
solute) and kinematic viscosity.
Results for aqueous solutions
of ethyl alcohol, lithium chlo-
ride and potassium formate are
given in Appendix 1. Results
for other solutions are given in Figure 3.16. Höppler Falling
4.2.3 in Melinder (1998a). ball Viscosimeter.

Fig 3.17. Höppler viscosimeter and measurement equipment

51
3.4.2 Measurements of specific heat and
thermal conductivity at Umeå university
Some laboratory measurements were made by personnel at Umeå
university when literature values were incomplete or were deemed
unreliable. This univ. has advanced technology and equipment to
make accurate measurements.
Measurements of specific heat were carried out with a Perkin-
Elmer differential scanning calorimeter (DSC-2). The principle for
the measurements of specific heat, cp, is among else described in
Int. Journal of Thermophysics (1994) [75]. Temperature calibration
was made with standard materials at heating. The rate of cooling
was 5 K/minute. The specific heat data obtained for samples were
determined during cooling.
The transient hot-wire technique used to determine the thermal
conductivity and volumetric heat capacity was developed at the
Department of Experimental Physics of Umeå University and is
among else described in Int. Journal of Thermophysics (1987) [74].
Measurements of the thermal conductivity, k (or O), and the volu-
metric heat capacity, U cp, of samples of aqueous solutions were
made at atmospheric pressure in the temperature range of 200-320
K. During measurements the temperature was decreased at a con-
stant rate of approximately 0.01 K/s.
Common tap water was used for the test experiments. Samples
were heated up to 320 K and degassed before experiments. The
largest discrepancy between experimental and reference data for k
(or O) is less than 2%, for U cp, about 5% and for cp about 3%.
These results confirm that convection effects are reasonably small
in the methods used. Both experimental and smoothed data of
thermal conductivity and specific heat outlined can be used for dif-
ferent kinds of engineering calculations, but the U cp data obtained
should be used only as qualitative results.
Some measuring results with certain aqueous solutions of ethylene
glycol, glycerol, calcium chloride and potassium acetate are in
Melinder (1998a) given as charts of specific heat and thermal con-
ductivity. Approximate crystallization or freezing point tempera-
tures are indicated graphically as a drastic change in the value of
thermal conductivity (See Figure 3.9).

52
4 M a k i n g t h e r m o p hy s i c a l
properties of aqueous
s o l u t i o n s a va i l ab l e f o r
computer use

4.1 Introduction
As mentioned in earlier chapters, aqueous solutions are used in
various types of indirect refrigaeration and heat pump systems. Re-
liable basic thermo-physical data is needed for choice of suitable
fluid and for technical calculations of the system. This chapter de-
scribes how the selection of property values results in tables and
charts intended for the industry. The chapter also describes the
process to transform these property values to a material suitable
for computer treatment (Melinder 2006).

4.2 Results of selection of property

values of aqueous solutions using
ethyl alcohol as example
Based on the work described in the previous chapter, property -
values have been selected, generally for even concentrations or
mass fractions of the aqueous solutions examined. With the help
of a Matlab program, coefficients are generated from these prop-
erty values and together with polynomial equations this material
can be a useful tool for computer treatment. This is described in
4.3.
To see how this process is made, aqueous solutions of ethyl alco-
hol, one of the most versatile secondary working fluids, is used as
example. This choice of fluid can also be seen as a test of this
method, as the basic thermophysical properties of aqueous solu-
tions of ethyl alcohol seem to present more challenges that most
other fluids examined. This is true about the freezing point curve,
density, specific heat and viscosity. Hence, the choice of ethyl al-
cohol is a good test case for the equations used for computer
treatment.

53
The results for aqueous solutions of lithium chloride and potas-
sium formate, salts that may be used at quite low temperatures, are
also given as these types have been added since Melinder (1997)
and Melinder (1998a) were compiled.
Based on the work described in chapter 3, Table 4.1 gives basic
thermophysical properties of aqueous solutions of ethyl alcohol.
The following symbols are used:
tF freezing point [°C]
cA additive concentration [ wt-% ]
t fluid temperature [°C]
U density [kg/m³]
cp specific heat [J/kg K]
k thermal conductivity [W/m K]
P dynamic viscosity [mPa s]

Table 4.1. Basic thermophysical properties of ethyl alcohol

Chemical formula: C2H5OH; M = 46.07
tF cA t U cp k P
[°C] [wt-%] [°C] [kg/m³] [J/kgK] [W/mK] [mPas]
0 0 40 992.2 4177 0.631 0.653
30 995.7 4177 0.615 0.798
20 998.2 4181 0.598 1.002
10 999.7 4192 0.580 1.306
0 999.9 4218 0.561 1.792
-2.09 5 40 983.1 4235 0.588 0.77
30 986.6 4235 0.576 0.96
20 989.5 4240 0.562 1.24
10 990.8 4270 0.548 1.70
0 991.4 4325 0.532 2.45
-2.09 991.2 4340 0.528 2.70
-4.47 10 40 974.9 4270 0.554 0.90
30 978.7 4270 0.542 1.16
20 982.0 4280 0.529 1.53
10 983.8 4310 0.516 2.17
0 984.9 4360 0.503 3.25
-4.47 985.1 4390 0.498 4.20
-7.36 15 40 966.8 4320 0.518 1.03
30 971.2 4310 0.507 1.35
20 975.2 4320 0.496 1.84
10 978.0 4340 0.485 2.70
0 979.9 4370 0.474 4.20
-7.36 980.5 4410 0.466 6.5
(cont.)

54
Table 4.1. Basic thermophysical properties … (cont.)
tF cA t U cp k P
-10.92 20 40 958.8 4315 0.484 1.16
30 963.8 4310 0.475 1.55
20 968.8 4310 0.466 2.16
10 972.5 4330 0.457 3.22
0 975.7 4360 0.448 5.15
-10 977.8 4385 0.439 9.7
-10.9 978.0 4390 0.438 10.3
-15.45 25 40 950.5 4310 0.454 1.27
30 957.0 4300 0.446 1.73
20 962.0 4290 0.438 2.45
10 967.0 4280 0.430 3.71
0 971.0 4270 0.422 6.10
-10 974.5 4260 0.414 11
-15.45 975.5 4254 0.410 17
-20.47 30 40 940.8 4240 0.423 1.36
30 948.0 4240 0.417 1.86
20 954.0 4230 0.411 2.68
10 960.0 4200 0.405 4.08
0 965.4 4170 0.398 6.80
-10 969.5 4120 0.391 12.2
-20 972.9 4050 0.385 25
-20.47 973.0 4045 0.385 26
-29.3 40 40 920.1 4120 0.369 1.47
30 927.7 4080 0.365 2.01
20 935.1 4030 0.361 2.90
10 942.5 3970 0.357 4.35
0 949.4 3900 0.352 7.05
-10 956.0 3810 0.348 12.5
-20 962.3 3710 0.344 24.5
-29.3 967.5 3610 0.340 50
-37.7 50 40 897.6 3950 0.322 1.48
30 905.8 3890 0.320 2.00
20 914.0 3820 0.317 2.84
10 921.3 3730 0.315 4.20
0 929.4 3630 0.312 6.60
-10 937.0 3510 0.309 11
-20 945.0 3370 0.307 20
-30 953.0 3230 0.304 41
-37.7 959.0 3110 0.302 75
-44.9 60 40 874.3 3690 0.280 1.44
30 882.7 3630 0.279 1.92
20 891.3 3550 0.278 2.65
10 899.3 3450 0.277 3.78
0 907.3 3330 0.276 5.75
-10 915.5 3200 0.275 8.85
-20 923.5 3050 0.274 14.7
-30 931.7 2880 0.273 26.5
-40 940.0 2710 0.272 55
-44.9 943.7 2630 0.272 85

55
4.3 Polynomial equations for computer
treatment
The purpose of this par of the chapter is to make basic thermo-
physical property values available for computer treatment. To
make technical calculations of refrigeration and heat pump systems
we need, for the chosen freezing point or concentration and the
fluid temperature, to know values for basic thermophysical proper-
ties such as density, specific heat, thermal conductivity, and dy-
namic viscosity. Such properties are in 4.2 presented for aqueous
solutions of ethyl alcohol. Other property values can then easily be
calculated with a method or procedure developed with the kind
help of Prof. Jesper Oppelstrupp and M. Sc. Thomas Oppelstrupp,
NADA, KTH. A mathematically suitable polynomial equation
form was chosen. A Matlab program was made to generate coeffi-
cients to the equations from a set of basic property values.
The basic polynomial equation form chosen for most of these
thermophysical properties is
f = 6 Cij˜(x-xm)i ·(y-ym)j (4.1)
where i+j d 5 and where Cij is the coefficient for each term.
However, the logarithmic equation form of the same polynomial
is used for dynamic viscosity:
log f = 6 Cij˜(x-xm)i ·(y-ym)j (4.2)
also with i+j d 5.
The freezing point temperature, tF, or concentration, cA, can be
chosen as input parameter x and the fluid temperature, t, as input
parameter y.
Mean values of all x and y values in a whole table, xm and ym, have
been calculated. In an effort to minimize errors (x-xm) and (y-ym)
appear in the polynomials given above.
Max values for i, j and i+j have for each table been chosen to
minimize the max deviation from the values in the table. Choos-
ing i d 5, j d 3 and i+j d 5 we can write the basic equation so that
the coefficients appear in the same order as in the tables:

56
 f = C00 + C01˜(y-ym) + C02˜(y-ym)² + C03˜(y-ym)³ +
C10˜(x-xm) + C11˜(x-xm) (y-ym) + C12˜(x-xm)˜(y-ym)² +
C13˜(x-xm)˜(y-ym)³ + C20˜(x-xm)² + C21˜(x-xm)² (y-ym) +
C22˜(x-xm)²˜(y-ym)² + C23˜(x-xm)2˜(y-ym)³ + C30˜(x-xm)³ +
C31˜(x-xm)³˜(y-ym) + C32˜(x-xm)³˜(y-ym)² + C40˜(x-xm)4 +
C41˜(x-xm)4˜(y-ym) + C50˜(x-xm)5 (4.3)
Table 4.2 gives coefficients of aqueous solutions of ethyl alcohol
obtained from the Matlab program. The freezing point, tF, is cho-
sen as x and the temperature, t, as y. Symbols from Table 4.1.

Table 4.2. Coefficients for properties of ethyl alcohol

i j cA U cp k P
0 0 3.00E+01 9.61E+02 4.19E+03 4.03E-01 1.48E+00
0 1 1.12E-04 -5.37E-01 2.83E+00 6.54E-04 -4.74E-02
0 2 -1.55E-05 -3.21E-03 -3.58E-02 3.48E-07 4.33E-04
0 3 6.60E-07 1.49E-05 5.22E-04 -2.11E-08 -3.03E-06
1 0 -1.04E+00 1.58E+00 2.30E+01 5.16E-03 -1.38E-02
1 1 -2.81E-05 2.17E-02 -5.87E-01 2.22E-05 -4.57E-05
1 2 -4.74E-07 -2.15E-04 3.67E-03 3.72E-08 5.97E-06
1 3 -2.45E-08 7.94E-07 5.66E-05 -2.01E-10 -1.03E-07
2 0 2.00E-03 -3.93E-02 -2.39E-01 1.39E-05 -8.21E-04
2 1 -5.01E-06 3.26E-04 -6.32E-03 3.48E-07 1.83E-05
2 2 5.43E-08 3.94E-07 3.37E-05 -5.85E-09 -1.63E-07
2 3 -2.33E-09 -6.60E-09 8.83E-07 2.16E-11 -3.68E-09
3 0 -5.19E-04 3.65E-04 -1.49E-02 2.64E-06 -2.74E-05
3 1 2.04E-07 -7.91E-06 5.73E-04 4.22E-08 7.24E-07
3 2 2.04E-09 1.81E-07 -1.20E-06 -2.79E-10 -2.64E-09
4 0 -2.42E-05 7.41E-05 -1.08E-03 1.80E-07 -1.54E-06
4 1 1.49E-08 -1.46E-07 1.77E-05 1.14E-09 2.37E-08
5 0 -4.65E-07 2.13E-06 -1.45E-05 3.80E-09 -2.58E-08

xm = -20.2024 -20.2024 -20.2024 -20.2024 -20.2024

ym = 8.1578 8.1578 8.1578 8.1578 8.1578

Max relative errors, absolute where tabulated data is zero */

cA U cp k P
Error 0,1690 */ 0,0006 0,0137 0,0044 0,0284
Row nr 5 */ 25 22 41 67

The error value marked */ for cA is an absolute error where for tF

= 0°C and t = 0°C, cA = -0.17, but should be 0. This and the other
rel. error values may be acceptable for technical calculations.

57
4.4 Comparison of basic values from
literature search with polynomial
equation values for ethyl alcohol
The basic polynomial equation and the coefficients obtained are
put into an Excel sheet that generates values of the basic proper-
ties and other prpperties as well. In order to see how well the Mat-
lab program predicts values for an aqueous solution, the series of
property values in Table 4.1 and values with even additive concen-
trations generated from the Matlab program and Excel sheet are
plotted in the same figures. Series from Table 4.1 are given with
full lines while series generated from the Matlab program are given
with bold dotted lines.This will give a good indication of how use-
ful this tool can be to calculate thermophysical properties of aque-
ous solutions used as secondary working fluids. Figures 4.1 to 4.4
give density, specific heat, thermal conductivity and dynamic vis-
cosity of aqueous solution of ethyl alcohol.
Discussion of Figures 4.1-4: The figures give basic properties of
ethyl alcohol – water and show a good agreement between the ba-
sic values (full lines) and the values based on the Matlab program
and the Excel sheet (dotted lines).
Deviations in Figure 4.1, giving density, such as between high and
low temperatures near curves from 10 to 25% concentration (tF =
-5°C to -15°C) are small (d 0.2% of the density values).
Deviations in Figure 4.2, giving specific heat, such as between high
and low temperatures between curves for 40% concentration and
tF = -30°C are rather small (d 1% of the specific heat values). The
peak between 10 and 15% also is handled rather well by the com-
puter program.
Deviations in Figure 4.3, giving thermal conductivity, such as be-
tween high and low temperatures between curves for 50% concen-
tration and tF = -40°C, are rather small (d 2% of the thermal con-
ductivity values).
Deviations in Figure 4.4, giving dynamic viscosity, are quite uni-
form and generally relatively small (d 2-3% of the viscosity values).
Note that the peak between 30 and 40% is handled rather well by
the computer program.

58
 1010
% by w t
Ethyl alcohol 0
1000
Density
5
990
10
.
980 15
20
Density [kg/m3] .

970 25
30
960

950
40
940

930
50
60
920

910
-50 -40 -30 -20 -10 0 10 20 30 40

Temperature [°C] Melinder 2007

Figure 4.1. Density of ethyl alcohol – water

4600
% by w t
4400 15
20 10
5
4200 0 .
30
Specific heat [J/kgK] .

4000 25
40
3800

3600

3400
50
3200
60
3000
Ethyl alcohol
2800
Specific heat
2600
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.2. Specific heat of ethyl alcohol - water

59
 0,65

Ethyl alcohol % by w t
0,60
Thermal conductivity 0

Thermal conductivity [W/mK] .

0,55 5
10
0,50
15
20
0,45
25
30
0,40

40
0,35
50
0,30
60

0,25
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.3. Thermal conductivity of ethyl alcohol - water

100
50 Ethyl alcohol
Dynamic viscosity
60 40
Dynamic viscosity [mPa·s] .

30

% by w t
10

20
15
10
5
0

1
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.4. Dynamic viscosity of ethyl alcohol - water

60
4.5 Presenting aqueous solution values
of ethyl alcohol
This part describes how selected property values, polynomial equa-
tions, Matlab program already described and an Excel program re-
sults in tables and charts intended for the industry. Using freezing
point and temperature as input values, the other property values
can be generated as sen in Table 4.3.

Table 4.3. Example with basic thermophysical properties of

aqueous solutions (here for tF = -20°C and t = -10°C)
tF t cA U cp k P
Water solution: [°C] [°C] [wt-%] [kg/m³] [J/kgK] [W/mK] [mPas]
Validity limits : 50<x<0 x<y<100
Ethylene glycol -20 -10 36,19 1058 3494 0,415 7,99
Validity limits : 50<x<0 x<y<100
Propylene glycol -20 -10 39,41 1045 3620 0,384 21,67
Validity limits -45<x<0 x<y<40
Ethyl alcohol -20 -10 29,73 970 4127 0,393 12,22
Validity limits -73<x<0 x<y<40
Methyl alcohol -20 -10 24,87 969 3914 0,431 5,42
Validity limits -40<x<0 x<y<40
Glycerol -20 -10 46,28 1129 3155 0,407 19,18
Validity limits -84<x<0 x<y<20
Ammonia -20 -10 13,46 953 4239 0,460 2,94
Validity limits 37,5<x<0 x<y<40
Potassium carbona-
te -20 -10 30,95 1322 2877 0,519 6,25
Validity limits -50<x<0 x<y<40
Calcium chloride -20 -10 20,82 1196 2992 0,530 4,48
Validity limits 33,5<x<0 x<y<40
Magnes. chloride -20 -10 16,59 1151 3192 0,505 5,58
Validity limits -21<x<0 x<y<40
Sodium chloride -20 -10 22,62 1182 3305 0,530 4,22
Validity limits -50<x<0 x<y<40
Potassium acetate -20 -10 27,91 1155 3232 0,467 5,67
Validity limits -50<x<0 x<y<40
Potassium formate -20 -10 28,94 1192 3121 0,506 3,55
Validity limits -75<x<0 x<y<40
Lithium chloride -20 -10 13,66 1083 3405 0,522 3,97

61
The results obtained can generate values for any freezing point.
Table 4.4 gives property values of ethyl alcohol for tF = -5°C,
-10°C, -15°C, -20°C, -30°C and -40°C. For symbols see Table 4.1.
Table 4.4. Basic thermophysical properties of ethyl alcohol
tF cA t U cp k P
[°C] [wt-%] [°C] [kg/m³] [J/kgK] [W/mK] [mPas]
-5 11.1 40 972.8 4287 0.546 0.92
30 976.9 4285 0.534 1.20
20 980.3 4310 0.522 1.60
10 982.7 4350 0.509 2.25
0 984.0 4396 0.496 3.48
-5 984.2 4419 0.490 4.51
-10 18.7 40 960.8 4313 0.493 1.12
30 966.0 4315 0.484 1.50
20 970.5 4333 0.474 2.09
10 974.1 4361 0.464 3.09
0 976.6 4390 0.455 5.00
-10 978.1 4415 0.445 9.10
-15 24.5 40 951.5 4298 0.456 1.26
30 957.5 4291 0.448 1.71
20 963.0 4292 0.440 2.42
10 967.7 4295 0.433 3.67
0 971.6 4297 0.425 6.05
-10 974.5 4291 0.417 11.16
-15 975.6 4284 0.413 15.89
-20 29.7 40 941.8 4259 0.425 1.37
30 948.5 4239 0.419 1.86
20 954.8 4219 0.412 2.65
10 960.5 4196 0.406 4.03
0 965.6 4166 0.399 6.67
-10 970.0 4127 0.393 12.22
-20 973.6 4074 0.387 25.24
-30 40.7 40 918.4 4119 0.366 1.47
30 926.0 4078 0.362 2.02
20 933.5 4024 0.358 2.88
10 940.8 3957 0.354 4.35
0 947.8 3877 0.349 7.03
-10 954.6 3783 0.345 12.4
-20 961.1 3675 0.341 24.0
-30 967.1 3553 0.337 52.2
-40 53.1 40 890.8 3862 0.308 1.45
30 899.0 3812 0.306 1.99
20 907.1 3740 0.305 2.81
10 915.2 3650 0.303 4.11
0 923.2 3542 0.301 6.33
-10 931.1 3419 0.299 10.4
-20 939.1 3280 0.297 18.6
-30 947.1 3129 0.295 36.5
-40 955.1 2967 0.293 79.9

62
Based on these basic thermophysical properties, other properties
can easily be obtained with simple relations. See Table 4.5.
Table 4.5. Example with thermophysical properties (for tF = -
20°C and t = -10°C) based on the basic properties in Table 4.4.
tF t cA Q Pr U cp
2 3
Water solution: [°C] [°C] [wt-%] [mm /s] [-] [kJ/m K]
Validity limits -50<x<0 x<y<100
Ethylene glycol -20 -10 36,19 7,55 67,2 3697
Validity limits -50<x<0 x<y<100
Propylene glycol -20 -10 39,41 20,73 204,1 3785
Validity limits -45<x<0 x<y<40
Ethyl alcohol -20 -10 29,73 12,60 128,4 4003
Validity limits -73<x<0 x<y<40
Methyl alcohol -20 -10 24,87 5,59 49,2 3795
Validity limits -40<x<0 x<y<40
Glycerol -20 -10 46,28 16,99 148,7 3562
Validity limits -84<x<0 x<y<20
Ammonia -20 -10 13,46 3,08 27,1 4041
Validity limits 37,5<x<0 x<y<40
Potassium carbonate -20 -10 30,95 4,73 34,6 3803
Validity limits -50<x<0 x<y<40
Calcium chloride -20 -10 20,82 3,75 25,3 3579
Validity limits 33,5<x<0 x<y<40
Magnesium chloride -20 -10 16,59 4,85 35,3 3674
Validity limits -21<x<0 x<y<40
Sodium chloride -20 -10 22,62 3,57 26,3 3908
Validity limits -50<x<0 x<y<40
Potassium acetate -20 -10 27,91 4,91 39,2 3734
Validity limits -50<x<0 x<y<40
Potassium formate -20 -10 28,94 2,98 21,9 3719
Validity limits -75<x<0 x<y<40
Lithium chloride -20 -10 13,66 3,66 25,9 3688

Input values: -20 -10

These tables are in the Excel program extended to make simple

calculations for heat transport, pressure drop and heat transfer.
This is described further in chapter 5. Similar tables can be gener-
ated for each of the aqueous solutions.

63
In the updated handbook the property values in Table 4.1 will be
presented in the same table with the values in Table 4.4. Values
from both these tables will also be plotted in the same diagrams.
The good agreement seen in Figures 4.1 to 4.4 can also be ob-
served between the basic series of values from Table 4.1 and the
series of values for even freezing points from Tables 4.4 obtained
from the Matlab program and the Excel program.
Discussion of Figures 4.5-8:
Here these series of values are plotted in the same figures, Figure
4.5 to 4.8 for aqueous solutions of ethyl alcohol. Series with even
concentrations from Table 4.1 are given with full lines. Series with
even freezing points from Table 4.4 are given with dashed lines.
The freezing points presented are: tF = -5°C, -10°C, -15°C, -20°C,
-30°C and -40°C.
Deviations are generally quite small: In Figure 4.5 d 0.2% of the
density values; in Figure 4.6 d 1% of the specific heat values; in
Figure 4.7 d 1% of the thermal conductivity values; in Figure 4.8
deviations are quite uniform and d 2-3% of the viscosity values.
This rather good agreement between the basic values (full lines)
and the values for even freezing points based on the Matlab and
Excel programs (dashed lines) for ethyl alcohol is a good indica-
tion that we can expect similar or even better agreement for the
other aqueous solutions. As earlier mentioned, the basic thermo-
physical properties of aqueous solutions of ethyl alcohol present
more challenges than most other fluids examined.
The moderate deviations are also a good indication that we can
expect any computer program based on the Matlab and Excel pro-
grams to represent the basic values well.
In 4.6 and 4.7, tables and figures with basic properties of aqueous
solutions of lithium chloride and potassium formate are presented
in a similar way.

64
 1010
% by w t
1000 Ethyl alcohol 0
Density 5
990
10
.
980 15
20
Density [kg/m3] .

970 25
30
960

950 40

940

930
50
920
60
910
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.5. Density of ethyl alcohol - water

4600
% by w t
15
4400 20 10
5
25 0 .
4200 30
Specific heat [J/(kg·K)] .

4000
40
3800

3600 50

3400
60
3200
Ethyl alcohol
3000
Specific heat
2800

2600
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.6. Specific heat of ethyl alcohol - water

65
 0,625
0,600 % by w t .
0,575 Ethyl alcohol 0
0,550 Thermal conductivity 5
Thermal conductivity [W/(m·K)] . 0,525 10
0,500
15
0,475
20
0,450
0,425 25
30
0,400
0,375
40
0,350
0,325 50
0,300
60
0,275
0,250
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.7. Thermal conductivity of ethyl alcohol - water

100
50
Ethyl alcohol
Dynamic viscosity
60
Dynamic viscosity [mPa·s] .

10
% by w t
40
20
15
10
5
0

1
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.8. Dynamic viscosity of ethyl alcohol - water

66
4.6 Presenting aqueous solution values
of lithium chloride
The critical evaluation of property values of aqueous solutions of
lithium chloride presented in 3.3.3 has resulted in the following
property values. Table 4.6 gives basic property values of aqueous
solutions of lithium chloride. In Figures 4.9 to 4.12, series with
even concentrations are plotted together with series with even
freezing points.

Table 4.6. Basic thermophysical properties of lithium chloride

Freezing Concen- Tempe- Density Specific Thermal Dynamic
Point tration rature heat conduct. viscosity
tF Conc. t U cp k P
°C wt-% °C kg/m³ J/kgK W/mK mPa·s
-1.72 2 40 1004 4080 0.625 0.69
30 1007 4080 0.609 0.86
20 1010 4080 0.592 1.07
10 1011 4080 0.575 1.39
0 1012 4090 0.557 1.90
-1.72 1012 4090 0.554 2.01
-4.86 5 40 1021 3920 0.617 0.78
30 1024 3910 0.602 0.94
20 1027 3910 0.585 1.18
10 1029 3910 0.569 1.55
0 1030 3915 0.552 2.06
-4.86 1030 3920 0.544 2.40
-10 8.6 40 1042 3766 0.607 0.89
30 1045 3748 0.592 1.08
20 1048 3731 0.577 1.34
10 1050 3715 0.561 1.72
0 1052 3702 0.545 2.32
-10 1052 3694 0.529 3.33
-12.6 10 40 1050 3710 0.603 0.93
30 1053 3690 0.589 1.13
20 1056 3670 0.573 1.41
10 1058 3650 0.558 1.84
0 1060 3630 0.543 2.45
-10 1060 3610 0.527 3.45
-12.6 1061 3610 0.523 3.80
-15 11.5 40 1059 3652 0.599 0.99
30 1062 3626 0.585 1.20
20 1065 3599 0.570 1.48
10 1067 3574 0.555 1.90
0 1069 3549 0.540 2.56
-10 1070 3528 0.525 3.67
-15 1070 3518 0.517 4.51

67
Table 4.6. Basic thermophysical properties … (cont.)
tF Conc. T U cp K P
-20 13.7 40 1071 3572 0.594 1.08
30 1075 3539 0.580 1.30
20 1078 3505 0.566 1.61
10 1080 3471 0.551 2.06
0 1082 3437 0.536 2.77
-10 1083 3405 0.522 3.97
-20 1084 3374 0.507 6.10
-23.5 15 40 1079 3530 0.591 1.12
30 1082 3490 0.577 1.37
20 1085 3450 0.563 1.70
10 1088 3410 0.548 2.19
0 1089 3370 0.534 2.91
-10 1091 3330 0.520 4.1
-20 1092 3290 0.505 6.3
-23.5 1092 3280 0.500 7.8
-30 16.8 40 1090 3472 0.587 1.21
30 1093 3426 0.573 1.47
20 1096 3379 0.559 1.82
10 1099 3332 0.545 2.34
0 1101 3284 0.531 3.14
-10 1102 3236 0.516 4.5
-20 1103 3189 0.502 6.8
-30 1104 3142 0.488 11.1
-40 19.1 40 1104 3400 0.582 1.33
30 1107 3343 0.568 1.62
20 1110 3287 0.554 2.01
10 1113 3230 0.540 2.59
0 1115 3173 0.526 3.47
-10 1117 3116 0.512 4.89
-20 1118 3058 0.498 7.29
-30 1119 3000 0.484 11.64
-40 1120 2941 0.471 20.03
-44.5 20 40 1109 3370 0.579 1.40
30 1112 3310 0.566 1.71
20 1116 3250 0.553 2.12
10 1118 3190 0.538 2.68
0 1121 3130 0.524 3.57
-10 1123 3070 0.510 5.02
-20 1124 3010 0.497 7.55
-30 1125 2950 0.483 12
-40 1126 2890 0.469 20
-44.5 1126 2860 0.463 26
Cont.

68
Table 4.6. Basic thermophysical properties … (cont.)
tF Conc. t U cp K P
-50 21.2 40 1116 3333 0.576 1.47
30 1120 3269 0.563 1.79
20 1123 3205 0.549 2.24
10 1126 3142 0.535 2.88
0 1128 3078 0.522 3.85
-10 1130 3014 0.508 5.38
-20 1132 2950 0.494 7.91
-30 1134 2886 0.481 12.3
-40 1135 2821 0.467 20.3
-50 1135 2755 0.454 35.9
-56 22.5 40 1123 3295 0.572 1.55
30 1127 3230 0.559 1.94
20 1130 3160 0.546 2.41
10 1133 3095 0.532 3.1
0 1136 3030 0.519 4.1
-10 1139 2960 0.505 5.7
-20 1140 2890 0.492 8.4
-30 1142 2830 0.479 13
-40 1143 2760 0.465 21
-50 1144 2690 0.452 36
-56 1145 2650 0.445 50
-60 23.3 40 1128 3273 0.569 1.66
30 1132 3204 0.557 2.01
20 1135 3135 0.544 2.51
10 1138 3066 0.530 3.23
0 1141 2998 0.517 4.33
-10 1143 2930 0.504 6.02
-20 1146 2862 0.491 8.75
-30 1147 2793 0.477 13.3
-40 1149 2725 0.464 21.3
-50 1150 2657 0.451 36.0
-60 1150 2589 0.438 64.4
*/ Chemical formula: LiCl; M = 42,40

Discussion of Figures 4.9-12:

Figure 4.9-12 show good agreement between the basic values (full
lines) and the values for even freezing points based on the Matlab
and Excel programs (dotted lines). This can be seen by the fact
that the curves are parallel to each other and The freezing points
presented are: tF = -5°C, -10°C, -15°C, -20°C, -30°C, -40°C -50°C
and -60°C. Deviations are generally quite small: In Figure 4.9 d
0.2% of the density values; in Figure 4.10 d 1% of the specific heat
values; in Figure 4.11 d 1% of the thermal conductivity values; in
Figure 4.12 deviations are quite uniform and d 2-3% of the viscos-
ity values.

69
 1150
% by w t
1140
-50°C
22.5
1130
1120 -40°C
20
1110
1100
Density [kg/m3] .

1090 15
.-20°C
1080
1070
1060 10
1050
1040
1030 5
1020
1010 Lithium chloride 2
Density
1000 0

990
-60 -50 -40 -30 -20 -10 0 10 20 30 40

Temperature [°C] Melinder 2007

Figure 4.9. Density of lithium chloride - water

4300
% by w t
4200 0
4100 2 .
4000
5
3900
3800
Specific heat [J/(kg·K)] .

3700 10
3600
3500
15
3400
3300
20
3200
22,5
3100
3000
2900
2800 Lithium chloride
2700 Specific heat
2600
-60 -50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.10. Specific heat of lithium chloride - water

70
 0,62
% by w t
0,60
0
2
0,58 5
10
Thermal conductivity [W/(m·K)] .

0,56 15
20
0,54 22,5

0,52

0,50

0,48

0,46 Lithium chloride

Thermal conductivity
0,44
-60 -50 -40 -30 -20 -10 0 10 20 30 40
Tem perature [°C] Melinder 2007

Figure 4.11. Thermal conductivity of lithium chloride - water

100

Lithium chloride
Dynam ic viscosity
Dynamic viscosity [mPa·s]

10

%by wt

22,5
20

15
10

5
0
1
-60 -50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] M elinder 2007

Figure 4.12. Dynamic viscosity of lithium chloride - water

71
4.7 Presenting aqueous solution values
of potassium formate
The critical evaluation of property values of aqueous solutions of
potassium formate presented in 3.3.4 has resulted in the following
property values. Table 4.7 gives basic property values of aqueous
solutions of potassium formate. In Figures 4.13-4.16, series with
even concentrations are plotted together with series with even
freezing points. Deviations are smaller than with ethyl alcohol.
Table 4.7. Basic thermophysical properties of potassium formate
Freezing Concen- Tempe- Density Specific Thermal Dynamic
Point tration rature heat conduct. Viscosity
tF Conc. t U cp k P
°C wt-% °C kg/m³ J/kgK W/mK mPa·s
-2.2 5 40 1023 4000 0.616 0.72
30 1026 4000 0.602 0.88
20 1029 4000 0.587 1.09
10 1031 4005 0.572 1.41
0 1032 4010 0.556 1.9
-2.2 1032 4010 0.552 2.05
-5 10.2 40 1054 3815 0.605 0.78
30 1057 3808 0.592 0.95
20 1060 3803 0.578 1.18
10 1062 3801 0.563 1.50
0 1064 3801 0.548 2.00
-5.0 1065 3803 0.541 2.35
-10 17.9 40 1102 3552 0.589 0.90
30 1105 3539 0.576 1.08
20 1108 3527 0.564 1.33
10 1111 3517 0.551 1.68
0 1114 3507 0.537 2.20
-10 1115 3500 0.523 3.02
-11.5 20 40 1114 3490 0.586 0.92
30 1118 3480 0.573 1.11
20 1121 3465 0.560 1.35
10 1124 3455 0.547 1.70
0 1127 3440 0.534 2.23
-10 1129 3430 0.521 3.08
-11.5 1129 3428 0.519 3.25
-15 24.0 40 1141 3356 0.577 1.01
30 1144 3340 0.565 1.21
20 1148 3325 0.552 1.47
10 1151 3309 0.540 1.84
0 1154 3294 0.527 2.40
-10 1157 3280 0.515 3.28
-15 1158 3273 0.508 3.91

72
Table 4.7. Basic thermophysical properties … (cont.)
tF Conc. t U cp k P
°C wt-% °C kg/m³ J/kgK W/mK mPa·s
-16.0 25 40 1148 3320 0.576 1.04
30 1152 3305 0.563 1.24
20 1155 3290 0.550 1.51
10 1159 3275 0.538 1.88
0 1162 3260 0.525 2.44
-10 1165 3245 0.513 3.32
-16.0 1166 3235 0.505 4.1
-20 28.9 40 1174 3207 0.568 1.12
30 1178 3190 0.555 1.33
20 1181 3173 0.543 1.61
10 1185 3155 0.531 2.01
0 1189 3138 0.518 2.61
-10 1192 3121 0.506 3.55
-20 1194 3103 0.494 5.14
-21,3 30 40 1182 3180 0,565 1,15
30 1186 3160 0,553 1,36
20 1190 3140 0,541 1,64
10 1193 3125 0,529 2,06
0 1197 3105 0,517 2,69
-10 1200 3085 0,504 3,66
-20 1202 3070 0,492 5,25
-21,3 1202 3065 0,490 5,50
-30 36,8 40 1229 2995 0,550 1,32
30 1233 2977 0,538 1,57
20 1237 2959 0,526 1,89
10 1242 2941 0,514 2,34
0 1246 2923 0,502 3,01
-10 1249 2905 0,490 4,11
-20 1252 2886 0,478 6,0
-30 1255 2866 0,466 9,5
-35,0 40 40 1252 2920 0,540 1,40
30 1257 2900 0,529 1,67
20 1261 2885 0,517 2,00
10 1265 2865 0,506 2,50
0 1269 2850 0,494 3,20
-10 1273 2830 0,482 4,40
-20 1276 2810 0,471 6,40
-30 1279 2795 0,459 10,0
-35 1280 2785 0,453 13,5

73
Table 4.7. Basic thermophysical properties … (cont.)
tF Conc. t U cp k P
°C wt-% °C kg/m³ J/kgK W/mK mPa·s
-40 42,9 40 1274 2846 0,531 1,50
30 1279 2827 0,520 1,79
20 1284 2808 0,509 2,16
10 1288 2790 0,497 2,68
0 1292 2772 0,486 3,46
-10 1296 2755 0,474 4,74
-20 1299 2737 0,463 6,98
-30 1302 2719 0,451 11,2
-40 1304 2700 0,440 20,0
-50,0 48 40 1316 2735 0,517 1,78
30 1321 2710 0,506 2,15
20 1326 2695 0,495 2,65
10 1330 2675 0,484 3,32
0 1335 2655 0,472 4,4
-10 1338 2640 0,461 6,1
-20 1342 2620 0,449 9,2
-30 1345 2605 0,438 14,8
-40 1348 2585 0,426 26,5
-50 1350 2570 0,415 57

*/ Chemical formula: KHCO2; M = 84.12

Discussion of Figures 4.13-16:

Figure 4.13-16 show good agreement between the basic values (full
lines) and the values for even freezing points based on the Matlab
and Excel programs (dotted lines). The freezing points presented
are: tF = -5°C, -10°C, -15°C, -20°C, -30°C, and -40°C.
As with lithium chloride, deviations for potassium formate are
rather small: In Figure 4.9 d 0.2% of the density values; in Figure
4.10 d 1% of the specific heat values; in Figure 4.11 d 1% of the
thermal conductivity values; in Figure 4.12 deviations are quite uni-
form and generally relatively small (d 2-3% of the viscosity values).

74
 1350
48 % by wt
1325
45
1300
1275 40
1250
35
1225
Density [kg/m3] .

1200 30
1175
25
1150
20
1125
1100
1075 10
1050 Potassium formate
5
1025 Density .
1000 0
975
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.13. Density of potassium formate - water

4400
Potassium formate % by w t
4200 0
Specific heat
4000 5 .
Specific heat [J/(kg·K)] .

3800 10

3600
20
3400
25
3200
30

3000 35
40
2800 45

2600 48

-50 -40 -30 -20 -10 0 10 20 30 40

Temperature [°C] Melinder 2006

Figure 4.14. Specific heat of potassium formate – water

75
 0,620
% by w t
0,600 Potassium formate
Thermal conductivity 0
0,580 5

Thermal conductivity [W/(m·K)] .

10
0,560 20
25
0,540 30
35
0,520
40
0,500 45
48
0,480

0,460

0,440

0,420
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2006

Figure 4.15. Thermal conductivity of potassium formate – water

10
Potassium formate
Dynamic viscosity
Dynamic viscosity [mPa·s] .

% by w t
48
45
40
35
30
25
20
10
0
1
-50 -40 -30 -20 -10 0 10 20 30 40
Temperature [°C] Melinder 2007

Figure 4.16. Dynamic viscosity of potassium formate – water

76
5. Comparison of aqueous
solutions as to heat
transpor t, pressure drop
and heat transfer

5.1. Introduction
Indirect systems with secondary working fluids are used exten-
sively in heat pumps and in supermarkets in Sweden. Using such
indirect systems make it possible to minimize the primary refriger-
ant charge and leakage which is important from environmental as-
pects. A secondary working fluid should have enough freezing se-
curity, good transport and heat transfer abilities and the viscosity
should not be too high at the operating temperatures, as it deter-
mines the type of flow occurring and the pressure drop.
This chapter compares a few types of aqueous solutions used as
secondary working fluids for two types of applications. The first is
bedrock heat pumps and the second is cooling cabinets in super-
markets. Low temperature applications are considered in PAPER 5
(Melinder 2004) and PAPER 7 (Melinder 2005).
The additive mass fraction or concentration of the aqueous solu-
tions is often higher than necessary for the application. How does
the additive concentration affect the thermal performance? As an
aid to find out, the aqueous solutions used in the examples are
compared with different concentrations, giving the freezing points
tF = -10°C, -15°C and -20°C for heat pumps and tF = -15°C and -
20°C for cooling cabinets.
Typical examples are used to compare the fluids as to volumetric
heat capacity, viscosity, Reynolds number, needed pumping power
as well as heat transfer that affect temperature differences in heat
exchangers. The result is presented in figures that also give indica-
tions how the type of flow influences these properties.
What equations are used in the comparisons chosen for heat
transport, heat transfer and pressure drop?

77
5.2 Equations for heat transport, heat
transfer and pressure drop
5.2.1 Heat transport and volumetric heat
capacity
The cooling capacity, Q [W] that can be transported through
pipes or tubes with a given secondary fluid flow rate and tempera-
ture change can for a fluid be obtained with the following rela-
tions:
Q = m’cp't = V U cp't (5.1)

In these equations, m’ [kg/s] is the mass flow rate, V [m3/s] the

volume flow rate, 't [K] the temperature change between fluid ex-
iting and entering the cooling object, U [kg/m3] the fluid density
and cp [J/kgK] its specific heat. The cooling capacity, Q , is then
proportional to the volumetric heat capacity, U cp.

5.2.2 Friction coefficient, pressure drop

and pumping power
In order to use a simple pump with moderate pumping power
the fluid should have low viscosity giving small pressure drop for
the system fluid flow especially with laminar flow. The pumping
power is
E P = 'pf ·V /KP (5.2)
In this equation, KP [–] is the efficiency of the pump (incl electric
motor) and 'pf [Pa] is the pressure drop due to friction for flow
in a tube. This pressure drop due to friction, 'pf, can be esti-
mated with the following equation, where f is the friction coeffi-
cient, w is the fluid velocity, L is the total tube length and d is the
inner tube diameter.
'pf = (f/2)˜U˜w2˜L/d (5.3)
The friction coefficient for laminar flow (Re d 2300) is:
f = 64/Re (5.4)
where

78
 Re = wd/Q = wdU/P (5.5)
In eqn 5.5 the P [Ns/m2] = UQ is the dynamic viscosity andQ
[m2/s] is the kinematic viscosity.
A commonly used friction coefficient equation for turbulent flow
(valid for 3000 < Re < 10000) is Blasius relation:
f = 0.316/Re0.25 (5.6)
A friction coefficient given in Gnielinski (1976) is valid for turbu-
lent flow as well as the transitional region (Re t 2300):
f = (0.79·lnRe -1.64)-2 (5.7)
In our comparison we will use equations 5.4 and 5.7 that together
cover all types of flow that are of interest here.
The basic thermo-physical property that has most influence the
friction coefficient is the viscosity. The influence of viscosity is
stronger with laminar flow than with turbulent flow as is indicated
by the equations above.

5.2.3 Heat transfer and temperature

differences
Good heat transfer is desired in order to get small temperature
differences between the liquid and inner tube wall in the collector
tube and also the evaporator. This temperature difference, -m,
can for a circular tube (with A = S·d·L) be estimated with
Q = D·A·-m (5.8)

The heat transfer coefficient, D [W/m2K], of a fluid with laminar

flow is here estimated with the equation by Sieder and Tate
(1956) that is valid for Re d 2300 and where the geometry has a
large influence:
Nu = 1.86˜(Re˜Pr)1/3˜(d/Lstr)1/3(P/Pw)0.14 (5.9)
Here the Nusselt number is defined Nu = Dd/k, the Reynolds
number Re = wd/Q and the Prandtl number Pr = Pcp/k where P
= QU is the fluid viscosity at bulk temperature and Pw at wall
temperature.

79
One equation valid for heat transfer at turbulent flow in circular
tubes is:
Nu = 0.023˜Re0.8˜Pr1/3 (5.10)
This equation is valid for flow in relatively long smooth tubes in
turbulent flow with 104 < Re < 2·105 but it can with some uncer-
tainty be extrapolated for use down to Re about 3·103.
The heat transfer coefficient of a fluid with non-laminar flow is
here calculated with the following equation by Gnielinski, that with
the friction coefficient in equation 5.7 can be used for the transi-
tion region as well as for turbulent flow, as it is valid for 2300 < Re
< 106 and Pr > 0.5 (Gnielinski, 1976):
( f / 8)(Re 1000) Pr
Nu (5.11)
1  12,7( f / 8)1 / 2 (Pr 2 / 3  1)
Tests at our department of the influence of Reynolds number on
heat transfer indicate the following: For Re < 3000 - 4000 the heat
transfer coefficient decreases rapidly due to transition to laminar
flow in harmony with what equation 5.11 gives (See Melinder
1989, Melinder 1998a,b, Granryd and Melinder 2005)

5.3 Comparison for a bedrock heat

pump application

5.3.1 Example used for comparison of

fluid types and concentrations
In order to make a comparison of the fluids for a domestic heat
pump application with rock as heat source the following tempera-
tures and other parameters are chosen. Typical operating tempera-
tures for the fluids differ at different places and are here set be-
tween +4°C to -4°C as indicated in Figures 5.1-5. In order to see
how the additive concentration and freezing point of the aqueous
solutions influence performance, aqueous solutions of propylene
glycol, ethyl alcohol and glycerol are compared with concentra-
tions giving the freezing points of tF = -10°C, -15°C and -20°C.
The reason for the comparison of the fluids at three different con-
centrations is the following. The freezing point temperature should
be below the operating temperature of the secondary fluid. Aque-

80
ous solutions do not freeze to solid ice even if the temperature is
somewhat lower than the freezing point except near eutectic con-
centration (See chapter 2.1). The fluid used should have enough
but not more than needed freezing security (5-10 K) as a higher
concentration of the freezing point depressant additive will mean
poorer thermophysical properties as we will see in this example.
A domestic heat pump with rock as heat source use circular collec-
tor tubes with the length L = 200 m and an inner diameter of di =
0.035 m. Let us assume that a heat flow of Q = 5000 W is drawn
from the ground and transported with the fluid having the velocity
w = 0.6 m/s. The pumping efficiency is set to KP = 0.4.
In a domestic heat pump with rock as heat source the tube parts
that are near the ground surface can be well insulated. The coldest
part would most likely be at the evaporator and the freezing point
of the fluid could be chosen near the evaporation temperature.

5.3.2 Heat transport ability and

volumetric heat capacity
The refrigerating capacity Q [W] that can be transported with a
given volume flow and temperature change is proportional to the
volumetric heat capacity, U·cp [kJ/m3K].
How do the various concentrations or freezing points of the three
aqueous solutions compare as to volumetric heat capacity that de-
termines the heat transport ability of the fluid?
Figure 5.1 shows that ethyl alcohol gives 5-20 % higher U·cp values
than propylene glycol and glycerol. The high values of ethyl alco-
hol at tF = -10°C and also -15°C are consistent with Figures 3.1, 4.2
and 4.6 showing a higher value of cp than water. The volumetric
heat capacity decreases with higher additive concentrations.

5.3.3 Viscosity and type of flow

The kinematic viscosity, Q = P/U, is a crucial property as it deter-
mines the type of flow we get, whether laminar or not. The Rey-
nolds number, Re = wd/Q = wdU/P, is inversely proportional to
the viscosity. A high viscosity makes it impossible to keep the flow
turbulent with a reasonable pumping power in a collector tube and

81
also in a heat exchanger with circular tubes. How do the chosen
fluids and concentrations compare?
5,0

4,5

4,0
Vol. heat capacity [MJ/m3K]

3,5

3,0
PG -20
2,5 EA -20
GL -20
2,0 PG -15
EA -15
1,5 GL -15
PG -10
1,0 EA -10
GL -10
0,5
Oper. t.
0,0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.1. Volumetric heat capacity as function of fluid temp.

18
PG -20
16 EA -20
GL -20
PG -15
14
EA -15
Kinematic visc. [mm2/s]

GL -15
12 PG -10
EA -10
10 GL -10
Oper. t.
8

0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.2. Kinematic viscosity as function of fluid temperature

82
Figure 5.2 shows that the viscosity of ethyl alcohol solutions is
much less affected by the additive concentration or freezing point
than propylene glycol and glycerol solutions. This is consistent
with the viscosity peak at | 40% seen in Figures 4.4 and 4.8. How-
ever, at concentrations giving tF = -10°C, the viscosity of glycerol
is even lower than that of ethyl alcohol.
What type of flow can we expect for the example of a domestic
heat pump using circular collector tubes with an inner diameter of
di = 0.035 m and a fluid velocity w = 0.6 m/s.
Figure 5.3 gives the Reynolds number, Re [ -], as function of the
fluid temperature for the given example. The figure shows that we
at operation temperatures near 0°C can expect laminar flow (below
Re = 2300) for propylene glycol and glycerol and freezing point tF
= -20°C. For propylene glycol and tF = -15°C as well as for ethyl
alcohol and tF = -15°C and -20°C, we may get transitional flow
(2300<Re <3000) but not for tF = -10°C. For the other freezing
points of the aqueous solutions we can expect turbulent flow.

5.3.4 Friction coefficient, pressure drop

and pumping power
In order to be able to use a pump with moderate pumping power
the fluid should have low viscosity giving small friction coefficient
and pressure drop for the system fluid flow. The pressure drop
due to friction, 'pf , for flow in a tube (in which no consideration
is taken for inlet and exit losses or pressure drop in bends and
valves etc) are estimated with equations 5.2-4 and 5.6.

Figure 5.4 gives pumping power as function of fluid temperature

for the four concentrations of the two aqueous solutions. From
the charts we see that the values go down 30-40% when the flow
turns laminar but increases quickly again with lower temperature
that gives higher viscosity. As we here consider only the pressure
drop due to friction, both pressure drop and pumping power are
proportional to the friction coefficient. The pumping power is
only between 45 and 75 W with the chosen pump efficiency.

83
 6000

5000

Reynolds number [ - ]
4000

PG -20
3000 EA -20
GL -20
PG -15
2000 EA -15
GL -15
PG -10
EA -10
1000
GL -10
Oper. t.
2300
0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.3. Reynolds number as function of fluid temperature

90

80

70
Pumping power, EP [W]

60

50
PG -20
40 EA -20
GL -20
PG -15
30 EA -15
GL -15
20 PG -10
EA -10
10 GL -10
Oper. t.
0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.4. Pumpinge power as function of fluid temperature

84
5.3.5 Heat transfer and temperature
differences
Figures 5.5 and 5.6 give heat transfer coefficient, D, and tempera-
ture difference between the liquid and inner tube wall in the collec-
tor tube, -m, estimated with equations 5.7-8 and 5.11 for the ex-
ample in 5.3.1.
Figures 5.5 and 5.6 give heat transfer coefficient and temperature
difference as function of fluid temperature for the three concentra-
tions of the three aqueous solutions.
From Figures 5.5 and 5.6 we see that laminar flow should be
avoided as it would make the temperature difference between the
liquid and inner tube wall in the collector tube more than 2K larger
which would be detrimental to the heat pump system. Transitional
flow, with Reynolds number up to Re = 3000 or even more could
have a similar effect though it is not so easy to know how large the
temperature difference then will be.

5.4 Comparison for a supermarket

cooling application
5.4.1 Example used to compare properties
of aqueous solutions for cooling
The following fluids are used in this comparison: Aqueous solu-
tions of ethylene glycol, ethyl alcohol, potassium acetate and po-
tassium formate with concentrations giving the freezing point tF =
-15°C and tF = -20°C. A non aqueous heat transfer fluid that has
been suggested for supermarket cooling, a hydrofluoroether with
low viscosity (HFE-7100) is here included in the comparison (Hor-
ton and Groll 2003). Operating temperatures are somewhere be-
tween -4 to -12°C. See also PAPER 7 (Melinder 2005).
In order to illustrate how the fluids differ with temperature, the
following example is used. Cooling cabinets with the cooling ca-
pacity Q = 2000 W have a total circular tube length Ltot = 20 m,
the inner tube diameter di = 0.015 m, each straight tube length Lstr
= 1.5 m and the fluid velocity is w = 0.8 m/s. The pumping effi-
ciency is set to KP = 0.4.

85
 1100
1000
900

Heat transfer coeff. [W/m2K] 800

700
600
PG -20
500 EA -20
GL -20
400 PG -15
EA -15
300 GL -15
200 PG -10
EA -10
100 GL -10
Oper. t.
0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.5. Heat transfer coefficient as function of fluid temp.

4,0
PG -20
EA -20
3,5
GL -20
PG -15
Temp. diff. at tube wall [K]

3,0 EA -15
GL -15
2,5 PG -10
EA -10
GL -10
2,0
Oper. t.

1,5

1,0

0,5

0,0
-20 -15 -10 -5 0 5 10 15 20
Temperature [°C]

Figure 5.6. Temperature difference as function of fluid temp.

86
5.4.2 Heat transport ability and
volumetric heat capacity
How do the two concentrations or freezing points of the aqueous
solutions and the non aqueous fluid compare as to volumetric heat
capacity that determines the heat transport ability of the fluid?
Figure 5.7 shows that ethyl alcohol gives 5-10 % higher U·cp values
than the other aqueous solutions and the volumetric heat capacity,
U·cp, decreases with higher concentrations of each fluid but the
differences are not too big. However, U·cp of the non aqueous
fluid in the comparison, hydrofluoroether, is less than half of U·cp
of the aqueous solutions. This means that more than twice as big
volume flow is needed to get the same cooling capacity.

5.4.3 Viscosity and type of flow

The aqueous solution with lowest viscosity in Figure 5.2, ethyl al-
cohol, has the highest viscosity of the fluids compared in Figure
5.8, followed by ethylene glycol. The salts, especially potassium
formate, have low viscosity. However, hydrofluoroether has very
low viscosity, which is its best property.
What type of flow can we expect for the example of circular tubes
in a cooling cabinet? Figure 5.9 gives Reynolds number, Re [ - ], as
function of the fluid temperature for the given example.
Figure 5.9 indicate that we at operation temperatures can expect
laminar flow (below Re = 2300) for ethylene glycol and ethyl alco-
hol for both freezing points. With potassium acetate we may get
transitional flow (2300<Re <3000) while potassium formate and
hydrofluoroether give turbulent flow.

5.4.4 Friction coefficient, pressure drop

and pumping power
Friction coefficient, pressure drop and pumping power are propor-
tional to each other. How do the fluids compare?
Figure 5.10 gives pumping power as function of fluid temperature
for the four concentrations of the two aqueous solutions. From
the charts we see that the values go down 30-40% when the flow
turns laminar but increases quickly again with lower temperature
that gives higher viscosity. As we here consider only the pressure
drop due to friction, both pressure drop and pumping power are

87
proportional to the friction coefficient. The pumping power is
only between 5 and 8 W with the chosen pump efficiency.
4,5

Vol. heat capacity [MJ/m3K] . 4,0

3,5

3,0
EG -20
2,5 EA -20
KAC -20
2,0 KFO -20
EA -15
1,5 EG -15
KAC -15
1,0
KFO -15
0,5 HFE-7100
Oper. t.
0,0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.7. Volumetric heat capacity, U·cp [kJ/(m3˜K)] as func-

tion of fluid temperature
18
EG -20
16 EA -20
KAC -20
14 KFO -20
Kinemiatic visc. [mm2/s] .

EG -15
12 EA -15
KAC -15
10 KFO -15
HFE-7100
8 Oper. t.

0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.8. Kinematic viscosity as function of fluid temperature

88
 6000
EG -20
EA -20
5000 KAC -20
KFO -20
EG -15
Reynolds number [ - ] .

EA -15
4000
KAC -15
KFO -15
HFE-7100
3000 Oper. t.
2300

2000

1000

0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.9. Reynolds number as function of fluid temperature

16
EG -20
EA -20
14 KAC -20
KFO -20
12 EG -15
Pumping power [W]

EA -15
10 KAC -15
KFO -15
HFE-7100
8 Oper. t.

0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.10. Pumping power as function of fluid temperature

89
 2000
EG -20
1800 EA -20
KAC -20
1600
KFO -20
Heat transfer coeff. [W/m2K] 1400 EG -15
EA -15
1200 KAC -15
KFO -15
1000 HFE-7100
Oper. t.
800

600

400

200

0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.11. Heat transfer coefficient as function of fluid temp.

5,0
EG -20
4,5 EA -20
KAC -20
4,0 KFO -20
Temperature diff. [K]

EG -15
3,5 EA -15
KAC -15
3,0 KFO -15
HFE-7100
2,5 Oper. t.
2,0

1,5

1,0

0,5

0,0
-20 -15 -10 -5 0 5 10
Temperature [°C]

Figure 5.12.Temperature difference as function of fluid temp.

90
5.3.5 Heat transfer and temperature
differences
Figures 5.11 and 5.12 give heat transfer coefficient, D, and tem-
perature difference between the liquid and inner tube wall in the
collector tube, -m, estimated with equations 5.7-8 and 5.11 for the
example in 5.3.1.
Figures 5.11 and 5.12 give heat transfer coefficient and tempera-
ture difference as function of fluid temperature for the four con-
centrations of the two aqueous solutions.
From the charts we see that laminar flow should be avoided as it
would make the temperature difference between the liquid and in-
ner tube wall in the collector tube more than 2K larger which
would be detrimental to the heat pump system. Transitional flow,
with Reynolds number up to Re = 3000 or even more could have a
similar effect though it is not so easy to know how large the tem-
perature difference then will be.

5.5 Discussion
For the chosen example we see that laminar flow can be expected
for the highest concentrations (lowest freezing point) of propylene
glycol. We may get transitional flow for both types of fluids except
for freezing points tF = -5°C and -10°C, when we can expect tur-
bulent flow. The influence on friction coefficient, pressure drop
and pumping power is not crucial with this example. However,
laminar and also transitional flow should be avoided because of the
added temperature difference between the liquid and inner tube
wall in the collector tube.

5.6 Nomenclature
The following nomenclature is used in chapter 5.
cp specific heat [J/kgK]
d tube diameter [m]
f1 friction factor [ - ]
k thermal conductivity [W/mK]

91
 Nomenclature, cont.
t temperature [°C]
w fluid flow velocity [m/s]
A heat transfer area [m2]
E P pumping power [W]
Q cooling capacity [W]
V volume flow [m3/s]
D heat transfer coefficient [W/m2K]
't temperature change [K]
P dynamic viscosity; P = Q˜U [Pa˜s]
Q kinematic viscosity [m2/s]
U density [kg/m3]
-m temperature difference [K]
Nu Nusselt number; Nu = D˜d/k [ - ]
Pr Prandtl number; Pr = P˜cp/k [ - ]
Re Reynolds number; Re = (w˜d)/Q [ - ]

92
6 . B a s i c t h e r m o phys ica l
properties of aqueous
solutions used as ice slur r y

6.1. Introduction
The most basic form of ice slurry is pure waterwith ice which is an
excellent secondary working fluid. However, when temperatures
below 0°C are required, freezing point depressant additives are
used. Most aqueous solutions used as single-phase secondary fluids
can also be used as ice slurry, a fluid consisting of liquid and ice
where small ice crystals are produced, usually with some type of ice
generator. The ice crystals are then transported to the cooling ob-
ject from which heat is removed when some or all of the ice crys-
tals melt. The great enthalpy change when ice melts requires less
temperature change in the cooling object and makes it possible to
reduce the volume flow rate for a given cooling capacity and also
to use smaller pipe dimensions in the system.
Some pioneers in the field of ice slurries focused on getting ice
slurry equipment working for various applications, such as air con-
ditioning, cooling cabinets in supermarket, district cooling, pre-
serving fish, shrimps and vegetables. (Paul 1995, Snoek 1993,
Wang and Goldstein 1996) A lot of research has since then been
made, focusing also on various aspects and details of ice slurry
properties, control, equipment, measurements etc. (Bel 1999, Ben
Lakhdar 1998, Egolf et al 1996, Fournaison et al 2001, Guilpart et
al 2005, Meewisse 2004, Melinder and Hägg 2005, Pronk et al
2005). Much of this research has been carried out in connection
with an IIR Working party that since 1999 has arranged workshops
that from 2006 were transferred into conferences on Phase
Change Materials and Slurries for Refrigeration and Air Condition-
ing. This work has resulted in an IIR Handbook on Ice Slurries
(Kauffeld, Egolf and Kawaji 2005).
The requirements that have to be fulfilled by an ideal secondary ice
slurry fluid are basically the same as for single phase fluids as out-
lined in chapter 1.3.2, except for the fact that ice slurries operate a
few degrees below the freezing point of the solution while single

93
phase fluids operate several degrees above the freezing point to
give sufficient freezing protection.
Ice slurry may be produced at nighttime and stored in an ice slurry
tank. The ice slurry may then be introduced in the secondary fluid
loop at daytime and ice may be melted when there is high need for
cooling. If lower electric cost is available at night time, this could
lead to significanlty lower operation costs. For design purposes it is
of interest to estimate ice mass fraction and enthalpy values and
enthalpy-phase diagrams and related tables can serve that purpose.
In order to choose a secondary fluid for ice slurry use and to make
correct technical calculations of the ice slurry system there is a
need to examine and evaluate thermophysical properties and other
aspects of ice and of the aqueous solution used.

6.2 Thermophysical properties of ice

In order to predict the thermophysical properties of ice slurries, it
is necessary to know the properties of the aqueous solution and of
ice by combining these properties in the correct ratio. The ice mass
fraction or ice concentration is here a crucial parameter. Also vis-
cosity, thermal conductivity and enthalpy of the ice slurry need to
be determined carefully as described in the following sections.
925
Plank, 1988
924
Raznjevic, 1976
923 Eqn. 6.1
Density of ice [kg/m3]

922

921

920

919

918

917

916
-50 -45 -40 -35 -30 -25 -20 -15 -10 -5 0
Temperature [°C]
Figure 6.1. Density of ice as function of temperature

94
Linear equations for density, thermal conductivity and specific heat
have been chosen based on data in handbooks (Plank 1988, Razn-
jevic 1976) and some other sources. See Figures 6.1-3.
The density of ice is lower than the density of water in the liquid
phase (ice floats on the liquid surface) but increases with lower
temperature. In Figure 6.1, the following equation for density of
ice, UI [kg/m3], as function of temperature -I [°C] is based on data
from Plank (1988) and Raznjevic (1976):
UI = 917 – 0.13·tI (6.1)
The specific heat of ice is much lower than the specific heat of
water in the liquid phase. Measured values of specific heat of ice by
Dickinson-Osborne, published in CRC Handbook of Chemistry &
Physics (1988-89) and a serie of measurements by Fransson (2002)
are plotted besides data in Plank (1988) and Raznjevic (1976). Note
that data from Plank (1988) show a much steeper slope than the
rest and there is a spread of values at or near 0°C.

2,15
Osborne, 1915
2,10 Fransson, 2002
Plank, 1988
Specific heat, cp [kJ/kgK]

Raznjevic,1976
2,05 Eqn. 6.2

2,00

1,95

1,90

1,85

1,80
-40 -35 -30 -25 -20 -15 -10 -5 0
Temperature [°C]

Figure 6.2. Specific heat of ice as function of temperature

Based on these data, the following equation for specific heat of ice,
cpI [kJ/kgK], as a function of temperature can be described by:
cpI = 2.12 + 0.008·tI (6.2)

95
The enthalpy of ice, hI [kJ/kg], at the temperature, -I [°C], can
then be calculated with the relation, where the heat of melting of
ice is set to 332.4 kJ/kg:
hI(tI) = –332.4 + tI (2.12 + 0.008·tI) (6.3)
This equation has been used to construct enthalpy phase diagrams
for a number of aqueous solutions.
The thermal conductivity of ice is much higher than that of wa-
ter in the liquid phase. This feature has a good influence on the
performance of ice slurries.
2,9
Plank, 1988
2,8 Raznjevic, 1976
Thermal cond. of ice [W/mK]

Eqn. 6.4
2,7

2,6

2,5

2,4

2,3

2,2
-50 -45 -40 -35 -30 -25 -20 -15 -10 -5 0
Temperature [°C]

Figure 6.3. Thermal conductivity of ice as function of temp.

Based mainly on data from Plank (1988) and Raznjevic (1976) as

shown in Figure 6.3, the following equation for thermal conductiv-
ity of ice, kI [W/mK], a function of temperature, tI [°C], is used:
kI = 2.21 – 0.012·tI (6.4)
Note that the equation by Bobkov (1977) kI = 2.22(1–0.0015·tI)
gives much lower increase (of kI) with lower temperatures.

96
6.3 Freezing point, ice fraction and
density of ice slurries in equilibrium
6.3.1 Freezing point and ice mass fraction
Figure 6.4 shows how the freezing point changes with the compo-
sition of the mixture when in equilibrium. The diagram can be
used for any aqueous solution (compare Figure 2.1). An increased
concentration of the freezing point depressant additive lowers the
freezing point temperature. This continues until we reach the con-
centration where the additive and water forms a eutectic solution.
With even higher concentration the freezing point is raised again.
Let us assume a solution with the additive mass fraction c0 and
note what happens when it is cooled down from A to the freezing
point F (c0, tF) and below that temperature. If the temperature de-
creases below the freezing point and the process takes place in
equilibrium, a separation of the mixture takes place. Ice crystals
that ideally consist of pure water freeze out and are separated from
the rest of the solution and the additive mass fraction in the re-
maining solution rises as a consequence.

Temp.
A
0
tF F

tS K S

tE E
Å lS ÆÅ lI Æ

0 c0 cS cE Additive mass fraction

mI ' m S

Figure 6.4. Freezing point diagram with seesaw balance

97
When the temperature has been lowered so that it corresponds to
K in Figure 6.4, there are ice crystals with the condition (c = 0, tS)
as well as solution with the condition (cS, tS).
The solution does not freeze to a solid mass when the freezing
point is passed, but ice slurry is formed. Only when the tempera-
ture goes down to the eutectic temperature will the solution fully
solidify. We should also keep in mind that sub-cooling phenomena
often occur in practice. The first ice crystals are often not formed
until the solution temperature falls to one or two degrees below
the equilibrium freezing point, but when some ice crystals have
been formed, the continuation of the process seems easier to take
place in equilibrium.
The concentration of the remaining solution is given by a mass
balance based on the condition that the mixture has an unchanged
total amount of additive. Let us refer to the mass of pure ice by mI
and the mass of solution by mS, the concentration of which is cS.
Then the amount of additive is mScS. In the original solution the
additive amount was c0(mI+mS). As the amount of additive has
not changed, it follows that mScS = c0(mI+mS), which also can be
written as mS(cS-c0) = mIc0 (Granryd and Melinder 2005).
The mass fraction of ice or ice concentration, cI, can then easily
be calculated from
cI = mI/(mI+mS) = (cS-c0)/cS (6.5)
The ice mass fraction is closely linked to the slope of the freezing
point curve between the temperatures considered. The amounts
(cS-c0) and c0 are in Figure 6.4 illustrated as the lengths lI and lS.
The correlations for mass balance and ice mass fraction can then
be expressed as
mI/mS = lI/ lS or cI = lI/(lI+lS) (6.6)
These correlations follow the so called seesaw principle. Imagine
that we place the mass mI at cA = 0, the mass mS at cA = cS and
that these represent weights on a seesaw that should be in balance
when supported at cA = c0. This helps us see that eqn. 6.6 is valid
(Granryd 2005, Melinder and Hägg 2005).
Extensive studies on ice slurry properties, such as those by Bel et
al (1996) and Guilpart et al (1999), present ice mass fraction as a
function of the temperature for various additive concentrations of

98
the freezing point depressant substance, usually ethyl alcohol-
water. Suitable equations for ice slurries are also given in Ben
Lakhdar (1998), Kauffeld (1999) and Lottin and Epiard (2000).
Figure 6.5 gives ice mass fraction as a function of temperature for
various additive concentrations of aqueous solutions of ethyl alco-
hol. Property values are taken from CRC Handbook of Chemistry
and Physics (1988-89) and the Matlab and Excel programs de-
scribed in chapter 4.
0,60
0,55
0,50
Ice concentration [kg ice/ kg tot.] .

0,45 Initial conc. of

additive [kg
0,40 add./ kg tot.]
0,35
0,05
0,30
0,10
0,25 0,15
0,20 0,20
0,15 0,25
0,30
0,10
0,05
0,00
-25 -20 -15 -10 -5 0
Ice slurry temperature [°C]

Figure 6.5. Ice mass fraction as function of temperature for

aqueous solutions of ethyl alcohol

6.3.2 Density and volumetric ice fraction

The density of ice slurries at equilibrium, UIS, may be estimated
with a mixing equation where UCS is the density of the concen-
trated aqueous solution (Melinder 2003a):
UIS = 1/[(cI/UI)+(1-cI)/UCS] (6.7)
In Figure 6.6 the density of ice slurry based on eqn. 6.7 is given as
function of temperature for six types of aqueous solutions. Density
of ice and three of the aqueous solution types as single phase are
also given in the figure. The aqueous solutions of salts have at sin-
gle phase density so high that they are outside the figure. The den-
sity of ice slurries is generally lower than the density of single-

99
phase fluids because of lower additive concentrations and the low
density of ice.
Aqueous solutions of ethyl alcohol have the lowest density, quite
near that of the ice particles, which may prevent stratification of
the fluid caused from the buoyancy force on the ice crystals. The
high density of the salts may cause separation between ice particles
and the rest of the solution. Besides those aspects that need con-
sideration for ice slurry applications, a high density generally im-
proves transport and transfer abilities of a fluid.
1080
Ice slurry, tF=-5°C
1060
Ethylene Glycol
Density of ice slurry [kg/m3] .

1040 Propylene Glycol

Ethyl Alcohol
1020
Sodium Chloride
1000 Potassium Acetate

980 Potassium Formate

Liquid only, tF=-15°C
960
Ethylene Glycol

940 Propylene Glycol

Ethyl Alcohol
920
Ice density
900
-12 -10 -8 -6 -4 -2 0
Temperature [°C]

Figure 6.6. Density as function of temperature for ice slurry and

single phase of some aqueous solutions

A comparison of some properties of aqueous solutions for liquid

only and ice slurry is made in PAPER 4 (Melinder 2003a). This
comparison includes density, volumetric enthalpy change, dynamic
viscosity and thermal conductivity.
In some equations for viscosity and thermal conductivity the vol-
ume fraction of particles is introduced, corresponding to a volume
fraction of ice, cIv defined as
cIv = cI/((cI+(1-cI)UI/UCS) (6.8)
with symbols or nomenclature as earlier given in this chapter part.

100
6.4 Heat transport ability and enthalpy
change of aqueous solutions
6.4.1 Heat transport ability, enthalpy
change and apparent specific heat
The secondary fluid flow rate required to achieve a cooling capac-
ity of Q in an indirect system with a single-phase fluid can be cal-
culated by the following relation:
Q = m cp 't = V U cp 't (6.9)
For an ice slurry system, it is more suitable to apply the relation:
Q = m 'h = V U'h (6.10)
Both sensible heat and latent heat of melting are included in the
enthalpy difference, 'h. In these relations, m  , is the fluid mass
flow rate, V , is the volumetric flow rate, 't is the temperature
change between the fluid entering and exiting the cooling unit, U is
the fluid density and cp is its specific heat.
The heat transport ability of ice slurries in equilibrium is propor-
tional to the enthalpy difference, 'h, obtained for a certain tem-
perature change between the fluid entering and exiting. It is much
higher for ice slurry than for mono-phase aqueous solutions, as
there is a benefit from the latent heat of melting of ice crystals,
which reduces the fluid flow rate required for a given cooling ca-
pacity and also the temperature change. This heat transport ability
depends partly on the specific heat of the fluid, but more on the
ice fraction of the ice slurry that enters the cooling unit, on the
amount of ice that melts (Melinder, 2002b).
This heat transport ability can also be expressed as an apparent
specific heat, cp,a that can defined as cp,a = 'h/'t.
This apparent specific heat includes sensible heat of the fluid and
latent heat of melting of the ice.
Figure 6.7 gives apparent specific heat of ethyl alcohol. Just below
the freezing point there are very high values especially at low addi-
tive fractions. Some works on ice slurry properties give apparent
specific heat as a function of the temperature for various concen-
trations of the freezing point depressant substance, usually for

101
ethyl alcohol – water (Bel and Lallemand 1999, Guilpart et al
1999). See also PAPER 6 (Melinder and Granryd 2005).

80
Additive
70 m as s
fraction
60
Apparent specific heat [kJ/kg]

0.05
0.10
50
0.15
0.20
40
0.25
0.30
30
0.40
0.50
20

10

0
-25 -20 -15 -10 -5 0
Tem perature, °C

Figure 6.7. Apparent specific heat as function of temperature

and additive mass fraction of ethyl alcohol - water

6.4.2 Heat transport ability of ice slurries

and the volumetric enthalpy change
The volumetric heat transport ability of a fluid, the cooling capac-
ity divided with the fluid volume flow rate, Q /V ·is equal to the
estimated volumetric enthalpy change for a certain temperature
change between the fluid entering and exiting the cooling object,
U'h for ice slurries and U·cp't for single-phase fluids (Compare
eqns. 6.9 and 6.10).
In Figure 6.8 a comparison is made of this volumetric enthalpy
change from t = –5qC for ice slurries with concentrations giving
the freezing point tF = –5qC and for single-phase fluids with con-
centrations giving the freezing point tF = –15qC. The figure shows
that the volumetric enthalpy change obtained for a certain tem-
perature change for cooling applications can be 7 to 10 times
higher for ice slurries than for single phase aqueous solutions, be-
cause of the large benefit from the latent heat of melting of ice
crystals.

102
 160 Ice slurry, tF=-5°C
Ethylene Glycol
Volumetric enthalpy change [MJ/m3] .

140 Propylene Glycol

Ethyl Alcohol
120 Sodium Chloride
Calcium Chloride
100 Potassium Acetate
Potassium Formate
80 Liquid only, tF=-15°C
Ethylene Glycol
60 Propylene Glycol
Ethyl Alcohol
40 Sodium Chloride
Calcium Chloride
20 Potassium Acetate
Potassium Formate
0
-10 -9 -8 -7 -6 -5 -4
Temperature [°C]

Figure 6.8. Volumetric enthalpy change from t = -5qC of ice

slurries compared to that of single-hase fluids

This volumetric enthalpy change depends only a little on the sensi-

ble heat of the fluid and much on the latent heat, the amount of
ice that will melt, i.e. on the ice fraction of the ice slurry that enters
the cooling object. This high volumetric heat transport ability of
ice slurries for cooling applications makes it possible to reduce the
size of piping in the system.
The enthalpy values used to construct the curves in Fig. 6.8 for ice
slurries are obtained from the type of enthalpy phase diagrams in-
troduced in 6.5 (Compare Kauffeld et al 2005, Guilpart et al 2006).
See also PAPER 2 (Melinder 2002a).

6.5 Enthalpy phase diagrams of

aqueous solutions
6.5.1 Constructing enthalpy phase
diagrams of aqueous solutions
A way to visualize the benefits from the latent heat or enthalpy dif-
ference in the melting process of ice may be achieved by con-
structing an enthalpy-phase diagram of the aqueous solution at at-
mospheric pressure in a similar way as Bošnjakoviý (1961) and
Rant (1961). The amount of heat exchanged in the freezing proc-
ess can be estimated from such enthalpy-phase diagrams with

103
some added features. Such diagrams can then help us estimate
what additive concentrations and temperatures give most benefit.
For calculation purposes we choose to view an ice slurry solution
as consisting of i. small ice crystals of pure water that freeze out,
and ii. the rest of the solution with a higher concentration of the
freezing point depressant additive. In order to estimate enthalpy
values of an aqueous solution with ice particles we need freezing
point data, values of specific heat as well as enthalpy values of ice
and heat of mixing of the solution. The heat of mixing can, in the
diagrams, be seen as the variation in enthalpy values with the con-
centration of a reference isothermal line in the liquid phase (such
as tR = 0°C).
Let us first as an example consider a basic enthalpy-phase diagram
of sodium chloride – water (Figure 6.9) that mainly considers the
area of interest for an ice slurry application. Besides isotherms it
gives lines for ice mass fraction and it is extended beyond the
eutectic point. An enthalpy-phase diagram of ethyl alcohol –water
is then presented that gives further added features (Figure 6.10).
The enthalpy of ice slurry, hIS, at point K in Figure 6.9, can be
expressed by the following equation:
hIS = h(K) = hI (tCS)cI + hCS(cCS, tCS)(1-cI) (6.11)
where hI(tCS) = hH - hN = hH-hP + (hP-hN) and
t ,CS
hCS(cCS, tCS) = h0,R + 'hM(cCS, tR) + ³ (cp·dt) (6.12)
t ,R

The enthalpy difference between N and P, the heat of melting of

water, is hN-hP = 332.4 kJ/kg. The enthalpy of ice at various tem-
peratures has been estimated using Equation 6.3.
The enthalpy of water at the reference temperature tR = 0°C, is de-
fined as: h0,R (0°C) = 0 kJ/kg. The enthalpy values for different
additive fractions for the reference temperature corresponds then
to the heat of mixing value, 'hM(cCS, tCS). The sources of heat of
mixing values, 'hM, and the reference temperature, tR, for each
diagram are as follows: Sodium chloride (tR = 0°C), 'hM from
Bošnjakovic (1961); Ethyl alcohol (tR = 0°C), 'hM from Landolt-
Börnstein, (1976). Heat of mixing values for other aqueous solu-
tions have also been taken from Christensen C. et al (1982) and
Christensen J. et al (1986).

104
 100
20°C 20°C
Liquid, t > t F
50 10°C 10°C
N 0°C A 0°C
0 F h,t F S -10°C
-2°C -10
-4°C -6°C -8°C
-10°C h,t F
-50 -14°C
c I = 0,2
Enthalpy, h [kJ/kg] .

-20°C E
-100
c I = 0,4 Ice/Liquid, t < t F
-150
K Eutectic
-200 cI = 0,6 area,
t E | -21°C
c I = Ice mass fraction
-250
c I = 0,8
t E | -21°C
-300 M
P
Solid phase
-350H t < -21°C
M elinder, 2006
-400
0 0,05 0,1 0,15 0,2 0,25
Additive mass fraction, c A [kg NaCl/kg solution]

Figures 6.9 Enthalpy phase diagram of sodium chloride

Added features in Figure 6.10 compared to Figure 6.9 are the en-
thalpy lines connecting the temperatures tF-1, tF-2, etc. to tF-10.
These lines correspond to 1, 2, etc. to 10K below the freezing
point curve. These enthalpy lines that are marked h,tF-1, h,tF-2, etc.
to h,tF-10, in the diagrams are made as an aid to estimate the heat
transport ability of each aqueous solution when used as ice-slurry.

6.5.2 Example of simple adiabatic ice

slurry processes in equilibrium
With an enthalpy-phase diagram with some added features it is
possible to estimate what happens during ice slurry processes, es-
pecially those that take place adiabatically in equilibrium. The way
these enthalpy-phase diagrams are constructed makes it also possi-
ble to generate the following basic diagrams, all as function of the
additive mass fraction of ethyl alcohol. A simple example of the
process in an ice generator is drawn in each of the figures.

105
 50
Ethyl alcohol - water

0
-1°C 5°C
cI=0,1 -2°C 0°C
-3°C
-50 -4°C h,t F
-5°C -5°C
cI=0,2 -6°C
-8°C -10°C
h,t F-1 -10°C
-100 cI=0,3 -12°C
Enthalpy, [kJ/kg] .

h,t F-2 -15°C

cI=0,4
h,t F-3
-150 cI=0,2
cI=0,5
h,t F-5
-20°C
-200 cI=0,6 h,t F-7
h,t F-10 cI=0,4 -25°C
cI=0,7
-250
cI=0,8
cI=0,6
-300
cI=0,8

-350
0,00 0,03 0,06 0,09 0,12 0,15 0,18 0,21 0,24
Additive mass fraction [kg/kg sol] Melinder, 2007

Figure 6.10. Enthalpy phase diagram of ethyl alcohol with added

enthalpy lines

Figure 6.11 gives first an enthalpy phase diagram. Figure 6.12 gives
ice slurry enthalpy, figure 6.13 gives ice mass fraction based on
eqn. 6.5, figure 6.14 gives ice slurry density based on eqn. 6.7. All
figures are given for additive mass fractions cA =0.0 to 0.21, corre-
spondding to freezing points tF = 0ºC to tF | -12ºC. Each figure
gives values for freezing point and for some temperatures below
the freezing point curve, tF-0.5 , tF-1 ,tF-1.5 to tF-7, etc.
Processes in an ice slurry plant where ice is neither removed nor
added (as from an external ice storage tank) ideally takes place ver-
tically in an enthalpy-phase diagram. In an ice generator heat is
removed from the fluid, the enthalpy decreases and the ice fraction
increases as the process goes vertically down in the diagram. In the
user unit (ie. a cooling cabinet battery) heat is taken up by the fluid,
the enthalpy increases and the ice fraction decreases as the process
goes vertically up in the diagram. The enthalpy change is obtained
by the refrigerating effect 'h = Q/m applied to the fluid.

106
Example 1: Let us as an example assume that an aqueous solution
of 11 mass-% ethyl alcohol, cA = 0.11 with 5 mass-% ice, cI =
0.05, entering the ice generator that has the refrigerating effect
Q/m = -90 kJ/kg applied to the fluid. What freezing point, tem-
perature, enthalpy, ice mass fraction and density can we expect to
get? (Compare also Guilpart et al 2005). The entering condition
corresponds to point A in Figure 6.11 and this point is indicated
also in Figures 6.12-14. The process in the diagrams is vertical as
the additive mass fraction is the same after the ice generator. Note
first from Figure 6.11 that the freezing point temperature of the
solution, tF = -5.0°C, is obtained by going vertically up to the
freezing point curve, marked h,tF.
Figure 6.14 gives the enthalpy value of the mixture when entering
the ice generator to hin =-70 kJ/kg. When exiting the ice generator
at B it is hex = -70 -90 = -160 kJ/kg and cI = 0.31. From isotherms
in the two-phase region we find the temperature to be tB = -8.0°C
(In Figure 6.14, B is on the line h,tF-3 , hence t = -5-3 = -8°C. Fig-
ure 6.13 gives densities as: UIS,in|979.5 kg/m3, UIS,ex|959.5 kg/m.
50
Ethyl alcohol - water
10°C
0
-1°C 5°C
cI=0,1 -2°C h,t F 0°C
-3°C
-50 -4°C
-5°C -5°C
cI=0,2 A -6°C
-8°C
h,t F- 1 D -10°C
-100
Enthalpy, [kJ/kg] .

cI=0,3
cI=0,2 h,t F- 2 -12°C
cI=0,4 E
C h,t F- 3
-150 -15°C
cI=0,5 B
cI=0,4 h,t F- 5
-200 cI=0,6
h,t F- 7 -20°C
cI=0,7 h,t F- 10
-25°C
-250
cI=0,8 cI=0,6

-300
cI=0,8
-350
0,00 0,03 0,06 0,09 0,12 0,15 0,18 0,21
Additive mass fraction [kg/kg sol] Melinder, 2007

Figure 6.11. Enthalpy phase diagram of ethyl alcohol – water for

examples 1 and 2

107
 0

-25
-50

-75 A
Ice slurry enthalpy [kJ/kg] D
-100

-125 E
C
-150
B t-tF=0
-175 t-tF=-0,5
t-tF=-1
-200 t-tF=-1,5
t-tF=-2
-225 t-tF=-2,5
t-tF=-3
-250 t-tF=-4
t-tF=-5
-275 t-tF=-6
t-tF=-7
-300
0,00 0,03 0,06 0,09 0,12 0,15 0,18 0,21
Additive mass fraction [kg add./kg sol.]

Figure 6.12. Ice slurry enthalpy of ethyl alcohol - water

0,60
t-tF=0
0,55 t-tF=-0,5
t-tF=-1
0,50 t-tF=-1,5
t-tF=-2
Ice mass fraction [kg ice/kg sol] .

0,45 t-tF=-2,5
t-tF=-3
0,40 t-tF=-4
t-tF=-5
0,35 t-tF=-6
B t-tF=-7
0,30

0,25

0,20

0,15

0,10

0,05 A

0,00
0,00 0,03 0,06 0,09 0,12 0,15 0,18 0,21
Additive mass fraction [kg add./kg sol.]

Figure 6.13 Ice mass fraction of ethyl alcohol - water

108
 990

985

980 A
975
Ice slurry density [kg/m3]

970

965

960 B t-tF=0
t-tF=-0,5
955 t-tF=-1
t-tF=-1,5
950 t-tF=-2
t-tF=-2,5
t-tF=-3
945
t-tF=-4
t-tF=-5
940 t-tF=-6
t-tF=-7
935
0,00 0,03 0,06 0,09 0,12 0,15 0,18 0,21
Additive mass fraction [kg add./kg sol.]

Figure 6.14. Ice slurry density of ethyl alcohol - water

6.5.3 Examples of simple adiabatic ice

slurry mixing processes
What happens if solutions with different additive concentrations
are mixed? If we can assume that the mixing takes place under
adiabatic conditions, then (what has been called) the seesaw prin-
ciple can be applied in an enthalpy-phase diagram and the mixing
will ideally take place on a straight line between the two conditions.
See Melinder and Hägg (2005), Granryd and Melinder (2005).
In Example 2 two ice slurry solutions with different additive con-
centrations are mixed. An ice slurry solution of ethyl alcohol – wa-
ter with cA = 0.16; t = -9°C (tF | -8°C; cI | 0.09) is mixed with
twice as much of an ice rich solution with: cA = 0.04; cI = 0.40 (t |
-2.9°C; tF | -1.7°C). See points D and C in Figs. 6.11 and 6.12.
The mixture will then get the condition in E: cA =
(2˜0.04+1˜0.16)/3 = 0.08; tF | -3.5°C; t | -5.2°C; cI = 0.29.
Assume in Example 3 that 25% ethanol – water (cA = 0.25) at t =
-12°C point G (above the freezing point curve in Figure 6.15) is
mixed with the same mass amount of ice – water (cA = 0; t = 0°C)
with the ice mass fraction cI = 0.6, in F. See Figure 6.15. What will

109
the ideal mixing condition be? According to the seesaw principle,
the mixing condition will be on a straight line between G and F, at
point H with cH = (0.25–0)/(1+1) = 0.0125. The freezing point of
this mixture (cA = 0.125) is tF | -5.8°C and the ice mass fraction is
cI | 0.26. The enthalpy of the mixture, hH, can be estimated by the
seesaw principle as hH | 1˜(-332.4˜0.6) + 1˜(-100) | -150 kJ/kg (see
Figure 6.15). This example illustrates how ice – water can be mixed
with a single-phase secondary fluid (that could be produced with-
out ice generator) at or near the freezing point ideally giving ice
slurry of suitable composition as a result.
0 10°C
5°C Ethyl alcohol - w ater
-2°C 5°C
c I,m=0,1 0°C
-4°C 0°C
-50
c I,m=0,2 -6°C -5°C -5°C
-8°C
- -10°C
-10°C G
-100 cI=0,3
-15°C
h,tF
-15°C
-20°C
Enthalpy, [kJ/kg] .

cI=0,4 h,tF-2
H cI=0,2 -20°C K
-150 -25°C
-25°C
cI=0,5 h,tF-5
cI=0,2 J
FcI=0,6 cI=0,4 M
-200 30 -30°C
h,tF-10
cI=0,4 L -
cI=0,7
-250 -35°C
cI=0,8
cI=0,6 I -40°C

-300

cI=0,8

-350
0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4
Additive mass fraction [kg/kg sol] Melinder, 2007

Figure 6.15 Enthalpy-phase diagram of ethyl alcohol – water for

examples 3 and 4.

In Example 4 ice slurry solutions with different additive concentra-

tions are mixed. An ice slurry solution of ethanol – water, cA |
0.296; t | -25°C; tF | -20°C; cI | 0.15 is mixed with the same mass
of an ice rich solution with: cA = 0.16; t | -25°C; cI = 0.54. See
points J and I in Figure 6.15. The mixture has the condition in
point L: cA = (1˜0.296+1˜0.16)/2 = 0.228; t | -25°C; cI = 0.34.

110
This is a higher mass ice fraction than in point J, but Figure 6.15
shows that in this case it is not possible to have larger enthalpy
change between the temperatures t | -25°C and t | -20°C from
point L (to M) than it is from point J (to K). In other words, it is
not possible to melt more ice between the given temperatures.
The enthalpy-phase diagram and the seesaw-principle show that
even if one mixes in ice or an ice rich solution, as from a storage
tank, it is not possible to melt more ice than at thermal equilibrium
between given temperatures, t and tF. Even though someone might
think that ice slurry becomes more effective if more ice is mixed in
than what is obtained at thermal equilibrium, this example shows
that this is not always the case. All these examples given show that
an enthalpy-phase diagram and other charts and tables may be of
value to visualize both possibilities and limitations of ice slurries at
various temperatures. See also PAPER 9 (Melinder 2006b).

6.6 Viscosity of ice slurries in

equilibrium
The shear viscosity of a concentrated suspension, taking into ac-
count the hydrodynamic interaction of particles, particle rotation,
collision between particles, agglomerate formation, and so on, can
be expressed by a theoretical equation in a power series (Thomas,
1965):
K * / K1 1  a1c  a2 c 2 r a3c 3 r ... . (6.13)

The constant a1 is assumed to have the value a1 2.5 determined

by Einstein (Kauffeld et al 2005). Based on many experimental in-
vestigations of the viscosity of suspensions it was shown that the
term 1  2.5 ˜ c  10.05 ˜ c 2 accounts for over 97.5% of the values
of the relative viscosity K * / K1 for volume fractions of particles, c,
below 25% ( c < 0.25). For values of c > 0.25 satisfactory results
were obtained by Thomas (1965):
K * / K1 1  2.5 ˜ c  10.05 ˜ c 2  ' (6.14)

where ' A ˜ e B˜c , A 0.0273, B 16.6, K * is shear viscosity of

the suspension, and for two-phase mixtures like ice-slurry the
background shear viscosity K1 is the viscosity of the liquid solu-
tion.

111
The Thomas equation that assumes a spherical form of particles
and Newtonian behavior has been used extensively as a reference
in ice slurry research, especially for low or moderate ice concentra-
tions. It can for ice slurries be written as follows
PIS=PCS(1+2,5 cI,v+10,05 cI,v 2 +0,00273·e16,6·cI,v) (6.15)
where cIv is the volumetric ice fraction as given in equation 6.7.
Viscosity values of many aqueous solutions can be obtained from
Melinder (1997). Some literature data for the shear viscosity of so-
lutions are also given in Tvsetkov (2001) and Vargaftic (1972).
Figure 6.16 shows the dynamic viscosity of ice slurry in equilib-
rium (based on eqn. 6.15) as function of temperature for six types
of aqueous solutions for ice slurries with concentrations giving the
freezing point tF = –5qC and for single-phase fluids with concen-
trations giving the freezing point tF = –15qC.
35
Ice slurry, tF=-5°C
Ethylene Glycol
30
Propylene Glycol
Dynamic viscosity [mPas] .

Ethyl Alcohol
25
Sodium Chloride
Potassium Acetate
20
Potassium Formate
Liquid only, tF =-15°C
15
Ethylene Glycol
Propylene Glycol
10
Ethyl Alcohol
Sodium Chloride
5
Potassium Acetate
Potassium Formate
0
-12 -10 -8 -6 -4 -2 0
Temperature [°C]

Figure 6.16. Dynamic viscosity as function of temperature for

ice slurry and single phase of some aqueous solutions

The figure shows how the viscosity of ice slurry increases rapidly
below the freezing point temperature as ice crystals are formed and
the ice fraction increases rapidly. The aqueous solutions with high-
est viscosity are in this case ethyl alcohol and propylene glycol.

112
Their viscosity is about double that of the salts, still not viewed as
too high to be used.

6.7 Thermal conductivity of ice slurry

The theory of the heat conduction phenomena through a random
suspension of spherical particles in a matrix of uniform thermal
conductivity has suggested that the volume fraction of the particles
is small enough to make all interactions between the spheres negli-
gible. For this kind of suspension an effective thermal conductivity
k * in the Maxwell approach can be represented by
k * k 1 1  3Ec (6.16)

where E (D  1) (D  2), D k 2 k1 , c is the volume fraction of

the particles (spheres), k1 is the thermal conductivity of a matrix,
and k 2 is the thermal conductivity of the spheres.
However, the validity of equation (6.16) is restricted to a very small
volume fraction of particles.
To represent the actual behaviour of the thermal conductivity in-
cluding the effect of interactions between spheres, the following
approximation of k * has been proposed (Jeffrey, 1972).
D 1
k * k 1 1  3Ec  3E 2 c 2 (1  2G ) (6.17)
D 2
The information about the coefficients G , E and D is summa-
rized by Oleg B. Tsvetkov in chapter 2 of IIR Handbook on Ice
Slurries (Kauffeld et al 2005). For the ice-slurry models the back-
ground conductivity k1 is the thermal conductivity of the liquid
water solution, and k 2 is the thermal conductivity of ice.
In Figure 6.17 the overall thermal conductivity of ice slurries in
equilibrium, kIS, is estimated by the following mixing equation (as-
suming spherical form of the ice slurry particles) given by Perry
(1973)
kIS = kCS(1+2cIv y)/(1-cIv y) (6.18)
where
y=(1- kCS/kI)/(2·kCS/kI+1) (6.19)

113
where kCS is the thermal conductivity of the concentrated aqueous
solution, kI is the thermal conductivity of ice and cIv is the volu-
metric ice fraction as given in equation 6.8.
Figure 6.17 shows the overall thermal conductivity of ice slurry in
equilibrium (based on eqn. 6.18) as function of temperature for six
types of aqueous solutions for ice slurries with concentrations giv-
ing the freezing point tF = –5qC and for single-phase fluids with
concentrations giving the freezing point tF = –15qC. The figure
shows that the thermal conductivity increases rapidly below the
freezing point temperature, as ice crystals are formed. This is due
to the high thermal conductivity of ice. The figure shows that the
thermal conductivity obtained for ice slurries is much higher than
for single phase aqueous solutions. Sodium chloride and potassium
formate have highest values as ice slurries with the freezing point
partly because of the high values of the single phase aqueous solu-
tions.
1,0
Ice slurry, tF=-5°C
0,9 Ethylene Glycol
Propylene Glycol
Thermal conductivity [W/mK] .

0,8 Ethyl Alcohol

Sodium Chloride

0,7 Potassium Acetate

Potassium Formate
Liquid only, tF=-15°C
0,6
Ethylene Glycol
Propylene Glycol
0,5
Ethyl Alcohol
Sodium Chloride
0,4
Potassium Acetate
Potassium Formate
0,3
-12 -10 -8 -6 -4 -2 0
Temperature [°C]

Figure 6.17. Thermal conductivity as function of temperature for

ice slurry and single phase of some aqueous solutions

114
6.8 Nomenclature
The following nomenclature is used in chapter 6.
cA additive mass fraction [kg add./kg sol.]
cp specific heat [J/(kg˜K)]
cI ice mass fraction [kg ice/kg sol.]
cIv ice volume fraction [m3 ice/ m3 sol.]
h enthalpy [J/kg]
k thermal conductivity [W/m K]
m mass [kg]
m mass flow rate [kg/s]
t temperature [°C]
Q cooling capacity [W]
V volume flow rate [m3/s]
'h enthalpy difference [J/kg]
'hM heat of mixing [J/kg]
't temperature change [K]
K* shear viscosity of the suspension [Pa s]
K1 is shear viscosity of liquid solution [Pa s]
P dynamic viscosity [Pa s]
U density [kg/m3]
Index:
A additive
CS concentrated solution
F freezing point
I ice
IS ice slurry
0 initial
R reference
in inlet
ex exit

115
Summary and conclusions
Aqueous solutions have long been used as single phase secondary
working fluids for cooling in supermarkets, ice rinks, heat recovery
systems, heat pumps and other applications. They are increasingly
used also for freezers in supermarkets and other applications in
low temperature refrigeration.
When comparing different secondary working fluids, reliable data
are needed of basic thermo-physical properties in order to choose
suitable fluid and to make technical calculations of an indirect re-
frigeration or heat pump system.
The main focus of this work has been to select property data of
freezing point, density, viscosity, specific heat, and thermal con-
ductivity with good or acceptable technical accuracy of a number
of aqueous solutions used by the refrigeration and heat pump in-
dustry, rather than focusing on a limited number of property val-
ues or scientifically very accurate measuring techniques.
A thorough literature search was made to find the most reliable
property values of aqueous solutions. Detailed literature references
are given for thermo-physical properties of the following aqueous
solutions: ethylene and propylene glycol, ethyl and methyl alcohol,
glycerol, ammonia, potassium carbonate, calcium, lithium, magne-
sium and sodium chlorides as well as potassium acetate and potas-
sium formate. Some laboratory measurements were made when
literature values were incomplete or deemed unreliable. Methods
used are briefly described and results are given.
The material in the Eng. Licentiate Thesis: Thermo-physical proper-
ties of liquid secondary refrigerants, A Critical Review on Literature
References and Laboratory Measurements (Melinder 1998a) forms
the basis for the charts and tables used in the IIR-publication
Thermo-physical properties of liquid secondary refrigerants (Melin-
der, 1997). The present Doctoral thesis reports on the update
made since 1998, including review work done on two additional
fluids not covered in Melinder (1998a).
The thesis describes how the selection of property values results in
tables and charts intended for the industry. Coefficients for poly-
nomial equations are generated from these property values using a
Matlab program and this material is intended as a useful tool for
computer treatment. Aqueous solution of ethyl alcohol is used as

116
example to see how this process is made. This choice of fluid can
also be seen as a test of this method, as the basic thermo-physical
properties of aqueous solutions of ethyl alcohol present more chal-
lenges than the other fluids examined.
A comparison is made of a few types of aqueous solutions used as
secondary working fluids for two types of applications. The first
example is bedrock heat pumps and the second is cooling cabinets
in a supermarket. An effort is made to see how the additive con-
centration affects the thermal performance.
Most aqueous solutions used as single-phase secondary fluids can
also be used as ice slurry, a fluid consisting of liquid and ice where
small ice crystals are produced, usually with some type of ice gen-
erator. The ice crystals are then transported to the cooling object
from which heat is removed when ice crystals melt. This results in
less temperature change in the cooling object and makes it also
possible to reduce the volume flow rate and to use smaller pipe
dimensions in the system.
In order to choose a secondary fluid for ice slurry use and to make
correct technical calculations of the ice slurry system there is a
need to examine and evaluate thermo-physical properties and
other aspects of ice and of the aqueous solution used. For dimen-
sioning purposes it is of interest to estimate ice mass fraction and
enthalpy values and enthalpy-phase diagrams can serve that pur-
pose. This thesis presents enthalpy-phase diagrams made by author
that besides isotherms contain lines with ice fraction and lines
connecting enthalpies at freezing point and 1, 2, etc. to 10K below
the freezing point curve. Examples are given to see how the dia-
grams can be used .
The intention of the author is now to finalize the updated publica-
tion with Thermophysical properties of secondary working fluids and
make it ready for the industry and others who may want to use it.
At the same time, with some funds from the Swedish Energy
Agency and support from many companies, work to produce a
handbook on indirect systems with secondary working fluids has
also recently started.

117
References
Aittiomäki, A., Lahti, A., 1997, Potassium formate as a second refrig-
erant, Int. Journal of Refrigeration, vol. 20, No 4, p.276-282
Aittomäki, E.A., Karkkiainen, S.V., 1999, Lowering temperature by
brines in ice banks, 20th Int. Congress of Refrigeration, IIF/IIR,
Sydney.
Aittomäki, A., Kianta, J, 2003, Indirect Refrigeration Systems Design
Guide Book, Tampere Univ. of Technology, Finland.
American Institute of Physics Handbook, 1972, 3rd edition
Andersson, O., 1996, 2002, Internal reports of measurements of ther-
mal conductivity and volumetric heat capacity with transcient hot
wire technique at Umeå univ.
Arias, J., Lundqvist, P., 2005, Modelling supermarkets energy usage,
IIF/IIR, Commercial refrigeration, Vicenza
ASHRAE Handbook, 2001, Fundamentals
Bel, O. et al.; 1996, Thermal study of an ice slurry used as re-
frigerant in a cooling loop; Applications for Natural Refriger-
ants, Aarhus
Bel, O., Lallemand, A., 1999, Study of a two-phase secondary refrig-
erant. 1. Intrinsic thermophysical properties of an ice slurry, Int.
Journal of Refrigeration, IIR/IIF, Paris, p.164-174
Ben Lakhdar, M.A., 1998, Thermal hydraulic behavior of a two phase
mixture: The ice slurry. Theoretical and experimental studies
(French), Doctoral Thesis, Lyon, France.
Ben Lakhdar, M.A., Melinder. Å., 2002, Facing the challenge to pro-
duce ice slurry for freezer application, 5th IIR Workshop on Ice
Slurries, Stockholm
Bobkov V.A., 1977, Production and Application of Ice, Foods Indus-
try Publ. House, Moscow, p. 232.
Bošnjakoviü, F., 1961, Technische Thermodynamik, II Teil, Band 12,
in Wärmelehre und Wärmevirtschaft in Einzeldarstellung,
Dresden/Leipzig
Bäckström, M., Emblik, E., 1965, Kältetechnik
Carrier, Handbook of Air Conditioning System Design, 1965
Christensen, C., Hanks, R.W., Izatt, R.M., 1982, Handbook of Heats
of Mixing
Christensen, J., Gmehling, J., Rasmussen, P., Weidlich, U., 1986.
Heats of Mixing Data Collection, 1. Binary Systems

118
Conde,M.R., 2004, Properties of aqueous solutions oflithium and cal-
cium chlorides: formulations for use in air conditioning equipment
design, Int. Journal of Thermal Sciences, Volume 43, p.367-383.
CRC Handbook of Chemistry and Physics, 69th edit., Weast, 1988-89
DOW, Organic Chemicals / The Glycols, 1974
Egolf, P. W., Brühlmayer, J., Özvegyi, F., Abächerli, F., Renold, P.,
1996, Properties of ice slurry, IIRConf., Applications for Natural
Refrigerants, Aarhus, p. 12.1-10.
Egolf, P.W., Kauffelt, M., 2005, From physical properties to indus-
trial ice slurry applications, Int. Journal of Refrigeration, Volume
28, Number 1, p.4-12, Paris.
Evenmo, K., 1998, A New Secondary Fluid, Natural Working Flu-
ids’98, IIR – Gustav Lorentzen Conf., Oslo.
Fennema, O.R. Powrie, W.D., Marth, E.H., 1973, Low-temperature
preservation of foods and living matter, Wisconsin
Field, B.S., Kauffeld M., 2003, Use of Ice Slurry in a Supermarket
Display Case, 21st Int. Congress of Refrigeration, IF/IIR, Washing-
ton DC
Florian, O,, 1997, Heat Transfer and Pressure drop of Liquid Secon-
dary refrigerants - Experiments and theory; Engineering Thesis
work, Energy Technology, KTH.
Fournaison, L., Chourot, J.M., Guilpart, J., 2001, Different ice mass
fraction measurement methods, Fourth IIR Workshop on Ice Slur-
ries, Osaka, Japan
Fransson, Å., 1996, 2002, Internal reports of measurements of specific
heat with differential scanning calometry at Umeå univ.
Fukusako, S., Kozawa, Y., Yamada, M., Tanino, M., 1999, Research
and Development Activities on Ice Slurries in Japan, First IIR
Workshop on Ice Slurries, Yverdon-les-Bains, Switzerland
Geiringer, P.L., 1962, Handbook of Heat Transfer Media, Reinhold
Publ. Co., New York
Gmelins Handbuch der Anorganischer Chemie, 1960, Lithium, 20
Gnielinski, V., 1976, New equations for heat and mass transfer in tur-
bulent pipe and channel flows, Int. Chem. Eng., 16, 359-368.
Granryd, E., 2002, Perspective on ice slurry, Fifth IIR Workshop on
Ice Slurries, Stockholm, May
Granryd, E., Melinder, Å., 2005, Secondary refrigerants for indirect
systems (pp.6:2-6), in Refrigerating Engineering, Granryd, E. et al,
KTH, Stockholm.

119
Guilpart J., Fournaison L., Ben Lakhdar, M.A., 1999, Calculation
method of ice slurries thermophysical properties-application to wa-
ter/ethanol mixture, Int. Congress of Refrigeration, IIR/IIF, Sydney
Guilpart J., Stamatiou, E., Fournaison L., 2005, The control of ice
slurry systems: an overview, Int. Journal of Refrigeration, Volume
28, Number 1, p. 98-107, Paris.
Guilpart J., Stamatiou, E., Delahaye, A., Fournaison L., 2006, Com-
parison of the performance of different ice slurry types depending
on the application temperature, Int. Journal of Refrigeration, Vo-
lume 29, Number 15, p. 781-788, Paris.
Hofmann, E., 1959, Verfahren der Kälteerzeugung und Grundlagen
der Wärmeübertragung,.
Horton, W.T., Groll, E.A., 2003, Secondary Loop Refrigeration in Su-
permarket Applications: A Case Study, 21st Int. Congress of Re-
frigeration, IIR, Washington DC
Hägg, C, Melinder, Å., 2002, Selecting fluids suitable for ice slurry
applications at low Temperatures, Zero leakage – Minimum
Charge, IIF/IIR, Stockholm
International Critical Tables of Numerical Data, Physics, Chemistry
and Technology, 1928-29.
Int. Journal of Thermophysics, 1987, Vol. 8, p.351
Int. Journal of Thermophysics, 1994, Vol. 15, p.165
Jeffrey D.J., 1972, Conduction through a random suspension of
spheres, Proc. Roy. Soc., London A, vol. 335: p. 355-367.
Jokinen, J., 2003, Betaine based heat transfer fluid as a solution for
toxicity and corrosion problems in heating and cooling systems,
21st Int. Congress of Refrigeration, IF/IIR, Washington D.C.
Jokinen, J., Willems, B., 2004, Betaine based heat transfer fluid as a
natural solution for environmental, toxicity and corrosion problems
in heating and cooling systems, Sixth Gustav Lorenzen Conference
on Natural Working Fluids, IIF/IIR, Glasgow, UK.
Kauffeld M., Christensen K.G., Lund, S., Hansen, T. M., 1999, Ex-
perience with ice slurry, 20th Int. Congress of Refrigeration,
IIR/IIF, Sydney
Kauffeld M., Kawaji M., Egolf P.W., 2005, Handbook on Ice Slurries
– Fundamentals and Engineering, IIF/IIR, Paris
Kruse, H., 2005, Commercial refrigeration – on the way to sustain-
ability, IIF/IIR, Commercial refrigeration, Vicenza
Landolt-Börnstein, Eigenschaften der Materie in Ihren Aggregat-
zuständen, 1960-1971.

120
Landolt-Börnstein, Numerical Data and Functional Relationships in
Science and Technology; 1977, Vol. 1b
Landolt-Börnstein, Numerical Data and Functional Relationships in
Science and Technology, 1976, Group IV, Vol. 2. Heats of Mixing
and Solution
Lange’s Handbook of Chemistry, 1985, 13th edition
Lindborg, A, 2003, Indirect refrigeration Systems, 21st Int. Con-
gress of Refrigeration, IIF/IIR, Washington DC
Lottin, O., Epiard, C. Thermodynamic properties of some currently
used water - antifreeze mixtures when used as ice slurries, Eighth
Int. Refrigeration, Purdue, USA, p.391-398, 2000
Lundqvist, P.G., 2000. Recent refrigeration equipment trends in su-
permarkets: Energy efficiency as leading edge?, Bulletin No
2000.5 of the Int. Institute of Refrigeration, IIR, Paris
Macmillan’s Chemical and Physical Data, 1992
Malter, L., 1996, Binary-ice generation and applications of pumpable
ice slurries for indirect cooling, Application for Natural Refriger-
ant, Aarhus
Mao, Y., Hrnjak, P.S., 1999, Defrost issues of display cabinets with
evaporating R404A and heat exchangers with secondary refriger-
ants, 20th Int. Congress of Refrigeration, IIF/IIR, Sydney.
Meewisse, J.W., 2004, Fluidized bed Ice Slurry Generator for En-
hanced Secondary Cooling Systems, Doctoral Thesis, Delft, The
Netherlands.
Melinder, Å., Berendson, J., Granryd, E., Kyrk, B., 1989, Köldbärare
för värmepumptillämpningar. Termodynamiska egenskaper och
korrosion för kaliumkarbonat samt jämförelse med andra köldbära-
re, Report R18:1989, Swedish Council for Building Research
Melinder, Å., 1997, Thermo-physical properties of liquid secondary
refrigerants. Tables and Diagrams for the Refrigeration Industry,
IIF/IIR, Paris
Melinder, Å., 1998, Thermophysical properties of liquid secondary
refrigerants, A Critical Review on Literature References and
Laboratory Measurements, (Engineering Licentiate Thesis), KTH,
Stockholm
Melinder, Å, 1998, A Critical Review on Thermophysical Properties
of Liquid Secondary Refrigerants, Natural Working Fluids’98, IIR
– Gustav Lorentzen Conf., Oslo
Melinder, Å., 2002a, Enthalpy-phase diagrams of aqueous solutions
for ice slurry applications, Fifth IIR Workshop on Ice Slurries,
Stockholm

121
Melinder, Å., 2002b, Influence of temperature on latent heat benefits
in ice slurry applications, Zero leakage – Minimum Charge,
IIF/IIR, Stockholm
Melinder, Å., 2003a, Comparing Properties of Aqueous Solutions for
Liquid Only and Ice slurry Applications of Indirect Systems, Proc.
21st Int. Congress of Refrigeration, IIR/IIF, Washington
Melinder, Å., 2003b, Update on indirect systems and secondary fluids
with focus on supermarket refrigeration, Int. Conf. on Fluid and
Thermal Conversion, FTEC2003, Bali, Indonesia
Melinder, Å., 2004, Secondary fluids for low operating temperatures,
Sixth Gustav Lorenzen Conference on Natural Working Fluids,
IIF/IIR, Glasgow, UK
Melinder, Å., 2005, Properties of various secondary fluids for super-
market applications, Thermophysical Properties and Transfer
Processes of Refrigerants, IIF/IIR, Vicenza, Italy
Melinder, Å., Granryd, E., 2005, Using property values of aqueous
solutions and ice to estimate ice concentrations and enthalpies of
ice slurries, Int. Journal of Refrigeration, Volume 28, Number 1,
p.13-19, Paris
Melinder, Å., Hägg, C., 2005, Enthalpy-phase diagrams and the see-
saw-principle – aids to describe various ice slurry conditions,
Thermophysical Properties and Transfer Processes of Refrigerants,
IIF/IIR, Vicenza, Italy.
Melinder, Å, 2006a, Updating and extending secondary flud data for
various refrigeration and heat pump applications, Seventh Gustav
Lorenzen Conference on Natural Working Fluids, IIF/IIR, Trond-
heim
Melinder, Å, 2006b, Equilibrium properties of ice slurry based on
aqueous solutions of ethyl alcohol, Seventh Gustav Lorenzen Con-
ference on Natural Working Fluids, IIF/IIR, Trondheim
Melinder, Å, 2006c, Equilibrium properties of ice slurry based on two
chosen types of aqueous solutions, Seventh IIR Conference on
Phase Change Materials and Slurries for Refrigeration and Air
Conditioning, Dinan, France
Melinder, Å, 2006d, Updating properties of secondary working fluids
for various refrigeration systems, Int. Conf. on Fluid and Thermal
Conversion, FTEC2006, Jakarta
Mellan, 1970, Industrial Solvents Handbook, New Jersey: Noyes Data
Corporation, 133 p.
Paul, J., 1995, Binary Ice as a secondary refrigerant; 19th Int Con-
gress of Refrigeration, Vol.4b, Den Hague

122
Paul, J., Jahn, E., 1997, Binary-ice generation and applications of
pumpable ice slurries for indirect cooling, The Inst. of Refrigera-
tion, UK
Paul, J., Malter, L., Münster, U., 1999, Generation and utilization of
liquid ice (binary ice), 20th Int. Congress of Refrigeration, IIR/IIF,
Sydney.
Pearson, S. F., 1993, Development of Improved Secondary Refriger-
ants; Int. Institute of Refrigeration, UK
Perry, R.H., Chilton, C.H., 1973, Chemical Engineers’ Handbook, 5th
edition, New York
Plank, R., 1988, Handbuch der Kältetechnik, VI/B (edited by Steimle,
F.).
Pronk, P., Hansen, T.M., Infante Ferreira, C.A., Witkamp G.C.,2005,
Time dependant behaviour of different ice slurries during storage,
Int. Journal of Refrigeration, Volume 28, Number 1, p.27-36,
2005, Paris.
Pronk, P., 2006, Fluidized bed heat exchangers to prevent Fouling in
ice slurry systems and industrial crystallizers, Doctoral Thesis,
Delft, The Netherlands.
Rant, Z. 1961, Diagramm-Mappe zu Band 13, Verdampen in Theorie
und Praxis, in Wärmelehre und Wärmevirtschaft
Raznjevic, K. 1976, Handbook of Thermodynamic Tables and-
Charts.
Schmidt, G.H., Design experience with Flo-Ice and binary ice as a
secondary refrigerant for refrigeration installations, First IIR
Workshop on Ice Slurries, Yverdon-les-Bains, Switzerland
Sherwood, G., 1995, Secondary Heat Transfer Systems and the Appli-
cation of a New Hydrofluoroether; The 1995 International CFC and
Halon Alternatives Conference
Sieder, E.N., Tate, G.E., 1956, Heat transfer and pressure drop of
liquids in tube, Int. Chem. Eng., 28, p.1429-1435
Snoek, C.V., Hampton, D., 1993, Ice slurry based district cooling sys-
tems, Proceedings, Int District Heating and Cooling Ass. 8th Ann.
Cooling Conference, Tulsa, Oklahoma
Snoek, C.W., 1995, District cooling using ice slurry as the heat
transfer fluid, Nouvelles technologies et sécurité dans le froid,
Journées Européennes du froid, p.157-184, AFF-lnterclima, Paris.
Stakelbeck, H., Plank, R., 1929, Ueber die Zähigkeit von
Chlornatrium-, Chlorkalzium- und Chlormagnesium-lösungen in
Abhägigkeit von Temperatur and Konzentration, Zeitschrift für Die
Gesamte Kälte-Industrie, p.105-112

123
Stamatiou, E., Meewisse, J., Kawaji, M., Ice slurry generation involv-
ing moving parts, Int. Journal of Refrigeration, Volume 28, Num-
ber 1, p. 60-72, 2005, Paris
Svensk Kylnorm, 1992, Faktablad 7, Etanolbaserade köldbärarsystem
Thomas, D.G., 1965, Transport characteristics of suspension, Journal
of Colloid Science Vol. 20, p. 267-277
Timmermans, J., 1960, The Physico-chemical Constants of Binary
Systems in Concentrated Solutions, Vol. 3 and 4
Tsvetkov O.B., Laptev Ju. A., Kolodiaznaia V.S., 2001, One-and two-
phase refrigerating liquid media, Kholodilnaja Teknika, no. 10: p.
8-12.
Ure, Z., 2003, Secondary refrigeration developments and experi-
ences in Europe, 21st Int. Congress of Refrigeration, IIF/IIR,
Washington DC
Vargaftic, N.B., 1972: Thermophysical properties of gases and liq-
uids. Handbook, Nauka Publ. House, Moscow
VDI Heat Atlas, 1993
VDI Wärmeatlas, 2002, Beräknungsblätter für den wärmeübergang,
VDI-GVC
Vervoerd, M., Liem, N., van Gerwen, R.J.M., 1999, Performance of a
frozen food display cabinet with potassium formate as secondary
refrigerant, 20th Int. Congress of Refrigeration, IIF/IIR, Sydney.
Wang, M.J., Hansen, T., Kauffelf, M., Christensen, K.C., Goldstein,
V., 1999, Application of ice slurry technology in fishery, 20th Int.
Congress of Refrigeration, IIF/IIR, Sydney
Wang, M.J., Goldstein, V., 1996, A novel ice slurry generation sys-
tem and its applications. 2nd Conf. on the use of non-artificial sub-
stances, Aarhus, DK, p. 12-19-12-23
Wimby, J.M., Berntsson, T.S., 1994, Viscosity and density of aqueous
solutions of LiBr, LiCl, ZnBr2, CaCl2 and LiNO3”, 1. Single salt
solutions, J. Chem. Engineering Data 39, p.68-72
Yang Y., 2001, Investigation of Deep-freeze Refrigeration Systems
in Supermarket Application, Masters Thesis, Energy Technol-
ogy, KTH, Stockholm.

124
APPENDIX
Table 2.3. Conversion chart of additive concentrations
Additive Additive conc., % by volume
conc., % Ethylene Propylene Ethyl Glycerol,
by weight glycol, EG glycol, PG alcohol, EA GL
0 0 0 0 0
5 4.4 4.8 6.24 4.2
10 8.9 9.6 12.39 8.5
15 13.6 14.5 18.47 12.7
20 18.1 19.4 24.47 17.0
25 22.9 24.4 30.39 21.2
30 27.7 29.4 36.18 25.5
35 32.6 34.4 41.83 30.2
40 37.5 39.6 47.33 35.0
45 42.5 44.7 52.66 39.5
50 47.6 49.9 57.83 44.1
55 52.7 55.0 62.84 49.3
60 57.8 60.0 67.69 54.5
65 62.8 65.0 72.38 59.5
70 68.3 70.0 76.91 64.5
75 73.6 75.0 81.27 69.7
80 78.9 80.0 85.46 75.0
85 84.3 85.0 89.46 81.2
90 89.7 90.0 93.25 87.4
95 95 95.0 96.79 93.7
100 100 100 100 100

Table 2.4. Index of refraction of aqueous solutions at 20°C,

n20°C, as function of the additive concentration
wt-% EG PG EA MA GL NH3
0 1,333 1,333 1,333 1,333 1,333 1,333
2 1,335 1,335 1,334 1,333 1,335 1,334
5 1,338 1,338 1,336 1,334 1,339 1,335
10 1,343 1,344 1,340 1,335 1,345 1,338
15 1,347 1,349 1,343 1,337 1,351 1,341
20 1,352 1,355 1,347 1,338 1,357 1,344
24 1,356 1,360 1,350 1,339 1,362 1,345
28 1,361 1,364 1,352 1,340 1,368 1,347
32 1,365 1,369 1,355 1,341 1,373 1,348
36 1,369 1,374 1,357 1,342 1,379 1,349
40 1,373 1,378 1,358 1,343 1,384 1,350

125
 wt-% EG PG EA MA GL NH3
44 1,377 1,382 1,360 1,343 1,390
48 1,381 1,387 1,361 1,343 1,395
52 1,385 1,391 1,362 1,343 1,401
56 1,389 1,395 1,363 1,343 1,407
60 1,393 1,400 1,364 1,343 1,413

wt-% K2CO3 CaCl2 MgCl2 NaCl LiCl

0 1,333 1,333 1,333 1,333 1,333
2 1,337 1,338 1,338 1,337 1,335
5 1,342 1,345 1,346 1,342 1,344
10 1,350 1,358 1,359 1,351 1,354
15 1,359 1,371 1,372 1,359 1,365
20 1,367 1,384 1,386 1,368 1,376
22 1,371 1,390 1,392 1,372 1,380
24 1,374 1,395 1,398 1,376 1,384
26 1,377 1,401 1,403 1,389
28 1,381 1,407 1,409 1,394
30 1,384 1,412 1,415 1,398
32 1,387 1,418
36 1,394
40 1,401

Table 2.5. Thermal volume expansion, TVE, of aqueous solu-

tions with concentrations giving tF = -15ºC and water.
Fluid Temperature of fluid[°C]
symbol -15 -10 0 10 20 30 40
EG 0,000 0,100 0,347 0,654 1,017 1,434 1,902
PG 0,000 0,120 0,420 0,795 1,237 1,738 2,291
EA 0,000 0,110 0,410 0,808 1,291 1,849 2,469
MA 0,000 0,060 0,238 0,493 0,820 1,218 1,683
GL 0,000 0,119 0,403 0,739 1,115 1,521 1,947
NH3 0,000 0,034 0,179 0,416 0,733 1,118
K2CO3 0,000 0,100 0,356 0,675 1,042 1,445 1,867
CaCl2 0,000 0,085 0,318 0,612 0,941 1,277 1,618
NaCl 0,000 0,163 0,506 0,872 1,261 1,672 2,106
KAC 0,000 0,113 0,368 0,663 0,995 1,364 1,767
KFO 0,000 0,077 0,287 0,552 0,852 1,167 1,476
LiCl 0,000 0,020 0,116 0,279 0,497 0,760 1,059
H2O 0,000 0,000 0,015 0,165 0,415

126
Results of measurements of viscosity
CALCIUM CHLORIDE - WATER (Melinder; 1983-1996)
Concentration (wt-%) 28 28 28 28 28
Number of measurem. 2 3 3 3 4
Mean temperature (°C) -18,9 -9,0 2,0 13,0 24,4
Max temp variation (°C) 0,1 0,1 0,2 0,1 0,1
Density (kg/m³) 1277 1273 1268 1263 1257
Ball constant 0,0700 0,0700 0,0700 0,0700 0,0700
Ball density (kg/m³) 2228 2228 2228 2228 2228
Aver falling time (s) 143,17 97,27 68,48 50,40 38,75
Dyn. visc. (mNs/m³) 9,53 6,50 4,60 3,40 2,63
Kin. visc. (mm²/s) 7,46 5,11 3,63 2,70 2,10
Concentration (wt-%) 28 28 28 28 28
Number of measurem. 3 4 5 5 5
Mean temperature (°C) -18,6 -9,1 0,4 9,9 20,0
Max temp variation (°C) 0,2 0,1 0,2 0,1 0,1
Density (kg/m³) 1277 1273 1269 1264 1259
Ball constant 0,0718 0,0718 0,0718 0,0718 0,0718
Ball density (kg/m³) 2411 2411 2411 2411 2411
Aver falling time (s) 113,87 78,98 58,44 45,67 36,09
Dyn. visc. (mNs/m³) 9,27 6,45 4,79 3,76 2,99
Kin. visc. (mm²/s) 7,26 5,07 3,78 2,98 2,37
Concentration (wt-%) 22 22 22 22 22
Number of measurem. 5 5 5 5 5
Mean temperature (°C) -21,4 -15,0 -6,8 -0,6 5,0
Max temp variation (°C) 0,1 0,1 0,1 0,1 0,1
Density (kg/m³) 1212 1211 1209 1207 1205
Ball constant 0,0718 0,0718 0,0718 0,0718 0,0718
Ball density (kg/m³) 2411 2411 2411 2411 2411
Aver falling time (s) 82,38 64,45 48,68 40,35 34,77
Dyn. visc. (mNs/m³) 7,09 5,55 4,20 3,49 3,01
Kin. visc. (mm²/s) 5,85 4,59 3,47 2,89 2,50
Concentration (wt-%) 22 18 18 18
Number of measurem. 6 1 2 3
Mean temperature (°C) 10,0 -6,95 -4,50 1,28
Max temp variation (°C) 0,1 0,1 0,1 0,1
Density (kg/m³) 1204 1168 1167 1165
Ball constant 0,0718 0,0718 0,0718 0,0718
Ball density (kg/m³) 2411 2411 2411 2411
Aver falling time (s) 30,04 39,92 36,89 30,87
Dyn. visc. (mNs/m³) 2,60 3,56 3,29 2,76
Kin. visc. (mm²/s) 2,16 3,05 2,82 2,37

127
LITHIUM CHLORIDE - WATER (Melinder, 1996)
Concentration (wt-%) 20 20 20 20 20
Measuring serie 1 2 3 4 5
Number of measurem. 4 4 4 4 4
Mean temperature (°C) -18,8 -13,4 -5,0 2,3 9,8
Max temp variation (°C) 0,1 0,1 0,1 0,1 0,1
Density (kg/m³) 1124 1123 1122 1120 1118
Ball constant 0,0700 0,0700 0,0700 0,0700 0,0700
Ball density (kg/m³) 2228 2228 2228 2228 2228
Aver falling time (s) 92,60 73,85 54,18 42,74 34,57
Dyn. visc. (mNs/m³) 7,16 5,71 4,20 3,31 2,69
Kin. visc. (mm²/s) 6,37 5,09 3,74 2,96 2,40

Concentration (wt-%) 20 20 20
Measuring serie 6 7 8
Number of measurem. 5 3 4
Mean temperature (°C) 15,0 22,3 32,5
Max temp variation (°C) 0,1 0,1 0,1
Density (kg/m³) 1117 1115 1111
Ball constant 0,0700 0,00918 0,00918
Ball density (kg/m³) 2228 2228 2228
Aver falling time (s) 30,19 198,45 165,92
Dyn. visc. (mNs/m³) 2,35 2,03 1,70
Kin. visc. (mm²/s) 2,10 1,82 1,53

Concentration (wt-%) 15 15 15 15 15
Measuring serie 1 2 3 4 5
Number of measurem. 4 4 5 5 2
Mean temperature (°C) -16,8 -13,0 -5,0 5,0 21,9
Max temp variation (°C) 0,1 0,1 0,1 0,1 0,1
Density (kg/m³) 1092 1091 1090 1089 1085
Ball constant 0,0700 0,0700 0,0700 0,0700 0,00918
Ball density (kg/m³) 2228 2228 2228 2228 2228
Aver falling time (s) 68,05 57,75 43,11 31,26 153,85
Dyn. visc. (mNs/m³) 5,41 4,60 3,43 2,49 1,62
Kin. visc. (mm²/s) 4,96 4,21 3,15 2,29 1,49

128
POTASSIUM FORMATE (HYCOOL) (Melinder, 1999)
Concentration (vol-%) 45 45 45 40 40
Density at 20°C (kg/m³) 1318 1318 1318 1294 1294
Concentration (vt-%) 47 47 47 44 44
Number of measurements 4 4 5 4 4
Mean temperature (°C) 5,8 10,5 20,3 2,8 7,1
Max temp variation (°C) 0,1 0,1 0,1 0,2 0,1
Density (kg/m³) 1326 1323 1318 1303 1301
Ball constant 0,0700 0,0700 0,0700 0,0700 0,0700
Ball density (kg/m³) 2228 2228 2228 2228 2228
Aver falling time (s) 51,34 45,29 36,27 48,92 43,56
Dyn. visc. (mNs/m³) 3,24 2,87 2,31 3,17 2,83
Kin. visc. (mm²/s) 2,44 2,17 1,75 2,43 2,17

Concentration (vol-%) 40 30 30
Density at 20°C (kg/m³) 1294 1248 1248
Concentration (wt-%) 44 38 38
Number of measurem. 5 5 5
Mean temperature (°C) 19,4 5,24 11,5
Max temp variation (°C) 0,1 0,1 0,1
Density (kg/m³) 1294 1255 1252
Ball constant 0,0700 0,0700 0,0700
Ball density (kg/m³) 2228 2228 2228
Aver falling time (s) 32,60 36,62 31,47
Dyn. visc. (mNs/m³) 2,13 2,49 2,15
Kin. visc. (mm²/s) 1,65 1,99 1,72

129
APPENDIX PAPERS:
PAPER 1: Melinder, Å., 1998, A Critical Review on Thermophysical
Properties of Liquid Secondary Refrigerants, Natural
Working Fluids’98, IIR – Gustav Lorentzen Conf., Oslo
PAPER 2: Melinder, Å., 2002, Enthalpy-phase diagrams of aqueous
solutions for ice slurry applications, Fifth IIR Workshop
on Ice Slurries, Stockholm
PAPER 3: Melinder, Å., 2003, Update on indirect systems and sec-
ondary fluids with focus on supermarket refrigeration, Int.
Conference on Fluid and Thermal Conversion,
FTEC2003, Bali, Indonesia
PAPER 4: Melinder, Å., 2003, Comparing Properties of Aqueous So-
lutions for Liquid Only and Ice slurry Applications of In-
direct Systems, Proc. 21st Int. Congress of Refrigeration,
IIR/IIF, Washington
PAPER 5: Melinder, Å., 2004, Secondary fluids for low operating
temperatures, Sixth Gustav Lorenzen Conference on
Natural Working Fluids, IIF/IIR, Glasgow
PAPER 6: Melinder, Å., Granryd, E., 2005, Using property values of
aqueous solutions and ice to estimate ice concentrations
and enthalpies of ice slurries, Int. Journal of Refrigera-
tion, Volume 28, Number 1, p.13-19, Paris
PAPER 7: Melinder, Å., 2005, Properties of various secondary fluids
for supermarket applications, Thermophysical Properties
and Transfer Processes of Refrigerants, IIF/IIR, Vicenza
PAPER 8: Melinder, Å, 2006, Updating and extending secondary
fluid data for various refrigeration and heat pump applica-
tions, GLC 2006, Trondheim
PAPER 9: Melinder, Å, 2006, Equilibrium properties of ice slurry
based on two chosen types of aqueous solutions, 7th Con-
ference on Phase Change Materials and Slurries for Re-
frigeration and Air Conditioning, 2006, Dinan, France

130