Abstract be found in various domestic and industrial appliances,

Abstract

The experiment was run to predict the
performance of a pre-heater which would be used in an incineration plant. This
was achieved by operating a test rig to find the physical and thermal
properties of the CFC, pentafluorobutane. The properties were used to the find
the experimental heat flux, which was

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

 and the maximum heat transfer coefficient,
which was

.
The theoretical heat flux was also found using the Zuber equation and came to
be

which
had a percentage error of 4.8 %. The results obtained were used to find the
time needed to boil 175 kg of the CFC which came to be 101 minutes. The initial
heat transfer coefficients for convective boiling were very low however when
the boiling regime changed to nucleate a rapid increase was seen in the coefficient
until a maximum was reached. Similarly, the heat flux was initially low however
it continued increasing throughout the boiling process. The heat flux increased
fastest during the nucleate boiling phase. The data collected suggested that
the pre-heater should be operated within the nucleate boiling phase, as the
most efficient heat transfer occurs leading to shorter heating times and
reduced energy expenditure.

 

 Introduction and Theory

Immersion heaters can be found in various domestic
and industrial appliances, which include kettles, central heating boilers and
process plant pre-heaters. Process plant pre-heaters can be used within high temperature
incineration plants, which are an available method of disposing of compounds
which are known to be hazardous or harmful to the environment. The liquid
investigated in this experiment, pentafluorobutane, is a chlorofluorocarbon
(CFC) and an example of a compound which would be incinerated due to its harmful
effects on the environment. Before the compound can be fed into the plant, the
physical properties must be determined. Properties such as the critical heat
flux and the heat transfer coefficient have to be determined to ensure the
plant is run safely as well as at optimum conditions. 

The aim of this experiment was to generate
results from a laboratory scale heat transfer which could be used to predict
the performance of an industrial-scale pre-heater. This was achieved by observing
the relationship between the physical and thermal properties of pentafluorobutane
as a function of temperature as well as finding the experimental and
theoretical critical heat flux.

When a heating element, such
as the copper heating element, is submerged within a liquid and has a temperature
which exceeds the saturation temperature of the liquid, heat is transferrred from
the element to the liquid. The rate of heat transfer and regime of boiling is
determined by the difference between the surface temperature of the element and
the saturation temperature of the liquid at the specified pressure (Foust,
1980).
There are 3 definitive regimes of boiling. Initially, convective boiling occurs
in which heat is transferrred from the element to the surface of the liquid via
convection due to the buoyancy effect (Collier, 1981). Once a high enough
temperature difference is achieved, the heat is transferred as the latent heat
of vaporization which begins the formation of vapour bubbles. This marks the
beginning of the next regime known as nucleate boiling. At first vapour bubbles
form in favourable nucleation sites which have been found to be areas with
surface imperfections (Foust, 1980). The resulting
bubbles increase the agitation and the heat transfer coefficient of the liquid therefore
a greater heat flux was expected to be observed during the nucleate boiling
phase (Foust, 1980). The final regime,
film boiling, occurs once the heating element becomes immersed within a vapour layer.
As vapour has a considerably lower heat transfer coefficient compared to liquid
the heat flux is expected to decrease (Coulson JM, 1964). The heat transfer
coefficient can be determined using the following equation (Spakovszky, 2007)

 

Eq 1

 

 

where

 is the heat flux (Wm-2), h is the
heat transfer coefficient (Wm-2K-1), Ts is the
temperature at the surface of the heat element (K) and Tsat is the
saturation temperature of the fluid at a given pressure. The equipment used
allowed the heat flux to be determined using the power transferred as thermal
energy and the surface area of the element in equation 2 (Spakovszky, 2007):

 

Eq 2

where

 is the power
transferred as thermal energy (W)  and A
is the surface area of the heating element (m2) .

Method

The experimental equipment consisted of a
sturdy glass cylinder which held the saturated liquid, a copper heating element
and a water-cooled coil which acted as the condenser. The condenser coil was in
the top region of the cylinder and was used to control the temperature and
pressure of the process using the cooling water flow rate whereas the heating
element was in the bottom region of the cylinder completely submerged within
the liquid. The heating element also modified the temperature and pressure
depending on the electrical input.

To begin the experiment the connections for the
cooling water flow rate and electrical supply were made. The cooling water flow
rate was set to a specific level and the ambient readings of pressure and the temperature
of the heating element and the liquid were taken. The power supply for the
heating element was increased in regular intervals of about 15 W and allowed to
stabilize for 2 minutes before the readings of power, temperature of heating
element and liquid, pressure and flow rate were recorded. During this process,
the development of the boiling regime was studied and noted. The process was
repeated until the power was raised to 360 W where the intervals were reduced
to 10 W. The experiment ended once the thermal safety cut-out occurred for the
equipment used.

Figure 1.
Schematic of boiling heat transfer laboratory test rig

Results
and Calculations    

 

Film
Nucleate
Convective
 

Boiling
regime data collected in:
 

Figure 2. Heat
transfer coefficient as a function of the temperature difference

The heat transfer coefficient appeared to rapidly
increase for small changes in the temperature difference until a maximum was
reached. After the maximum, the heat coefficient decreased at a rate which was
lower than the rate at which it increased before the maximum.

The heat flux was calculated using equation 2,
therefore the surface area of the heating element had to be calculated. The
heating element was cylindrical with only one face exposed to the fluid,
therefore the following equation was used to find the surface area:

 

Eq 3

where A is the surface area (m2), d
is the diameter of the heating element (m) and r is the radius of the heating
element (m).

 

Dimensions for heating element:

Length, L=0.042 m

Diameter, D =0.0127 m

 

Sample calculation of heat flux using equation
2 and a power output of 90 W:

 

 

Figure 3.  Heat flux as a function of temperature
difference

The relationship between heat flux and
the temperature difference was initially hard to distinguish however for heat
fluxes above

 the heat flux was shown to increase at a
decreasing rate as the temperature difference increased.

The critical heat flux was found by marking
out the maximum flux at which nucleate boiling occurred in Figure 3:

 

The maximum heat transfer coefficient was
found by observing the highest value obtained for heat transfer coefficient in
Figure 3:

 

 

 

 

The programme ASPEN Properties was used
to determine the density of pentafluorobutane as a liquid and vapour at ambient
conditions (17 °C) for the calculation
of the theoretical critical heat flux using the Zuber equation  (Bath
University Department of Chemical Engineering, 2017):

 

1289.27 kgm-3

188.19 kJkg-1

6.56648 kgm-3

0.013392
Nm-1
 

 

 

 

Values for the latent heat of
vaporisation (

 and surface tension (

 were taken from refrigerator tables (Laboratory,
2017).

 

Comparison
of the experimental and theoretical critical heat fluxes:

 

 

 

Time
taken to boil 175 kg of the CFC

Assuming:

·     
Operation
at 80% of critical heat flux (

 

·        
Area
available, A= 0.032 m2

·        
Latent
Heat of Vaporisation, hfg= 188.19 kJkg-1

The
energy required to vaporise liquid, q, was found using the latent heat of
vaporisation,

(kJkg-1), and the mass of the
CFC,

 (kg):

 

 

The time taken, t, was found by finding
the rate of heat transfer using the heat flux and the heating element surface
area and dividing the energy required to vaporise the liquid by the rate:

 

Discussion

A relationship to the temperature difference
was hard to deduce for both the heat transfer coefficient and the heat flux for
the initial data collected as shown by Figures 2 and 3 respectively. This was a
result of the large temperature difference between the heating element and the liquid.
The initial data was collected whilst the boiling mechanism was seen to be convective
as little to no ebullition was observed, however convection currents were
visible at the heating element surface. The large disparity in temperature
readings for the heating element and the liquid may have been due to the limited
heat transfer as well as the distance between the thermometer and the heated
liquid. Error was introduced in the liquid temperature reading as the
thermometer was not able to record an average temperature for the first few
results due to the poor mixing preventing the increase in thermal energy being
observable. An electric heat sensor could be introduced to monitor the
temperature of the liquid as the thermometer became difficult to view as the
bubble formation increased and became more vigorous.

The change in boiling regime to nucleate
boiling was marked by the formation of small vapour bubbles. The first few nucleation
sites were located around surface imperfections such as where the surface of
the heating element was disturbed by protruding wires. Despite knowledge on the
analytic prediction of the formation of nucleation sites being limited, the
location of the nucleation sites has been found to be in pits and grooves along
surfaces which corresponds to the results of this experiment (Eckert &
Drake, 1972).
Figure 2 showed the change to nucleate boiling corresponded with a large
increase in heat flux whilst Figure 3 revealed a rapid increase in heat
transfer coefficient, which was expected due to similar results being derived
in literature (Foust, 1980).
The increase in the heat transfer coefficient may have been caused by vapour
bubbles which condensed before reaching the surface of the liquid therefore
releasing energy into the bulk liquid (Roshesnow, 1964). As the temperature difference was
increased, the bubble formation became more vigorous leading to turbulence which
caused the mixing within the fluid to increase (Roshesnow, 1964). The increase in
heat flux may have been a result of the combination of the increased mixing and
the condensation of vapour bubbles within the bulk fluid.

 After
the heat transfer coefficient reached a maximum of

,
Figure 2 shows the value of the coefficient gradually decreased as the
temperature difference increased at a rate which was lower than the initial
increase. This may have been due to the additional bubble formation and growth
reducing the heating elements contact with the liquid (Foust, 1980).
As vapour generally has a lower heat transfer coefficient compared to liquid,
the bubbles caused a resistance to heat transfer, which in turn reduced the
total observable heat transfer coefficient (Coulson
JM, 1964).
The effects of the increased bubble production were also revealed in Figure 3,
as a decrease was observed in the rate at which the heat flux increased at a
temperature difference slightly less than 20 ?C, which was the same point at
which the heat transfer coefficient reached a maximum and began decreasing. The transition to film boiling was marked by the
heating element being engulfed in a vapour bubble. This coincided with the
equipment reaching the cut-out point as well as a considerable decrease in the
heat transfer coefficient as described by Figure 2. This was due to the heating
element having no contact with the liquid, therefore the heat transfer was
slowed by the low heat transfer coefficient of vapour as mentioned before (Coulson JM,
1964).
Whilst data was being collected, a minimal change in the temperature of the
liquid was observed which may have contributed to the low heat transfer
coefficient calculated. The small changes in temperature may have been due to
the poor insulation provided by the glass cylinder causing heat to be transferred
to the surroundings. To improve the experiment an insulated material could be
used for the container with a small viewing glass to allow the regimes of
boiling to be observed to reduce heat dissipated to the surroundings.

The experimental critical heat flux was found
to be

 however the theoretical value was higher at

 which gave a percentage error of 4.8% between
the two values. Due to the small error margin the theoretical critical heat
flux could be used as a prediction however the experiment provides a basis for
the suitability of the approximation. An experiment could be run on a pilot
scale to verify whether the approximation would be safe to use for an
industrial scale plant. Operating at the critical heat flux can be dangerous as
it can cause the heating element to increase above a safe working temperature
which can lead to the equipment being damaged (Collier,
1981).
Therefore when boiling, the aim is to operate at a high heat flux which is
still lower than the critical heat flux to enable an efficient but safe
process. The example for finding the time taken to boil the CFC was an example
of this as the operational heat flux was 80%, which was a compromise between a
high heat flux and a safe value in respect to the critical value. When scaling
up to industrial scale operating at 80% of the critical heat flux would be
advisable and the boiling regime should be nucleate, therefore a temperature
difference of about 30 ?C should be implemented. The selected temperature
difference has the added benefit of a reasonably high heat transfer
coefficient. An attempt to induce the nucleate boiling regime faster could be
made by using a heating element and pre-heater surfaces rough therefore
reducing the time taken to boil the CFC and the energy cost of running the
heater.

 

Conclusion

The test rig allowed the determination of the
experimental critical heat flux, which was found to be

, and the maximum heat
transfer coefficient which was discovered to be

.
Both values can
be used to predict the performance of an industrial scale pre-heater as they
allow an optimal temperature difference to be chosen to insure the pre-heater
is operated at a safe heat flux level whilst insuring a high heat transfer
coefficient. The theoretical critical heat flux was calculated to be

which
was within the same magnitude as the experimental heat flux. However as the
theoretical heat flux was higher than the experimental one, using the
theoretical value as a basis may lead to a heat flux that is too high for the
system being used therefore experimentation was needed to define the safety
limits for the process. The results of the experiment were limited by the
equipment used as the glass cylinder holding the liquid was a poor insulator
leading to heat being dissipated to the surrounding and the temperature reading
on the thermometer was obstructed when the ebullition became vigorous.
Therefore the experiment could be improved by using an insulated material to
reduce heat loss and an electric heat sensor to monitor the temperature. The data collected from
the experiment suggested that operating whilst the boiling regime was nucleate
boiling, and more specifically at a temperature difference around 30 ?C provided an
efficient but safe method of heat transfer.