Upload
paulo-fraga
View
220
Download
0
Embed Size (px)
Citation preview
7/28/2019 11 destilao amonia (1)
1/7
Distillation column configurations in ammoniawater
absorption refrigeration systems
Jose Ferna ndez-Seara*, Jaime Sieres, Manuel Va zquez
Area de Maquinas y Motores Termicos, Escuela Tecnica Superior de Ingenieros Industriales, Universidad de Vigo,
Campus Lagoas-Marcosende, N 9, 36200 Vigo, Spain
Received 29 October 2001; received in revised form 30 May 2002; accepted 7 June 2002
Abstract
In ammoniawater absorption refrigeration systems a purification process to reduce the water content in the vapour
leaving the generator is required. During this process the water content in the vapour must be reduced to a minimum,
otherwise it tends to accumulate in the evaporator and strongly deteriorates the efficiency of the system. The vapour
purification can be carried out by partial condensation, by establishing a liquidvapour counter flow or by combining
both methods. In systems with partial condensation, the distillation column can be composed of one or more rectifiers
using different cooling mediums, and the rectifying and stripping sections. In complete condensation systems only the
rectifying and stripping sections can be used. Therefore different distillation column arrangements should be con-
sidered. This paper presents a study of several distillation column configurations for single stage ammoniawater
absorption refrigeration systems with partial and complete condensation. In order to evaluate and compare the dif-
ferent configurations, a parameter that indicates the ratio of the ammonia vapour concentration increase in each partof the column to the total ammonia purification has been defined. The analysis has been based on the system COP.
Finally the efficiency in each part of the column has been calculated to estimate its design requirements.
# 2002 Elsevier Science Ltd and IIR. All rights reserved.
Keywords: Refrigerating system; Absorption system; Ammoniawater; Distillation column; Design
Configurations des colonnes de distillation des syste` mes
frigorifiques a` absorption a` ammoniac/eau
Mots cles : Syste` me frigorifique ; Syste` me a` absorption ; Ammoniac-eau ; Colonne de distillation ; Conception
1. Introduction
This work is circumscribed to ammoniawater absorp-
tion systems. This technology, applied to refrigeration,
dates its first practical applications around 1860. From
this time on the understanding of the processes involved
in these systems has been increasing, to the point where
today the applied technology is considered well known.
However, some pitfalls must be avoided in designing
and building a prototype or an industrial system in
order to obtain an effective behaviour and performance.
Some of these drawbacks can derive from a deficient
ammonia vapour purification process of the regenerated
vapour from the system generator.
0140-7007/03/$20.00 # 2002 Elsevier Science Ltd and IIR. All rights reserved.
P I I : S 0 1 4 0 - 7 0 0 7 ( 0 2 ) 0 0 0 3 7 - 3
International Journal of Refrigeration 26 (2003) 2834
www.elsevier.com/locate/ijrefrig
* Corresponding author. Tel.: +34-986-812605; fax: +34-
986-812201.
E-mail address: [email protected] (J. Ferna ndez-Seara).
http://www.elsevier.com/locate/ijrefrig/a4.3dmailto:[email protected]:[email protected]://www.elsevier.com/locate/ijrefrig/a4.3d7/28/2019 11 destilao amonia (1)
2/7
This paper deals with the ammonia purification process
in ammoniawater absorption refrigeration systems. In
this type of systems the vapour pressure of water (sor-
bent) is not negligible compared to the vapour pressure
of ammonia (refrigerant). Therefore the vapour leaving
the generator always contains a small part of water. The
water content in the vapour must be reduced, because it
tends to accumulate in the evaporator. The presence ofwater in the evaporator diminishes the evaporation pres-
sure and strongly deteriorates the efficiency of the system.
Ignoring the water content in the regenerated vapour
constitutes one of the major pitfalls in designing an
ammoniawater refrigeration system, as pointed out by
Bogart [1].
On the other hand, the processes used to reduce the
vapour water content always involve the condensation
and return of a vapour fraction back to the generator.
This means that part of the heat input in the generator is
used to generate vapour which does not produce refrig-
eration. Consequently, the system efficiency is reduced.
The effectiveness of the purification process has become
a crucial issue in order to guarantee a reliable and effi-
cient system operation.
The vapour purification process can be carried out by
partial condensation, by establishing a liquidvapour
counter-flow or by combining both methods. In partial
condensation the vapour is cooled, so that a small partwith a high water content condenses. The condensate is
called reflux. The cooling process can be carried out
using different cooling mediums. The heat exchanger
where the partial condensation takes place is usually
named as rectifier or reflux cooler. The counter flow
processes are established in a distillation column. The
vapour stream from the generator enters at the bottom of
the column and rises in counter-flow to a liquid stream,
so that heat and mass transfer processes between both
phases are enabled.
The liquid stream can be produced by condensation of
the vapour at the column top, or can be the strong solu-tion (or a fraction) from the absorber. The liquid stream
at the column top can be obtained by partial condensa-
tion or by returning part of the condensed liquid from the
condenser. The last is usually denominated complete
condensation. The contact zone between the vapour and
the reflux is known as the rectifying section. When the
liquid stream is the strong solution, then the vapour
liquid contact is established below the column feed entry
point. This contact zone is known as the stripping section.
If the distillation column is composed of rectifying and
stripping section, then the liquid stream in the stripping
section is the mixture of the reflux from the rectifying
section and the strong solution from the absorber.
In systems with partial condensation, the distillation
column can be composed of one or more rectifiers and
the rectifying and stripping sections. In complete con-
densation systems only the rectifying and stripping sec-
tions can be used. Therefore the ammonia purification
process can be carried out with different distillation
column configurations.
In this paper a single stage ammoniawater absorp-
tion refrigeration system has been studied. Different
possibilities of distillation column configurations have
been analysed for partial and complete condensation. In
partial condensation the rectifier cooling medium can beeither an external fluid or the strong solution when a
single stage cycle with rectifier heat integration is con-
sidered. The evaluation of the different column con-
figurations has been based on the system COP.
Moreover the efficiency in each part of the column is
calculated in order to estimate its design requirements.
2. System description
The single stage ammoniawater absorption refrig-
eration systems with partial and complete condensation
Nomenclature
COP coefficient of performance
Cp specific heat (J kg1 K1)
Gr temperature gradient (K)
h specific enthalpy (J kg
1
)m:
mass flow (kg s1)
P pressure (Pa)
%Pur purification ratio (%)
Refl internal reflux ratio
Q:
heat flow (W)
T temperature (K)
x ammonia mass concentration (kg kg1)
Greek symbols
" mass transfer efficiency
energy efficiency
SubscriptsA absorber
b bomb
C condenser
E evaporator
G generator
l liquid
R refrigerant
rect rectifying section
Rss strong solution cooled rectifier
Rw water cooled rectifier
ss strong solution
strip stripping section
v vapourw water
ws weak solution
J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834 29
7/28/2019 11 destilao amonia (1)
3/7
examined in this paper are depicted schematically in
Fig. 1. The components of the distillation column in the
partial condensation system are the stripping and the rec-
tifying sections, a heat integrated rectifier with the strong
solution as coolant and a water cooled rectifier at the top
of the column. For the complete condensation system
only the rectifying and stripping sections are considered(dashed line in Fig. 1).
3. Mathematical model
A lumped steady state model has been developed in
order to study the vapour purification process in single
stage ammoniawater absorption refrigeration systems
with partial and complete condensation. The mathema-
tical model is based on the application of global mass,
species and energy balances and heat and mass transfer
equations.In formulating the model the following assumptions
have been made: heat losses to the environment are
negligible, pressure drops are considered only between
the evaporator and absorber, the condensed liquid from
the condenser and the weak solution leaving the gen-
erator are saturated, as well as the liquid and the vapour
currents inside the column. The irreversibilities in the
absorber, solution and liquidvapour heat exchangers
and in the solution pump are defined by means of their
efficiencies. State equations used for the ammoniawater
equilibrium and thermodynamic properties have been
calculated from Ziegler and Trepp [2].
The mathematical model has been used to determine
the thermodynamic state at every representative point of
the cycle, outside of the distillation column, for the
partial and complete condensation systems. As a result,
the conditions of the streams entering and leaving the
column are known. In order to determine the liquid and
vapour state conditions inside the column an appro-
priate parameter has been introduced. It has been
named ammonia purification ratio (%Pur) and is
defined as the quotient of the ammonia vapour enrich-ment achieved in any component of the column to the
ammonia concentration difference between the vapour
leaving the column and the vapour produced in the
generator.
The equations applied to the rectifying and stripping
sections of the column and to the rectifiers are reported
in the following sections. The model equations of the
remainder system components are not presented here
for the sake of simplicity and brevity, but they are
described in detail in Refs. [3,4].
3.1. Rectifying and stripping sections
Applying global mass, species and energy balances
over the rectifying section the corresponding equations
are obtained. The vapour ammonia purification ratio
for the rectifying section (%Purrect) is calculated using
Eq. (1), according to its definition and nomenclature in
Fig. 1.
%Purrect x19 x17
x23 x131
For the stripping section the corresponding equations
for the global mass, species and energy balances and for
the ammonia purification ratio have been derived. At
the feed entry point, it is assumed that the liquid and
vapour fractions of the feed flow mix with the down-
ward liquid from the rectifying section and with the
upward vapour from the stripping section, respectively.
Thus, the liquid and vapour conditions at the top of the
stripping section can be determined.
3.2. Rectifiers
Global mass, species and energy balances over the
water cooled rectifier are also obtained. For the watercooled rectifier, the purification ratio is defined in Eq.
(2).
%PurRw x23 x21
x23 x132
In the literature concerning mass transfer operations,
the liquid and vapour phases leaving a rectifier are
usually considered to be in thermodynamic equilibrium
and therefore the rectifier is assumed to be equivalent to
a theoretical stage [5]. However, in this paper this
assumption is not made and an efficiency term () thatFig. 1. Schematic diagram of the distillation column with par-
tial or complete condensation.
30 J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834
7/28/2019 11 destilao amonia (1)
4/7
expresses the vapour concentration deviation from
thermodynamic equilibrium conditions is taken into
account. For the water cooled rectifier, the efficiency
used is given in Eq. (3).
"Rw x T23 T22 x21
x23 x21
3
In mass transfer operations, deviations from conditions
of a theoretical stage are generally expressed by the Mur-
phree efficiency [6], which is defined as the inverse of Eq.
(3). However, in a rectifier the liquid temperature at the
outlet is frequently greater than the vapour temperature
leaving the rectifier and, consequently, the Murphree
vapour efficiency exceeds unity. Therefore, the efficiency
expression in Eq. (3) has been chosen. For the strong
solution cooled rectifier corresponding equations to the
water cooled rectifier have been derived.
4. Results and discussion
The mathematical model described in the previous sec-
tion has been implemented in a computer program using
Fortran 90. The implemented model has been used to
simulate and analyse the performance of absorption
refrigeration systems considering partial and complete
condensation processes to purify the generated vapour.
The analysis has been performed varying the distillation
parameters one at a time while keeping the others para-
meters that define the operation of the absorption
refrigeration system at a constant value. The operating
conditions are indicated in Table 1, where the rectifier
efficiencies are given only for the partial condensation
system.
4.1. Partial condensation systems
The results presented here for the partial condensa-
tion systems examine the effect of diverse configurations
of the distillation column on the system COP, as well as
the required components efficiencies. This has been done
by varying the vapour purification ratio in the rectifiers
and in the stripping and rectifying sections, considering
also the cases of systems where some of these components
are not used.
4.1.1. Effect of vapour purification ratios distribution on
the system COP
Fig. 2 shows the system COP as a function of the
ammonia vapour purification ratio carried out in therectifiers. The rectifiers purification ratio was varied
from 0 to 90% of the total ammonia purification pro-
cess in the column (from x13 to x23) while a constant
value of 10% was established for the rectifying section
(%Purrect=10%). Thus, the special cases when one of
the rectifiers is not used (%PurRw=0% or%PurRss=
0%), none are used (%PurRw=0% and%PurRss= 0%)
or the stripping section is not used (%PurRw+%
PurRss=90%) are included in this figure.
Fig. 2 reveals that solutions can not be obtained for
the whole range considered. This means that for the
operating conditions in Table 1, the specified ammoniaconcentration can not always be attained. For a fixed
value of the ammonia purification ratio in the water
cooled rectifier %PurRw (different curves in Fig. 2) the
existing solutions are limited for small %PurRss values
since, in this case, the liquid reflux generated from the
condensation of the vapour stream in the rectifiers is not
enough to achieve the desired ammonia concentration
specified in the rectifying section (x19x17). However, if
%PurRw increases, solutions can be found for smaller
%PurRss values since the required reflux is generated in
the water cooled rectifier. On the other hand for a given
%PurRw value, the available solutions are also limited
to the %PurRw value for which the sum of the vapour
purification ratios in all components is 100%.
Fig. 2 shows that increasing %PurRss for a constant
%PurRw value decreases the system COP. The same
result is observed when increasing %PurRw for a con-
stant %PurRss value. However, in this case COP values
Table 1
Operating conditions
TE (K) 258 P (Pa) 10 000
TG (K) 398 "A 0.7
Tw (K) 293 "l-v 0.7
TC (K) 8 "ss-ws 0.8
TA (K) 10 "b 0.5
QE (W) 5000 "Rss 0.7
xR (kg kg1) 0.999 "Rw 0.7
Tevap (K) 10
Fig. 2. Partial condensation system COP as a function of the
purification ratio of the strong solution cooled rectifier for dif-
ferent values of the purification ratio of the water cooled recti-
fier and with a value of 10% for the purification ratio of the
rectifying section.
J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834 31
7/28/2019 11 destilao amonia (1)
5/7
are lower since the rectifier heat is rejected to a heat sink
outside the system (cooling water) instead of being used
within the absorption system [7].
If the rectifying section is not considered (%Purrect=
0%), the range of available solutions increases, since
restrictions are only imposed by the stripping section
efficiency which must be between 0 and 1. However, theCOP obtained is the same than in Fig. 2, since the rec-
tifying section is replaced with a larger stripping section
and both of them operate adiabatically. From the pre-
sented results, an appropriate configuration can be chosen
for a given application, based on the system COP.
4.1.2. Rectifying and stripping sections efficiencies
The achievement of a high purification ratio in a dis-
tillation column component usually means high efficiency
requirements. This implies that the vapour current con-
tacts a large surface area, resulting in a component of
higher dimensions. Thus, in order to complete the pre-vious analysis the efficiencies of the stripping and recti-
fying sections have also been studied when varying the
purification ratios of the column components.
The stripping and rectifying section efficiencies are
defined as the quotient of the ammonia vapour con-
centration enrichment value to the theoretical maximum
value for the given vapour and liquid inlet conditions. Eqs.
(4) and (5) represent the efficiencies of the rectifying and
stripping sections, according to nomenclature in Fig. 1.
"rect x19 x17
x T19T20 x174
"strip x15 x13
x T15T16 x135
Fig. 3 shows the efficiency of the rectifying section nee-
ded to achieve a purification ratio of 10% (%Purrect=
10%) versus the strong solution rectifier purification ratio
(%PurRss) and for different %PurRw values. The recti-
fying section efficiency is rather low, which suggests that
for the given conditions a higher vapour ammonia con-
centration could be obtained. However, the vapour
enrichment is limited for decreasing values of %PurRss
and %PurRw because the reflux solution is low and,consequently, the efficiency of the rectifying section
based on the liquid phase concentration which is defined
in Eq. (6) is close to unity.
"rect;l x20 x18
x20 x T18T17 6
Fig. 4 shows the necessary stripping section efficiency
for different values of %PurRss and %PurRw for the
most adverse case corresponding to not using the recti-
fying section (%Purrect=0%). Some of the curves in
Fig. 4 extend from zero to unity efficiency values, limit-ing %PurRss possible values. Since the liquid flow is
much higher than the vapour flow, it is usually possible to
achieve high purification ratios with a stripping section of
moderate dimensions.
4.1.3. Rectifier efficiencies
Fig. 5 shows the strong solution cooled rectifier effi-
ciency when %Purrect= 0%. The efficiencies are defined in
Eqs. (7) and (8) and considering the relation between the
heat capacity ratio of both fluids and the nomenclature
shown in Fig. 1.
Rss Q:
Q:
max
m:
7 h8 h7
min m:
19 h19 m:
21 h21 m:
20 h20 T21T7 ;
m:
7 hT8 T19 h7
7
Fig. 3. Rectifying section efficiency as a function of the purifica-
tion ratio of the strong solution cooled rectifier for different values
of the purification ratio of the water cooled rectifier and with a
value of 10% for the purification ratio of the rectifying section.
Fig. 4. Stripping section efficiency as a function of the pur-
ification ratio of the strong solution cooled rectifier for different
values of the purification ratio of the water cooled rectifier and
with the purification ratio of the rectifying section being zero.
32 J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834
7/28/2019 11 destilao amonia (1)
6/7
Rw m:
w Cpw Tw;out Tw;in
min m:
21 h21 m:
23 h23 m:
22 h22 T23 Tw;in ;h
m:
w Cpw TTw;outT21 Tw;in i
8
The efficiency of the water cooled rectifier increaseswhen increasing %PurRw but remains constant when
varying the purification ratio of the strong solution rec-
tifier (%PurRss), because the vapour inlet conditions are
set by %PurRw. However, Fig. 5 reveals that the strong
solution rectifier efficiency increases when increasing
%PurRss and when decreasing %PurRw, since in this
case, the condensate solution leaving the water cooled
rectifier enters the strong solution cooled rectifier and its
thermodynamic conditions vary with %PurRw.
As stated before, the system COP decreases when
increasing the vapour enrichment carried out in the
water cooled rectifier. Therefore, avoiding the use of thisrectifier seems a practical option attending to the system
COP results. However, according to Fig. 5 for %PurRw=
0% values, the required solution rectifier efficiency can be
as high as 0.95, hence for specific applications it would be
interesting to use the water cooled rectifier in order to
avoid the needs for a high efficiency in the solution rectifier.
4.2. Complete condensation systems
The parameters being considered in the analysis of the
purification process by means of complete condensation
are the internal reflux ratio and the purification ratios in
the rectifying and stripping sections. However, since the
column operates adiabatically only the reflux ratio
influences the system COP. The internal reflux ratio is
defined in Eq. (9), according to nomenclature in Fig. 1.
Reflm:
24
m:
23
m:
24
m:
1 m:
24
9
4.2.1. Effect of the internal reflux ratio on the system COP
Fig. 6 shows the system COP versus the reflux ratio,
for different purification ratios in the rectifying andstripping sections. When increasing the reflux ratio, the
vapour leaving the distillation column and the liquid
returned from the condenser increase. Thus, more liquid
returns to the generator and more vapour must be pro-
duced, which implies that the generation heat load must
increase and consequently the system COP decreases.
The COP values obtained are lower than for the partial
condensation system. This is due to the extra heat rejected
for cooling the column liquid reflux in the condenser,
where no vapour purification is performed.
4.2.2. Rectifying and stripping sections efficienciesAs mentioned, the purification ratios do not affect the
system COP, but rather influence the components effi-
ciency and the range of possible solutions. Results
obtained from the model developed have shown that the
ammonia refrigerant concentration specified in Table 1
can be achieved with a specific COP and a given reflux
ratio with different column configurations.
Fig. 7 shows the rectifying efficiency versus the reflux
ratio for different purification ratios of the rectifying
section (%Purrect). For large %Purrect values available
solutions are limited by small reflux ratio values. This
accounts for the small liquid flow returned to the column,
which is not enough to enrich the vapour current to the
desired ammonia concentration. Under these conditions
the liquid may come close to equilibrium with the vapour
and consequently the rectifying efficiency in terms of
liquid composition approaches unity.
Fig. 7 reveals that fewer solutions were found for
small purification ratio values, since according to Fig. 8,
the stripping efficiency approaches unity, meaning that the
vapour current is close to equilibrium with the liquid
Fig. 5. Strong solution cooled rectifier efficiency as a function
of the purification ratio of the strong solution cooled rectifier
for different values of the purification ratio of the water cooled
rectifier and with the purification ratio of the rectifying section
being zero.
Fig. 6. Complete condensation system COP as a function of
the internal reflux ratio.
J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834 33
7/28/2019 11 destilao amonia (1)
7/7
reflux. For large values of the reflux ratio, the liquid mass
flow returned from the condenser increases significantly,
thus, reflux conditions dominates in the stripping sec-
tion instead of the feed point conditions, which explains
the changes in the slope of the curves of Fig. 8.
Figs. 7 and 8 also show that when decreasing the
reflux ratio, fewer purification ratio combinations are
possible. The minimum reflux ratio is found when onlyone possible solution is obtained and the stripping effi-
ciency and the rectifying efficiency in terms of liquid
composition are unity [8] (extreme point of dashed curve
in Figs. 7 and 8).
5. Conclusions
This paper presents a study on the different distilla-
tion column configurations to carry out the ammonia
vapour purification process considering partial and
complete condensation for a single stage ammonia
water absorption refrigeration system. The analysis is
based on a mathematical model applied to different dis-
tillation column configurations. The model includes a
new parameter named as ammonia purification ratio. The
different column configurations are evaluated and com-pared based on the ammonia purification ratio carried out
in each column component and on the system perfor-
mance obtained. Moreover the mass and heat transfer
efficiencies in each part of the column are calculated in
order to evaluate their design requirements.
In the partial condensation system the liquid reflux is
generated in two rectifiers located at the top of the col-
umn over the rectifying section. Results have shown that
using the water cooled rectifier decreases significantly
the system COP. Therefore this rectifier should be avoi-
ded or reduced as much as possible. Moreover, the rec-
tifying section was shown to have low efficiency in termsof vapour concentration due to the small liquid reflux
from the rectifiers. Hence, the use of the rectifying sec-
tion could be avoided in some applications, given that
an appreciable benefit is not expected, which would also
simplify the column design. The complete condensation
system leads to simpler column designs, since the recti-
fiers are not used. However, the COP values predicted
are lower than for the partial condensation system.
References
[1] Bogart MJP. Pitfalls in ammonia absorption refrigeration.
Int J Refrigeration 1982;4:2038.
[2] Ziegler B, Trepp Ch. Equation of state of ammonia-water
mixtures. Int J Refrigeration 1984;7:1016.
[3] Ferna ndez-Seara J. Disen o, simulacio n, construccio n y
evaluacio n de un prototipo de sistema de refrigeracio n por
absorcio n con NH3-H2O para la produccio n de fro en
barcos de pesca, accionado con la energa te rmica residual
recuperada de los gases de escape. PhD thesis, A rea de
Ma quinas y Motores Te rmicos, E.T.S.I. Industriales y
Minas, Vigo University, Spain, 1999.
[4] Ferna ndez-Seara J, Sieres Atienza J, Va zquez Va zquez M.
Modelo de ca lculo para el estudio y ana lisis del compor-
tamiento de un prototipo de sistema de refrigeracio n porabsorcio n con NH3-H2O ya construido. Recliem2000, La
Habana, Cuba, July 2000.
[5] Henley EJ, Seader JD. Equilibrium-stage separation
operations in chemical engineering. New York: John
Wiley & Sons; 1981.
[6] Sherwood TK, Pigford RL, Wilke CR. Mass transfer.
McGraw-Hill; 1975.
[7] Herold KE, Radermacher R, Klein SA. Absorption chil-
lers and heat pumps. CRC Press; 1996.
[8] Bogart MJP. Ammonia absorption refrigeration in indus-
trial processes. Gulf Publishing Company; 1981.
Fig. 8. Stripping section efficiency as a function of the internal
reflux ratio for different values of the purification ratio of the
rectifying section.
Fig. 7. Rectifying section efficiency as a function of the internal
reflux ratio for different values of the purification ratio of the
rectifying section.
34 J. Fernandez-Seara et al. / International Journal of Refrigeration 26 (2003) 2834