Molecular sieve abstract
The bed tester includes nitrogen and helium sources, a scale, temperature
and pressure readouts, and a computer program for determining the
available adsorption capacity or activity of molecular sieve beds.
Activity is a measure of the condition of the molecular sieve and
is defined as the ratio of the weight of nitrogen adsorbed in the
bed under test to the weight of nitrogen adsorbed by an equivalent
weight of activated molecular seive. To determine the weight of
nitrogen adsorbed for the bed under test, a chamber or plenum of
known volume is filled with helium, and expanded into the bed. Then
the helium is evacuated, and the bed is pressurized with nitrogen.
Measured values of pressure, temperature and weight, and known values
of volume and density, are used with the ideal gas law (PV=MRT)
and other equations in calculations to arrive at the true weight
gain due to nitrogen adsorption. Determining the weight of nitrogen
adsorbed by an equivalent weight of activated molecular sieve is
accomplished by using a set of pure crystal isotherm parameters
for 5A and 13X zeolites, which are stored in a table. These parameters
were determined by collecting pure crystal-N 2 isotherm data for
the zeolite crystals over the temperature range of 14-44 C. The
data were then fit to a Sips equation by a least squres technique.
The weight percent of water of the molecular sieve is calculated
based on correlations with activity determined by a least square
technique.
Molecular sieve claims
What is claimed is:
1. A bed tester for testing a bed of molecular sieve oxygen concentrator
for determining activity, said bed having molecular sieve pellets
with beta cages for adsorption of nitrogen, comprising:
bed test apparatus which comprises a helium system with a chamber
of known volume, weighing means for measuring the weight gain of
the bed due to nitrogen adsorption, pressure measuring means for
measuring the pressure in the helium system and in the bed, temperature
measuring means for measuring the temperature in the helium system
and in the bed;
valve means having positions for selectively connecting the helium
system to a vacuum source, for connecting the helium system to a
source of helium under pressure, for connecting the helium system
to the bed, for connecting the bed to the vacuum source, and for
connecting the bed to a source of nitrogen under pressure;
so that the helium system may be filled with helium and the pressure
and temperature measured, the helium system may be connected to
the bed for the helium to expand and the pressure and temperature
measured, and the bed may be evacuated and weighed and then pressurized
with nitrogen, and the pressure and temperature measured, and the
bed weighed.
2. A bed tester according to claim 1 which further includes a
computer program which includes:
means for entering the measured initial values of pressure and
temperature in the helium system filled with helium, final values
of pressure and temperature of the helium after expansion into the
bed, values of nitrogen pressure and temperature after the bed is
pressurized with nitrogen, and a value of weight gain after the
bed is pressurized with nitrogen;
means for calculating the weight Q.sub.X of nitrogen adsorbed for
the bed under test, using said values of pressure, temperature and
weight gain, and values of volume and density from a table, using
the ideal gas law (PV=MRT) and other equations in calculations to
arrive at the true weight gain due to nitrogen adsorption;
means for determining the weight Q.sub.M of nitrogen adsorbed by
an equivalent weight of activated molecule sieve using a set of
pure crystal isotherm parameters, which are stored in a table and
were determined by collecting pure crystal-N2isotherm data for the
zeolite crystals over a given temperature range and fit to a Sips
equation by a least squares technique;
means for calculating the activity as the ratio of the weight of
nitrogen adsorbed in the bed under test to the weight of nitrogen
adsorbed by an equivalent weight of activated molecular sieve (Q.sub.X
/Q.sub.M).
3. A bed tester according to claim 2 wherein said computer program
further includes means for calculating the weight percent of water
of the molecular sieve based on correlations with activity determined
by a least squares technique.
4. A bed tester according to claim 2 wherein said means comprising
said computer program include equations as follows:
the nomeclature in the above equations being as follows with the
units in the last column:
5. A bed tester according to claim 4 wherein said computer program
further includes a list of molecular sieve types which includes
type MG 3 and type 5AMG, means for selecting one of these types,
the values from the table for void volume of .beta.-cages (V.sub.x),
bulk density of pellets (.rho.), and the parameters (a, b and c)
from a Sips equation being functions of the molecular sieve type;
means for calculating the weight percent of water (Y) of the molecular
sieve based on correlations with said activity (X) by executing
the equation
if the selected type of molecular sieve is type 5AMG, and alternatively
executing the equation
if the selected type of molecular sieve is type MG3.
6. A bed tester according to claim 1 wherein
said pressure measuring means comprises a pressure sensor electrically
coupled to a pressure readout unit;
said temperature measuring means comprises a temperature sensor
electrically coupled to a temperature readout unit;
said valve means comprises
a first valve having a "He SUP." position connecting
the pressure sensor to the source of helium, a "N2SUP."
position connecting the pressure sensor to the source of nitrogen,
a "He SYS." position connecting the pressure sensor to
the helium system and a "BED" position connecting the
pressure sensor to the bed;
a second valve having a "N2" position connecting the
bed to the source of nitrogen, a "VAC." position connecting
the bed to the vacuum source, a "He" position connecting
the bed to the helium system, and a "VENT" position connecting
the bed to a vent;
a third valve having a "Vac." position connecting the
helium system to the vacuum source, a "He" position connecting
the helium system to the source of helium, and an "OFF"
position closing off the helium system from the vacuum source and
the source of helium;
and fourth, fifth and sixth valves which are ON-OFF valves for
the source of helium, the source of nitrogen, and the vacuum source
respectively;
and said weighing means is a scale on which the bed is placed.
7. A bed tester according to claim 6 which further includes a
computer, a keyboard coupled to the computer for entries by a user,
and a program for loading in the computer, said program having references
to data in table and instructions for interacting with the user
and for performing calculations;
instruction means having a first list of molecular sieve types
and a second list of concentrator types, directions to the user
to select a type from each list, and means for entering the type
of each from responses at the keyboard;
wherein the data in tables includes values for the universal gas
constant R, a volume V.sub.c of connecting tubing to the bed, a
volume V.sub.hi for the helium system, bulk densities .rho. for
pellets and volumes V.sub..chi. for beta cages for the types of
molecular sieve on the first list, bed volumes V.sub.b for the types
of concentrators on the second list;
wherein the data in the tables further includes values of pure
crystal isotherm parameters, which were determined by collecting
pure crystal nitrogen isotherm data over a given temperature range
for the types of molecular sieve on the first list, then fit by
a least squares technique to a Sips equation ##EQU3## where, q=amount
adsorbed
p=pressure
a, b, and c=parameters determined by least squares analysis and
stored in the table;
instruction means for initializing the bed tester apparatus, directing
the user to operate the first valve to the "BED" position,
the second valve to the "VENT" position, the third valve
to the "OFF" position, the fourth, fifth and sixth valves
to OFF positions, to place the bed on the scale and to connect it
to the bed tester, and to not install the temperature sensor on
the bed;
instruction means directing the user to operate the fourth, fifth
and sixth valves to ON positions;
instruction means directing the user to operate the first valve
to the "He SUP." position and the "N2 SUP."
in turn, and to verify that the pressure readout at each position
is within stated limits;
instruction means directing the user to operate the first valve
to the "He SYS." position and the third valve to the "VAC"
position and to wait until the pressure readout indicates a stated
value (vacuum pressure);
instruction means directing the user to operate the third valve
to the "He" position (which pressurizes the hellium system
from the source of helium);
instruction means directing the user to operate the third valve
to the "OFF" position.
instruction means directing the user to enter the pressure readout
and the temperature readout values and to enter them in turn, and
means for entering these values from responses at the keyboard as
an initial helium system pressure P.sub.hia in psia and an initial
helium system temperture T.sub.hia in degrees C.;
instruction means directing the user to operate the first valve
to the "BED" position and the third valve to the "VAC"
position and to wait until the pressure readout indicates a stated
value (vacuum pressure);
instruction means directing the user to install the temperature
sensor at the bed midpoint;
instruction means directing the user to operate the second valve
to the "He" position (which permits the helium to expand
into the bed);
instruction means directing the user to enter the pressure readout
and the temperature readout values and to enter them in turn, and
means for entering these values from responses at the keyboard as
a final helium system pressure P.sub.hfa in psia and a final helium
system temperature T.sub.hfa in degrees C.;
instruction means directing the user to operate the second valve
to the "VAC" position and to wait until the pressure readout
indicates a stated value (vacuum pressure);
instruction means directing the user to place the bed on the scale
and to press a tare switch;
instruction means directing the user to operate the second valve
to the "N2" position (which pressurizes the bed with nitrogen
from the source of nitrogen);
instruction means directing the user to enter the pressure readout
and the temperature readout values and to enter them in turn, and
means for entering these values from responses at the keyboard as
a final bed pressure P.sub.na in psia and a final bed temperature
T in degrees C;
instruction means directing the user to enter the weight indicated
on the scale, and means for entering this value from a response
at the keyboard as a weight gain .DELTA.W.sub.a in grams after nitrogen
adsorption;
instruction means for shutting down the bed tester apparatus, directing
the user to operate second valve to the "VENT" position,
the fourth, fifth and sixth valves to OFF positions, to remove the
bed from the scale, to remove and store the temperature sensor,
and to press the tare switch;
instruction means for calculating the weight W.sub.x of activated
molecular sieve as the product (W.sub.x =V.sub.b..rho.) of the bed
volume V.sub.b for the selected type of concentrator on the second
list and the bulk density .rho. for pellets of the selected type
of molecular sieve on the first list;
instruction means for executing the equations
and
to convert pressure values to atmospheres, temperature values to
degrees Kelvin, and volume values to liters;
instruction means for executing the equation
using the ideal gas law to calculate the moles M.sub.n of helium
in the gas phase;
instruction means for executing the equation
using the ideal gas law to calculate the total system volume V.sub.t
;
instruction means for executing the equation
to calculate the void volume V.sub.vh of the bed including the
beta cages;
instruction means for executing the equation
to calculate the volume V.sub..beta. of the beta cages;
instruction means for executing the equation
to calculate the void volume V.sub.v of the bed excluding the beta
cages;
instruction means for reading the pure crystal isotherm parameters
(a, b and c) from the table, interpolating if the final bed temperature
with T.sub.n is between values in the table, and using these parameters
in the Sips equation
to calculate the amount Q.sub.n of nitrogen adsorbed
instruction means for executing the equations
and
using the ideal gas law to calculate the moles M.sub.n of nitrogen
in the gas phase and the corresponding weight W.sub.n in grams;
instruction means for executing the equations
to calculate the true weight gain due to nitrogen adsorption first
in grams (.DELTA.W.sub.b), converting to moles (.DELTA.W), then
to ml at standard temperature and pressure;
instruction means for executing the equation
to calculate the amount of nitrogen in ml per gram at standard
temperature and pressure, corrected to account for a twenty weight
percent binder content of the molecular sieve pellets; and
instruction means for executing the equation
to calculate the activity (X) in the bed in percent.
8. A bed tester according to claim 7 wherein said first list of
molecular sieve types includes 16.times.40 mesh type MG3 and type
5AMG;
wherein said program further includes instruction means for calculating
the weight percent water (Y) of the molecular sieve by executing
the equation
if the selected type of molecular sieve is type 5AMG, and alternatively
executing the equation
if the selected type of molecular sieve is type MG3.
9. The method of testing a bed of a molecular sieve oxygen concentrator
for determining activity, said bed having molecular sieve pellets
for adsorption of nitrogen, said method comprising the steps:
filling a chamber having a known volume V.sub.hi with helium, measuring
the initial helium pressure P.sub.hi and temperature T.sub.hi, and
employing the ideal gas law to calculate the mass of the helium
M.sub.h, using a value for the universal gas constant R from a table,
as
connecting the chamber to the evacuated bed and allowing the helium
to expand into the bed, measuring the final helium pressure P.sub.hf
and temperature T.sub.hf in the bed, and employing the ideal gas
law to calculate the total void volume V.sub.t of the bed and chamber,
based on the mass of the helium M.sub.h, as
determining a void volume V.sub.vh of the bed by subtracting the
chamber volume V.sub.hi and a tester connection tubing volume V.sub.c
from the total void volume V.sub.t ;
calculating the cumulative volume V.sub..beta. of the molecular
sieve beta cages based on a parameter V.sub.x from a table for the
void volume of the beta cages for the crystal type, and calculating
a corrected void volume V.sub.v for adsorption by subtracting the
volume V.sub..beta. of the beta cages from the void volume V.sub.vh
of the bed;
removing the helium from the bed;
weighing the bed to determine a tare value;
pressurizing the bed with nitrogen;
measuring the final pressure P.sub.na and temperature T for the
bed filled with nitrogen;
weighing the bed for the weight gain .DELTA.W.sub.a of the bed
after pressurization with nitrogen;
calculating the amount of nitrogen Q.sub.n adsorbed per unit weight
of pure crystal based on a Sips equation ##EQU4## where, q=amount
adsorbed, p=pressure, and a, b, and c are pure crystal isotherm
parameters from a table, which were determined by collecting pure
crystal nitrogen isotherm data for the crystals over a given temperature
range, then fit by a least squares technique to said Sips equation;
calculating the weight W.sub.n of nitrogen gas filling the bed
void volume based on the final pressure P.sub.nb and temperature
T.sub.n for the bed filled with nitrogen, the corrected void volume
V.sub.v, and the molecular weight of nitrogen, using the ideal gas
law in the form
calculating the true weight gain .DELTA.W.sub.b of the bed due
to nitrogen adsorption during pressurization by subtracting the
weight W.sub.n of nitrogen gas filling the bed void volume from
the weight gain .DELTA.W.sub.a of the bed after pressurization with
nitrogen, and then dividing the result by the molecular weight of
nitrogen to obtain the weight gain .DELTA.W in moles;
calculating the total amount of nitrogen Q.sub.X adsorbed by the
bed, based on said weight gain .DELTA.W in moles, using the ideal
gas law in the form
calculating the total amount of nitrogen Q.sub.M adsorbed for an
identical weight of activated molecular sieve, as the product of
the weight W.sub.x of molecular sieve pellets within the bed and
the amount of nitrogen Q.sub.n adsorbed per unit weight of pure
crystal, corrected to account for a twenty weight percent binder
content of the molecular sieve pellets,
calculating an activity parameter X of the molecular sieve in the
bed under test by dividing the total amount of nitrogen Q.sub.M
adsorbed for an identical weight of activated molecular sieve into
the total amount of nitrogen Q.sub.X adsorbed by the bed (X=(Q.sub.X
/Q.sub.M).100).
10. The method according to claim 9 which further includes the
step of calculating the weight percent of water of the molecular
sieve based on correlations with activity determined by a least
squares technique.
11. The method of testing a bed of a molecular sieve oxygen concentrator
for determining activity, using a bed tester having a plurality
of valves and tubing connected to a helium source, a nitrogen source,
a vacuum source, a helium chamber, a pressure sensor, and a temperature
sensor, said method comprising the steps:
providing tables containing values for the universal gas constant
R, a volume V.sub.c of connecting tubing to the bed, a volume V.sub.hi
for the helium chamber, bulk densities .rho. for pellets of a number
of types of molecular sieve crystal, volumes V.sub..chi. for beta
cages for the types of molecular sieve crystal, bed volumes V.sub.b
for a number of concentrator types;
providing a table of pure crystal isotherm parameters, which were
determined by collecting pure crystal nitrogen isotherm data for
the types of crystals over the a given temperature range, then fit
by a least squares technique to a Sips equation ##EQU5## where,
q=amount adsorbed
p=pressure
a, b, and c=parameters determined by least aquares analysis and
stored in the table;
selecting a type of molecular sieve from a list;
choosing a type of concentrator from a list;
assigning a value for a bed volume parameter V.sub.b selected from
a table based on the type of concentrator chosen, V.sub.b being
the volume occupied by molecular sieve pellets within the bed;
initializing the bed tester;
connecting the bed tester to the bed;
checking the pressure of the helium source;
checking the pressure of the nitrogen source;
evacuating the helium chamber to a pressure of at most 0.1 psia;
filling the helium chamber from the helium source to a pressure
between 55-60 psia and then closing the connection between the helium
chamber and the helium source;
measuring and entering the initial helium pressure P.sub.hi ;
measuring and entering the initial helium temperature T.sub.hi
;
evacuating the bed under test is to a pressure of at most 1.0 psia;
connecting the helium chamber to the bed and allowing the helium
to expand into the bed;
measuring and entering the final helium pressure P.sub.hf ;
measuring and entering the final helium temperature T.sub.hf in
the bed;
removing the helium from the bed;
weighing the bed to determine a tare value;
connecting the bed to the nitrogen source so that the previously
evacuated bed is pressurized with nitrogen to 55-60 psia;
measuring and entering the final pressure P.sub.n for the bed filled
with nitrogen;
measuring and entering the final temperature T.sub.n for the bed
filled with nitrogen;
weighing the bed and entering the weight gain .DELTA.W.sub.a of
the bed after pressurization with nitrogen;
shutting down the bed tester;
calculating the weight W.sub.x of molecular sieve pellets within
the bed based on the bulk densities from a table for the molecular
sieve type and the bed volume parameter V.sub.b ;
calculating the total void volume V.sub.t of the bed and helium
chamber, based on the initial and final helium pressures P.sub.hi
and P.sub.hf, and the initial and final helium temperatures T.sub.hi
and T.sub.hf, employing the ideal gas law to first calculate the
mass of the helium M.sub.h and then the total void volume V.sub.t,
using the equations
determining a void volume V.sub.vh of the bed is by subtracting
values from the table for the helium chamber volume V.sub.hi and
a tester connection hose volume V.sub.c from the total void volume
V.sub.t ;
calculating the cumulative volume V.sub..beta. of the molecular
sieve beta cages based on the parameter V.sub..chi. from the tables
for the void volume of the beta cages for the selected crystal type,
and calculating a corrected void volume V.sub.v for adsorption by
subtracting the volume V.sub..beta. of the beta cages from the void
volume V.sub.vh of the bed;
calculating the amount of nitrogen Q.sub.n adsorbed per unit weight
of pure crystal for the selected molecular sieve type based on said
Sips equation using said parameters a, b and c from said table of
pure crystal isotherm parameters for said final temperature T.sub.n
for the bed filled with nitrogen;
calculating the weight W.sub.n of nitrogen gas filling the bed
void volume based on the final pressure P.sub.nb l and final temperature
T.sub.n for the bed filled with nitrogen, the corrected void volume
V.sub.v, and the molecular weight of nitrogen, using the ideal gas
law in the form
calculating the true weight gain .DELTA.W.sub.b of the bed due
to nitrogen adsorption during pressurization by subtracting the
weight W.sub.n of nitrogen gas filling the bed void volume from
the weight gain .DELTA.W.sub.a of the bed after pressurization with
nitrogen, and then dividing the result by the molecular weight of
nitrogen to obtain the weight gain .DELTA.W in moles;
calculating the total amount of nitrogen Q.sub.X adsorbed by the
bed in ml at standard temperature and pressure, based on said weight
gain .DELTA.W in moles, using the ideal gas law in the form
calculating the total amount of nitrogen Q.sub.M adsorbed for an
identical weight of activated molecular sieve, as the product of
the weight W.sub.x of molecular sieve pellets within the bed and
the amount of nitrogen Q.sub.n adsorbed per unit weight of pure
crystal, corrected to account for a twenty weight percent binder
content of the molecular sieve pellets,
calculating an activity parameter X of the molecular sieve in the
bed under test by dividing the total amount of nitrogen Q.sub.M
adsorbed for an identical weight of pure crystal into the total
amount of nitrogen Q.sub.X adsorbed by the bed (X=(Q.sub.X /Q.sub.M).100).
12. The method according to claim 11 further including the step
of calculating the weight percent water Y of the molecular sieve
based on the activity parameter X, and correlations with activity
which were determined by a least squares technique for molecular
sieve types 5AMG and MG3 respectively as follows:
Molecular sieve description
BACKGROUND OF THE INVENTION
The present invention relates generally to a bed tester for molecular
sieve oxygen concentrators.
Molecular sieve oxygen concentrators have become increasingly popular
for the production of high purity oxygen (up to 95%) because of
their simplicity, reduced energy consumption, and low operating
costs. Portable units are now widely used to produce medical oxygen
for patients requiring oxygen therapy. Molecular sieve oxygen concentrators
are also in use aboard military aircraft for the production of an
oxygen enriched breathing gas to prevent hypoxia. In addition, future
military aircraft will have oxygen breathing systems employing molecular
sieve oxygen concentrators. Oxygen concentrators may have from two
to six beds filled with molecular sieve. For further background
relating to molecular sieve oxygen concentrators and zeolites, see
a copending patent application Ser. No. 07/151383 filed Feb. 2
1988 now Pat. No. 4813979 issued Mar. 21 1989 to G. W. Miller
and C. F. Theis, "Secondary Oxygen Purifier for Molecular Sieve
Oxygen Concentrator", and the following papers: D. M. Ruthven,
Sec. 1.4 on "Zeolites" in Principles of Adsorption and
Adsorption Process, pages 9-16 John Wiley and Sons, New York, N.Y.
(1984); G. W. Miller, Dr. K. G. Ikels, and P. A. Lozano, "Chemical
Contamination Studies on a Molecular Sieve Oxygen Concentrator (MSOC):
Comparison of MG3 and 5AMG Molecular Sieves", Safe Journal,
Vol. 16 No. 4 (1986); D. E. W. Vaughan, "The Synthesis and
Manufacture of Zeolites", Chemical Engineering Progress, February
1988 pages 25-31; D. M. Ruthven, "Zeolites as Selective Adsorbents,
Chemical Engineering Progress, February 1988 pages 42-50; G. W.
Miller, K. S. Knaebel, and K. G. Ikels, "Equilibria of Nitrogen,
Oxygen, Argon, and Air in Molecular Sieve 5A", AIChE Journal,
February 1987 Vol. 33 No. 2 pages 194-201; G. W. Miller, "Adsorption
of Nitrogen, Oxygen, Argon, and Ternary Mixtures of These Gases
in 13X Molecular Sieve", American Institute of Chemical Engineers
Symposium Series, Vol. 83 No. 259 (1987) pages 28-39.
At present, most molecular sieve oxygen concentrators use 16.times.40
mesh type 5AMG or MG3 molecular sieves, having zeolite 5A and 13X
crystals, respectively. The crystal structure has voids in the form
of .alpha. cages and .beta. cages, as described in the above papers.
Both nitrogen and oxygen are adsorbed in the large .alpha. cages
of these zeolites, however, these crystals have a greater affinity
for nitrogen due to its slight molecular polarity. Nitrogen and
oxygen do not enter the smaller .beta. cages. Due to its small molecular
size and nonpolarity, helium adsorbs in negligible quantities, and
hence, enters the entire void volume of the zeolite crystals (.alpha.
and .beta. cages).
The concentrator's performance or oxygen enriching ability is directly
related to the activity of the molecular sieve. Further, the activity
of a molecular sieve bed can be degraded by exposure to certain
chemical species (principally water) resulting in a reduction in
system performance. There is a need for a means for testing the
molecular sieve beds to ensure they meet accepted standards of activity.
In the prior art, there are two methods used to determine the activity
of a molecular sieve bed. The first method involves reactivating
several samples of molecular sieve which have been removed from
the bed. The second method involves determining the bed washout
pattern using nitrogen and oxygen. Both methods have limitations
and disadvantages which are discussed below.
Using the activation method one must remove several samples (3-5)
of molecular sieve from the bed. Each sample must be heated to 350
C. for a period of at least four hours at a pressure of approximately
one Torr. Based on the weight change of the sample one can calculate
the amount of water removed, and therefore, arrive at a value for
the weight percent water contained by the sample. The bed weight
percent water is determined by averaging the results for all samples.
Because bed activity is generally a function of the weight percent
water one can arrive at a value for the activity. The limitations
and disadvantages of this method are listed below.
1. The activation method requires disassembly of the molecular
sieve bed for the removal of several samples. Disassembly and reassembly
can be time consuming and must be performed by a skilled technician
to ensure the bed is properly reassembled.
2. This method is labor and time intensive. In general, an activity
test using this method would require approximately 6-8 hours per
bed.
3. If the samples are not taken randomly, this method can give
inaccurate results. These inaccuracies may occur because generally
only 1-2% of the molecular sieve in the bed undergoes the test.
A schematic diagram of the apparatus required for the washout pattern
technique is shown at FIG. 1 (See K. G. Ikels and C. F. Theis, Aviation,
Space, and Environmental Medicine, 56: 33-6 1985.). Using this
technique, the molecular sieve bed is first flushed with oxygen
via valves V1a, V3a, and V4a. Confirmation of a thoroughly flushed
bed would be a 100% oxygen signal at a mass spectrometer 1a. The
gas flow is then switched from oxygen to nitrogen via a valve V3a,
and the oxygen washout pattern is recorded on a strip chart recorder
2a. A waveform of the nitrogen front exiting the bed is recorded
and used to determine the activity of the molecular sieve. The lower
the activity of the molecular sieve in the bed the shorter the time
required for the nitrogen front to appear. Because washout time
is a function of bed activity one can arrive at a value for the
bed activity, if one has defined this relationship for the particular
bed under test. The limitations and disadvantages of this technique
are presented below.
1. Ideally this technique requires a mass spectrometer to analyze
the concentration of nitrogen and oxygen in the flow. The cost of
this unit is approximately $45000-60000. Hence, the cost of an
apparatus for testing bed activity based on the washout pattern
technique would be expensive.
2. The possibility of obtaining inaccurate values for the bed activity
is likely due to the dynamic nature of the washout pattern technique.
The results are highly dependent on:
a. The pressure upstream of valve V4a.
b. The steady-state flow setting.
c. The geometry of the particular bed under test.
d. The atmospheric pressure.
e. The diameter of the piping.
f. The response time of the mass spectrometer (if a unit other
than a Perkin-Elmer MGA-1100 is employed).
Hence, reproducibility of the data between two apparatuses could
be a problem.
3. Use of this technique would require the user to establish a
relationship between the washout pattern and activity for each type
of molecular sieve bed tested. This relationship would have to be
accomplished by a skilled technician.
4. The washout pattern technique also requires a skilled technician
to interpret the washout patterns.
United States patents of interest include No. 4725293 to Gunderson,
which relates to automatic control for a pressure swing adsorption
system which fractionalizes air to recover a high purity component.
This patent discloses a preferred embodiment in which comparator-controllers
are implemented by a microprocessor based programmable controller
using software provided therewith which includes Proportional-Integral
and Derivative (P-I-D) control algorithms. See, for example, col.
8 line 44 et seq., col. 12 line 4 et seq. and appendix A.
Pat. No. 4648888 to Rowland relates to an oxygen concentrator
and discloses a controller having a microprocessor which may be
programmed to change the sieve bed and/or surge tank charge times
to maintain desirable oxygen conentrations in the product gas. Similarly
see Pat. No. 4561287 to Rowland.
Pat. No. 4627860 to Rowland relates to an oxygen concentrator
and test apparatus having means for selecting any of the functions
monitored by the microprocessor. The test apparatus is connected
to the concentrator and displays the selected monitored functions
for diagnosing performance levels and component problems or failures.
Pat. No. 4404005 to Hamlin et al relates to a breathable gas supply
for aircrew in a pressurized cabin, comprising a control system
based upon a microprocessor which can incorporate a self-test facility.
Pat. No. 4272265 to Snyder describes apparatus for generating
oxygen by the pressure swing method. The apparatus is comprised
of a plurality of vessels each having a molecular sieve bed.
SUMMARY OF THE INVENTION
An object of the invention is to provide a quick, accurate, and
cost-effective means for determining the available adsorption capacity
or activity of molecular sieve oxygen concentrator beds, for ensuring
that the molecular sieve beds meet accepted standards of activity.
Activity is a measure of the condition of the molecular sieve and
is defined herein as, ##EQU1##
The invention provides an apparatus and method for determining
the activity, using a bed tester unit (which includes nitrogen and
helium sources, a helium chamber of plenum, a scale and temperature
and pressure readouts) and a computer program.
Determining the weight of nitrogen adsorbed for the bed under test
includes pressurizing the bed from a pure nitrogen source, and weighing
the bed before and after to provide a measure of the weight gain.
Nitrogen is adsorbed in the large alpha cages of the zeolite crystals.
In addition nitrogen fills the void volume of the molecular sieve
pellets. Hence, the weight of the gas in the void volume must be
subtracted to find the true weight gain due to adsorption. The ideal
gas law (PV=MRT) is used a number of times in the calculations to
find the weight of the nitrogen gas in the void volume. First, a
chamber or plenum of known volume is filled with helium, and expanded
into the bed. Then the helium is evacuated, and the bed is pressurized
with nitrogen. Measured values of pressure, temperature and weight,
and known values of volume and density, are used with the gas law
and other equations in calculations to arrive at the true weight
gain due to nitrogen adsorption.
Determining the weight of nitrogen adsorbed by an equivalent weight
of activated molecular sieve is accomplished by using a set of pure
crystals isotherm parameters, which are stored in a table. These
parameters were determined by the applicant by collecting pure crystal-N2
isotherm data for the zeolite crystals over the temperature range
of 14-44 C. The data were then fit to a Sips equation by a least
squares technique.
The weight percent of water of the molecular sieve is calculated
based on correlations with activity determined by a least squares
technique. These correlations were determined by the applicant.
The bed tester and method according to the invention could be used
in the following ways:
1. Molecular sieve beds could be tested as they leave the manufacturer's
production line. This would ensure molecular sieve beds meet an
acceptable activity specification before installations on newly
manufactured concentrators.
2. Molecular sieve beds already in service could be tested periodically
(possibly during aircraft phase inspections) to ensure they meet
minimum specifications. This testing would also provide a convenient
means for tracking the activity of the bed during the molecular
sieve's life.
3. Molecular sieve replacement beds taken from storage could be
tested before installation to ensure that the activity of the molecular
sieve had not degraded in storage.
ADVANTAGES AND FEATURES
Advantages and features of the invention include the following:
1. The activity and equivalent weight percent water of a molecular
sieve bed is determined based on the bed weight change during nitrogen
adsorption.
2. The apparatus can be constructed for less than $4000.
3. The bed activity can be determined in 6-15 minutes.
4. The activity of any type of concentrator bed can be determined.
The only information required for each type of bed is the volume
occupied by the molecular sieve pellets.
5. Correlations technique developed by the applicant (G. W. Miller)
are used to determine the equivalent weight percent water.
6. The apparatus and computer program for practicing the invention
can be operated by individuals with litte or no training.
7. The activity test is conducted on the entire contents of the
molecular sieve bed.
8. The adsorption capacity of pure molecular sieve crystals is
the standard used for determination of activity.
BRIEF DESCRIPTION OF THE DRAWING
FIG. 1 is a block and schematic diagram showing apparatus required
for a prior art technique to determine a washout pattern;
FIG. 2 is a flow chart; and
FIG. 3 is a block and schematic diagram showing bed tester apparatus
according to the invention.
DETAILED DESCRIPTION
In general, the activity of a molecular sieve bed degrades upon
exposure to certain chemical species (especially water). In most
cases this degradation in activity occurs at a slow rate, however,
if the molecular sieve were exposed to liquid phase water, due to
condensation or feed air water separator failure, the degradation
could be significant. A degradation in molecular sieve activity
results in a reduction in concentrator performance and contaminant
removal ability. The use of the method and apparatus according to
the invention provides reproducible information on the exact condition
of the molecular sieve material. With this information one can set
minimum acceptable specifications for bed activity, and hence, use
this specification as a guideline for bed replacements, thereby,
ensuring consistent performance of oxygen concentrators.
The basic steps and elements relating to the invention are illustrated
by a flow chart of FIG. 2. Bed tester apparatus (which includes
nitrogen and helium sources, a scale and temperature and pressure
readouts) is used in steps shown by blocks 102-118. A computer program
is preferably used to perform calculations shown in blocks 130-152.
Predetermined values for data and constants are shown in the chart
as tables in blocks 120 and 122 which are preferably included in
the computer program. Broadly speaking, the tables and formulas
for the calculations could be simply provided on paper.
As stated above, the basic principle of the invention is to determine
the activity of molecular sieve beds by finding the ratio of the
weight of nitrogen adsorbed for a bed under test, shown in block
138 as Q.sub.X, to the weight of nitrogen adsorbed for an equivalent
weight of activated molecular sieve, shown in block 144 as Q.sub.M.
The ratio Q.sub.X /Q.sub.M is shown as being formed in block 150.
Determining the weight of nitrogen adsorbed for the bed under test
includes pressurizing the bed from a pure nitrogen source, and weighing
the bed before and after to provide a measure of the weight gain.
Nitrogen is adsorbed in the large alpha cages of the crystals and
gas phase nitrogen fills the void volume of the molecular sieve
pellets. The weight of this gas in the void volume needs to be subtracted
to find the true weight gain. The ideal gas law (PV=MRT) is used
a number of times in the calculations to find the weight of the
nitrogen gas in the void volume. First, referring to blocks 102-110
a chamber or plenum is filled with helium, the initial pressure
and temperature are measured, the helium is expanded into the bed,
the final helium pressure and temperature are measured, and the
helium is evacuated. As shown at blocks 110-118 the bed is now
weighed for a tare value, it is pressurized with nitrogen, the pressure
and temperature are measured, and the bed is weighed again to determine
the weight gain.
The ideal gas law is used first at block 130 to calculate the mass
of the helium (M=PV/RT) using the initial values of pressure and
temperature measured at step 104 with the chamber volume and the
gas constant from the table 120. This value of mass is used at block
132 to calculate the total system volume (V=MRT/P) using the final
helium pressure and temperature measured at step 108. At block 134
the void volume is determined by subtracting values from table 120
for volumes of the chamber, some connecting tubing, and the beta
cages from the system volume. This void volume is used at block
136 along with the nitrogen pressure and temperature measured at
step 116 to calculate the weight (M=PV/RT) of nitrogen in the gas
phase. At block 138 this weight is deducted from that measured at
step 118 to obtain the true weight gain due to nitrogen adsorption.
Determining the weight of nitrogen adsorbed by an equivalent weight
of activated molecular sieve is accomplished by using a set of pure
crystal isotherm parameters, which are stored in a table read at
block 122. These parameters were determined by applicant (G. W.
Miller) by collecting pure crystal-N2 isotherm data for 5A and 13X
zeolite crystals over the temperature range of 14-44 C. The data
were then fit to a Sips equation (shown below) (see Sips, R. J.,
J. Chem. Phys., 16 490 (1948)) by a least squares technique. ##EQU2##
where, q=amount adsorbed
p=pressure
a, b, and c=parameters to be determined by least squares analysis
To read the isotherm parameters at step 122 the value of nitrogen
temperature measured at step 116 is used.
The weight of activated molecular sieve is calculated at block
140 using values of bed volume and density from the table at block
120 and using the product of these values. At block 142 the Sips
equation is used with the isotherm parameters a, b and c read from
the table at step 122 to calculate the amount of nitrogen adsorbed.
At block 144 the maximum amount of nitrogen adsorbed for the equivalent
weight of activated molecular sieve is calculated using the results
from steps 140 and 142. The activity and weight percent of water
are now calculated at steps 150 and 152.
The weight percent of water of the molecular sieve is determined
based on correlations with activity which were developed by applicant.
The correlations for 5AMG and MG3 were determined by a least squares
technique and are presented below.
where,
Y=weight percent water
X=activity
Calculations
The calculations of blocks 130-152 are given below as a set of
equations. The nomenclature or definition of terms used in the equations
is listed first with the symbol used in the equations in the first
column, a FORTRAN name in the second column, the definition in the
third column, and the units in the fourth column. The units of pressure
are pounds per square inch (psia), atmospheres (atm), and torrs
(mm of Hg). The units of temperature are degrees centigrade (C.)
and degrees Kelvin (K.). The units of volume are millimeters (ml)
and liters. The units of weight and mass are grams (gm) and gram
moles (moles). The units for some amounts are millimeters per gram
at standard temperature (zero degrees C.) and pressure (one atmosphere)
(ml STP/gm). The FORTRAN names for pellets and crystal are given
in the second column only for type 5AMG pellets and 5A crystal. |