N O T I C E

THIS DOCUMENT HAS BEEN REPRODUCED FROM MICROFICHE. ALTHOUGH IT IS RECOGNIZED THAT

CERTAIN PORTIONS ARE ILLEGIBLE, IT IS BEING RELEASED IN THE INTEREST OF MAKING AVAILABLE AS MUCH

INFORMATION AS POSSIBLE

t-

[ `C

Y S

(NASA-Ch- 160917) BREADBOARD SOLID Alf E N81-18660WATER DESORBED CO2 CONTROL SYSTEM TestReport, 1 Sep. 1979 30 Jun. 1980 (Hamilton'Standard, Windsor Locks, Conn.) 59 p Unclas j

aC A04/MF 1101 CSCL 06K G3/54 41.460 i, !Test Report .,

µ'

Breadboard Solid ► US Water Casorb :d

Arlo Co Control System

By

Arthur K. Colling andMark M. Hultman

Prepared Under Contract No NAS 9-13624

ByyHamilton Standard

^ Division of United Technologies CorporationWindsor Locks, CT Ii

For

National Aeronautics and Space AdministrationLyndon B. Johnson Space Center

rHouston, TX

November,-1980

1

l

F

M AP .

,$1r

RECEIVED

"Am STI FAMINAMESS DEPT., . +

i

MAM^LI^N sTANO^RD0 amp'"

SVHSER 7226

Test Report

Breadboard Solid Amine Water Desorbed

CO2 Control System

By

Arthur K. Colling andMark M. Hultman

Prepared Under Contract No. NAS 9-13624

By

Hamilton StandardDivision of united Technologies Corporation

Windsor Locks, CT

For

National Aeronautics and Space AdministrationLyndon B. Johnson Space Center

Houston, TX

November, 1980

HAN U^Olii sTANOrARpODIWndvlw * SVHSER 7226

'rest Report

Breadboard Solid Amine Water Desorbed

CO2 Control System

By

Arthur K. Colling andMark M. Hultman

Prepared Under Contract No. NAS 9-13624

By

Hamilton StandardDivision of United Technologies Corporation

Windsor Locks, CT

For

National Aeronautics and Space AdministrationLyndon B. Johnson Space Center

Houston, TX

November, 1980

i

MA 1r1M:TANaAAp Dhowd • SVIISER 7226

ABSTRACT

A regenerabee CO removal system is being developed for poten-tial use on the Shuttle as an alternate to the baseline lithiumhydroxide (LiOH) system. The system is called the solid aminewater desorbed, SAWR, CO removal system. It uses a solid aminematerial to adsorb CO fiom the atmosphere. The material isregenerated by heatinj it with steam from a zero gravity waterevaporator.

A full sized, thermally representative breadboard canister and apreprototype water evaporator were built and tested to shuttlerequirements for CO control. The test program was utilized toevaluate and verify 2 the operation and performance of these twoprimary components of the SAWD system.

ii

HAMILTON sTANQARp %°i"° ^ SVHSER 7226

FOREWORD

This report has been prepared by Hamilton Standard, Division ofUnited Technologies Corporation, for the National Aeronauticsand Space Administration's Lyndon B. Johnson Space Center inaccordance with Contract NAS 9-13624, "Breadboard and FlightPrototype CO and Humidity Control Systems." The report coverswork accomplished on the breadboard test program for the SolidAmine Water Desorbed CO Control System between September 1,1979 and June 30, 1980.2

Appreciation is expressed to the Technical Monitor, Mr. FrankCollier of the NASA, Johnson Space Center, for his guidance andadvice.

This program was conducted under the direction of Mr. Harlan F.Brose, Program Manager, and Mr. Albert M. Boehm and Mr. ArthurK. Colling, Program Engineers, with the assistance of Mr. MarkM. Hultman, Analysis and Mr. Philip B. Noll, Test Engineering.

i ii

NA^^OM sTA1A11p °ri SVHS ER 7 2 26

TABLE OF CONTENTS

Title Page No.

Summary 1Introduction 3Objectives 4Conclusions 5Recommendations 7Description 8

SAWD Test System 8Instrumentation 8Air System Leakage Test 8Test Canister Background 13Testing Approach 14

Discussion 16Bed Differential Pressure Tests 16Water Evaporator Tests 16Baseline Performance Teats 23Parametric Tests 30

Test Results Discussion 35CO Adsorption on Solid AmineBea Drying During Cyclic Operation

3541

CO 2 Desorption from Solid Amine 41

iv

HAMILTON CAI O °"'^`"' ^ S VHS ER 7 2 2 6

LIST OF FIGURES

Figure No. Title Page No.

SAWD Test Facility Block DiagramSAWD Test System SchematicSAWD Test CanisterSAWD Test System CO LeakageSAWD Baseline Testik With Various Beds at 0.4% CO2Solid Amine Hed Pressure DropWater Evaporator Cone;-*.Zero Gravity Water evaporato r Differential PressureSAWD Cyclic Test Runs S-27 S-36.BAWD Cyclic 'Pest Runs S-37 - S-46SAWD Cyclic Test Rums S-47 - S-60SAWD Cyclic Test Runs S-61 - S-65Typical SAWD Test Breakthrough CurveCO Removal Efficiency Versus TimeSAk Cyclic Test Runs S-67 - S-80SAWD Cyclic Test Runs S-81 - S-93SAWD Cyclic Test Runs S-94 - S-101SAWD Cyclic Test Runs S-1.02 - S-115SAWD Cyclic Test Runs S-116 - S-122Bed Temperature During AdsorptionTypical SAWD Test Breakthrough CurveSolid Amine Bed Capacity as a Function of CO2 Partial Pressureante Cycle TimeEffect of Desorb Temperature on Adsorption BreakthroughEffect of Desorption Temperature on Adsorption PerformanceSAWD Canister Weight Change During AdsorptionEffect of CO Loading on Desorption Flow and TimeDesorption Miss Characteristic with High CO2 LoadingDesorption Steam Requirements as a Function of Bed MoistureLevelDesorption Time Versus Bed Water LoadingPercent Moisture Versus Desorption TimeDesorption Thermal Characteristic With Low CO LoadingDesorption Thermal Characteristic With High C82 Loading

Figure 1Figure 2FigureFigure 4FigureFigure 6Figure 7Figure 8Figure 9Figure 10Figure 11Figure 12Figure 13Figure 14Figure 15Figure 16Figure 17Figure 18Figure 19Figure 20Figure 21Figure 22

Figure 23Figure 24Figure 25Figure 26Figure 27Figure 28

Figure 29Figure 30Figure 31Figure 32

394142434546

4748495051

v

HAWLTON as --- Awl 0-" toSVIISER 7226

LIST OF TABLES

Table No. Title Page No.

Table I Instrunentation List

12Table 2 Water Evaporator Characteristics

19

Vi

M^MILT^l1 STAi110lARt^^4,,ay.Ynof SVHSDR 7 226VNM.

SUMMARY

The Solid Amine Water Desorbed (SAWD) system test program wasdivided into the major tasks of test system setup, leakagechecks and instrumentation calibration; water evaporator tests;solid amine bed differential pressure tests; SAWD system base-line performance tests; and SAWD system parametric tests.

The initial phase of the test setup included building the breadboardsolid amine canister and the preprototype zero-gravity water evapora-tors. The canister was conservatively sized for a four-man metabolicCO load based on previous small scale test results. It was designedto2be thermally representative of a flight unit. Four preprototypewater evaporators were constructed. They consisted of tubular elec-tric heating elements inserted into stainless steel tubes. These fourunits allowed testing of two different heater sizes and four differentdiametral clearances between the heating elements and the tubes. TheSAWD system components were installed in the same test system usedpreviously for HS-C system testing. The baseline and parametric SAWsystem cgclic tests were conducted using a closed loop and a 29.31 m(1035 ft ) chamber as a simulated shuttle cabin volume. Test systemleakage tests and instrumentation calibrations were conducted priorto SAWD system testing.

Four zero-gravity water evaporators were tested separately priorto the SAWD system tests. Steady state operation under simulatedzero-gravity conditions was demonstrated for various steaming rateswithin the power capability of the evaporators. Power requirementsand differential pressure measurements were taken during the steadystate testing. A startup method, which prevents liquid water dropletcarryover and limits steam temperature overshoot, was developed.Considering differential pressure through the evaporators, steadystate performance, and startup performance, diametral clearancesbetween the heating element and 'tube of 0.254 to 0.635 mm (0.010to 0.025 inch) were optimum.

A solid amine bed differential pressure map was generated over arange of flow rates for various bed moisture contents between 14and 41 percent by weight. For a given flow rate there was onlya slight increase in bed differential pressure as moisture contentincreased within the design range of 20 to 35 percent.

Baseline performance tests established 3 that a 9.53 kg (21.0 lbm)dry Freight solid gamine bed with 0.99 m /min (35 CFM) of air flowwould maintain an average CO partial pressure of 3.0 mmHg witha four-man crew. Cyclic equilibrium bed moisture content for thebaseline cabin relative humidity is 24 percent by weight for the56 minute test cycle used.

1

MA^LTON ^TAI^AAp ^""^^ SVHSSR 7226

Parametric tests were conducted to establish the effects of highand low cabin humidity conditions on the BAWD system. Using afixed operating cycle, with consistently high or consistentlylow cabin relative humidiq, the solid amine bed gradually at-tained a higher or lower cy >;lic equilibrium moisture content.CO performance of the soli6 amine did not degrade, if bed mois-tuie content was maintained between 20 and 35 percent by weight.If the bed moisture content exceeded these limits, CO perfor-mance degraded. However, performance recovered when Aoisturecontent was restored to the 20 to 35 percent range.

2

DhowidSVIISER 7226

INTRODUCTION

A regenerable CO control system, called the SolidDesorbed (SAWD) System, is being developed for useas an alternate to the baseline lithium hydroxideThe system uses a solid amine sorbent material tothe cabin atmosphere. The solid amine material isheated with steam to desorb the CO2.

Amine Wateron the shuttle

( G qHl ) system.aasorb Co fromnubsequeAtly

The solid amine material is a polystyrene-divinybenzene copoly-mer aminated with diethylenetriamine. The porous substrate ex-poses an extremely large surface area to the cabin atmosphere.

Per Contract Modification 255 of Contract NA,S 9v-13624, this testprogram was designed to evaluate and verify the operation andperformance of the two primary components of the BAWD system, thebreadboard solid amine canister and the preprototype zero-gravitywater evaporator,

3

r..

IiA t^ON=TAI^AIIp Dwand • SVHS ER 7226

OBJECTIVES

The primary objectives of this test program were to evaluate andverify the operation and performance of the two primary compon-ents of the SAWD system, the preprototype water evaporator andthe breadboard solid amine canister.

The test program was divided into five parts:

Leakage and Instrumentation CalibrationBed Differential Pressure TestsWater Evaporator TestsBaseline Performance TestsParametric Tests

The objectives of each test phase are listed below:

Leakage and Instrumentation Calibration- Measure air system leakage

. Bed Differential Pressure Tests- Map bed pressure drop versus air flow rate for variousmoisture contents

Water Evaporator TestsTest operation at various steam flow ratesDemonstrate steadystate operation and controlEstablish and test methods for start-up control

- Determine the effect of different diametral clearancesbetween the tube and heater element

- Determine power requirements versus flow rate- Establish the relationship between inlet pressure andoutlet temperature

- Establish gravity independence of operation

. Baseline Performance Tests- Determine required air flow rate for the baseline

conditions- Demonstrate bed water content equilibrium and systemCot equilibrium For the baseline conditions.

Parametric TestsEvaluate the effects of extremes of cabin temperatureand humidity on CO 2 performance and moisture conditioningof the bed.

4

NAMILI^ON:TAMOiAIIp SVHSER 7 226

CONCLUSIONS

1. A 9.53 kq (21.0 lbm) dry weight bed of solid amine with anadsorption flow of 0.99 m 3/min (35 CFM) and a total cycle timeof 96 minutes maintains average cabin Co t partial pressure at3.0 mmHg with a four-man crew.

2. The solid amine material exhibits consistent CO adsorptionperformance when bed water content is maintained between 20and 35 percent by weight.

3. When the solid amine bed water content is returned to the 20to 35 percent range after exceeding this range, Co t adsorptionperformance returns to normal.

4. For the baseline case relative humidity of 50 percent, thesolid amine bed cyclic equilibrium water content is 24percent.

5. If inlet relative humidity during adsorption is continuallymaintained at the extremes of 27 or 70 percent, the bed cyclicequilibrium water content exceeds the 20 to 35 percent rangefor optimum CO performance. However, numerous- cycles at anextreme relative humidity condition are required before watercontent exceeds this range.

6. With a fixed steaming rate, the desorption time for a solidamine bed is a strong function of bed water content. There-fore, bed water content for a given SAWD system can bedetermined from desorption time.

7. During desorption the steam progresses through and heats thesolid amine bed in a well-defined wave.

8. During desorption all of the CO which had been ansorbed onthe bed is readsorbed and concentrated in the last portionof the bed. When the steam reaches this region of the bed,the CO is pushed out in a high purity wave. The maximumrate ai which CO 2 leaves the bed is dependent on the steamingrate.

9. When desorption is performed with the solid amine bed at lessthan atmospheric pressure, the reduced saturated steam temper-ature causes some residual CO 2 to be left on the bed.

10. Solid amine bed differential pressure increases only slightlyas bed water content varies from 20 to 35 percent by weight.

11. The zero-gravity water evaporator operatad well during bothstart-up and steady state conditions. Insensitivity togravity was demonstrated during testing by operating thewater evaporator in three orthogonal positions.

5

NAMILl^OIi sTAL^laARp ^,f°"'^°^•

SVHSER 7226

12. The diametral clearance between the tubular heating elementand the tube of a zero-gravity water evaporator should be inthe range of 0.254 and 0.635 mm ( 0.010 and 0 . 025 inch).

13. Two control methods for steady state operation of the zero-gravity water evaporators were developed.

14. The startup of the zero-gravity water evaporators can be con-trolled without experiencing any liquid water carryover.

6

NMMIL'rON a►TANaARp k ""." .SVHSER 7226

RECOMMENDATIONS

1. Based on the successful results of the breadboard SAWD sys-tem test program, a prototype SAWD system should be builtand tested.

2 The water evaporator should be attached to the inlet headerof the solid amine canister to minimize condensation in thebed inlet header during desorption by preheating the canisterheader metal.

3. A two bed SAWD system should be investigated. Two bedswould provide redundancy and would help to level desorptionpower usage peaks and minimize variations in cabin CO2parLal pressure and relative humidity.

NA ^ A p ioo'.d SVHSER 7226

DESCRIPTION

SAWD Test System

The SA4D system test components were ;located in the same testfacililty used for HS-C system testing. A block diagram k)f thefacility is shown in Figure 1. The SAWD test system portion ofthe block diagram is shown schematically in Figure 2.

Air from the simulated shuttle cabin was ducted to Rig 88 wherea blower provided the head for the entire plumbing loop. Theair then passed through a series of heat exchangers and electricreheaters, which provided the temperature and humidity condition-ing. Since the complete range of desired dry bulb temperaturesand dew points was not obtainable with the test rig, the air wasconditioned to an equivalent relative humidity. Before the airwas directed to the SAWD canister, CO partial pressure, tempera-ture, dew point, and flow rate were measured. A. venturi andvalve arrangement in Rig 88 provided for flow measurement andcontrol. A bypass line allowed conditionv to be establishedwithout flowing through the canister. Excess moisture in theair leaving the SAWD canister was removed by a heat exchangerbefore the air returned to the simulated shuttle cabin. Also,CO at the metabolic production rate was fed into the air streambetiore it returned to the cabin volume.

The breadboard SAWD test canister is shown in Figure 3. Theinside diameter is 40.64 centimeters (16 inches), and the maximumbed depth is 22.86 centimeters (9.0 inches). This size accommo-dated the bed size requirement estimated from earlier smallscale testing. The breadboard canister is thermally representa-tive of a flight unit.

Instrumen tation

The parameters measured during testing are shown in the SAWDtest system schematic of Figure 2. The instruments used aredefined in Table 1.

All instrumentation was calibrated prior to testing by theHamilton Standard Instrumentation and Metrology Department.Additionally, the CO analyzers and hygrometer were calibratedLafore each day of t^sting.

Air System Leakage Test

An air system leakage test was conducted to establish the leakagerate of the overall air circuit, including the breadboard systemair circuit, Rig 88, the simulated shuttle cabin volume, and allinterconnecting plumbing. The results were used to adjust theCO2 feed rate to account for losses due to leakage.

8

MAN^TON 3TAI^A1lp °i^^^. SVHS ER 7226

CO 2 FEED &

MONITORING

SIMULATED SHUTTLE

CABIN

29.3 M3(1035 FT

RIG 88FACILITY SUPPORT

THERMAL CONDITIONINGHUMIDITY CONDITIONINGAIR CIRCULATIONMONITOR:

PCO2

TEMPERATUREDEW POINTAIRFLOW RATE

SAWD

TEST SYSTEM

I

Figure ISAWD TEST FACILITY BLOCK DIAGRAM

9

MAMN.TVN 3TA^ARO WMLMns SVHSER 7226

COOLANT

CO2 F TOT

OUT T I

SAWD jT

0%) CANISTER

H2O

DRAIN

I SCALE I

VENT -4—

RELIEF

T

WATER

EVAP,

F TFLOWRATE

2)ENSING .AIR TO

HX SIMULATEDCABIN

BYPASS ----y- H2O

VALVE

T PCO2 DPQ v v AIR FROM

RIG 88

INSTRUMENTATION

d'T - TEMP*ERATURE<P PRESSURE<P - DELTA PRESSURE<1PCO 2 - CO2 PARTIAL PRESSURE

<DP DEW POINT

Figure 2SAWD TEST SYSTEM SCHEMATIC

I 10

j

11AMILt^ON sTANI'aARp °"""'^• SVHSER 7226

AIR IN

CO

r ♦ • ( n r,r.. rn • r rV (1 ♦ !a C? a eJ A n V ^r 1 e)f1

4? O C) ^' F '!^ • Io6 O D to^ r M ~1' Q G C^ 4A J SOLID AMINE RED nn C, ..TIC'S

. ^ +..., ,

r ^ ro" a' r, n . ^^''rr

FILL

r

.r+. ^c^ ^'^^ ('^`^. rei] ^i^ Ai i At+ e 6•.^nr^C>^?n

INNER SHELL40.6 CM I.D.(16 IN.)

INSULATION

OUTER SHELL

STEAM --P►

STEAM COMPRESSORS -4--►-

DRAIN -rt--,

t DRAIN

—o.- AIR OUT

Figure 3SAWD TEST CANISTER

SVHSER 722.6HAMILTON $TANaARp ynn ^.

Table 1

Instrumentation List

Parameter Units Instrument Range

Rig 88 and SAWD Flow CFM Venturi/Mancmeter 0 to 55 cfm

Makeup CO2 Flow lbmAr Flag Meter 0 to 1.2 lbm/hr

SAWD Canister in. H2O U-tube Mancmter 0 to 20 in. H2ODifferential Pressure

SAWD Canister Inlet OF Hygrometer -40 to 120 °FDew Point

SAWD Canister volume Infra-red Analyzer 0-3%Inlet PCO2

SAWD Canister volume Infra-red Analyzer 0-3Outlet PCO2

SAWD Canister Weight lbm Scale 0-300 Ihn

Makeup CO2 Weight Im Scale 0-300 lbm

SAWD Canister psia Gage 0-30 psiaOutlet Pressure

Water Evaporator psia Gage 0-30 psiaInlet Pressure

Temperatures OF Thenwcouples -328 to 752°FCanister InletCanister OutletWater Evaporator OutletWater Evaporator InletWater Evaporator (9)SAWD Canister (12)

Water Evaporator lbm/hr Flow Meter 0-22 lbn/hrFeed Flow

t .

12

MA^MLTOM sTANaARp ^. "'"" ^. SVIISIR 7226

The leakage test was conducted by establishing a known CO2concentration in the test system and monitoring the decay inCO concentration while the air circuit was in operation.FiiUre 4 shows the decay in CO concentration during the test.With an average CO concentration of 3.0 mmHg, 0.4 percent byvolume, CO leakagg was 0.01331 kg (0.02934 1bm) per 96 minutetest cyclet The SAWD canister was bypassed during this test toensure that no CO would be adsorbed on the bed. The smallportion of the system containing the canister had previouslybeen shown to be free of leakage by a vacuum drop test.

Test Canister Background

The baseline removal capacity for a SAWD system must be 0.254 kg(0.56 lbm) of CO for a four-man crew during a 96 minute orbit.The required solid amine bed size evolved to the final value of9.53 kg (21 lbm) dry weight during development testing. Figure5 summarizes this evolution. Based on testing with a 1.81 kg (4lbm) bed prior to August 1979, the baseline system design was a14.97 kg (33 lbm) dry weight constrained bed. Airflow was upwardthrough the bed at 1.42 m /min (50 CFM), and steam flow was alsoupward at 4.31 kg/hr (9.5 lbm/hr). The SAWD test canister wasbuilt to these requirements. However, since it was initiallyfilled with solid amine containing 10 percent moisture, the testcanister actually held 14.06 kg (31 lbm) of dry solid aminematerill. As seen in Figure 5, the 14.06 kn (31 lbm) bed at1.42 m /min (50 CFM) had significantly greater removal capacitythan required. Scaling down directly, the new required bed sizewas 11.34 kg (25 lbm), of dry solid amine and flow direction waschanged to down flow to prevent potential channelling. Thechange in 3 flow direction resulted in improved performance, andat 1.42 m /min (50 CFM) removal capacity was sti113significantlygreater than required. A flow reduction to 0.99 m /min (35 CFM)gave some reduction in bed loading, but a second scaling down ofbed weight was necessary for correct sizing. The final systemdesign for the SAWD test program was a 9.53 kg (21 lbm) dryweight, loose filled bed. Air flow was downward at 0.99 m /min(35 CFM), and steam flow was upward at 1.36 kg/hr (3.0 1bm/hr).

Testing Approach

During the SAWD test program. 122 test cycles were run on the9.53 kg (21 1bm) dry weight bed. The cycles were 96 minutes inlength to correspond to one orbit. The target for adsorptiontime was 52 minutes. However, since required desorption timeusing a fixed steam flow varied with bed moisture content,adsorption time was made equal to 96 minutes minus desorptiontime. The first 26 cycles were used to establish and optimizethe test procedure. Also bed differential pressure tests andreduced pressure desorption tests were performed during thisperiod. Test cycles 27 to 66 were used to establish the baselinesystem performance, and cycles 67 to 114 were used for parametrictesting. Baseline performance was tested for repeatabilityduring the final eight cycles.

13

.^_A

SV

'HS

ER

7226

tnLn

W0wW 1

0co 0

w

C7

uUVRV 0

t0 Co

H

U !

oO o

to co N

•0

M a

+-iO

!M

r-! r•'!Z

N o H

O2

HZ

O

2•tn

x3

OoN

• W W

UN

Cn

EiA

to—

in

v

m

N

r-r

L3

O

O

O

CD

O

Z00 SNSOHad

Figure 4

SAWD TEST SYSTEM CO2 LEAKAGE

o°

OWEHH

I

^I 14

HAMILTON WANDARD 4owwo

f sv)tsrA 7226

.45

i1.0

.3G

c^

ia . 27:

(.6cswR.

t.a

wsr:

ti t,2)

Cl

a

11.3 KG Ben, 1.42 M3/MIN

(25 LAM BED, 50 CFM)14.1 KC nRn, 1.42 M3,/MTN

(31 CmM RED, 50 cFm)

11.3 KG BED, 0.99 M3/MIN(25 1,BM BED, 35 CFM)

! —SAWD Col. LOADING PER CYCLE

MIN. MAX,SAWD BAWD

Ile-"YCLE CYCLE

9, a 3 K(: Arm, Q.C)6 via "ktl(21 LAM BFn{ 34 01"M)

10-- wK ^., "v BO

TIME MIN

Rigure 5SAWD BASELINE TESTING WITH VARIOUS

BEDS AT 0.4% CO,,

,

15

i

MAMIL'IiONiTA1Al1p WAw^. SV'HSER 7226

DISCUSSION

Bed Differential Pressure Tests

Tests were conducted to map bed differential pressure versus airflow rate over a range of beg water contents. Air flow rate wasvaried from 0.566 to 1.331 m /min (20 to 47 CFM). Six bed watercontents were tested between 14 to 41 percent. The range of 20to 35 percent bed moisture content is of particular interest,since the solid amine material must be within these limits forproper Cot performance. The results are shown in Figure 6.

There is an increase in bed differential pressure as moisturecontent increases. However, since the solid amine materialswells as water is adsorbed, the inter-particle void space growsin proportion to water loading. This tends to limit the beddifferential pressure increase with higher water loading.

Water Evaporator Tests

Preprototype water evaporators were performance tested prior toinstallation into the SAWD test system. The zero-gravity waterevaporator concept is shown in Figure 7. The evaporator consistsof a tubular electric heating element inserted into a stainlesssteel tube. The diametral clearance between the heating elementand the tube is 0.254 - 0.635 mm (0.0:10 - 0.025 inch). The heaterand tube lengths are typically 2.54 - 5.08 meters (100 - 200 inches)depending on the required steam production rate. However, once theheating element has been inserted into the tube, the assembly can bebent into a convenient shape. A temperature sensor is installed incontact with the heating element near the steam outlet to provide acontrol signal. Feedwater to the evaporator is supplied through aflow meter by a variable stroke, piston type metering pump.

Four preprototype zero-gravity water evaporators were performancetested. Three of the evaporators consisted of 1.5875 centimeter(0.625 inch) outside diameter stainless steel tubes with 1.260centimeter (0.496 inch) diameter tubular electric heating elementsinside. Three different wall thickness tubes resulted in threedifferent diametral clearances between the tubes and heating ele-ments. The fourth evaporator was°.,.,instructed with a 0.9525 centi-meter (0.375 inch) outside diameter tube with a 0.6604 centimeter(0.260 inch) diameter heating element. A summary of the waterevaporator characteristics is given in Table 2.

i16

I.'A

SVIISER 7226

50.(20)

0

38.

25. ^

P, (10)0$Xim

12.

(5)

0

I I ICANISTER

AMINE - 9.53 KG (21 mm) DRYHEIGMT - 17.8 Cm (7 IN.) AT 35% H20

AREA 0.13 M 2 (1.4 FT2H

2 0 CON

38,

Oil

0 .283 .566(10) (20)

.850 1.13(30) (40)

1. 4^(50

rENT

'41%

29%

23%

19%

14%

AIRFLOW - M 3/MIN (SCFM)

FIGURE 6SOLID AMINE BED PRESSURE DROP

I

17

HAMILTQM STANOIARp U °rM cl

- wr

f^1

I

I BEND AS RODAFTER ASSY

-0.635MM _ 0,254MM(0.025 INCH-0.010 INCH)DIAMETRAL CLEARANCE

SVHSER 7226

WATERINLET

TUBULARHEATINGELEMENT

ELECTRICALTERMINAL(EACH END)

TOTALLENGTH(100 TO200 IN.)2,84 TO5.06M

`—COMPRESSIONHEATER SEAL(EACH END)

SUPERHEATEDCONTROL STEAMTEMP. SENSOR

Figure 7WATER EVAPORATOR CONCEPT

18

HA

MILT

ON

iTA

M01A

1lO %

/nr..S

VH

SE

R 7

22

6

s^

jj1

tt1 N

N

vv

vv

^

^

a ^

^i

N.^.i

N^-iCN

ai••zz

^^ C

J^

NN

N

Q^ .^.

rNINi

C4

CT

c`lM

,N

Q

t3

In

Ch

to

in

to

NNo

=i

Caao

^

bN

,..m

tnm

A c,s ^

oa

o0

CN

iaJ.

CT

0%

s^

ca

o

R! A

koCVwN

wCNgoo

E••

O u

^,-^

^O

ao

00

U.-i UPI

^O

O^.

"

yQ

1to

t n

Ln

r?v

lxlv

I'-v

Q O O G

NN

_10

1C

Nt0

C^M

•,^

OO

OO

°=

(!L

ntn

Lnco

coco

N•

o

19

I_.A

HAMILTON STOJO- SVHSCR 7226

The test syste^ is shown schematically in Figure 2. Supportequipment for the evaporator consisted of a metering pump, aflow meter, an electrical controller, a pressure gage, and athermocouple readout. Tine metering pump was a Fluid MeteringIncorporated model RP-D-0. It was a reciprocating piston pumpwith a variable stroke. Thermocouples were positioned incontact with the outside of the stainless steel tube at ninelocations along the evaporator. Two thermocouples were placedin contact with the heating element. One was located near themiddle of the evaporator, and the other was near the steamoutlet. This latter thermocouple was used as a controllerinput. Since ti,e evaporator outlet was vented to atmosphere,t;`^e pressure gage at the inlet: indicated pressure drop throughthe evaporator.

Figure 8 shows the relationship between pressure drop through theevaporators and steam flow rate. Pressure drop is a function ofboth evaporator diameter and diametral clearance between the heat-ing element and the tube, since both of these factors affect theflow area. Four evaporators with diametral clearances between theheating element and tube ranging from 0.1651 min inch) to0.7493 min (0.0295 inch) were tested. As expected, larger clear-ances resulted in smaller differential pressure drops through theevaporators. Steady state performance was satisfactory for allevaporators tested, except the 0.1651 min inch) clearanceevaporator had a pressure drop which was too high for safe opera-tion with the test system. 'rhe medium clearance evaporator,0.3937 min inch), was easier to control during startupwithout experiencing liquid water carryover.

During the remainder of the SAWD test program the 0.9525 centi-meter (0.375 inch) diameter, two KW evaporator was used for steamdesorption of the SAWD beds, since its size matched the steam flowrates needed. Over 200 hours of operating time was completed withthis evaporatorwith no performance degradation.

Water evaporators were successfully tested over the range of flowrates xrom 0.59 kg/hr (1,30 lbm/hr) to 4.20 kg/hr (9.25 lbm/hr).These limits are of no particular significance, except that theywere within the capability of the test equipment. Water evaporatorsof this design can be built for any steam flow rate correspondingto available heater element and tube size combinations.

20

L

HAMMON s /- 01-0 1

SVHSER 7226

345(50)

Ha

a 276(40)

wD

207a (30)aHHzww 138w (20)HA

69(10)

414(60)

00

1.5875 CM DIA.(0.0625 INCH)0.3937 MM.(0.0155 INCH)DIAMETRALCLEARANCE

'105875 CM DIA.(0.625 INCH)0.7493 MM(0.0295 INCH)DIAMETRALCLEARANCE

X0.9525 CM DIA.i (0.375 INCH)

0.2159 MM(0.0085 INCH)DIAMETRALCLEARANCE

.45 .91 1.4 1.8 1.3 1.7 3.2 3.6(1) (2) (3) (4) (5) (6) (7) (8)

STEAM FLOW - KG/HR (LLCM/HR)

Figure 8

ZERO GRAVITY WATER EVAPORATORDIFFERENTIAL PRESSURE

4.1 4.5 5.0(9) (10) (11)

1.

21

HAMIL 0 STAIrO ^, 1>~01 SVHSER 72 2 6

Steady state operation at numerous steaming rates was demonstrated.With proper power control, there was no liquid water carryover, andthe degree of superheat was easily controlled. Two control methodswere successfully tested for steady state operation. With the firstmethod a fixed heater power was cycled on and off based on a selectedcontrol sensor temperature. With the second method voltage to theheater was adjusted based on a selected control sensor temperature.

The objective of startup control was to prevent liquid water dropletsfrom being carried with the steam during evaporator heat-up. If thepump and heater were energized simultaneously, some liquid water was.forced out with the initial steam flow. If the heating element waspreheated before the pump was started, water carryover was initiallyprevented, but there was a temporary instability in steam temperature.Typically, steam temperature first overshot and then undershot thedesired temperature before stabilizing. Often some water dropletcarryover was experienced during the low temperature swing. Theabove problems were eliminated with the following startup method.The heating element was preheated to a specified temperature. Whenthe pump was started, the heater was deenergized until outlet steamtemperature peaked. Then, heater power was returned to its operat-ing level.

For all evaporators tested average power consumption was 0.761KW1d/kq (0.345 K1111/lbm) of steam. This value corresponded to a5.6 percent inefficiency due to heat losses to ambient. Since heatloss to ambient was independent of steam flow rate, the inefficiencypercentage was slightly less at higher steaming rates.- Power forthe metering pump was not included in these values.

Inlet pressure and outlet temperature were not directly related.Inlet pressure was primarily a function of diametral clearancebetween the tube and heating element and steam flow rate. Outlettemperature was primarily a function of input power.

Operation of the water evaporators was demonstrated to be insensi-tive to gravity orientation. Evaporators were operated in bothstartup and steady state modes in three orthagonal positions. Nodifferences were seen in either peformance or operating charac -teristics. Therefore, the performance should not be affected byzero-gravity.

NA^ML7^ON s'TANaAROers• SVHSER 7226

Baseline Performance 'rests

The breadboard SAWD system was tested to verify baseline perfor-mance under the following conditions:

Parameter

Inlet Air TemperatureInlet Air Dew PointAir Flow RateCO Flow RateCO 2 Level (Max. Average)Adsorb and Desorb Cycle

Requirement

21.11°C (70°F)10°C (J0 °F)0.99 m /min (35 CFM)0.159 kg/hr (0.35 lbm/hr)3.0 mmHg96 minutes

Since the required temperature and deli point could not be obtainedwith the test system, it was agreed with the NASA technical monitorthat temperature and dew point could be flexible, if the correctrelative humidity was maintained in the canister inlet air flow.The CO flow rate used in the test corresponds to the nominal four-man metabolic rate of 0.957 kg (2.11 lbm) of CO per man-day. Theactual CO, flow rate used during the test was increased by 0.0083kq/hr (0.8183 lbm/hr) to account for air system leakage. CO feedrate was constant during each test cycle, and the test chamber CO2level was allowed to vary. The test plan specified varying airflow rate as necessary to maintain an average test chamber CO levelof 3.0 mm[Ig. Thus, the required baseline system air flow couldbe determined. Once the baseline air flow rate was established,test cycles were run using the established air flow value untilCO performance and bed water content equilibrium could- be demonstated.Figures 9 through 12 provide a summary of the system performanceduring the baseline test cycles, S-27 through S-65. The data fortest cycle 5-66 was not included due to a rig problem. The figuresgive test conditions, test chamber: CO 2 partial pressure, and bedmoisture content. It should be noted that since the test volumewas only approximately half the size of an actual shuttle cabin,the CO partial pressure variation during each test cycle wastwice is large as the variation in an actual shuttle vehicle.

As seen in Figures 9 through 12, an average test chamber CO par-tial pressure oS 3.0 mmHg, 0.4 percent by volume, was maintainedwith the 0.99 m /min (35 CFM) air flow rate. Consequently, noair flow rate variations were necessary to establish the baselinevalue. The test chamber and canister outlet CO 2 partial pressuresare shown in Figure 13 for a typical adsorption cycle. Carbondioxide removal efficiency for the same cycle is shown in Figure14.

23

... ,m

Lza

0

— ^ 1.9

HAMILT0

Apk N

osold

%Pmr ss

SV

HS

ER

7226

IIE-4

N ^^

E-4 vy

0co11

E-4 IIE-4 itE-4

rl-1

0

m

co

0(N

N

H

r-i^o

EVIG'um Eulvm Gas *

O

C)0

(W) Z

OO

sua

ssau

d q

Vilu

va

cc

C,11

1E-04 l'u,

0 co

co N

111

00 t--

Lnr- N

111

E-4

ri

I

Pioure 9

SA

WD

CY

CL

IC T

ES

T R

UN

S S

-2

7 - S

-36

24

a.

SNIOVor l HaLMI GaEl

MFig

ure

10SAWD CYCLIC TESTS RUNS S-37 - S

-46

SV

HS

ER

722

6

qr kD

Ln r-

11if

1 :14 E-4

ImV w

aH Ln r

^

Ul r-

11 11a E-4

Q

--:r kDm

to r-

-t:r

1111

1

44 E-4

V)

ev Ln r4

L'I r--

E-4tn

aU:Hrr to

Uto 1

44 E-4

A-T Ln 0

Ln fl-

1111

1C4 E-4

tn

-T Ln

CN

to rl-

04 E-4

'14' toco

Ln r-E-4

Ln

,; to

toCw E-4

ur

0~

1 cd

25

r.

0

0

E-4

( ) DR

IQV

M E

a.1

w Q

a

MA

MILI^O

N sT

AN

D^A

RQ

°i®"

^^

SVHSSR 7226

aww

^o

..—W°

WcT

tn

W WoD1t^f l l

1i ;E^

1(

H QWA

NU1

Ln

Ln

HlD

vw

I-Ln

1t^WaU

m

CO

N

H

^oJCO

tn

A H

I

co

OLnLO

00

m

COr,0VI t!1

r-

AlO

tf1^T

r%

CV O

O

O

O

O

O

O%

SO

D 3

Sn

SS

3U

d 'I

KIlH

Vd

Figure 11

SAWD CYCLIC TEST RUNS 5-47 - 5-

60

26

}1

Ln co

W HALn co

►n r

Ln O

A^

n OL

LAqr00

Lnr^I

A E^

WauU

1^I

liAM

ILl^ON

STA

MD

iARp ""^^.

SVHSER 7226

MOMN

kDMrriN

NMNNIn

MNrnri

NL

oLn

o

^oON

NN

($ ) D

NIC

Ival u

azvm

QS

S

w^wO

9RhII

HA

L^sr

M

NO

OO

OO

M

z00 sansssxa gVilava

Fig

ure

12

SAWD CYCLIC TEST RUNS S-61 - S-65

27

cy^

0

low

NCO l

G AN O

0.5

0.4

HazHE, 0.

NOUdw

0.2

0.1

11AMILTnN a'TAN^ARp °"'^"^. SVHS8R 1226

50 60I 6U J 4U

TIME INTO ADSORPTION (MINUTES)

Figure 13

:!'YPICAL SAWD TEST BREAKTHROUGH CURVE

7

28

OONDITIgNS: BED WEIGHT - ' 9.53

AIR FLOW - 0.991

KG (21

M3/MIN

IM)

(35 CflH)

f «

to0 10 20 30 40 50 60 7a

aO

9UwHU

gHGtr

GtrW

a 7'

0w 6xN

51

41

MAMILTOM:TANOMIIp amulg °"^^^. SVHSER 7226

TIME INTO ADSORPTION MIN

Figure 14CO2 REMOVAL EFFICIENCY VERSUS TIME

29

lIAMILt^ON sTAN0iAR0amwvld s SVHS SR 7 2 2 6

Bed moisture content cyclic equilibrium for the baseline condi-tions was reached by test cycle S-43. This equilibrium value,approximately 24 percent by weight, was maintained for the re-maintj ,,t); 22 cycles of the baseline tests. Minor moisture contentvariations from cycle to cycle are attributable to similar vari-ations in inlet air relative humidity.

Parametric Tests

The effect of off-design cabin conditions on the SAWD system wastested per the following table:

Test Air T2W. Air .D .Air Flow* CO, Flowa C_( °F) m''/min (CFM) kg i ( r)

high R.H. 21.11 (70) 15.56 (60) 0.99 (35) 0.159 (0.35)low R.H. 26.67 (80) 6.11 (43) 0.99 (35) 0.159 (0.35)

*The air flow rate established during baseline testing was used.

These tests were utilized to determine the effects of extremes incabin temperature and humidity on the CO, performance and moistureconditioning of the bed. The actual CO `flow rate used during thetests was increased by 0.0083 kg/hr (0.8183 lbm/hr) to account forair system leakage. As in the baseline performance tests, thetemperature and dew point of the inlet air was flexible, if thecorrect relative humidity was maintained.

Figures 15 through 18 provide a summary of the system performanceduring the parametric test cycles, S-67 through S-115. The figuresgive test conditions, test chamber CO 2 partial pressure, and bedmoisture content.

Cyclic testing with high relative humidity inlet air was conductedduring test cycles S-67 to S-90. Bed moisture content graduallyincreased to 35 percent by weight by test cycle S -80. CO perfor-mance did not degrade during this moisture content transi8nt. Aftertest cycle S-80, due to the high bed water content, desorption timewith a fixed steaming rate exceeded 44 minutes. Therefore, tomaintain the 96 minute cycle time, adsorption time was reduced below52 minutes. The shorter adsorption times reduced drying ability,and the bed moisture content increased to 42 percent by cycle S - 90.Also as seen in Figures 16 and 17, CO performance began to degradewhen bed moisture content increased above 35 percent after cycleS •-80. Inlet air relative humidity was reduced for test cyclesS-92 and S-93. Bed moisture content started to decrease; however,numerous cycles would have been necessary to reduce the moisturecontent significantly. Therefore, the bed was dried to 30 percent

30

IND -sw/-D—

'O'S

VH

SE

R7226

co

000

0M N

ML

E

Ln

H040

CAI

2XM

N N

W^ ^

04 Ha

Q4 N

N114M

11

q

co

tn r-

Ln

ifa

P+

04 H

Mcn 04

trw

I'lIf

If

04

E-4

U

M

>4

if

r- U

c^

C14

Of

NLn r-

04

E-'I

T

00

.-1

r-4

Ln r-E-4

tri

C'4aM C4

0CD

w n

r-

C`

11 11

P4 E-4

1

CN

Mw11

P, E-^

Ln

CO M

00

•tn t

o

C%3

a4 E-4

U)

9z

00 00

r-

Ln

Ln >o

11 11

w

CL E-4

1

tpC14

co

cf

wLn

It:r m

04

r-4

MW

V'IIUVd*

miaval usaw Use

Figure 15

SA

WD

CY

CL

IC T

ES

T R

UN

S S

-67

- S-8

0

31

0L4

UH

U

8 8

8 8

8 8

8 8

W L7 0

0 sa

nssm

id 7

vllu

vd

-cr

1^rm

m

fn

W oNiavm omm ass

SV

HS

ER

72

26

0-%

Figure 16

SAW

D C

YC

LIC

TE

ST

RU

NS

S-8

1 - S

-93

t

32

N

-1

O

M

M

mM

CN

ICV

CII ?

ZjV

M 0

38

SV

EIS

ER

7226

r^--w0

P+

N

^-iO

A

cn

In co o

011

(

i A

H

A

w t•-

cm^ m

o E

- i

t` m h

o

^^ O

N,.A H it;

are m

aN

r`G

it II

1a H

U0

o cv b

ro t°- Q

ss+E- 4

I

^

w

^n^r -

m

N

^

O

O

O

OO

O

O

O

w za p

^xnss^a 'I^

tIs^3^ a

t

I`w H

,^w

r^aN

iu

acn

Figure 17

SAWD CYCLIC TEST RUNS 5-94 - S-101

33

MA

MIO

OM

sT

AN

aA

Rp

^• ow

wnd

SVHSER 7226

rua

WaUU

o

Ln

o

M JN

IGv

og

'dS

IVM

QS

S

w

30

; Z

UnS

SE

Ud ` V

I.Lld^d

Pigure 18

SAWn CYCLIC TEST RUNS S-102 - 5-115

34

,,_A

SVHSER 7226a

water content with a continuous dry air flow before the next testcycle. Test cycles S-94 to S-101 were run using near baselinetest conditions. Bed moisture content fluctuated near 30 percent.Figure 17 shows that CO performance returned to that experiencedduring baseline testing? when the bed was dried below 35 percentwater content.

Cyclic testing with low relative humidity was conducted duringtest cycles S-102 to S-114. Refer to Figure 18. The bed watercontent was reduced to 12 percent by weight using dry air flowprior to test cycle S-102. During test cycles S-102 through S-104,canister inlet air relative humidities averaging 28 percent wereused, and bed water content remained between 11 and 12 percent.The CO feed was intentionally retarded during these tests. Itwas de h red to maintain a reasonable system CO level of approx-imately 3 to 4 mmHg, and the bed breakthrough irofile showedpredictable degradation at this low moisture level. Nigher inletrelative humidities were used for the next five test cycles, S-105to S-109, and bed water content was increased to the 18 to 19percent range. However, CO performance remained degraded duringthese tests. For test cycles S-110 to S-114 average canister inletrelative humidities of 35 percent caused bed water content to remainstable in the 18 to 19 percent range. Carbon dioxide performancecontinued to be slightly degraded with an average test chamber COpartial pressure of 3.8 mmHg, 0.5 percent CO by volume. For the 2final eight cycles of the test program, average canister in letrelative humidity was 60 percent, which caused bed water contentto increase to 24 percent by weight. As seen in Figure 19, base-line CO performance was restored after bed water content increasedabove t9e 20 to 21 percent range.

Test Result, Discussion

CO2 Adsorption On Solid Amine

Adsorption of CO on a solid amine bed begins immediately when ahot bed is returned to adsorption after being steam desorbed. Therapid cooling in the front of the bed, shown in Figure 20, allowsadsorption to begin. This is shown by the typical adsorptionbreakthrough curve of Figure 21.

Solid: amine capacity for CO increases significantly as CO partialpressure increases. Figure 2 22, which was generated by a c^mputersimulation based on available data points, shows the cyclic COloading on solid amine versus CO partial pressure for various 2adsorption cycle times.

II

35

MA

MILT

hM s'T

AN

QA

RO

C"M

cd

SV

IISE

R 7226

wo r.

--w0qHq

WW

Waua^

Oa

aa

OZoo S

unssau

d 7

VIlu

vd

Figure 19

SAWN CYCLIC TEST RUNS S-116

S-122

OMs1^N

U1C1

NN

MMNN

OrnN

rnNN

-4Nlf1

ON

[f1

^f'

C+')

N ''i

N N

N N

N

t%DNICIVOU

HS

ILV

M C

Iss

Or

36

93.3(200)

82.2(180)

71.3(160)

0

U0 60.0i (140)

wx

E' 48.(120)

P4

wH

37.8

15.6(60)0

CONDITIONS

FLOW -- 0.99 N 3/MIN (35 CFM)BED - 9.5 KG (21 LAM)TEST NUMBER - S-13PCO 2 - 0.4%

RFD INLETOman

(10(',)

26.,(80)

,^- TOP^^-- OUTLET

MIDDLEBOl."

10 20 30 40 50

TIME MINI

MAMN.TON STAMDiARD t^' """ors SVIISER 7226

Figure 20

BED TEMPERATURE DURING ADSORPTION

37

NCO 2

co

0 10 20 30 40 50 Fi

0.5

0.4

N

zH

F 0..

N0Udp

0.2

0.1

HAMILTON Dkv"co • SVHSER 7226

TIME INTO ADSORPTION (MINUTES)

Figure 21

TYPICAL SAWD TEST BREAKTHROUGH CURVE

CONDIT IONS: 1.56 M Z/MIN (55 CFM)TEST WEIGHT (DAY): 6.66 KC (19.1 LBM) CYCLE101.4 kPa (14. . 7 PSIA) TOTAL PRESSURE TIME

(MIN)70I6050

40

30

V^-0^00-000

zo

D 1 a a w) W

CO 2 PARTIAL PRESSURE ( MMIIG)

I I

HAMILTON sTAND^AR^ '^ °i" ^. SVHSER 7226

w

xp

N .040UWma

W .o;H

Naax .ory0U

US:

NF►rUdaaU0ww

Figure 22

SOLID AMINE BED CAPACITY AS A FUNCTION OFCO2 PARTIAL PRESSURE AND CYCLE TIME

39

NALT^N0-Mns

SVHSER 7226

During some of the initial SAWD test cycles, steam desorption wasperformed with the bed at a partial vacuum, so that less steamwould be required to desorb the bed. Since the bed is heated tothe saturated steam temperature, the partial vacuum reduces thetemperature to which the bed is heated. Figures 23 and 24 showthe adsorption breakthrough and integrated CO performance curvesafter desorpt,ons at various temperatures. Ii is clear that re-duced desorption pressures/temperatures cause higher residual COloadings on the solid amine. Consequently, adsorption performanceis degraded. For desorption at 62.05 kPa (9.0 psia) and 87.22°C(189 0 F), the cyclic CO capacity of the solid amine is reduced by7.3 percent compared t8 101.35 kPa (14.7 Asia) and 100 0C (212°F)desorption. However, even with the increased bed size to accountfor this reduced performance, the amount of steam required fordesorption is less than that required with 100°C (212°F) desorp-tion.

Bed Drying During Cyclic Operation

Bed drying during the adsorption phase of cyclic operation occursin two phases. The solid amine bed is an effective heat exchanger.Therefore, during the first few minutes of adsorption, air leavesthe bed saturated with water at the average bed temperature. Thisresults in rapid drying and cooling of the bed. The heat of vap-orization for the water is drawn from the solid amine material andthe bed structure. When the bed temperature approaches the temper-ature of the inlet air, the second phase of drying begins. Waterevaporation rate from the bed is nearly constant and depends on therelative humidity of the inlet air. Sensible heat from the inletair and heat released by adsorption provide the heat of vapori-zation for the water. The

Co,bed temperature becomes nearly constant.

slightly above the wet-bulb temperature of the inlet air. Figure20 shows temperature profiles at various locations in the bed duringa typical adsorption cycle. Figure 25 shows the weight --Flange ofthe bed during an adsorption cycle. The two phases of drying duringcyclic SAWD system operation are shown by the bed weight curve with-out adsorbed CO weight in Figure 25. It should be noted that ifadsorption air blow is continued indefinitely, the bed reaches anequilibrium water content, which is significantly less than thecyclic equilibrium water content with a fixed adsorption cycle.

CO 2 Desorption From Solid Amine

Solid amine is regenerated by heating it with steam. As the steamis injected into the bed, it condenses releasing its latent heat tothe solid amine material and the supporting structure. In this way,;.he heat progresses through the bed in a well-defined wave. The COis desorbed from the heated portion of the bed and readsorbed in thicool downstream.portion of the bed. This readsorption can occur,because the CO partial pressure is much greater than it had beenduring the adsorption cycle. The solid amine has a significantlygreater capacity for CO 2 , when the CO 2 is introduced at a higher

40

STANaARp l,,mum^i...^

SVHSER 72261

.5

.4

.3

NUa

.2

.1

C

INLET CO

70.56°C(1590 F) om^-

CC75.56°C(168°F)86.67°C(188°F)

100.0°C(212°F)

I ^^ooC^ 10 : G 3 C 4(, 5(. i

LEV

12

TIP-r. (NINUTE.S )

Figure 23

EFFECT OF DESORB TEMPERATUREON ADSORPTION BREAKTHROUGH

41

HAMILTON sTAN0^AR0 ^"^^•

SVIISER 7226

.3.(.71

.27(.6)

_ .22

y (•5)a

U ,18( . 4)

ccoMx .14cr,c

(.3)

N.09

(.2)

.05(.1)

0

SAWD!cEOWREMENT

DESnPTEMPERA

M o

(°F

870C'(189076 0C'

(1,69°F0^710C

(160°F

SAWDCYCIaE

oe

3 10 20 30 40 50 6C

i

BTIRE

)

TIME - MIN

Figure 24

THE EFFECT OF DESORPTION TEMPERATUREON ADSORPTION PERFORMANCE

42

HxN

Ho

Wu

3FO

E+E

''H

'H

Cc+

O W

'Ln

as

^ a

/N^' u

1• N

H yN

aC

,

U -0M

dP

of

I

sr •

.w

—ao

—o

,-.N

—v4 .-.

^iN

cr

cc•

koL

r) co

00oo tic

as tr L

^r

er

v o

0-410

sr o

1-r o

HH

(WB

U)

DX

- IHD

I3M

US

SS

INV

OIS

SZ

` VIO

I

0r0w0Lo0N0

er a

ZH1

wHC E

M

HAM

ILTON

s'TANO

MNp

"^ °"^"^.

SVHSER 7226

Figure 25

SAWD CANISTER WEIGHT CHANGE

DURING ADSORPTION

43

11At STAp a^.y,dSVHSER 7226

partial. pressure. This process continues until the remaining un -heated section of the bed is fully loaded with COWhen the steamprogresses into this portion of the bed, the desoibing CO cannotbe readsorbed and is pushed out of the bed. Figure 26 shows twoCO desorption profiles from earlier testing. These desorptionpr8files are from the same solid amine test bed, and they show theeffect of different CO loadings. For the high CO loading case,the end of the bed bec8mes saturated sooner, and t9erefore, CO ispushed from the bed sooner. However, the maximum CO flow rats isindependent of CO loading and depends on steaming rite. The smallflow during the first part of the desorption is ullage air beingdisplaced by steam from the inlet header and from between the solidamine beads. Figure 27 shows that the rate of bed weight changeduring desorption is equal to the steaming rate minus the rate atwhich CO2 is driven from the bed.

The steam requirement for desorption is a function of the dryweight of the bed, the Co loading, and the water content of thebed. For a given system the dry weight of the bed is fixed.Therefcre, the steam requirement only varies with CO loading andwater content. For a typical test cycle with a bed titer contentof 24 percent by weight, there is 2.29 kg (5.04 lbm) of water and0.254 kg (0.56 lbm) of CO on the bed. With these values theenergy necessary to heat i-he water from 21.11°C (70 0 F) to 1000C(21.2°F) is twice the energy required to desorb the COThus, themost significant factor in the variation of the desorption steamrequirement is water content on the bed. Figure 28 verifies thisconclusion by showing the strong correlation between the mass ofsteam required for desorption and the initial bed water content.Similar data is plotted in Figure 29. However, the mass of steamrequired has been converted to a desorption time using a fixedsteaming rate. Figure 30 is a more useful plot of this informa-tion. Using Figure 30, the percent of moisture on the bed can bedetermine] from desorption time with a fixed steaming rate.

Further insight into the desorption process was gained by observingthe axial temperature distribution of the bed during desorption forlow and high CO loading cases. For the low CO loading conditionFigure 31 shows 2 the expected rapid rise in temperature as the steamreaches each location in the bed. This data shows that the steamprogresses through the bed in a well-defined wave. With high CO2loading, Figure 32 shows that the temperature rise at each bedlocation, except near the bed inlet, begins sooner than it didwith the low CO loading case. The earlier temperature riseoccurs because Reat of adsorption of CO is released as the COfrom the upstream portion of the bed is 2readsorbed. As steam 2reaches each location in the bed, the temperature rise continues.However, at approximately 49°C (120°F) the rate of increase isreduced significantly as most of the heat added provides heat ofdesorption for CO 2 . Once most of the CO has been desorbed fromone location, the temperature rise continues as in the low CO2loading case. This effect is more pronounced near the end of thebed, because the CO desorbed closer to the front of the bed hasbeen readsorbed and 2concentrated near the end.

44

c^u --

vac a

m ^o

^c c^ri

•p

o ^-

NU ^Ix

14 CC

^a

Ur Lr

N.p

rl tf1ri V

O M

O C

VO

rlO •

O

•O

•p

•p

(wao)

Niw

lcw

- Mo

ga S

HO

SS

U

U OtnNON ZHiW

Ln ^-a

F--q E-

0OO

MAM

^TOM

:TAIMM

a"""d

SVHSER 7226

Fig

ure

26

EF

FE

CT

OF

CO

2 LO

AD

ING

ON

DESORPTION FLOW & TIME

45

2.(5

I.(4

a (3

c^xHx

0.9

3 (2

0.4(1

HAMILTON s'TAI^AIIp ^°"'^° ^ SVIISER 7226

x

TIME - MIN

Figure 27DESORPTION MASS CHARACTERISTIC WITH HIGH CO 2 LOADING

46

OwwAmd SVHSER 7226

1.361(3.0)

waNn

x

H0.901

a (2.0)W4

w0

a0x 0.454

(1.0)

!0

0

0O0 0 0

O0

Q DATA POINTRUNS 5-25

FROM- 5-122

L'^- 1.361 1.914 2.260 2.721 1.175 3.629 4.082

(3.0) (4.0) (5.0) (6.0) (7.0) (010) (9.0)

KG (LBM) OF WATER, ON 13F.R) P.: START OP CESORB

Fiqure 28

DESORPTION STEAM REQUIREMENTS AS A FUNCTIONOF BED MOIS'T'URE LEVEL

47

:a

5

i01 zl%or,%

DESORB TIME = 2.1207 [LBSNH2O + 29.7(MIN)N BEDBASED ON FIXED STEAMINGOF 4.5 LBM/HR

lZIRATE

4

4

4

4,

zN

4:w

NH 4;z0NE+ 4aa0mA 4(

39

3E

37

HAMILTON ODPA

ind s SVHSER 7226

,.* /-' 9 -

i.81 2.27 2.72 1.18 3.63 4.08(4) (5) (6) (7) (8) (9)KG (LBM) OF WATER ON BED AT START OF DESORB

Figure 29

DESORPTION TIME VS.BED WATER LOADING

48

inow

Z co

aw

0

5-4

U) La

tn

CO

rn "crIn C)

E-4

0VE-4

E-4

qvqw

14

ir

0II

0-4

rr

A

Q

Ln

C)Ln

1^

rn

04

N

r-i

H-31vm JmoutAd

Piq

ure

30P

ER

CE

NT

MO

IST

UR

E V

ER

SE

S D

ES

OR

PT

ION

TIM

E

IIAi^1^ON

iTA1^A1lp "'""^.

SV

IISE

R 7

226

49

A*, tIpwovid

%Fvmffwws*

SVIISER 7226

121,1(250)

10.0(50)

0

21.6 CM

10.2 CM

2.5 ol 17.8 CMOUTLET

FROMINLET

RED = 14.1 KG (31 LSM) DRYLENGTH = 22.9 CM (9")LOADING = 0.032 KG (0.07 LRM) CO 2

93.3(200)

ew0

u 65.60 (150)

w

37.8a4

(100)r^

col

E-4

0 10 20 30 40

50

TIME - MIN

Figure 31

DESORPTION THERMAL CHARACTERISTICWITH LOW CO 2 LOADING

50

10 20 30 40 50 60

TIME - MIN

121.](250)

00

2.5 CM

FROMINLET

10.2 CM 17.8 CM 21.6 CM

UTLET

AED = 14.1 KG (31 tRM) DRYLENGTH = 22.9 CM (911)LOADING = 0.50 KG (1.1 LBM) CO2

93.3(200)

w0

U 65.60 (150)

xH

376E(100)

wH

10.0

(50)

HAMILTON STANOiARp ^° °r"°"°'

SVHSER 7226

Figure 32DESORPTION THERMAL CHARACTERISTV' WITH HIGH CO 2 LOADING

51

IA