EPA-650/2-74-006



December 1973
Environmental Protection Technology Series


-------
                               EPA-650/2-74-006
  DEVICE FOR COLLECTION
               AND
ASSAY OF  AMBIENT GASES
                 by

              Peter Tsang
        Bendix Research Laboratories
             Bendix Center
         Southfield, Michigan 48076
          Contract No. 68-02-0657

         Project Element No. 1AA010



    EPA Project Officer:  Dr. Eugene Sawicki

      Chemistry and Physics Laboratory
    National Environmental Research Center
  Research Triangle Park, North Carolina 27711



              Prepared for

   OFFICE OF RESEARCH AND DEVELOPMENT
  U.S. ENVIRONMENTAL PROTECTION AGENCY
         WASHINGTON, D.C. 20460

             December 1973

-------
This report has been reviewed by the Environmental Protection Agency and




approved for publication.  Approval does not signify that the contents




necessarily reflect the views and policies of the Agency, nor does




mention of trade names or commercial products constitute endorsement



or recommendation for use.
                                11

-------
                            TABLE OF CONTENTS
SECTION 1 - INTRODUCTION                                             1

SECTION 2 - COLLECTOR SYSTEM DESIGN                                  5

2.1   Collector System                                               5
2.2   Collector Unit                                                 5
2.3   Collector Cartridge                                           10

      2.3.1   Cartridge Geometry                                    10
      2.3.2   Cartridge Charging                                    H

2.4   Preparation of Absorbent Packings                             11

      2.A.I   Treatment of Celite                                   11
      2.4.2   Coating of TEA-HC1                                    11
      2.4.3   Cobalt Oxide (Co203)                                  12
2.5   Vacuum Pump (Cast Model 1531)                                 12
2.6   Power Supply                                                  15
2.7   Timer                                                         15
2.8   Operation/Maintenance                                         15
2.9   Interface Device                                              16
2.10  Parts List                                                    16

SECTION 3 - DEVELOPMENT OF COLLECTION SYSTEM                        21

3.1   Introduction                                                  21
3.2   Identification and Characterization of Solid Chemical
      Absorbents                                                    21
      3.2.1   Technical Approach                                    21
      3.2.2   Evaluation Parameters                                 21
      3.2.3   Test Setup                                            22
      3.2.4   Evaluation Method                                     24

3.3   Evaluation of Candidates                                      28
      3.3.1   Triethanolamine and Supports                          28
      3.3.2   Thermal Instability of TEA Packings                   33
      3.3.3   Amino-Alcohol Screening                               33
      3.3.4   Triethanolamine Hydrochloride (TEA-HC1)               34
      3.3.5   Further Work with Celite                              39
      3.3.6   Recovery by Thermal Method (TEA.HCl/Celite)           39
      3.3.7   The N02 Collector:  TEA-HCl/Celite                    44
      3.3.8   NO Absorbent                                          48
      3.3.9   Calibration of Peak Areas                             57
3.4   Interference Studies                                          59

-------
3.5   Development of Collector System

      3.5.1   First-Generation Collector Cartridge and
              Field Collection System                               64
      3.5.2   Development of Collector Unit                         69
3.6   Sampling Pump                                                 75
3.7   Collection System  (Air Sampler)                               76
3.8   Interface Device                                              76
3.9   Preliminary Field Collection and Assay of N02                 77

      3.9.1   Experimental Procedure                                77
      3.9.2   Discussion                                            78

SECTION 4 - SUMMARY AND RECOMMENDATIONS                             81

4.1   Summary                                                       81
4.2   Recommendations                                               82
      4.2.1   N02 Collector with TEA-HC1                            82
      4.2.2   Cobalt Oxide                                          83
      4.2.3   TEA-HC1 Packing                                       84
      4.2.4   Cobalt Oxide Packing                                  85
      4.2.5   Collector Configuration                               86

APPENDIX A - TYPICAL RECORDINGS OF COLLECTOR MATERIALS
             EVALUATION                                             A-l

APPENDIX B - DIRECTIONS FOR ASSEMBLING COLLECTOR FIELD
             HOUSING                         .                       B-l
ii

-------
                         LIST OF ILLUSTRATIONS
Figure No.                       Title                             Page

     1        Collector System                                       4
     2        Collector System Housing                               7
     3        Collector Unit Parts                                   8
     4        Specifics of Collector Unit                            9
     5        Pyrex Glass Connector Cartridge                       10
     6        Gas Pump and Collector Unit                           13
     7        Connection for Collector Unit/Filter to Pump/
                Orifice                                             14
     8        Details of Interface Device                           17
     9        Interface Device and Proportional Temperature
                Controller                                          18
    10        Schematic of Test Setup for the Evaluation of
                Absorbents                                          23
    11        TEA/Chromosorb AW DMCS                                26
    12        Celite (AW)                                           28
    13        Pure TEA/Chromosorb G DMCS                            30
    14        TEA'HCl/Chromosorb G AW DMCS                          37
    15        Celite, Specially Treated (Acid-Washed, Neutralized
                with Distilled H20, Dried and Fired at 1200°C
                for 48 hr)                                          40
    16        Blank, TEA-HCl/Glass Wool (By Heat)                   43
    17        TEA.HCl/Celite                                        45
    18        Performance of N02 Collector TEA.HCl/Celite
                (S02-Treated, 1000 ppm S02/Air, 1 £/min for
                20 min)                                             47
    19        Total NOx Collector (Aluminum Cartridge),
                Co203/Chromosorb G                                  51
    20        NOx Collector, Co203/Chromosorb G 80/100 Mesh         52
    21        NOx Collector (Aluminum Cartridge, Co203/Chromo-
                sorb G 45/60 Mesh), Carrier Air Contains 92%
                Humidity                                            54
    2.2        Estimation of Quantity of Sample Injected with
                Calibration Sample Moisture                         58
    23        Effect of H2S (20.2 ppm) on TEA-HCl/Celite,
                Carrier Air Contains 96% Humidity                   60
    24        Effect of S02 (1000 ppm/air) on TEA-HCl/Celite,
                Carrier Air Contains 90% Humidity                   62
    25        Recovery from an N02 Collector, TEA'HCl/Celite
                (S02-Treated), Which was Injected with
                20 cc NOx (see Fi§- 1& f°r area count),
                Stored in a Lab Drawer for 48 hr                    64
    26        First-Generation Collector Cartridge                  66
    27        First-Generation Dual Path Collector/Pump
                Assembly                                            68
                                                                    iii

-------
 Figure Ho.                       Title

     28        Second-Generation Aluminum Cartridge (00263)           70
     29        Second-Generation Pyrex N02 Collector Cartridge
                 Shown with Aluminum Cartridge and Collector
                 Unit Subassembly                                    71
     30        Collector Unit Assembly with Glass H02 Collector
                 and Aluminum Collector Cartridges                   72
     31        Details of the Collector Unit Tube                    73
     32        Luer-Coupler (for Interfacing the Second-Generation
                 Co203 Collector Cartridge to the Analyzer)           75
    A-l        General Material Evaluation, Showing the Inertness
                 of (1) Pyrex Tube,  (2) Silane-Treated Glass Wool,
                 (3) Chromosorb T,  and (4) Chromosorb W AW DMCS     A-3
    A-2        Chromosorb W AW DMCS                                 A-4
    A-3        TEA/Chromosorb W AW DMCS                             A-5
    A-4        TEA/Regis GasPak FS                                  A-6
    A-5        TEA/Chromosorb T                                     A-7
    A-6        TEA/NEAT (Liquid)                                    A-8
    A-7        TEA Borate                                           A-9
    A-8        THEED                                                A-10
    A-9        Quadrol/Chromosorb W AW DMCS                         A-ll
    A-10       NiS04                                                A-12
    A-ll       CoS04                                                A-13
    A-12       CoO                                                  A-14
    A-13       CuO                                                  A-15
    A-14       Aluminum Cartridge (After  Heating Cycle to
                 200°C and Cooled Down to Room Temperature)          A-16
    A-15       Teflon-Coated Aluminum  Surface                       A-17
    B-l        Folded View of  Collector System Housing              B-2
    B-2        Assembled Collector Unit                             B-2
Table No.
    1         Parts List for the Collection System                  19
    2         Candidate Absorbents for NOX Collection               20
    3         Materials Evaluated                                   35
    4         Summary of Experimental Results                       79
    5         Comparison of Ambient N02~Concentration by
                Technicon and Bendix Monitors (Date:  12-8-73)      80
iv

-------
                              SECTION 1
                             INTRODUCTION

      Most current methods for collection and assay of ambient pollutant
gases involve liquid media in bubblers or impingers.  These techniques
suffer from generally poor collection efficiency, limited sampling capa-
city, tedious or cumbersome operation, and special handling requirements.
They are also troublesome for storage and transportation of reagents,
equipment, and collected samples.  These drawbacks have provided the im-
petus for developing better, simpler collection devices.  Ambient gas
collectors based on solid absorbents offer the most attraction for de-
velopment, since they would possess the essential physical characteristics
to provide simplicity, ruggedness, large capacity for air sampling, sta-
bility, and wide selection of analytical methods.
      Under EPA Contract No. 68-02-0657, Bendix Research Laboratories
developed such a solid-state collection device and transfer interface
device for nitrogen oxides.  These devices have been evaluated under
laboratory conditions and are believed to be highly promising.  However,
more rigorous field testing is necessary to validate their merits in
actual conditions.  Presently only very limited field evaluation has
been conducted.  The scope and main objectives of this development pro-
gram were as follows:
      (1)  Design, fabricate, and evaluate small, simple inexpensive
           collection devices with solid chemical absorbents that

-------
     quantitatively  collect atmospheric gases under  realistic



     field  conditions.  The device should require a  minimum of



     non-technical training for proper use and should survive



     mail shipment without damage or loss of sample.  Devices for



     collection and  transportation of nitrogen dioxide  (N0_) and



     nitric oxide (NO) have first priority, with development of a



     prototype device for these gases the primary goal.  The phy-



     sical  design of the collection device should be such that it



     can adapt readily to collection of lower priority  gases such



     as sulfur dioxide (SO ), carbon monoxide (CO),  formaldehyde



     (HCHO), and more complex organic compounds.



 (2)  Develop an interface device that quantitatively transfers the



     gas sample, or its derivative, from the collector  to an appro-



     priate analytical instrument.  The interface device should



     operate as simply as possible consistent with quantitative



     sample transfer to specific or multi-pollutant analyzers.



     The analyzers could range from inexpensive manual intruments



     to automated, multicomponent analyzers.



The extensive program effort resulted in the  following:



•  A long list of candidate solid chemical absorbents for NO,



   NO , or both NO and NO  (NO )  collection and their evaluation.
     ^                   ^    X


•  Selection of triethanolamine hydrochloride (TEA'HCl), m.p.  177-



   179, for selective NO  collection and cobalt oxide for NO and



   NO  collection.

-------
      •  A collector unit which is



         •  Capable of selective and quantitative collection of NCL and



            NO




         •  Easy to assemble and disassemble



         •  Lightweight (<350g) and rugged




         •  Suitable for mailing



         •  Adaptable through an interface to a chemiluminescence NO
                                                                    X


            analyzer, a gas chromatograph, or a wet chemical analyzer.



         o  Easily installed in an air sampler, requiring no special




            tools or skills.



         •  Reusable.



      •  An air sampling system capable of continuous, unattended opera-



         tion using 12V DC or 110V AC power.



      •  A weatherproof shelter to house collector unit, sampling system,




         timer, and batteries.



      •  An interface device for recovering pollutant samples from col-



         lection cartridges for subsequent analysis.



      The details of the different phases of the program are discussed



in the following sections.  Section 2 contains the instrument description,



while Section 3 presents experimental techniques and data analysis.  Sum-



mary and recommendations are given in Section A.

-------
Figure 1 - Collector System

-------
                              SECTION 2
                       COLLECTOR SYSTEM DESIGN

2.1   COLLECTOR SYSTEM
      The collector system consists of the following major components:
      »  Collector units
      o  Vacuum sample pump
      •  Power supply
      o  Timer
      o  System housing
The overall configuration of the system is shown in Figures 1 and 2.  The
collector pump and collector unit are housed on the top shelf; the second
shelf contains the timer and power supply and thus is the control shelf;
the bottom shelf holds the battery.  A locked door is provided in the
front for access to the control switches and battery.  The housing is made
from 16-gage aluminum with wooden shelves; it has a slanted roof to drain
rain or snow.  Side holes provide ventilation and ports for a sampling
line.  The entire housing is hinged so that the housing will fold toget-
her into a flat package for easy transporting when the collector system
components are removed.  Instructions for assembly of the housing are
given in Appendix B.
2.2   COLLECTOR UNIT
      The collector unit consists of a cylindrical aluminum outer case
which houses two Pyrex glass cartridges.  These cartridges, individually
protected by thin foam plastic covers, are joined tight with a Beckman

-------
Teflon connector (/M06).  The assembled cartridges are secured into the




aluminum case by inserting the inlet and outlet cartridge ends into Beck-




man bulkhead Teflon fittings (/M27) attached to each threaded end-plate




of the case.  The details of construction are illustrated in Figures 3




and 4.




      For actual assembly, one of the collector cartridges is first in-




serted into the Teflon bulkhead fitting (#427) on the left end-plate




(Figure 3) and is secured by finger-tightening the nut on the fitting.




The second cartridge is joined to the first via the Beckman Teflon con-




nector (#406) with finger-tight sealing.  Next, the bulkhead fitting




(#427) is removed from the right end-plate.  The free end of the second




cartridge is inserted into the removed fitting and secured by tightening




the nut.  This subassembly (cartridges and fittings) with foam plastic




covers is placed into the case, after which the right -end-plate and its




washer are tightened to the case.  Finally, the bulkhead nut and the




screw nut are tightened onto the end of the fitting protruding from the




right end of the case.  This completes the assembly of the collector unit.




As assembled, the unit weighs less than 350 g.




      A Millipore Swinex 13 //I filter assembly made of plastic (catalog




no. SXG S0130S) is attached to each end of the collector unit.  This




filter employs Gelman type-A glass fiber filters cut to size.  The fil-




ter assembly has Luer-lock terminals and is attached to the bulkhead




Teflon connector with a nylon female Luer-lock adapter.  This adapter




is made by cutting in half the Hamilton female/female Luer-lock connector

-------
                     SIDE PANELS
                    OPEN AS SHOWN
                        1/4-20 x 1/2
                                    9 CM HASP
             FRONT DOOR
           OPEN AS SHOWN
t
I
3
r> f
30 <-
u
r-
1

i
DOOR HEIGHT
i
3
i*
«
I
' \
t
1
i :
i t
a
r
' 1
i
3
1
i
' 1


f
U
O>


«
E
«
c
«
i_


13 MM
SHELF
3 PLACES
	 t_.
7"
38 MM
ANGLE ALUM.
SHELF FRAMES


fr
3
ft-
D
»
1


50
•<- ALUM.
4COF
___^--
i
)
T!
1 u
) »
oi
M
I-
k
>
3
3
r

                        PIANO
                        HINGE
                                                       VENT HOLES
                                                        5 MM DIA.
                                                                      FOLDED
Figure  2 - Collector  System Housing

-------
00
                                               Figure  3 - Collector  Unit  Parts

-------
                                                      SIDE PANELS
                                                     OPEN AS SHOWN
                                                         1/4-20 x 1/2
                                                                     9 CM HASP
            I
DOOR HEIGHT
s  §
00  U
   ID
   r-
      en  £
      in  u
         8

(«
                                 13 MM
                                 SHELF
                                3 PLACES
                  38 MM
               ANGLE ALUM.
               SHELF FRAMES
                         62 CM
                                             FRONTDOOR
                                            OPEN AS SHOWN
                                                 50 MM
                                              ALUM. ANGLE
                                               4 CORNERS
                                           T
                                              1
                                                               1/4-20
                                                               PIVOT
                                                  si
                                                              PIANO
                                                              HINGE
                                                                                         VENT HOLES
                                                                                          5 MM DIA.
                                                                                                         FOLDED
                                Figure 2  - Collector System Housing

-------
OC
                                               Figure 3 - Collector  Unit  Parts

-------
               END PLATE
                                                                                       SCREW NUT
                                                                                               1

                                                                                  BULKHEAD NUT
            STRAIGHT TEFLON
            BULKHEAD FITTING
             (BECKMAN #4271
TEA/CELITE IN
 GLASS TUBE
BECKMAN TEFLON
 CONNECTOR 406
                                                                          (a)  Collector Unit
Co2O3/CHROMOSORBG  45/60
     AW DMCS
    PYREX TUBE
FEMALE LUER
   LOCK
                                                      0.396
                                                                    1.905
                                                                                 - 0.432
                                                                        (b)   Optional Filter Adapter
                                                        Figure A  - Specifics of Collector Unit
VO

-------
 (Hamilton catalog no.  86505);  it is inserted into the Teflon connector
 and tightened with the screw nut.  (The adapter can also be made by taper-
 ing the inside of 1.9-cm long Teflon tubing from 0.396 cm ID to 0.432 cm
 as  shown in Figure 4.)  The other end of the collector unit is also at-
 tached to a Millipore  filter through a similar adapter and a Hamilton
 nylon male/male Luer-lock connector (Hamilton catalog no. 86506) which
 inserts into the female terminal of the filter.  This filter is connected
 to  the limiting critical orifice installed to the vacuum pump.  This
 filter assembly, however, needs to be evaluated for its inertness to NO
                                                                        X
 interconversion.
 2.3   COLLECTOR CARTRIDGE
       2.3.1   Cartridge Geometry
               The collector cartridges for the two absorbents are iden-
 tically constructed from Pyrex glass.   Their dimensions are given in
 Figure 5.   The inlets  and outlets are  0.635 cm OD Pyrex tubing which
 readily inserts into the finger-tight  Beckman Teflon connectors. In ana-
 lysis,  Beckman Teflon  connectors are  also  used to connect the cartridge
 to  the analytical instrument.
0.635






	


1.575
d 	 »>
/f- -^
^ •**
>x ^
^ ^
d 	 7 B7 	 to.

	
	

1.575





               Figure 5 - Pyrex Glass Connector Cartridge
 10

-------
      2.3.2   Cartridge Charging




              In charging the cartridge, the packing is first prepared




as directed.  One end of the tube is packed with a silanized glass wool




plug (Applied Science #14502).  The prepared packing is then transferred




into the cartridge with a clean spatula.  The tube is constantly vibrated




by hand while it is being packed.  When the tube is filled, the open end




is plugged with silanized glass wool.  The difference in weights of car-




tridge before and after charging is the weight of packing material.  It




is important that the cartridge be carefully packed to ensure consistency




in performance.  A small difference in packing weight and pressure drop




of charged cartridge is an indication of consistent packing.




2.4   PREPARATION OF ABSORBENT PACKINGS




      2.4.1   Treatment of Celite




              The Celite 15/30 mesh was obtained from Johns-Manville as




a free sample.  It is soaked in concentrated HC1 for over 72 hr, drained,




and washed to litmus paper neutral with distilled water.  The washed Celite




is then dried at 50-80°C, and brought to 250°C for 2 hr prior to firing




at 1200°C for 12-48 hr.  The Celite treated in this fashion is inert to




NO.  It absorbs N0« at room temperature, but releases it at 170-200°C.




The cooled Celite is stored in a clean, capped jar placed in a dessicator.




      2.4.2   Coating of TEA'HCl




              About 20 g TEA-HC1 (Eastman 1916) is dissolved in 80 cc




distilled water.  Then 40 g of treated Celite is added to the solution,




which is stirred and allowed to stand for 30 min.  The solution is evapor-




ated to dryness at 80°C in a hood; during drying, the solution is occa-




sionally shaken.  The dried TEA-HCl/Celite is then stored in an oven set
                                                                     11

-------
at  110-120°C.  Packing prepared in this manner is ready for charging  into



the collector when cool.  Each charged collector contains approximately



6g  of packing.  Before use, the charged collector needs to be  conditioned



at  160-170°C with either N0 or air flow at 100-120 cc/min until no NO
                          L                                          X


evolution is observed from the collector.  The collector can then be



cooled in a clean atmosphere, capped, and stored for future use.  CAUTION:



the conditioning temperature should not at any time exceed 170°C; beyond



this temperature, decomposition may occur, thereby impairing the perfor-



mance of the collector.



      2.4.3   Cobalt Oxide
              The cobaltic oxide  (J. T. Baker 1688) is a fine grayish-



black powder  (^200 mesh).  It is  first packed into a large Pyrex tube,



and heated at 500-550°C for 48 hr under nitrogen flow at 20-100 cc/min.



When cooled,  the powder is physically mixed well with Chromosorb G AW



DMCS 45/60 mesh (Applied Science Lab) in approximately 1:2 by volume.



This mixture  is then used to charge the NO  absorbent cartridge.  The
                                          A


charging procedure is the same as that for the NO  absorbent cartridge.



Prior to collection, the prepared cartridge must be conditioned at 420-



450°C under air flow at 20-120 cc/min either overnight or until it no



longer releases NO  (as indicated by an NO  analyzer).  After cooling
                  A                       A


in a clean environment, the collector cartridge is ready for use, or



should be capped for storage.



2.5   VACUUM PUMP (Cast Model 1531)



      This pump (obtained from Cast Manufacturing Corp., Benton Harbor,



MI  49022) is a 12V DC-operated Rotary Vane Vacuum/Compressor pump.  It
 12

-------
Figure 6 - Gas Pump and Collector Unit

-------
     has a free air flow of  42  £/min.   The pump draws 11.3 A  current and

     can be powered by a 12  V auto battery or by a 12 amp-rate battery charger

     (110V AC).  Figure 6 shows  the pump and collector unit assembly.

           The collector system  flow rate is determined by the pressure drop

     exerted by the collector unit and filters.  In order to  eliminate the

     flow rate variations caused by uncertainty in the uniformity  of each

     packing, a critical limiting orifice is placed between the pump and the

     collector outlet filter.  The orifice provided with the  unit  gives a

     flow rate of 4.86 &/min.  Figure  7 shows the quick-connection of  the col-

     lector, pump, and orifice.
                                          PIPE TOSWAGELOK
                                              ADAPTOR
         BECKMAN TEFLON
          CONNECTOR (406)
                                                                   \
                                                                  SWAGELOK THREAD
                                                                      TO PUMP
                                 SWAGELOK
                                   FITTING
      BENDIX CRITICAL
     ORIFICE (BENDIXESD)
      CATALOG NO. 7011
   FEMALE
  LUER-LOCK
  TO ACCEPT
CASSETTE/FILTER
  TERMINAL
PIPE THREAD
                           SPECIALLY DRILLED ORIFICE                     P.87.8(

        Figure 7 - Connection for  Collector Unit/Filter to Pump/Orifice
    14

-------
2.6   POWER SUPPLY




      Either 110V AC or a 12V DC battery can be used as the power source.




With 110V AC the collector unit is powered by the output of a 12V, 12A,




battery charger (Model 126, R. N. Industries, Dearborn, MI  48126), which




is in turn supplied by the 110V AC power line. A conventional 12V car bat-




tery provides a maximum of 12 hr continuous operation; a heavy duty 280 A




truck battery can extend the operation time over 24 hr0




2.7   TIMER




      The timer supplied with the collection system is a 110V AC-operated




Intermatic Model T103 (International Register Company, 2624 W. Washington




Blvd.,  Chicago 12,  Illinois).  This timer enables the collection unit to




sample  the air for  a preset collection time interval of 1-24 hr when




operated in the AC-mode.




2.8   OPERATION/MAINTENANCE




      In actual operation, the operator carries the collector system to




the chosen site and connects the collection unit to the pump orifice by




finger-tightening the Beckman Teflon connector.   (No special skills or




tools are necessary.)  Then the operator merely turns on the switch (either




the battery or the  timer).  At the end of collection, the operator comes




back to the sampler, turns off the switch,  removes  the collector unit,




and caps the ends.   (Either the Hamilton male Luer  plug or a piece of




0.63 OD solid Teflon rod can serve as a cap for the collector unit.   The




plug goes readily to the filter adapter,  while the  Teflon rod can cap




the unit terminals  by replacing the filter  adapter.)  He can then carry




the unit to the laboratory or send it out by mail for analysis of the




collected oxides of nitrogen at a central laboratory.
                                                                     15

-------
       Practically no maintenance of the collector system is required,




  other than changing the filter paper periodically.   The critical  orifice




  may be air-jet cleaned to remove any trapped particulates  if  it is  ac-




  cidentally contaminated.   The pump requires  no  lubrication.




  2.9  INTERFACE DEVICE




       The interface device accepts one collection cartridge and is capable




  of  providing sufficient heat  to bring the  collector  to  the temperature




  necessary to release absorbed nitrogen oxides for analysis.   The  device




  itself is a Kaowool-insulated aluminum block with holes drilled for the




  collector cartridge, a 250 W  Hotwatt cartridge  heater,  a 0-500°C  platinum




  temperature sensor (RFL 24928,  100 H),  and a thermocouple.  The specif-




  ications of the  interface are given in  Figure 8.  The sensor  and  heater




  are operated by  an RFL Model  71 Proportional Temperature Controller (RFL




  Industries,  Inc.,  Boonton, N.  J.  07005), which  is powered  by  a 110V AC




  line.   Figure 9  is a photograph of  the  interface  device and the tempera-




  ture controller.   A 16-cm long  rod  is provided  on the device  for  conven-




  ience  in setting  it up on an  iron  stand.




       In  operation,  the interface device is set  at the desired temperature




  (160-170°C  for the N02 collector;  402°C for  the NO collector); then the




  proper  collector  cartridge is  inserted  for analysis.  The  collector car-




  tridge,  of  course,  should be  connected  to  an  appropriate analyzer prior  to




  its  insertion into  the interface device.




  2.10  PARTS  LIST




       A  parts list  for  the collection system  is given in Table 1.
16

-------
              19.304 MM REAM THRU
           SILVER
           SOLDER
   vv
   -16 CM-
     3.17 MM
    50 MM DEEP
(FOR THERMOCOUPLE)

      .9.525 MM
      DRILL THRU
 (FOR CARTRIDGE HEATER)
ALUMINUM
                                          4.76 MM
                                         50 MM DEEP
                                       (FOR PT-SENSOR)
                                                        KAOWOOL INSULATION
                    i
                                                                                   BRASS
                                                                                   TUBE
                                                                                  /
|     flit»-
T	
I
1	
                                                                      4WV
                                             •T
                                              I
                                              I

                                                       i.
                                                       L
            L
                           -9CM-
                                                   u
                                                   i-.
                                                   IV."
                           Figure 8 - Details of Interface Device

-------
00
                             Figure 9 - Interface Device and Proportional Temperature Controller

-------
               Table 1 - Parts List for the Collection System
Item         Description, Model No.
  1    Collection System Housing Prototype

  2    Collector Unit (Aluminum case)
  3    Collector Cartridges (Pyrex tubing)
  4    Plastic Filter, Millipore Cat.  No.
        SXG S0130S, Swinex 13,  No. 1 size
  5    Beckman Bulkhead Teflon-Connectors
        Cat.  No. 427
  6    Beckman Teflon Connectors Cat.
        No. 406
  7    Hamilton female/female Luer-lock
        connector No. 86505
  8    Hamilton male/male Luer-lock
        connector No. 86511
  9    Teflon adaptor for filter with
        Luer-lock terminals
 10    Bendix Critical Limiting Orifice
        (for  flow) Cat. No. 7011
 11    Cast Vacuum Pump, Model  1531,
         12V  DC Motor
 12    Timer, Intermatic, Model T 103

 13    R.  H.  Battery Charger
        Model 126
 14    Proportional Temperature
        Controller and Platinum Sensor,
        RFL Model 71
 15    Cartridge Heater, Hotwatt, 250W

 16    Interface Device
         Manufacturer
Bendix Research Laboratories
Southfield, MI  48076
Bendix Research Laboratories
Bendix Research Laboratories
Millipore Filter Corp.
Bedford, MA
Beckman, Inc.
Southfield, MI  43075
Beckman, Inc.

Hamilton, Box 7500
Reno, Nevada
Hamilton

Bendix Research Laboratories

Bendix Environmental Science
Division, Baltimore, MD  21204
Cast Manufacturing Corp.
Benton Harbor, MI  49022
International Register Co.
Chicago 12, Illinois
R. N. Industries
Dearborn, MI
RFL Industries, Inc.
Boonton, NJ  07005

Bendix Process Instruments Div.
Ronceverte, W. Va.
Bendix Research Laboratories
                                                                        19

-------
N>
O
                                                     Table  2  -  Candidate Absorbents  for NO   Collection
                    GC phases/sorbent
Inorganic Bases
             Alcohols, amines, amlno-alcohols
       Inorganic Salts
                 DuPonc FCX-330

                 Halocarbon 11-14

                 Halocarbon 13-21

                 Fluorolube MD-10

                 Aroclor 1232

                 Chlorinated Biphenyl-32

                 Silicone Oil  GE,  SF96

                 Armeen SD

                 Hi-EFF-lOB

                 Molecular Sieve
                  5A,  13A, 13X

                 Activated Carbon

                 Alumina
NaO-CaO

NaOH
LiOH-H20
Triethanolamine

Triethanolamine hydrochloride

2-(2-butoxyethoxy)ethanol

1,1',l"-nitrilo-tri-2-propanol

Triethyl  N-tricarboxylate

Triethanolamine Borate

N ,N'-(2-hydroxyethyl)-Piperazine

N,N-Bis(2-hydroxyethyl)-o-toluene-sulfonamide

2,2-Bis(hydroxymethyl)-2,2',2"-nitrilotriethanol(Bis-tris)



2,2',2",  2"'-(ethylenedinitrilotetraethanol) (THEED)

2,2',2",  2"'-(ethylenedinitrilotetra-2-propanol) (Quadrol)

Polyoxyethylene(8) Ethylenediamine

Tr is-(hydroxymethyl)aminomethane

Tr is-(hydroxye thy Daminome thane

1,1,1-Tris-(hydroxymethy1)ethane

6-amino-l-hexanol

1.5-pentanediol
NaClO. /Alumina

(Cr03+H3P04)/Quartz

(Na2Cr20? + HjSO^) /Glass Fiber
                                                                                   CuO
                                                                                   CoO
                                                                                   C°2°3

                                                                                   Fe2°3
                                                                                   FeSO
                                                                                                                                PbO

                                                                                                                                Pb02

                                                                                                                                CuSO,

-------
                              SECTION 3


                   DEVELOPMENT OF COLLECTION SYSTEM





3.1   INTRODUCTION



      The total development program included five major efforts:  (1)



identification and characterization of solid chemical absorbents and



supports, (2) development of collector cartridge and collector unit, (3)



fabrication of interface device, (4) survey and selection of air sampling



pump, and (5) assembly of collection system.  Each of these efforts is



discussed in the following subsections.



3.2   IDENTIFICATION AND CHARACTERIZATION OF SOLID CHEMICAL ABSORBENTS



      3.2.1   Technical Approach



              The effort of this phase centered around the rapid screening



of candidate materials to be used as absorbents for the oxides of nitrogen.



Those absorbents could be either of two forms:  liquids that could be



conveniently coated onto an inert solid support, or solids that could



be either used by themselves or physically mixed with some suitable inert



support.  Candidates were chosen from literature references or from ex-



amination of chemical structures.  They included:  (a) inorganic bases,



(b) gas chromatography stationary phases or solid sorbents, (c) organic



bases, and (d) inorganic salts.  Table 2 summarizes the candidates chosen



for evaluation.



      3.2.2   Evaluation Parameters



              Candidate absorbents were screened on the basis of



      »  Ability to completely absorb either NO, N0_, or both (NO )
                                                   £             X
                                                                      21

-------
      •  Inertness to NO ,  so that the absorbent surface does not
                        x*


         catalyze NO+NO  interconversion.



If an absorbent met these requirements, it was evaluated for its ability



to quantitatively release collected NO  upon heating.
                                      X



      3.2.3   Test Setup



              The absorbents were evaluated using the simple test fixture




shown in Figure 10.  This fixture has a one-liter mixing bulb for sample



preparation which accepts the gas flows from the NOX tank and from the



diluent nitrogen tank.  Appropriate flow controllers and needle valves



are provided.  The pressure of the bulb can be read from the pressure



gage connected to it.  The outlet of the bulb goes to a 50 cc buffering



volume through a valve.  This volume is open to a vacuum vent line or



to the sample inlet of the absorbent to be evaluated.



              The sample inlet branches, going  to the NOX analyzer



(Bendix Chemiluminescence NOV Analyzer) directly, or through the collector
                            A


first and then to the analyzer and pump.  A Pyrex cold-trap (ice water)




is placed between the analyzer and the collector.



              A septum port is installed in the line between the mixing



bulb and the buffering volume for withdrawing prepared sample.  Another



septum port placed at the ambient inlet line is for sample injection.



Of course, when the valves before and after the buffering volume are



both open, the prepared  sample is continuously drawn, either directly  or



via the collector path,  into  the NO  analyzer. The nitrogen line and the
                                   A


vacuum vent allow the gas preparation system to be purged.  All the gas
22

-------
N)
OJ
              AMBIENT AIR

                  V
                     SCRUBBER
                    INJECTION
                     SEPTUM
                                             FOR SAMPLE
                                            WITHDRAWAL
                                              SEPTUM
                          NEEDLE
                          VALVE
                                             50 CC BUFFER
                                               VOLUME
                                   VACUUM
                                  VENT LINE
                                                                                  DIRECT PATH (BYPASS)
                                                                              CONNECTION FOR
                                                                              COLLECTOR TO BE
                                                                                EVALUATED
                                                                                                              FLOW
                                                                                                            CONTROL
                                                                                                                  ICE
                                                                                                                 COLD
                                                                                                                 TRAP
                                 Figure  10 - Schematic of Test Setup for  the Evaluation of Absorbents
 TO NOX-
ANALYZER
AND PUMP

-------
lines are Teflon, and connectors are either Beckman Teflon-type or stain-



less steel.  NO  is supplied from a tank of calibrated mixture prepared
               A


by Cryogenic Sales, Inc.  The sample prepared by this system has a con-



centration range of 1 ppm in N0_ and 0.1 ppm in NO.



              For evaluation, absorbent material was packed into a 2.5-



cm OD Pyrex tube with 0.64-cm OD inlet and outlet.  Silanized glass wool



was used for end plugs.  The tube had a ground-glass joint to allow easy



packing.



      3.2.4   Evaluation Method



              The Bendix Chemiluminescence NOV Analyzer gives readouts
                                             A.


of  [NO  ] and [NO] in ppm.  When a steady concentration of the oxide is
      X


present in the sample gas, the analyzer readout is a straight line (con-



tinuous mode).  If the sample concentration has a short lifetime or is



pulsed, a peak readout appears.  The latter occurs when a sample is in-



jected  into the ambient air stream.



              In evaluating a candidate as a potential absorbent, a



steady  concentration was first established in the gas system through



the "direct path".  The readouts were recorded as  [NOX]D and [NOjp.



Then the sample stream was routed to the "collector path" with a can-



didate  absorbent in an inert cartridge.  The readouts were recorded as



[NO..].,  and [NO]  .  The difference between  [NOV] and [NO] represents the
   ALL                             A


concentration of [NO-].  That is,






                         [N02]D -  [N°X]D -  [N°]D
                         [N°2]C =  [N°X]C '  [N°]C

-------
When  [NO ]  = 0, the candidate material  totally absorbs or  removes  the
        A L»


nitrogen oxides.  If [NO  ]  =  [NO]  , the material absorbs or  removes
                        AC       C


N0_.   [NO]  = [NO]  means that the candidate absorbent causes no  change



in the concentration of NO, thus implying the absence of N0?  -»• NO conver-



sion.  More concisely,
               [NOV]., = 0       Total removal of NOV
                 A C                              X





               [NOVK = [NO]    Total removal of N0_
                 A C       C                      ^




               [NO]  = [NO]     No conversion of NO,  to NO.






When these are not the case, the absorbent either removes NO,  incompletely



and/or catalyzes its conversion to NO.  The percentages of NO, removal  and



conversion can be calculated as follows:
                         [N°2]D =  [N°X]D -  [N°]D
                         [N°2]C =  [N°X]C -  [N°]C
                                           '  [N°2]C
                                 A  [N00]      .
                     i        -,        2 removal    ......
                     - removal = 	,   ,	 x 100
       % N02 absorption = (% N02 removal) - (% NO  conversion)





The evaluation of triethanolamine (TEA) coated onto Chromosorb W DMCS,



as shown in Figure 11, serves as a sample calculation.
                                                                      25

-------
              INCOMPLETE I\I02 - REMOVAL

                CONVERTS N02-" NO
                   (NO] c = 0.090
Since,
                   Figure  11  - TEA/Chromosorb AW DMCS
                  [N0]n = 0.025 ppm,  [N0y]_ = 0.50 ppm,
                      JJ                  A U
[N0]c =0.09 ppm,
                                             =0.12
                                                    ppm
26

-------
           [NOJ_ =  [NOV]_ -  [NOl  = 0.50 - 0.025 = 0.475 ppm
             £• U      A JJ       U
                 =  [NOX]C -  [NOJC = 0.12 - 0.09 = 0.03 ppm
Therefore
          A N02 =  [N02]D -  [N02]c = 0.475 - 0.03 = 0.445 ppm
          A NO = [NO]  - [NO]  = 0.090 - 0.025 = 0.065 ppm
                     C       D
Thus
  %  [N02+N0] = (A NO/[N021D) 100 = (0.065/0.475) 100 = 11.6%





  %  [N00] ,       .   = 82.1%
       2 'absorption




For the injection mode of operation,  the same basic calculations are used,



except that the areas of the generated peaks are used.  Each area is ob-



tained by triangulation of a given peak.



              In the quantitative recovery study, the elution of the oxides



of nitrogen from the collector is also in the form of a peak.  For the per-



cent recovery estimate, the peak area of the eluant from the collector is



counted as A        .  The average area of the peaks obtained by injecting



the same sample through the direct path is calculated as A..    .   The
                                                          direct
ratio A        /Adi   t Denotes c^e effectiveness in recovery.  When this



ratio is equal to 1, a 100% recovery is achieved.  The analytical accuracy



of the approach was somewhat limited by the triangulation method;  however,
                                                                      27

-------
it should suffice for the present application.  An electronic integrator




would definitely increase the accuracy of analysis.




3.3   EVALUATION OF CANDIDATES




      3.3.1   Triethanolamine and Supports




              The recent report by Levaggi et al. on the superior per-




formance of triethanolamine  (TEA) on Celite-22 fire brick prompted us




to attempt to reproduce their work.  Unfortunately, we observed persistent




N09 to NO conversion.  Since Celite washed with concentrated nitric acid




exhibited extensive conversion of N02 to NO  (Figure 12), we decided to
                        Figure 12 - Celite (AW)
28

-------
 study  other  support materials:  Corning special textured glass beads;




 Chromosorbs  G and W, both of which are AW-DMCS (acid washed and silane




 treated);  Chromosorb T  (Teflon); and Regis Gas Pak FS  (Teflon-coated




 diatomaceous earth).  Absorbent packing made by coating TEA on these




 supports activated the  conversion of N02 to NO, despite the fact that




 the supports themselves were found to be inert.  These puzzling results




 are illustrated in Figures A-l  through A-5 of Appendix A.




              We speculated that evaporating the methanol solution of TEA




 during the coating of the support might have introduced impurities or




 damaged the inert surface of the support.  Finally, after repeated failure




 to obtain an inert packing on any of these supports (including an attempt




 to coat TEA onto the support without solvent), it occurred to us that TEA




 itself might be impure.  This postulation was later confirmed by the ob-




 servation that TEA itself, without a support, gave a 10% conversion of




N02 to NO  (Figure A-6).




              A pure sample of TEA (clear liquid)  was then obtained and




stored in the dark.   A packing was made by dissolving 5 g of this sample




in methanol and evaporating the solution onto 20  g Chromosorb G AW DMCS




80/100 mesh.   After conditioning at 80°C for 4 hr  with 40 cc/min helium




flow,  this packing was then evaluated in a polypropylene tube.   As illus-




trated in Figure 13(a) and (b), this packing completely and selectively




absorbed N02  and was  inert to NO.




              In this experiment,  the original sample concentration was:






                           [NOX1D = 1.2 ppm
                                                                     29

-------
u>
o
                  INDIRECT x '
                     A - 408
                                                                                                      INOI0- [N0,lc- INOIC
                                                                                                       CONVERSION ABSENT
                                                                                                       10O* NO2 ABSORPTION
                                                                                                                   '"""COLLECTOR" '
                                                                                                                       A-38.5
                                         Figure  13(a)  -  Pure  TEA/Chromosorb G  DMCS -  Injection Mode

-------
                     UJ
                     a.
IN°! DIRECT
   0.145
                           [NOX] DIRECT
                              1.2 PPM
                                               [IMO]C=[NOX]C=[NO]D
                                               ..  100% N02- REMOVAL
                                 ,   5MIN.
                                 I «       >•
                                                 NO,
•NO CONVERSION
 ABSENT
                                                                 [NO]COLLECTOR
                                                                    0.14 PPM
                     Figure 13(b)  - Pure TEA/Chromosorb G DMCS -  Continuous mode

-------
                         [NO]D = 0.14 - 0.145
After passing the collector,





                         [NOY]_ = 0.14 - 0.145
                            A L*




                         [N0]c = 0.14 - 0.145





That is,
                 [NOX]C = [N0]c = [NO]D, or
                 [N0?] = 0 after passing the collector.
These values indicate 100% N0? removal and no NO-^NO conversion.



              Similarly, for the injection mode,




                                    • 109.8;
                             [NOY]
                                 D
                           A[NO]D
l[NOY]
                                   = 39.6;

                                 c
That is,
                               =  [N0]c =  [NO]D
which represents 100% NO- removal and no NO-^NO conversion.
32

-------
       3.3.2    Thermal  Instability of TEA Packings



               The  recovery of  the absorbed N02 from the TEA/Chromosorb G



 collector  by  elution into the  NO  analyzer was attempted using  thermal
                                A


 methods.   Evaporation  of TEA occurred above 80°C, and  the color of  the



 packing changed  to orange above 150°C,  thus indicating possible decom-



 position of the  packing.  An NO  elution peak was obtained, however, at
                               A


 this temperature.   This heat-treated packing subsequently converted NO-



 to NO  and  had  lost  its capability for NO- removal.  A  new collector was



 made by packing  the TEA/Chromosorb G into a stainless  tubing which was



 then subjected to  gas  chromatographic evaluation for bleeding.  Persistent



 bleeding occurred  at and above 75-80°C, despite extensive conditioning



 at this temperature.



              These results clearly indicated that materials having higher



 boiling points or melting points than TEA should be sought to improve



 thermal stability.  Solid candidates were particularly attractive.  Most



 of the candidates shown in the amines,  alcohols,  and amino-alcohol column



 of Table 2 were chosen on the basis that their chemical structure resembled



 that of triethanolamine and their melting points  were higher.   Triethano-



 lamine hydrochloride (m.p.  177-179°C),  and Bis-Tris (m.p.  103-104°C) are



 typical examples of this kind.



      3.3.3   Amino-Alcohol Screening



              Quite extensive but rapid screening of the  available amino-



alcohols resulted in the identification of triethanolamine hydrochloride



 (TEA-HC1)  as a potentially  reusable NO- collector material.  All the
                                                                    33

-------
other amino-alcohols either converted NO- to NO or failed to absorb N02

or NO.  Figure A-7 to A-9 in Appendix A are recordings from tests of some

of the amino-alcohols.  Because of the time constraint, each material

was rapidly screened for its affinity for N02 and/or NO, and its surface

activity toward NO, to NO conversion.  Although not every recording trace

was presentable, the necessary data were recorded for calculation.  The

results of the evaluation for all materials are summarized in Table 3.

      3.3.4   Triethanolamine Hydrochloride (TEA'HCl)

              The TEA'HCl was a fine white powder, fluffy in texture.

It was physically mixed with Chromosorb G AW-DMCS 80/100 mesh in a 1:1

ratio by volume, and then packed into a Pyrex tube for evaluation.  The

performance of TEA'HCl /Chromosorb G AW-DMCS, illustrated in Figure 14 (a)

and  (b), showed 100% N02 removal and the absence of N02 to NO conversion
                 [NOX]C =  [N0]c =  [HO]D = 0.14 - 0.15;
                        [N°]
              The preparation of TEA'HCl/Chromosorb G packing was somewhat

difficult due to the  fluffy nature of TEA'HCl.  Moreover, although TEA-

HC1  is very soluble in water, Chromosorb G, W, T, etc. are all hydrophobic.

Therefore, some other support should be sought. TEA'HCl can be transfor-
                                                    r     n
med  into larger particles.  Cold-pressing at 21 x 10  kg/m  for  20 min

produces a cake that will break into chips.  These chips, however, did

not  have adequate mechanical stability, since  they gradually broke down

into a powder.
34

-------
                                                    Table  3 - Materials  Evaluated  (1 of  2)
                                                         Material
                                                                                         Results and Remarks
                    Support
                                           1.  Celite 22
                                           2.  Corning textured glass bead


                                           3.  Chromosorb G AWDMCS

                                           4.  Chromosorb T (Teflon)

                                           5.  Chromosorb W AWDMCS

                                           6.  Regis Gas Pack FS (Teflon coated
                                                 diatomaceous earth)

                                           7.  Silane treated glass wool
                                       Specially acid  washed  and  fired  sample  absorbs
                                       N02t  is  inert  to  NO at room temp.,  gives  off
                                       absorbed NC>2 above  150°-200CC, chosen for TEA-
                                       HC1 support.
                                       Inert to NO, N02, has  low  surface  area  and capa-
                                       city  for coating  material.

                                       Inert to N02 and-  NO, hydrophobic;  Chrom.  G 45/60
                                       mesh  can be best  mixed with the  Co203 powder to
                                       afford a decent gas flow,  chosen for
                                       support.
                                       Inert to N02,  NO
                    Container
1.  Pyrex glass

2.  Aluminum

3.  Teflon coated aluminum
Inert to NO, N02, chosen for both lab test and
collector cartridge construction.
After going through heating cycle to 200°C,
converts
                                                                                  Inert to N02, NO
                    Absorbent Candidates
                       (Inorganic)
1.  Co203/Chromosorb G


2.  NiS04
3.  CuO (wire)

4.  Cu20/Chromosorb G
5.  CoS04

6.  CoO
Absorbs NOX completely up to 350°C.  Releases
the absorbed quantitatively above 390°C.
Does not absorb N02, converts N02-|'NO
Slightly absorbs N02.  Slightly converts N02-N0
Converts
Converts
                                                                                  100% NOX removal
                                                                                                    Does not absorb N02-
u>

-------
                                                 Table 3 - Materials  Evaluated (2  of 2)
u>
                 Absorbent Candidates
                    (Organic)
                                                      Material
1.  1,5-pentanediol/Celite


2.  THEED/Chromosorb W
3.  Quadrol/Chromosorb W

4.  Polyoxyethylene(8)ethylene
      diamine/Chromosorb T
5.  TEA-Borate
6.  2-(2-butoxyethoxy)ethanol/
      Celite
7.  1,1',l"-nitrilo-tri-2-
      propanol/Celite

8.  Impure TEA coated on
    a.  Chromosorb G
    b.  Chromosorb W
    c.  Chromosorb T
    d.  Regis Gas Pak FS
    e.  Without support

9.  Pure TEA/Chromosorb G
                                        10.  Triethanolamine hydrochloride
                                             (TEA-HC1)

                                        11.  Bis-Tris
                                        12.  N,N'-Bis-(2-hydroxyethyl)-
                                               piperazine

                                        13.  N,N'-Bis-(2-hydroxyethyl)-p-
                                               toluenesulfonamide

                                        14.  Phenyldiethanolamine succinate
                                               (Hi-EFF 10B)
                                        15.  Tris-(hydroxymethyl)-
                                               aminomethane
                                        16.  1,1-tris-(hydroxymethyl)-ethane
                                                                                      Results and Remarks
61% N02 removal, 11%. NO removal
<10% HO-, removal, 50% N02->-NO conversion
Does not remove N02 to any significant extent
Significantly converts N02-»-NO (70%) .

Does not remove N02
80% N02 removal, converts 13% N02-*NO

78% N02 removal, slightly converts  N02-»-NO.

100% N02 removal   Converts N02-"NO(^50%)


4% N02-»-NO conversion
16%
17%
40%
10%
0% conversion,  100% N02 removal    Poor in
thermal stability

0% conversion,  100% N02 removal,  chosen  N02
absorbent
Inert at room  temp.
N02 removal up  to 62%, converts N02->NO slightly


Does not absorb N02

Inert at room  temp.

<10% N02 removal,     2% conversion N02~NO.

<10% N02 removal,     2% conversion N02"NO.

-------
                                      INOIC INO,IC - IMO|Q
                                      .. 10O* NO2 - REMOVAL
                                        THERE IS NO
                                        CONVERSION
                                                                                      -\	h

                                     Figure  14(a)  - TEA-HCl/Chromosorb  G AW  DMCS  - Continuous  Mode
00

-------
         J°-COLllCIOB *
                                  |NOIO INOIC INO.IC
                                   10(7^ NO7 ABSORPTION
                             INO,ICOLL£CIOR x
                               S, • 72 6
             t ***** H
Figure  14(b)  - TEA-HCl/Chromosorb  G AW DMCS -  Injection Mode

-------
      3.3.5   Further Work with Celite



              Since Celite was the only available support that was not



hydrophobia, we decided to re-examine it as a possible support.  The



previous acid treatment used concentrated HNO., and a short soaking time.



In the retest, Celite 22 firebrick (from Johns Mansville) was soaked in



concentrated HC1 for over 72 hr, washed with distilled water until neutral



to litmus paper, dried at 80°C, heated to 250°C for 2 hr, and fired at



1200°C for 48 hr.  Figure 15 shows that Celite treated in this manner



was inert to NO and absorbed N0_ at room temperature up to 45%.  The



absorbed N02 was released at 150-200°C.



      3.3.6   Recovery by Thermal Method (TEA-HCl/Celite)



              Blank



              TEA'HCl was tested to determine whether it would decompose



to give off NOV.  For this test, a tube was packed with TEA'HCl and silane-
              A.


treated glass wool.  Figure 16 indicates that the TEA'HCl did give off



some NOY at a temperature of 160-170°C.  The evolution was most likely
       A


the N0_ absorbed by the TEA-HC1, since another heating cycle up to 200°C



did not cause further evolution.



              TEA-HCl/Celite



              TEA'HCl was weighed and dissolved in distilled water.  Pre-



weighed, acid-treated Celite was added and the solution was allowed to



stand an hour.  The water was then evaporated at 80°C with occasional



stirring; the dried packing was then transferred into the Pyrex cartridge.



Prior to evaluation, this collector cartridge was conditioned at 100-120°C



for 4 hr with 120 cc/min air flow, then heated to 170°C to check for
                                                                     39

-------
CELITE ABSORBS
                                                     CELITE DOES NOT
                                                    CONVERT NO2-NO
                                      INDIRECT     'N°ICELITE
                                         0.31          0-32
      -4-
-4-
•4-
-4-
-4-
                                                 INOI DIRECT
                                                0.30 • 0.31 PPM
                                                                             	1	1	1	

Figure 15(a)  - Celite,  Specially Treated  (Acid-Washed, Neutralized with Distilled H20,
                 Dried  and Fired at  1200°C  for 48 hr):  Inert to  NO, absorbs N02 at
                 Room Temperature
                                                                                     TO DIRECT
                                                                                       PATH

-------
                           0% ABSORPTION
                              AT 200° C
                  [NOX]CEL|TE 200° C
                      1.02-1.03
                          IN°X) DIRECT X 2
                           1.02- 1.03 PPM
       [-•-5MIN.-*-]
                TOCELITE
                AT 200° C
                                                               +*»J
Figure 15(b)  -
Celite,  specially  treated (Acid-washed, neutralized with
distilled  H20, Dried  and Fired at  1200°C for  48  hr):  at
200°C, does not absorb N0?

-------
•C-
fo
                                                                 CELITE CAN BE CLEANED

                                                                 ABSORBED NO2 CAN BE

                                                                 RELEASED BELOW 200 C
                                                                                                               5MIN.
                                                                                                          |NOXI X 1
                    Figure  15(c) - Celite,  Specially Treated  (Acid-Washed, Neutralized with  Distilled

                                    Dried  and  Fired at 1200°C  for 48 hr):  Releases Absorbed
NOX When  Heated

-------
THETEA-HCI

WAS CLEANED

BY HEAT CYCLE
                         O
                         >
                         u
           I     I
              Figure 16 - Blank,  TEA-HCl/Glass  Wool (by Heat)

-------
release of NO  .  Its performance at room temperature is given in Fig-
             A.



ure 17(a).  TEA-HC1 absorbed no NOY at 165-170°C as indicated in Fig-
                                  A



ure 17(b).




       3.3.7    The NO- Collector:  TEA-HCl/Celite
              That TEA'HCl/Celite was a true NO- collector was further



confirmed by the test results given in Figure 18.  This figure shows



clearly that the packing not only removed NO- from the sample air without



disturbing NO at room temperature, but also completely released the NO-



at an elevated temperature.  This experiment utilized the injection mode;



the volume used for each injection was 5 cc.  The collection efficiency



was estimated:
Therefore
                                         = 323

-------
                              o
                                                         o

                                                          I
                                                               tr
                                                               o
                                                               t-
                                                               u
                                                               8
o

 I
Figure 17(a) - TEA-HCl/Celite at room temperature,  inert  to NO,  absorbs

               N02 completely

-------
                                                                Q.

                                                                ID
                                                                O
Figure 17(b) - TEA-HCl/Celite at 165-170°C, N02  completely  passes  through
 46

-------
 |NOIC • INO.]C - INOID
 lOOfc ABSORPTION OF NO2
 (H. CONVERSION
 'in,.'"RECOVERY
 10OXRECOVERY


 NO. X I

AHfCOVERY 12"01 a "N021
     Figure  18 - Performance of  N02 Collector  TEA-HCl/Celite  (SOo-Treated,  1000  ppm SOo/Air
                  1  4/min  for 20  min)    .

-------
Thus,
               A[NQ j ^ 0  and   [N0]c ^  [NO^ ^  [NO]D
              A total of 20 cc was introduced into the collector in 4



sequential injections.  The recovery peak between 120-170°C had a cal-



culated area of 1240 counts, that is, about 4 times the area for each 5



cc injection of NCL.  In other words, TEA'HCl/Celite released its absorbed



N09 completely:
               A         = 1240 = 4 ArMA ,  = 4 x 323
                recovery             [NO-]-
The above collector material went through many temperature cycles before



these measurements [Figure 17(b) for example] and it also underwent an



SCL treatment in which 1000 ppm S09 in air was passed through the collector



for 20 min at 1 £/min flow rate.  Furthermore, the ambient air carrier



used in these experiments contained up to 95% humidity.  Nonetheless, the



material retained its performance as a good N0~ collector.



      3.3.8   NO Absorbent



              An effort was made to identify an NO absorbent or a total



NO  absorbent to complement the N00 absorbent TEA-HCl/Celite.
  X                               2,


              Since it was known that inorganic oxides can interact with



NOY (for example, Pb00 forms a nitrate with nitrogen oxides), the follow-
  A                  Z


ing inorganic  compounds were chosen for further study:  CuO, Cu90, CoSO,,



Co90_, CoO, NiSO,.  The cobaltous and cobaltic oxides proved to be ex-



cellent total NOV absorbents, whereas the other salts evaluated showed
                A
48

-------
negative results.  Figures A-10 to A-13 show the results obtained from



some of these oxides.



              The capability of cobaltic oxide to absorb NOX is clearly



depicted in Figures 19 and 20.  In these experiments the sample NO  was



quantitatively injected into the collector where it was completely ab-



sorbed and later released without loss at 395-420°C.  The packing was



prepared by first heating Co-O. (grayish-black powder) in a Pyrex tube



at 500-550°C with air flow for 48 hr.  When cool, the oxide was physically



mixed with Chromosorb G AW-DMCS 80/100 mesh (or 45/60 mesh) in 1:2 ratio



by volume, then packed into a Pyrex tube or an aluminum collector cart-



ridge.  Prior to evaluation, the collector was conditioned at 420°C over-



night or until it was free from any pre-absorbed NOV.  During the eval-
                                                   A


uation of the collector with an NO  analyzer, the ambient air carrier gas
                                  A.


contained 92% humidity.



              As shown in Figure 20(a), the cobalt oxides completely ab-



sorbed the NO at 250°C.  The area count for each injection was. established



by injecting samples into the collection system with the absorbent at



390-400°C at which temperature no absorption occurred.
                          A[NO]. .
                               injected
                         A1[NO]        „
                               recovered
                         A2[NO]        „ = 104
                               recovered
These values indicate 100% recovery.
                                                                    49

-------
              Figure 20(b) depicts  the  quantitative  recovery of  NO  as
                                                                   A


follows:
     An = Arwn  1         = 1224    (for  each  inJection  at  405-410°C)
      U     INUXJinjected
     A-  (recovery from 1 injection) = 1360 ^ A






     A,,  (recovery from 2 injections) = 2640 ^ 2A






     A   (recovery from 3 injections) = 4468 ^ 3A






     A,  (recovery from 4 injections) = 5472 ^ 4A






and,





               A2/A1 = 1.94, A3/A1 = 3.28, ^/^ = 4.02   .





              Results represented in Figure 21 were obtained on  the  col-



lector Co-O./Chromosorb G AW-DMCS 45/60 contained in an aluminum cartridge.



This figure also demonstrates that the collector absorbed and released



NOV quantitatively as shown below:
  A


      (a)  For each 1 cc injection of NO  sample, the average area
                                        A


           count was A.. = 545 (1 cc through a direct path).



      (b)  Injection of a 1 cc sample into a hot collector at 430°C



           afforded an area count of A_ = 560
50

-------
                                      REPEAT
                                  IN°X) DIRECT = °-70
  2.5 MIN.
         x COLLECTOR =
       .'. 100% ABSORPTION
                                                               'N°x' DIRECT X 1 = 0.65 PPM

                                                                r
                                                "^DIRECT*1
                                                  0.13 PPM
Figure 19  - Total  NOX  Collector  (Aluminum cartridge),  Co.O^/Chromosorb  G

-------
Ul
to
                                                                                                     IN°)COLLECTOR AT
                                                                                                        390 - 400° C

                                                                                                   A = 97 (~ 0% ABSORPTION)
                          Figure 20(a)  - NO  Collector,  Co203/Chromosorb  G 80/100 mesh - Quantitative
                                          Collection and  Recovery of NO

-------
                                                                                        5  i
                    Figure  20(b)  -  NOX  Collector,  C0203/Chromosorb G 80/100 Mesh - Quantitative Collection
                                   and  Recovery  of NOX
01
UJ

-------
                    A3cc= 1627.5
                                               |N°«JDIRECTX6
                                                  A,, = 1008
                               . = 54.5
Figure 21(a)  -  NOX Collector  (Aluminum Cartridge, Co203/Chromosorb G 45/60
                raesh) - Area Estimate of Sample Injected,  Carrier Air
                Contains 92% Humidity
  54

-------
                          Figure  21 (b) - NOX Collector  (Aluminum Cartridge,  Co203/Chromosorb G 45/60
                                        mesh) - Quantitative  Recovery of  2  cc  Sample Injected
Ln

-------
AABSORBED ; 2^560 (B)

ARECOVERY = 102°
 .. - 100% RECOVERY
ABSORPTION = 4 X 560 (See B)
        = 2240
ARELEASED =2569.6
 .  100% RECOVERY
       Figure 21(c)  - NOx Collector  (Aluminum  Cartridge,  Co203/Chromosorb  G 45/60
                        mesh)  - Quantitative Recovery of  2  cc and  4  cc Sample Injected

-------
       (c)  A 2 cc sample injected  into the collector  at ^300°C was



           totally absorbed; when  the collector  temperature was  raised



           to 430°C, a peak was obtained with an area count of
             A, = A          .    _     1100 = 2 A", =  2 A.
              3    recovery  from 2cc             1       2
                            = 2.0; A/A  = 1.96
       (d)  A. = A                    2569.6 •*. 4 x A.
            4    recovery from Ace                 2
           A, = A                  o 1020 = 4 x A..
            5    recovery from 2cc               1
It was therefore clear that an NO  absorbent had been identified.  More-
                                 A.


over, this cobalt oxide could serve as an  NO absorbent with a precolumn



such as TEA or TEA'HCl to first remove N02.  This tandem arrangement was



therefore chosen for the design of the NO  , NO collection unit.



      3.3.9   Calibration of Peak Areas



              A calibrated sample of 1.1 ppm NO (Matheson Gas Products,



Inc.) was used for the area/mass equivalent calculation.  Various volumes



of this sample were injected to establish an average area count.



              Each cubic centimeter of calibrated sample containing 1.4



ng produced an area count of 4, (Figure 22) making each area count equiv-



lent to 0.35 ng NO.  Based on the calculation, each cubic centimeter of



sample prepared by the laboratory test setup contained on the order of



10   g.  This sample size was injected into collectors being evaluated



to test their absorption efficiency.
                                                                    57

-------
Ui
oo
                                      LAB SAMPLE
                                        (X 05)
                                     0 tec
                                                         > 0 35 r»9'AREA COUNT
                                                         > FOR AN INJECTION OF
                                                          LAB SAMPLE. EACH cc
                                                          CONTAINS - 2 X ID-7 CM
                                                                            CALIBRATION SAMPLE 11.1 PPM NO)
                                                                                  I1.4 rt
                                                                                    NT IN AREA
    INO.I x 5
LABORATORY PREPARED
    SAMPLE
 INJECTION VOLUME
   Ice A-634
                          Figure 22  -  Estimation of Quantity of Sample Injected with  Calibration  Sample Mixture

-------
3.4   INTERFERENCE STUDIES



      The effects of moisture, hydrogen sulfide (H2S),  and sulfur dioxide



(S0_) on the performance of the collectors were investigated.  Moisture



alone up to 92% did not affect the performance of the cobalt oxide and



TEA'HCl/Celite.  Likewise the combination of H2S (20.2 ppra in N2 from a



permeation tube) and up to 96% humidity had no synergistic effect on



TEA'HCl/Celite (see Figure 23).  Passing 1000 ppm S02 in air at a flow



rate of 1 «,/min for 20 min through the TEA-HCl/Celite had no effect.



The support retained its superior performance.  The presence of S02 and



moisture did not change the concentration of  [NO] or [NO.J in the ab-



sence of TEA'HCl/Celite.  However, an effect was noted when  [NO ]  and
                                                               A L»


[N0]r were compared with or without the presence of both moisture and
    L*


SO .  A 6 to 30% reduction of NO passing through the collector was noted



in the presence of both 1000 ppm SO  and moisture (see Figure 24).  Thus



NO might have been converted to N0» which was in turn absorbed.  However,



this effect was produced by SO™ in a concentration 100 times that of the



nitrogen oxide in the sample (1 cc NO  sample with an area count about
                                     A
                                 _Q

90 was equivalent to about 3 x 10   g, while 1 cc of 1000 ppra SO- con-



tained about 3 x 10~  g).  This high a concentration is unlikely to be



encountered in actual field conditions.



      Figure 25 shows the recovery from  the SO -treated TEA'HCl/Celite



collector which had been injected with 20 cc NOV (see Figure 18 for area
                                               A


count), then stored in the lab drawer for 48 hr prior to thermal release.



The area count of the recovery peaks obtained from this collector was



more than equivalent to the 20 cc injection.  (A         = 1596.  A0_    =
           ^                                    recovery           /U cc


4 x 323 = 1292 per Figure 18),  indicating a 100% release from the



collector.




                                                                    59

-------
                                                                          §

                                                                          I*
                                                                          s
                                                                           <  j;
                                                                           tu  ID
  INO,ID
WITHOUT H2S
  A • 70.5
                   INO,I0
                   • 1cc
                   H2S
                   A =70
  |NO,]0
WITHOUT H2S
  A = 7«
   Ibl

INOXIDWITH
 1 ccHjS
 A. = 73.6
                                                                         So" 7  S
                                                                         5£ K"
                                                                                             t

          Figure 23(a)  - Effect  of H2S  (20.2 ppm) on  TEA-HCl/Celite, Carrier Air  Contains
                           96% Humidity - Direct Path  (Bypassing  Collector),  Injected
                           Volume:   1 cc

-------
I N°l DIRECT PATH
 WITHOUT H2S
    A = 89
             .'. |NOID- |NOIC-90
             .'. THERE IS NO CONVERSION
             .. |NO.Ic=(NOIci91.8
             .. 100% NO2 - ABSORPTION
              100% NO • PASSING THROUGH
INOl COLLECTOR
  » I cc HjS
   A =90
                                                   A/A,- 1.02
                                                   .. NO EFFECT ON
                                                    NO2- COLLECTION
INO,I COLLECTOR
 WITH 1ccH2S(20.2PPM)
   A = 91.8
                                                                                                            INO.I COLLECTOR
                                                                                                             WITHOUT HjS
                                                                                                               A = 93.8
          Figure 23(b)  - Effect  of  H2S  (20.2 ppm) on TEA-HCl/Celite, Carrier Air Contains
                             96% Humidity -  l^S  Does Not Affect  the  Performance  of  the  Collector

-------
ro
                                  WITHOUT SOZ
                                    A«5I9
SO2 HAS NO EFFECT ON
THE CONCENTRATION
OF MO, IN THE ABSENCE
OF COLLECTOR SURFACE
WITHOUT SO2
[NO 1 X 5
"DIRECT
A = 519
                                                           WITH SO2
                                                                                                                WITHOUT SO2
                                                                                                                INOI DIRECT XZ
                                                                                     WITH 1 cc SO2 11000 PPM}

                                                                                       INDIRECT X2
                                                                                         A -90
                        Figure 24 (a)  - Effect of S02  (1000 ppm/air) on TEA-HCl/Celite,  Carrier  Air Contains
                                        90%  Humidity - Direct  Path

-------
      ^WITHOUT SO2 " 8Si8
      SWITH S02 ° 80'8 (1ccl

      A'WITH S02 ' 61 (2 ccl

      .'. S02 LOWERS THE INOJ
78
                                                                          WITH 2 cc
                                                                           SO2   WITH 1 cc SO2
                                                                            J      I
                         WITHOUT SO
                                                                                         INOXICOLLECTOR X 2
                                                                                           WITHOUT SO2
      Figure 24(b) - Effect of  S02 (1000  ppm/air)  on TEA-HCl/Celite, Carrier Air  Contains
                      90%  Humidity - [802]  Lowers  [NOjJ  Passing  Through  the Collector

-------
         AREA COUNT FOR 20 cc = 1292
         AREA COUNT OF RECOVERY = 1596
         100% RECOVERY (THE EXCESS MIGHT
         BE DUE TO THE "PICK-UP" FROM THE
         LAB-AIR)
                                               (N°xl RECOVERY X
                                            COLLECTOR INSERTED
                                            INTO INTERFACE
                                            HEATER AT 170° C
                                                              '      I0*»
 Figure 25 - Recovery from N02 Collector, TEA-HCl/Celite (S02~treated),
             which was injected with 20 cc NO^ (see Figure 18 for area
             count), stored in a lab drawer for 48 hr.

3.5   DEVELOPMENT OF COLLECTOR SYSTEM

      3.5.1   First-Generation Collector Cartridge and

              Field Collection System

              The collector cartridge is required to be:  (a)  rugged,

(b) lightweight, (c) good in heat transfer, (d)  easily  packed and un-

packed, and (e) able to interface with a suitable analytical  instrument

for analysis.

              The first three requirements can be easily satisfied by

choosing an aluminum shell for the cartridge.   The last two are met by
64

-------
having a Luer-Lock terminal on the cartridge for syringe needle adaption



(to a gas chromatograph, for example), and a stem that can be easily



opened at one end for charging and discharging the cartridge.  Such a



design is shown in Figure 26.  In this first-generation cartridge, the



barrel can be opened near one end.  A V-clamp holds the barrel tight.



This clamp is tightened by an Allen-head screw.  Both terminals are



tapered to allow Luer-Lock connection.  As shown in Figure 26(b), one



of the Luer terminals is made by inserting a specially made Luer-end



screw into the barrel.



              The inside surface of the cartridge was coated with Teflon



to ensure its inertness, because inertness was essential for selective



NO- collection.  The surface was evaluated for its catalytic activity



in converting NO- to NO before and after the Teflon coating.  The results



showed that bare aluminum catalyzed conversion, whereas the Teflon-coated



surface was inert (Figures A-14 and A-15 in Appendix A).  For NO  col-



lection, however, the conversion is not a problem.  Therefore, two identical



aluminum collector cartridges were made, one of which was internally



Teflon-coated for the NO- collection absorbent, and the other uncoated



for the total NOV absorbent..
                A


              Parallel or dual-path collections could be employed, with



one collector for NO- and the other for NOV.  The difference between
                    2.                     A


the two would represent the concentration of NO.  This concept is reflected



in Figure 27 which illustrates the dual-path collector/pump assembly.  The



collector cartridge was supported by an aluminum tube lined with soft pad-



ding.  A filter was located in front of the cartridge and another behind
                                                                     65

-------
Figure 26 (a) - First-Generation Collector Cartridge - Aluminum Cartridge

-------
                 10 MM
  6.350x907 THRO
    BOTH ENDS
1 s — ""
in ^ 1.17MM 	 »-
J i
, 	 f

	 6. 5 MM 30°
127 MM
X1
V.
\
fc.
^
\
«/*
\
"*
^
^
Y
1/10 MM
6.3
S
o
i-
I

1
5 MM
1
I

                                256 SCREW
                               (ALLEN HEAD)
                2.80 MM DRILL
                                                           6.60 MM
                                                                        P 8 7.803-3
Figure  26 (b) - First-Generation Collector Cartridge - Details of  the
                Cartridge

-------
ffl
oc
                               Figure 27 - First-Generation Dual Path  Collector/Pump  Assembly

-------
it.  The filters were actually Beckman Teflon Connectors (#504) containing



a piece of Gelman Type-A glass filter paper cut to size.  A terminal



cut from a Hamilton female/female Luer connector was inserted into the



Teflon connector to enable the filter to be attached onto the cartridge.



A limiting flow critical orifice was located between the pump and the



filter.



              The drawback to this approach is that the NO concentration



is found indirectly by subtraction, and two collector units are required



each time.  A more direct method is desirable.



      3.5.2   Development of Collector Unit



              The development of a collector unit was based on a tandem



arrangement of an NO- collector and an NO  collector, with the former
                    L.                    X


serving as the precolumn.  When air is being sampled, the first collector



removes N00, leaving the NO to be taken up by the NO  collector.  This
          2.                                         A


arrangement thus afforded a direct method of sampling NO- and NO.  Only



one collection unit was necessary when these two collectors were pack-



aged in tandem.



              This collector unit should house the two collector cartridges



sturdily, should be easily assembled or disassembled, and should be easily



attached to the pump for field collection.  Figures 28 to 31 represent



a design possessing such features.  In this design, the NO- collector



cartridge is a disposable Pyrex tube (for the non-reusable TEA), which



has an 0.63 OD inlet on one end, and a Luer groundglass taper (Kontes



Glass Co, Vineland, N. J. //K-663500-0244) on the other.  The glass taper



readily fits into the aluminum cartridge (containing Co-O-) through the



inlet hole in its end.
                                                                      69

-------
        0.432
                                           ALL DIMENSIONS

                                           IN CENTIMETERS
                                                                                     1.27 TPF TYPICAL
              - 1.905 -
0.940
                                (a) CARTRIDGE BODY AND END PLUGS
                                     (0.635 OD 28/2.54 cm THREAD)

                                            (1/428THRD)
                                                                                           -—^^ — - 1.80
1 1
1 	 1 	
	 r- '
i 	 i.


t
- 1 80

0 94Q — ^

-- *—





.4

             0.279 DIA.


/

0.432 -»•
p-i

—




0.058
i^ v n mr
(SLOT

*- 0.792 -»
1.575 	 »•
* -•"•'
DEEP
1 /

t \
0.396
                                       (b) LUER - END SCREW




                   Figure 28 -  Second-Generation Aluminum Cartridge
                                                                                                           n
                                                                                                           o

-------
Figure 29 - Second-Generation Pyrex NC>2 Collector Cartridge shown with
            aluminum cartridge and collector unit subassembly

-------
ro
                                        TEA TUBE
                FOAM PLASTIC
                  SHIELD
                                                                                     OXIDE TUBE
                                                                                                       ALUMINUM HOUSING
            STRAIGHT TEFLON
            BULKHEAD FITTING
 TEA/CELITE
IN GLASS TUBE
CO203/CHROMASORB G
  IN ALUMINUM TUBE
SPONGE RUBBER
 PRESSURE PAD
                           Figure 30  -  Collector Unit  Assembly  with Glass N02  collector and  aluminum
                                         collector cartridges.

-------
                    2.54 OD, 20 THREAD/2.54 CM
                           (1-20 THRD)
                                                   ALL DIMENSIONS
                                                   IN CENTIMETERS
\
')
—



1.27
„



	 	 	 26.67


1.27

f
.54
1
	 L.

                (a) ALUMINUM TUBE
                                   2.54 OD, 20 THREADS/2.54 CM
                                         (1-20 THRD)
\
fr
Xs
0.305 	 +•
~\
1^
r /
F=-~-~
1
•4
W
1.27

r— 2.997
1
	
P-87-803-3
A =



0.953 OD, 24/2.54
CM THRD
1.27 DRILL
                  (b)  END-PLATE
Figure  31  - Details of  the collector unit tube
                                                              73

-------
               The  aluminum  cartridge with  Co_0_  packing  is  assembled




by  pressing  the  end plugs into  the  tube.   These  end  plugs are  removable.




The outlet of  the  aluminum  cartridge has a Luer-Lock terminal  provided




by  the Luer-end  screw  (Figure 28).  Though not shown in  the sketch, the




Luer-end screw has a Luer-Lock  ring attached onto  the slot  cut for  it.




In  assembly, the Pyrex  tube is  first inserted into the Beckiaan bulkhead




Teflon connector on the end-plate and  tightened; this then  attaches to the




aluminum cartridge via  the  Luer  ground taper.  The resulting subassembly




is  then carefully  placed into the collection tube  and the end-plate is




tightened.   The  assembled unit has  the Luer-Lock end of  the NOX collector




as  one of the  terminals.




               For  analysis, the  unit is disassembled,  and the  cartridges




are inserted separately into the interface device  for recovery of the




collected nitrogen oxides.  A special  Luer coupler shown in Figure  32




is  designed  for  the aluminum cartridge.  This coupler readily  goes  into




the  hole at  one  end of  the  cartridge and serves  as the terminal for




analysis.




               Although  this design  appears to be very attractive, the glass




Luer taper unfortunately restricts  the  sample flow and introduces mechanical




weakness because of its thin, narrow,  fragile neck.   To alleviate these




problems, a  design with two identical  Pyrex tubes joined by Beckman Teflon




connectors fitting into a unit housing  is  recommended.  This design has




been described in detail in Section 2.  One unit of  such design was




mailed from  Bendix to EPA and arrived  at the destination safely.
74

-------
                     ALL DIMENSIONS
                     IN CENTIMETERS
                                       0.305 DRILL ,
                                        THROUGH
 Figure  32 - Luer-coupler  (for  interfacing the second-generation
             collector  cartridge  to  the  analyzer)
3.6   SAMPLING PUMP

      Since  the sampling pump  is  one  of  the  key  components of the sampl-

ing system,  currently available pumps were surveyed.   Desirable features

for this pump are:   (1) low power DC- or 110V  AC-operated;  (2)  capability

for continuous 24 hr unattended operation; (3) capability  for providing

high air flow rates, greater than 10  £/min if  possible.

      Literature was requested from over thirty  manufacturers of pump and

air samplers.  The majority do not make  DC motors  for  the  pump.   Most of

the DC-operated vacuum pumps on the market are low in  air  flow  rate (less

than 10 £/min) and are not designed for  continuous operation with any

restrictions other than particle filter  paper  (<100 mm Hg) .   The most

promising pumps that can most closely meet the requirements  are  Model

107C DC20 from Thomas Industries Inc. (Sheboygan,  Wise.  53081)  and Model

1531 from Cast Manufacturing Corp.  (Benton Harbor, MI  49022).   Both

pumps deliver 35-42 Jl/min free air flow,  and remain operative against

500-600 mm Hg vacuum; they both can be powered by  12V  DC source.
                                                                     75

-------
      The Cast pump was chosen for the present program since its manu-




facturer had a better delivery time than Thomas Industries.   This Cast




Rotary Vane Pump weighs 8 lb., has a 1/10 HP power requirement, and




draws 11.3 A for operation.  A 12V battery or 12V battery charger (12




amp-rate) can serve as power supply for this pump.




3.7   COLLECTION SYSTEM (AIR SAMPLER)




      The system is required to collect samples continuously up to 24 hr




in the field with either DC or AC power.  This capability is provided




by the installation of an interval timer, and 12V battery charger to serve




as power supply and rectifier, or a 280 amp hr truck battery.




      The development effort was concentrated on the design of a portable,




weatherproof and tamper-proof housing for the collection unit, vacuum pump,




interval timer, power supply  (battery charger) and battery.   This has been




accomplished with a housing made of aluminum; it contains three shelves




for respectively the collector unit and pump; timer and battery charger




controls; and battery.  Locks are provided for the housing.   For dis-




assembly and transportation,  the housing can be easily detached and folded.




The details have been given in Section 2.




3.8   INTERFACE DEVICE




      When thermal release was established as a satisfactory means of




recovering the trapped oxides from the collectors for analysis, the inter-




face design became a relatively simple matter.  Two approaches were tried.




      The first approach utilized an aluminum tube to serve as the oven




body which readily accepted the collector cartridge.  Thermal power was




supplied from a thin integral heater/sensor  foil  (Thermofoil heater) manu-




factured by Minco Products, Inc.  The Thermofoil heater was easily wrapped
76

-------
around the aluminum heating tube with pressure-sensitive adhesive.  The



heater took either DC or AC power source; its heating power was limited



by the resistance of the heater foil, the power supplied, and the mass



and geometry of the heat sink.  Repeated trials, however, failed to



produce an oven capable of bringing the collector to the desired temper-



atures.  Custom manufacturing would be necessary to accomplish the task



and was too expensive to pursue.



      The second approach resulted in the interface design depicted and



described in Section 2.  This device had a variable temperature setting



ranging from ambient to 500°C; this range was made possible by a propor-



tional temperature controller and platinum sensor.



3.9   PRELIMINARY FIELD COLLECTION AND ASSAY OF N02 with TEA-HCl



      This was only a preliminary and limited experiment.  It was designed



to check the performance of the collection material under a more realistic



field condition.  A continuous chemiluminescence NO -monitor, and a con-
                                                   A


tinuous Saltzman NO -monitor  (Technicon Autoanalyzer) were used for



comparative study.



      3.9.1   Experimental Procedure



              This experiment was conducted at Wayne County Department



of Health, Air Pollution Control, Detroit.  Air sample was taken from a



sampling manifold leading into the laboratory from the rooftop.  One mani-



fold line was led to a Bendix chemiluminescence NO  monitor, and another
                                                  A


to a Technicon Wet Chemistry Analyzer for comparative, continuous NO,



N0? monitoring.  From the same manifold, air samples were drawn into
                                                                     77

-------
 the  collector by  a  small  diaphragm  pump  for  24 hours  continuous NO--




 collection.  The  collectors were  tubes packed with  Triethanolamine hydro-



 chloride mixed with Chromosorb  G  AWDMCS  (45/60 mesh)  in  1.5/1  ratio.



 Approximately 6 grams  of  packing  was used  for each  collector.  For assay,




 the  same elution/area  count technique as previously described  was used



 as in  the  screening tests with  a  Bendix NO -analyzer.  The experimental
                                           A


 results are  summarized in Table 4.  Table  5  gives the hourly  [NO  ] average



 read outs  by Technicon and chemiluminescence analyzers.



       3.9.2   Discussion



              The experimental  results indicate that the present ambient



 and  assay  method  with  solid absorber is promising when the flow rate



 is kept around 100  cc/min.  The comparison data were obtained  by a con-



 tinuous Saltzman  technique using  a  Technicon Autoanalyzer, and by a



 chemiluminescent  analyzer running at the same time  on the same sample.



As shown by  the attached  Table  5, the hourly averages of the two tech-



niques do  not agree satisfactorily  and neither do the 24 hr averages.



The  present  collection  and assay  technique show consistently higher



results than both other methods.  While it is possible that these pre-



liminary data contain systematic errors due  to variations in flow, the



lack of agreement of the  two other  techniques suggests the presence of



inadequacies in the methods employed.   The possibility definitely exists



that the present  collection and assay technique gives  correct results,



in which case the other two techniques  would give results seriously in



error on the low side.   It would be worthwhile to remove these uncer-



tainties and obtain consistent results  both in controlled laboratory



conditions  and in the field.
78

-------
                                Table 4 - Summary of Experimental Results
Collector
Number
2
3
16
Flow
Rate
1 lit/min.
1 lit/min.
105 cc/min.
Collection
Time
24 hrs.
24 hrs.
24.2 hrs.
Daily Average [N02]
(By Technicon and
Chemiluminescence)
0.035 ppm
0.025 ppm
0.032 ppm
Measured
Collected
Quantity
11.99 Mg
4 Mg
(a) by area count
quantity 10
Collected
(Calculated
Daily Ave
90.7 Mg
64.8 Mg
8.8 Mg
17
80 cc/min.  for 4 hrs.
75 cc/min.  for 20.2 hrs.
24.2 hrs.
0.032 ppm
    12 ug
(b)  by weighing
    13.85 pg

(a)  by area count
    7.6 ug
(b)  by weighing
    7.4 Mg
                                                                                                6.29 Mg

-------
           Table  5  -  Comparison of Ambient NC^-Concentration by
                     Technicon and Bendix Monitor*  (Date:  12-8-73)
.me (hr.)
0-1
1-2
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23-24
Technician
ppm
0.050
0.050
0.040
0.040
0.030
0.030
0.030
0.040
0.040
0.030
0.030
0.030
0.020
0.020
0.020
0.020
0.020
0.030
0.030
0.030
0.030
0.030
0.030
0.030
Bendix
ppm
0.035
0.035
0.030
0.030
0.020
0.020
0.025
0.030
0.030
0.025
0.025
0.020
0.015
0.020
0.015
0.015
0.020
0.025
0.025
0.020
0.025
0.020
0.020
0.020
Tech: Bendix
Ratio
1.43
1.43
1.33
1.33
1.50
1.50
1.20
1.33
1.33
1.20
1.20
1.50
1.33
1.00
1.33
1.33
1.00
1.20
1.20
1.50
1.20
1.50
1.50
1.50
80

-------
                             SECTION 4
                    SUMMARY AND RECOMMENDATIONS

4.1  SUMMARY
     Simple collection and assay of NO and N02 using solid chemical ab-
sorbents was proven feasible.  The selective collection of these oxides
is possible with a tandem arrangement of two absorbent cartridges:  The
first, containing triethanolamine hydrochloride packing, completely absorbs
NO- from the air sample without affecting the NO concentration.  The sec-
ond, filled with cobalt oxide packing, quantitatively removes NO.  These
collected oxides are later eluted and analyzed separately.
     The interface device developed to transfer the sample for analysis
accepts one collector cartridge and heats it to a temperature which re-
leases the nitrogen oxide.  At 160-170°C, the absorbed N02 is quantitively
released from the TEA-HC1 collector as NO; cobalt oxide gives up its cap-
tured NO completely at 400-420°C.  When the collector is coupled via this
interface to a suitable instrument, such as a chemiluminescence NO  ana-
                                  '                               x
lyzer, the NO and N0_ are easily measured.  When cooled, the collector
cartridges are ready for reuse in sample collection.
     For convenient handling, a special collector unit was developed.
This unit is easy to assemble and to install in a collector system; more-
over, it is easy to disassemble.  Its rugged, lightweight construction
enables its safe and inexpensive transportation within the postal system.
                                                                      81

-------
      A simple weatherproof collection system was built for field collec-




tion, based on the concept of a collector unit containing the solid chem-




ical absorbents.  Its Cast vacuum pump and interval timer enable the sam-




pler to operate up to 24 hr continuously without attention.  This operation




can be powered either by a 12V DC high capacity battery or by a 110V AC




power supply with a 12V DC output.




      Although the collectors developed under this contract met the per-




formance requirements in the laboratory, extensive field testing may lead




to possible design improvements.  The major effort of the program was in




the screening and evaluation of chemical absorbents, so the collector unit




represents an experimental prototype.  Additional work will be required




to rigorously test and further improve the designs.  Recommendations for




further work are detailed in the following subsections.




4.2   RECOMMENDATIONS




      4.2.1   NO- Collector With TEA-HC1







              TEA-HC1 (m.p. 177-179) selectively collects N02 and is inert




to NO at room temperature.  At higher temperatures, its affinity for NO-




decreases and it catalyzes the conversion of N02 to NO.  Its desirable




property as an NO  absorbent, however, returns when the temperature is




decreased.  In use, the TEA-HC1 cartridge is heated to 160-170° to recover




the absorbed N0«, thereby regenerating the absorber.  If the temperature




exceeds 170°C, however, some discoloration of the absorbent occurs, after




which N0« to NO conversion is noted with the cartridge at room temperature.




Limiting the temperature to 170°C, therefore, is a necessary precaution




with this cartridge.
82

-------
              The ultimate lifetime of the TEA'HCl absorbent should be




determined.  Laboratory tests confirmed its performance up to ten heating




cycles, but more extensive cycling should be evaluated.  If the absorbent




has a long lifetime, it can be encased in a more permanent material, such




as Teflon-coated aluminum instead of the present disposable Pyrex.  The




metal case has the obvious advantage of nonbreakage.




              Absorbent capacity also should be further evaluated.  The




present collector cartridge was designed in a size convenient to pack




and handle without regard to the ultimate absorbent capacity.  When the




absorbent's maximum capacity has been determined, the optimum size for




the cartridge can be established.




      4.2.2   Cobalt Oxide




              Since both cobaltous and cobaltic oxides absorb NO  com-
                                                                X



pletely, conversion of NO- to NO is not a problem.  The total absorption




capacity of each oxide should be measured, however, to design an optimum




collection cartridge.




              Preliminary tests indicate that cobaltic oxide quantitatively




absorbs and releases NO independently from the N0_ absorption.  If this




is true, one could use cobaltic oxide alone for the collection and assay




of the nitrogen oxides.  The analyzer, such as a gas chromatograph, would




have  to differentiate  the NO- from NO, since they would evolve simultan-




eously from the collector.  Further investigation of this phenomenon




could  lead to a single absorbent collector.
                                                                      83

-------
              The extensive and combined effects of possible interfering




species, such as moisture, SO-, peroxide, and C02, on collector perfor-




mance could be examined.  It has been established that moisture present




in air up to 96% humidity does not affect the performance of the collectors.




The TEA-HC1 absorbent was subjected to 1000 ppm S02 at a flow rate of U/




min for 20 min and still retained its performance.  Nonetheless, evalua-




tion of collection efficiency in the presence of both moisture and S02




in more common concentrations should be systematically investigated.




The preliminary data indicated that the presence of highly concentrated




S02 and moisture (96%) did lower the concentration of NO passing through




the TEA'HCl.  This may be caused by the conversion of NO to N02, the




latter then being absorbed.  The conversion may be attributed to the




combined effect of SO-, moisture, and the surface of the absorbent.  In




the absence of the absorbent, no change in concentration was observed.




This indicates the need for a systematic study of the effect of interfer-




ences on both absorbents under realistic field conditions.




              The nature of bonding of the nitrogen oxides to the absor-




bent surface was not studied for either absorbent.  This phenomenon should




be investigated using IR and DTA techniques to determine what inherent




properties of the absorbents might degrade their performance.  Such an




investigation might also reveal what other types of compounds could serve




as NO and N0_ absorbers.




      A.2.3   TEA'HCl Packing




              This absorbent is a fine, white powder which, when used by




itself as a collector packing, introduces great flow restriction and re-




sults in a poor flow rate.  Physical mixing of such a fine powder with an
 84

-------
 inert  support  such  as  Chromosorb G AW DMCS 45/60 does not produce an




 especially  good packing.  The best packing is produced by solution-coating




 the  TEA'HCl (in distilled water) onto a support such as Celite.  This




 technique gives a high-capacity packing with desirable flow characteristics,




               Two points with respect to this collector packing are worth




 further consideration.  First, the Celite surface should always be com-




 pletely covered by  the absorbent.  Celite absorbes N0_ at room temperature




 and, although  it gives up the captured N0_ at 170-200°C, the efficiency




 of the absorbent is better preserved and controlled when no extra foreign




 surface is  present.  Second, the collection efficiency of this absorbent




 at flow rates higher than 120 cc/min has not been evaluated.  Since field




 collection  may involve flow rates > l£/min, the absorbent efficiency under




 such conditions should be determined.




      4.2.4   Cobalt Oxide Packing




              Commercial cobalt oxide is a grayish-black fine powder




 (^200 mesh).  This powder must also be mixed with some inert, more coarse




 support (Chromosorb G, 45/60) to afford a reasonable flow rate.  Other




means of making packings of lower flow restriction that should be further




investigated are




      •  Hot-pressing or cold-pressing the oxide into a cake, firing the




         cake at high temperature,  then crushing the cake into chips for




         packing.




      o  Solution-coating Co(NO_)   onto a support  such as Celite,  drying,




         then coverting to Co20  at 200-250°C in air.
                                                                     85

-------
With the first method, the packing's capability of retaining its shape




without breaking down to a powder is the major question.  Moreover, the




firing temperature chemically converts the Co20_ into CoO and Co^,




which may affect the desirable properties as an absorber.  The second




method gives good flow characteristics as evidenced by its use with the




TEA-HCl/Celite packing; however, removing the residual nitrate from the




packing is quite difficult.  In fact, conditioning of packing prepared




by this method for over a week at 450°C in air did not result in a clean




absorbent.  A systematic evaluation of these packings should be undertaken.




      4.2.4   Collector Configuration




              The present design of the collector unit contains two Pyrex




cartridges.  Although these cartridges are ruggedly constructed, they can




be broken during assembly or disassembly.  When the necessary tests are




performed and the reusability and lifetime of the absorbents are fully




established, a nonbreakable cartridge design can be considered.  Experi-




mental work showed that aluminum cartridges coated inside with Teflon




did not affect N0_ collection.  The Teflon coating is required for the




TEA'HCl/Celite cartridge, since any NO™ to NO conversion results in ana-




lytical error.




              The final cartridge design will still require a precaution




with respect to overheating, since, extremes of temperature will induce




decomposition of the packing and the coating.  A longer range program




may prove worthwhile, therefore, to develop a collection material of




higher thermal stability.
86

-------
          APPENDIX A

    TYPICAL RECORDINGS OF
COLLECTOR MATERIALS EVALUATION
                                                A-l

-------
                             APPENDIX A




        TYPICAL RECORDINGS OF COLLECTOR MATERIALS EVALUATION






     This appendix contains a collection of recordings representing the




evaluation of various absorbents and container materials.  These record-




ings are given in Figures A-l through A-15.
A-2

-------
Figure A-l - General material evaluation, showing the inertness of  (1)
             Pyrex tube, (2) silane-treated glass wool,  (3) Chromosorb
             T, and (4) Chromosorb W AW DMCS

-------
P>J
o
CM
X
[NO] 2 = [N0]c
[NOX]2=[NOX]C
THE SUPPORT IS
INERT
                                                        P-8 7-803-3










1
0
n
0
o
s
00
cc
8
o
0
cc
X
o
0
z
^


o
8
6
X
t—
^
HI
cc
o
o
2
I
^- ^^_
          Figure  A-2  - Chromosorb W AW DMCS

-------
                    ..  INCOMPLETE NO2 - REMOVAL
                      CONVERTSNO2- NO
IN°) DIRECT
0.020
            IN°I
               COLLECTOR
           u
                      X! COLLECTOR
                                          'N°lCOLLECTOR
                                          0.08
                                        1	1	1	1	h


                                Figure A-3 - TEA/Chromosorb W AW DMCS

-------
   THERE'S NO APPARENT

    ABSORPTION OF NO2;


SIGNIFICANTLY CONVERTS NO2
                      INOICOLLECTOR X0'5

                            0.315
                                                        Q
                                                     i   UJ

                                                     I   &
                                                       o
                                                                      INOI DIRECT 0-MPPM
                                             Figure A-4 -  TEA/Regis  GasPak FS

-------
. INCOMPLETE
 N02 - ABSORPTION

 EXCESSIVE N02 - NO
 CONVERSION
            Figure A-5 - TEA/Chromosorb  T
                                                             A-7

-------
  [NOx]TEA(,jq)
      0.22
                                0.12
                                                                  ^DIRECT
                                                                     0.020
                       Figure  A-6 - TEA/NEAT  (liquid)
A-8

-------
 DIRECTX
2.3 PPM
               1    O
                   CM
                   X
                    cc
                    O
                    O CM
                    O
                    o
[NOX] DmECT X5
      2.24
                       DIRECT* 2
                        0.08
                                             IN°X]COLLECTOR X 5
                                                     0.28
            Figure A-7  - TEA Borate

-------
                                                     f *
Figure A-8 - THEED

-------
        I [N0]c
          0.30



        9*
INSIGNIFICANT

N02-REMOVAL
	      i
             Figure A-9 - Quadrol/Chromosorb W AW DMCS
                                                       A-11

-------
to
                                  DOES NOT ABSORB NO2. BUT
                                  CONVERTS NO2-NO
                                                                                IN°) COLLECTOR X °'5
                                                                                       0.02
                                                                                                           [NO
                                                                                                             IN°) DIRECT X0'5
                                                                                                                                     H	1-
                                                                   Figure  A-10 -  NiSO,

-------
        DOES NOT ABSORB
               NOo
                [N°1 COLLECTOR X0-5
                      0.09 PPM
X1
[NOX] COLLECTOR
    0.22 - 0.24
                                                                      n
                                                                      o
                                                                      oo
                        Figure A-ll - CoSO.
                                                               A-13

-------
                                          IN°] DIRECT X1
                                             0.22 PPM
                              Figure  A-12 - CoO
A-14

-------
IN°.1CU20
s
                                                 0.69
                                               0.32
    H	H
                    H	1	1-
                                                                                             INDIRECT
                                                                                              0.30 PPM
                                                                                                                           -\	1-
                                                       Figure A-13 -  CuO

-------
                 INOX]
             0.59 ALUMINUM
                  [NOX1 DIRECT
                    0.60 PPM
                                    (4-5MIIM.-*)
                               IN01ALUMINUM
                               INCREASES WITH
                               TIME INDICATING
                               CONVERSION
                                                  • CONVERTS NO2
                                                    TO NO

                                                  • DOES NOT ABSORB
                                                    NO2
                              \
                                               INO'DIRECT
                                                0.38 PPM
                                         t
                                          TO ALUMINUM
   Figure A-14  - Aluminum  cartridge  (after heating cycle to  200°C and
                  cooled down to room  temperature)
A-16

-------
                                        
-------
                              APPENDIX B
                      DIRECTIONS FOR ASSEMBLING
                       COLLECTOR FIELD HOUSING
 (1)  Place folded housing in flat position with back side down (large
      door with hasp up).

 (2)  Open door and remove legs.

 (3)  Close door and turn  unit over (front down).

 (A)  Remove wing nut and  screw holding cover in folded position.

 (5)  Unfold housing by pulling upper section so it swings upward  and
      toward bottom.

 (6)  Stand housing upright with cover in full open position.

 (7)  Remove five remaining wing nuts holding sides in folded  position

 (8)  Swing sides into position running studs into appropriate holes
      and fastening with wing nuts.

 (9)  Lay the housing on one side and install legs on the other side
      with 1/2" long 1/4-20 bolts;  match one leg with clearance hole
      to stud head.

(10)  Turn housing onto other side and install remaining two legs.
                                                                    B-l

-------
Figure B-l - Folded View of Collector
             System Housing
Figure B-2 - Assembled Collector Unit

-------