oERA
            United States
            Envirofvnentat Protection
            Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
                         EPA-600 2-79-028
                         February 1979
            Reaeardi and Oevelopment
Optimized
Chemiluminescence
System for
Measuring
Atmospheric
Ammonia

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development. U S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has  been assigned  to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to  repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                                EPA-600/2-79-028
                                                February 1979
     OPTIMIZED CHEMILUMINESCENCE SYSTEM FOR

          MEASURING ATMOSPHERIC AMMONIA


                       by
Ralph Baumgardner, W.A. McClenny and R.K. Stevens
   Atmospheric Chemistry and Physics Division
   Environmental Sciences Research Laboratory
  Research Triangle Park, North Carolina  27711
   ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION AGENCY
  RESEARCH TRIANGLE PARK, NORTH CAROLINA  27711

-------
                                DISCLAIMER

     This report has been reviewed by the Environmental Sciences Research
Laboratory,  U.S. Environmental Protection Agency, and approved for publica-
tion.  Mention of trade names or commercial products does hot constitute
endorsement or recommendation for use.

-------
                                 ABSTRACT





     This report describes the optimisation and testing of a continuous



measurement system for the analysis of ammonia (NH3) at concentrations found



in the atmosphere ranging from 0 to 20 parts per billion (ppb).   The measurement



system combines an ultrasensitive chemiluminescence nitric oxide (NO) detector,



with a thermal converter for NH- to NO conversion and an acidic  scrubber for



alternate removal of NH3 from the sample stream.



     The chemiluminescence monitor with a high temperature converter will



give a combined measurement of NO, nitrogen dioxide (ML), and NH~.   Other



nitrogen containing compounds, if present, may also be converted.   By placing



an acidic scrubber in the sample line, the NFL will be removed and a combined



measurement of (NO  [MO and N0?]) will be obtained.  By taking the signal
                  A           £


difference when the scrubber is in or out of the sample line,  the concentration



of NH3 can be determined.  Other "basic" nitrogen containing compounds present



in concentrations of greater than 1 ppb would, if thermally converted to NO,



appear as a portion of the ammonia signal.



     In the optimization of the measurement system it has been determined



that a second NO  signal, independent of the signal generated  with or with-
                A


out the scrubber, aids in validating the signal difference observed is



ammonia.  Different approaches for generating signal response  from the



system are discussed, namely the use of two detectors or one detector and a



switching valve.
                                      m

-------
     The use of two separate detectors was tested in  the laboratory.   The
detector for measuring the chemiluminescence of NO and ozone  (CL)  was  optimized
to allow for analysis of concentrations of less than  1 ppb of NO,  NOp, or
NH-.  An optimum configuration for the acidic NhU scrubber was also achieved.
     Performance tests were conducted on the total measurement system  to
verify its capability to measure  NO,  N02 and NH3 under laboratory  conditions
as well as ambient monitoring conditions.   A dynamic  calibration system
capable of generating 1 ppb of NHo was used for the testing.   The  calibration
system is a double dilution permeation system.
     Testing in the laboratory and monitoring of ambient air  indicate  that
the optimized measurement system is capable of measuring atmospheric con-
centrations of NH3 on a continuous basis.
                                      IV

-------
                                  CONTENTS
Abstract	iii
Tables 	   y
Figures	vi

     1.   Conclusions  	   1
     2.   Introduction 	   2
     3.   Discussion	   6
     4.   Experimental	19
     5.   Results	29

References	36
                                   TABLES

Number                                                                Page

  1   System Response to Humidified Samples of Nitric Oxide and
        Nitrogen Dioxide Using Two High Temperature Converters ....  35

  2   System Response to Humidified Samples of Nitric Oxide and
        Nitrogen Dioxide Using a High and Low Temperature
        Converter	   35

-------
                                   FIGURES
Number                                                                Page
  1    Flow diagram of NH3 measurement system 	  12
  2   Diagram of NO detector 	  13
  3   Diagram of sensitive chemiluminescent NO monitor  	  14
  4   Response of optimized NO detector  	  15
  5   Single spiral reactor  	  16
  6   Diagram of Thermo Electron  model  14T, NO.  NO  ,  NH, monitor  ...  17
                                                 A     O
  7   Diagram of Columbia Scientific Industries  NO  monitor  	18
                                                 /\
  8   Flow diagram of Aerochem chemiluminescence NO monitor  	  22
  9   Diagram of NH3 scrubber  	  23
 10   Diagram of NH3 scrubber system 	  24
 11   Diagram of NFL measurement system using two separate  detectors  .  25
 12   Diagram of modified NO, N0? calibration system  (gas phase
        titration)	26
 13   Diagram of Research Triangle Institute (RTI)  double dilution
        permeation system  	  27
 14   Function diagram, Research Triangle Institute permeation
        system	28
 15   Response of the measurement system to NH~  with  and without
        NH3 scrubber	32
 16   Ammonia response using different converters with  the  NH3
        scrubber system	:  ....  33
 17   Measurement of ambient concentrations of NH3  	  34

-------
                                  SECTION I
                                 CONCLUSIONS
     Test results and evaluation of the ammonia (NHL) measurement system show
that the design of the optimized system allows for measurement of ambient
ammonia concentrations as low as 1 ppb level.   Nitric oxide (NO) and nitrogen
dioxide (N02) at this low concentration can also be measured with the
measurement system described.
     The thermal converter in the measurement system has been shown to give
quantitative conversion of NH3 and NOp to NO at low concentrations under
varying humidity conditions.  The NH~ scrubber has been demonstrated to
quantitatively remove NHg in concentrations ranging from 0 to 50 opm while
passing N02 and NO.
     To measure ambient NHL concentrations, different system designs are
possible as long as the following criteria are met:
       •  The chemiluminescence detector must have a minimimun detectable
          limit of 0.5 ppb with a signal to noise ratio of one.
       •  The system must have two signal outputs, one for NO  and one
          for NOV plus NHV
                A        0
       •  Thermal converters must quantitatively convert N09 and/or NhL
          to NO.                                                      J
       •  The acidic scrubber must quantitatively remove NhL while
          quantitatively passing NO and ML.
     The total NhL measurement system including chemiluminescence detector,
thermal converters, and NhL scrubber provides a sensitive method for the
detection of atmospheric NFL.
                                      1

-------
                              .   SECTION II
                                INTRODUCTION

     Ammonia (NH3) is one of a number of nitrogen containing compounds found
in the atmosphere.  This natural  constituent plays an important part
in the total biological nitrogen  cycle.   Anthropogenic sources of ammonia
also exist from combustion of oil  and coal, which contribute minor amounts.
The impact of man-made sources however,  on the total  atmospheric cycle is
small (1).  In the future, urban  concentrations of NH- may be affected by
the number of catalytic converter equipped automobiles which under certain
operating conditions can produce  NH3 in  the exhaust.
     Unlike NOX, NH3 has no health related effect at  concentrations found in
the atmosphere.  Its importance in air pollution is limited to its function
in the neutralization of sulfuric and nitric acids. Ammonium sulfate and
ammonium nitrate, the end products of.these neutralization processes, have
been implicated in health related problems (2).
     To provide a more complete understanding of atmospheric aerosol forma-
tion accurate up-to-date data on  rural  and urban concentrations of atmo-
spheric NH3 must be obtained.
     Data on background concentrations of atmospheric NH3 has been reported
in studies by Ounge (3), Pate and Lodge  (4), Healy et al. (5), and Breeding
et al. (6).  Comparison of the data from these studies indicate that back-
ground concentrations of NH3 range from  less than 1 ppb to 20 ppb depending
on geographical location and time of year.  Breeding et al. and Junge have
                                      2

-------
shown that atmospheric NHL concentrations range from 1 to 10 ppb (.69 to 6.9
yg/m ) over much of the United States.  In reviewing the data on atmospheric
NH3 concentrations it becomes apparent that the measurement methodology used
to obtain the data had a great amount of uncertainty in the range of NhL
                                                 o
concentration found 1 to 20 ppb (.69 to 13.8 yg/m ).  Since static cali-
bration was used to standardize these methods rather than dynamic generation
of known NH, concentrations in the range of measurement, sample integrity in
the sampling portion of the measurement techniques could not be evaluated.
     Measurement methodology for ambient NH^ determination has generally been
limited to two manual colorimetric techniques in which the sample is collected
in bubblers.  The technique first developed and evaluated for atmospheric
monitoring was the Nessler colorimetric technique (7,8).  This technique
requires the collection of NH3 in an acidic solution, and subsequent reaction
with the Nessler (alkaline potassium iodide/mercuric chloride) reagent to
form a colored complex.  The NH3 concentration is determined spectrophoto-
metrically by measuring the absorbance of the colored complex at 440 nm.
The method has been shown to have interferences from hydrogen sulfide and
formaldehyde, and the stability of reagents is also a problem (9,10). The
minimum detectable concentration has been found to be 15 ppb (10.3 yg/m ).
     The other colorimetric method which has been more recently optimized is
the indophenol method (11,12,13).   Ammonia collected in an acidic solution is
reacted with phenol and alkaline sodium hypochlorite to form the indophenol
dye.  The sample is analyzed colorimetrically at 630 mm.  The advantages of
the indophenol method compared to the Nessler method are fewer potential
interferences and more stable reagents.  The range of the method has been
                                                    o
reported to be from 25 to 1000 ppb (17.2 to 697 yg/m ).  Errors in the

-------
                                                        o
technique can be as high as ±30% below 25 ppb (17.2 yg/m ).   By using an
autoanalyzer and modifications to the technique as described by Georgii  et
                                                            o
al.  (14), measurements can be made below .025 ppm (17.2 yg/m )  although
errors of ±10% remain at the 2 ppb level.
     The use of a pyridine-pyrazolone reagent for analysis  of NH3 was first
described by Kruse and Mellon (16) who used bubblers to collect the NH~.
The NH- collected in bubblers was reacted with the pyridine-pyrazolone
reagent and the absorbance of the solution was measured at  450 nm with a
spectrophotometer.  In 1971 Okita and Kamamori (15) reported a technique
combining pyridine-pyrazolone reagent with the sample collection using
filters impregnated with sulfuric acid rather than bubblers.  Okita found
that formaldehyde was not an interferent in this method as  with the Nessler
procedure, however no increase in sensitivity was determined.  Other recently
developed analytical procedures for analysis of atmospheric NH3 include  a
colorimetric bubbler technique using o-(benzenesulfonamide-p-benzoquinone
reagent (17) and an analysis procedure using a Ring Oven after NH3 collection
on a filter impregnated with oxalic acid (18).  David et al. (19) described a
measurement system using an optical waveguide chemically coated and sensitized
for NH3.  Light transmission through the waveguide changes  as a function of
the amount of NH3 absorbed in the reactant surface per unit time.  The
response of the system was found to be flow independent but was affected by
humidity.  The sensitivity of the method was reported to be 5 ppb with an
error of ±10%.  Mulik et al. (20) recently described a sensitive and selec-
tive method for the assay of ammonium ion in ambient aerosols using the
relatively new technique, ion chromatography.  Preliminary experiments in-
dicated gaseous NH3 could be collected in a 3% oxalic acid solution in a
bubbler or coated on a filter and  then analyzed  with the ion chromatograph.

-------
     In the measurement methodology described above, the sample is first
collected and analyzed at a later time.   None of these methods can answer the
need for a real-time continuous monitoring method which has specificity and
can accurately detect atmospheric NH~ levels of 1 ppb or less.  A measurement
system combining a chemiluminescence nitric oxide monitor (21,22) and a
thermal NH~ converter coupled with an NH3 scrubber has been reported which
would potentially have the sensitivity and specificity needed for accurate
NH~ measurement at the 1 ppb level.

-------
                                 SECTION III
                                 DISCUSSION

     Measurement of gaseous NH~ using a chenriluminescence nitric oxide
monitor with a thermal converter has been discussed previously by Breitenbach
and Shelef (23), Sigsby et al.  (24) and Hodgeson et al.  (25). The work by
Breitenbach and Shelef, and Sigsby et al. centered on measuring NHU in
auto exhaust where sensitivity was not critical, and the NH3 to NO converter
materials selected for this application were not thoroughly tested for use in
atmospheric monitoring.  The measurement system described by Hodgeson measured
atmospheric NO  and NFL concentrations.  However, the thermal converter
material used in this study, was gold wool, which subsequently was found to
have widely variable conversion properties.
     Laboratory evaluation of the chemiluminescence system as described by
Hodgeson verified its potential as an atmospheric NH3 monitor but indicated
that considerable optimization of the system would be necessary.  Components
of the measurement system which would need optimizing included the chemi-
luminescence NO detector, the NH3 to NO converter, and the NH3 scrubber.
     Evaluation of commercially available chemiluminescence NO monitors used
in ambient monitoring of NO  showed a sensitivity of 5 to 10 ppb to NO.
                           /\
Early development work on the chemiluminescence  NO detector by Fontijn et al.
(21) and Hodgeson et al. (25) report that a sensitivity of 1-ppb is possible
with the detector.  With reported ambient NH3 concentrations in the range

-------
of 1 to 10 ppb, a detection system with a sensitivity of less than 1 ppb



would be necessary to achieve the resolution for accurate monitoring of



NH3.



     Both Sigsby and Hodgeson approached the measurement of NhU as a sub-



tractive process using one chemiluminescence detector, a thermal converter



and an acidic scrubber.   The accuracy of the measurement of NH- depends on



the ability of the scrubber to completely remove the NH3 while quantitatively



passing the NOp and NO under widely varying atmospheric conditions of high



NO, NOp and relative humidity.  Variations in NO and NOp concentration in the



atmosphere make NHL determinations difficult.  Even with sample integration,



NO and NOp concentrations can vary 5 to 10 times the ambient NhL concen-



trations.   A different approach would therefore be needed to successfully



separate the signal  due to NH^ from the NO and NOp signal.



     A method for separating the signal generated by NH- from that generated



by NO and N09 would be to generate separate and continuous  signals for N0¥
            £                                                            /\


and for NOV plus NH,.  With a continuous signal for both NOV and for NOV plus
          X        O                                       XX


NH3, fluctuations of NO and NOp could be monitored by one signal while the



NH3 was being alternately measured and scrubbed in the second channel.



Different approaches have been demonstrated for generating  two continuous



signals from the same measurement system.  One approach would be to use two



chemiluminescence detectors, two phototube assemblies and two signal processors



with a split sample stream to give separate signals.   Another technique



would be to use three detectors, one phototube, and an optical chopper to



look at each detector alternately, with a portion of the sample stream going



to each detector.  A third possibility employs one detector, one phototube



and mechanically switches the sample stream from one flow arrangement to

-------
another, alternately looking at flow through one converter, then another.
Each of these approaches has been integrated into a viable chemi luminescence
monitor for NO and NCL.  All of these detection systems use a split sample
stream, and the placement and type of converter for each part of the sampling
system becomes important.
     An example of signal separation using two chemi luminescence detectors
and two converters is given in Figure 1.  The sample stream is split with
part of the sample passing through a low temperature converter to convert N02
to NO and part of the sample passing through a high temperature converter
where both N02 and NFL are converted.  If the conversion efficiency for N0?
is equal in both converters, the signals would be equal except for the
signal generated by the NH^.  By placing the NHg scrubber system in the
channel with the high temperature converter, verification of the NH3 can
be obtained when a difference in signals is observed.  When the sample is
flowing through the NHL scrubber, the NO  fluctuations can be monitored in
                      0                 X
the second channel and signals should be identical.  A different scrubber
arrangement with the two detector system would have two high temperature
converters which would convert both NO^ and NHo.  One portion of the sample
stream would be split and an NH3 scrubber system would be placed upstream
of one converter.   When the scrubber system was in the by-pass mode
the two signals would be identical; with the scrubber in line, the difference
would be the NH- scrubbed out in one channel.  Either of the two methods
would require a monitoring system with two continuous signal outputs:
one for NO . and one for NOV plus NHo.
          X                X        »5
Detector Design
     The sensitivity of the system can be enhanced either by modifying the NO
                                     8

-------
detector cell described by Hodgeson et al.  (25), or complete redesign of the
detector.  Optimization and modification of previous cell designs was attempted
to increase sensitivity.  The cell used is shown in Figure 2.  A diagram of
the total system is shown in Figure 3.  The cell as shown was positioned
within 2" of an EMI, S-20, extended-red photomultiplier tube. The tube was
cooled to -20°C.  The signal from the detector was sent to a Keithley Model
417 picoammeter.  A Spectrum Model 1021 filter was added to the output of a
picoammeter prior to being sent to a recorder.  Ozone for the NO-ozone re-
action was produced using a Model 100 Thermo Electron ozone generator.
Optimum flow conditions for the cell were an ozone partial pressure of 5 torr
and sample flow of 10 torr (150 cc/min).  Figure 4 shows a typical response
for the detector.  Maximum sensitivity of 5 ppb full scale with a 2% noise
was possible with this design.
     Independent of work within the Environmental Protection Agency on NO
detector cell modification, Aerochem Research Labs designed and fabricated an
NO reaction cell which differed from previous designs (26).  Using a spiral
reaction cell to increase residence time for the N0-0g reaction near the face
of the photomultiplier tube and high sample flow, the sensitivity of this
detector was determined to be 0.5 ppb.  A diagram of the reaction cell is
shown in Figure 5.  For field measurement for ambient NH3> the Aerochem
detector has the advantage of a high sample through-put to minimize sample
loss through the system and near ambient pressure operation to reduce instru-
ment size.
     Newly available commercial NO, NO  monitors which have the necessary
                                      /\
design features along with sensitivity to NO of <5 ppb include the Thermo
Electron Model  14T and the Columbia Scientific  Industries Model 1600.  The

-------
Thermo Electron monitor uses one phototube, three detectors and a light
chopping technique to look at each reaction cell alternately.  The sample
stream is split with part of the flow going to each cell.   (See Figure 6.)
The CSI monitor has one phototube, one cell and a valve which rapidly switches
different portions of a split sample stream into the detector as shown in
Figure 7.  The design of each of these analyzers would allow for separate
signals for NO  and NO  plus NH~.
              XX        0
Ammonia Converter
     Thermal conversion of NH- to NO using metal surfaces  has been reported
by a number of researchers (22,23,24).  Stainless steel heated to 800°C has
been shown to quantitatively convert NH~ to NO.  Data reported was generally
for concentrations higher than atmospheric.  Testing of stainless steel
indicated an initial quantitative conversion of NHo in the range of 1 to
20 ppb; however, prolonged heating of the stainless steel  tubing resulted in
a decrease in conversion efficiency.  Repetitive tests with different types
of stainless steel tubing gave the same results.  A different converter
material was necessary to assure adequate conversion efficiency for NH, at
atmospheric concentrations.
     Previous work by Bell (personal communication) indicated that platinum
gauze heated to a temperature greater than 300°C would convert NH3 to NO.
Evaluation of a thermal converter using a platinum catalyst showed that at a
                                                                         o
temperature of 800°C, NH-, concentrations of 1 to 50 ppm (.69 to 34.5 yg/m )
could be quantitatively converted to NO.  Repetitive tests using this con-
verter indicated sufficient lifetime (6 months) for use in a continuous
monitoring system.
                                     10

-------
Ammonia Scrubber
     The use of a chemical scrubber to remove NH-, from an air stream was
previously discussed by Sigsby et al.  (23) and Hodgeson et al. (24).  The
scrubber used by Hodgeson relied on diffusion of NHU from a sample stream to
the wall of a glass tube coated with phosphorous acid.  Durham (27) charac-
terized the conditions necessary for removal of sulfur dioxide by diffusion
from an air stream using a chemically coated tube.  The same equations can be
used to determine the optimum conditions for removal of NH3 using a coated
tube.  For removal of NH3 in an air stream of fixed laminar flow, the tubing
diameter and tubing length can be set to assure that NH, is quantitatively
removed.
                                     11

-------
ro

                             NO,CONVERTER
          SAMPLE IN-
                SCRUBBER

                SYSTEM
            BYPASS
AMMONIA
SCRUBBER
                                     SAMPLE-
                                                           DETECTOR 2
                                                            DETECTOR 1
CONVERTER
                                                                •SAMPLE-
                                                                            VACUUM
                                                                            -^
                                                                             PUMP
                                                                    OZONE IN
                              Figure 1.  Flow diagram of ammonia measurement system.

-------
                      600 r,m CUT OFF FILTER
OZONE X
SAMPLE A"
     EXHAUST pg^^'''''-'''"'-''^^^
     (VACUUM)'
                                 -8 cm
                         GLASS CELL
7
                                                           COOLED PHOTOTUBE HOUSING
                                                              PHOTO MULTSPLIER TUBE
                             Figure 2. Diagram of NO reaction cell.

-------
           02
             OZONE
           GENERATOR
CHEMILUMINESCENCE REACTION CELL
             SAMPLE
             SOURCE
      COOLING
       WATER
       BATH
                                           COOLED
                                      P.M. TUBE HOUSING
                                     ooooooooo
          A
         V
PHOTO-MULTIPLIER
     TUBE
                                  I O O O O O O O  O O O
VACUUM
                                            HIGH
                                          VOLTAGE
                                           POWER
                                           SUPPLY
                                                                        PICO-
                                                                      AMMETER
                                                                       FILTER
                                                                     RECORDER
            Figure 3.  Diagram of sensitive chemiluminescence NO monitor.

-------
CO


o
a.
CO
LLJ
QC

LU
                           I

                 5 ppb NITRIC OXIDE
                          15
    10

(TIME, minutes
                           Figure 4.  Response of optimized NO detector.

-------
 SPIRAL REACTION CHAMBER
       GOLD COATED
CUTOFF FILTER (600 nm)
                                       CROSS SECTION
                              Figure 5.  Single spiral reactor.
                                        16

-------
                         SAMPLE IN
AIR IN

i

NOXNH3
CONVERTER

s-t-
p
•

NOX

CUNVtH.bR VACUUM
PUMP

wtttitiniti)

i ;[-
fy////////////<
B^\ I
j{
V////M////,
OZONb J | ;TJ
GbNbKATOR
Ci ..... •• ,,,,,, fj
/{((({(//f//f//t

c
PHOTO MULTIPLIER
TUBE
LIGHT CHOPPER
                                         v TRIPLE
                                       REACTION CELL
                                                                                               NO    NOX   NH3
                         Figure 6.  Diagram of Thermo Electron Model 14T NO, NOX, NH3 Monitor.

-------
              DRY
              AIR
                                                                  HEATED AND
                                                                  REGULATED
                                                                     R.C.
          EXHAUST-*
CO
           SAMPLE
            GAS
           INLET
           ^»-  r«n i ii*Uhn i *•	,   Dcp
                  FILTER          /  |
                                               CAPILLARY FLOW
                                              CONTROL(OZONE)
                                                CAPILLARY
                                               FLOW CONTROL
MOTORIZED
  3 WAY
  VALVE
                                          THERMOELECTRICALLY
                                        /COOLED AND REGULATED
                                            PHOTO MULTIPLIER
                                                 TUBE

                                        OPTICAL
                                    |   FILTER

HIGH
VOLTAGE


ELECTROMETER
   CIRCUIT
                                                                                         NO RECORDER
                                                                                            OUTPUT
                                                                      N02RECORDER
                                                                         OUTPUT
              NOX RECORDER
                 OUTPUT
                                                                                       0.01 VDC^I.O VDC  0.01 VDCJ 1.0 VDC 0.01 VDC| 1.0 VDC
                                                                                           0.10VDC         0.10VDC        0.10VDC
                                                    Figure 7.  Diagram of CSI Model 1600 NOX Monitor.

-------
                                 SECTION IV



                                EXPERIMENTAL





     A prototype NO , NH~ monitor was fabricated by Aerochem and delivered to
                   X    «j


EPA under Contract 68-02-2424.   The monitor is equipped with two of the



spiral reaction chambers discussed earlier in this report.   (See Figure 8.)



Sample is pulled into the instrument through a common inlet and is split as



shown in the diagram.  The sample flow into each reaction chamber is 2.5



1/min.  Sample passes through two high temperature platinum converters heated



to 1000°C.  Air is pulled through the ozonator at a flow of 360 cc/min and



then is split to go to each reaction chamber to give an ozone flow of 180



cc/min.  Each detector uses an  extended-red trial kali photomultiplier cooled



by a thermoelectric cooled housing to 5°C.  The monitor has signal output



ranges of 0-5 ppb, 0-20 ppb, 0-50 ppb, 0-200 ppb, 0-500 ppb for NO  and NO
                                                                  X       X


plus NH3.



     An NH_ scrubber was designed and fabricated under contract to Research



Triangle Institute to be used with the NO -NH3 monitor built by Aerochem.



Quantitative removal of NH3 at  a total sample flow of 5 1/min would require a



coated glass tube 300 cm in length and 1/4" I.D.  In order to make a more



compact unit for field use, a bundle of 10 tubes 30 cm in length and 1/4"



I.D. was used as shown in Figure 9.  The ends of the bundle were placed in a



teflon frame which was clamped to an outer cap.  The scrubber was placed in a



box that was constructed as shown in Figure 10. Two teflon solenoid valves



were used to direct the sample flow through the scrubber or by-pass.  The



                                     19

-------
solenoid valves were connected to an electronic timer to automatically switch
the sample through the scrubber or the by-pass. This interval  could be varied
from 2 sec to 60 min.
     The total NhU measurement system was assembled as shown in Figure 11.
Sample lines upstream of the NH-, converters were heated to prevent loss of
NH3.
     Two calibration systems were fabricated to deliver known  concentrations
of NO, N02> and NH3 over a range of 1 to 500 ppb.  The calibration system
for delivering known concentrations of NO and N02 was designed similar
to that described by Rehme et al. (28).   Figure 12 is a flow diagram of the
system modified to introduce water vapor to regulate humidity.  An EGG Model
680 Dew Point Hygrometer was used to measure the relative humidity of the
calibration gases.
     The second calibration system was a double dilution permeation system
built by Research Triangle Institute (RTI) (29).  This system was fabricated
to meet the need for accurately generated ammonia concentrations as low as 1
ppb.  A flow diagram of the system is shown in Figure 13.  A supply of zero
air flows into the system where it is split and routed to two separate
automatic flow controllers.  One controls air over the permeation tube, the
other controls the flow of dilution air.  Air flow over the permeation tube
can be regulated between 200 cc/min and 10 1/min.  A splitter sends 100
cc/min of the effluent from the permeation tube chamber to a chamber where it
is mixed with the dilution air stream.  Dilution air flow can be varied from
200 cc/min and to 10 1/min.  The permeation tube is housed in a constant
temperature oven, and controlled to within ±.01°C over a ±10°C ambient tempera-
ture change.  The temperature of the oven can be varied from 20°C to 30°C.  A

                                     20

-------
functional diagram of the system is shown in Figure 14.  A digital panel
meter allows direct readout permeation flow, dilution flow and oven tempera-
ture.
                                     21

-------
ro
       SAMPLE
                                                                          —AIR
                                   NOX
                                CONVERTER
                                                          VSINGLE
                                                            SPIRAL
                                                           REACTOR
                                                                                                     ELECTRONICS
                                                                                                  I
                                                                                                  I
                                                                                                  I
                                                                                                 NO,
   I

NOX + NH3
                                    Figure 8.  Flow diagram of Aerochem chemiluminescence monitor.

-------
OJ
           6 mm 1.0.
        TEFLON SEALED
                                                                    -30cm-
                                                            10 TUBES,
                                                       6mm DIA.x30cm. LONG
                                                          PRESS FIT TO
                                                        TEFLON END PIECE
                                                     _Q
                                                      \
                         50 mm JOINT, W/"0" RING SEAL
                     CLAMPL =50 PINCH TYPE W/SCREW LOCK
                                                                                  60mm DIA.
                                                                                                       8 mm
                                                         Figure 9. Diagram of ammonia scrubber.

-------
SAMPLE
 INLET
                                 THREE-WAY SOLENOID
                                       VALVES
                   TIMER/
                 CONTROLLER
              I
                  POWER
                 SUPPLY
                                                      RELAY
                                                      SWITCH
                                                     FUNCTION
                                                      SWITCH
SAMPLE
OUTLET
                Figure 10. Diagram of ammonia scrubber system.
                                       24

-------
             SAMPLE
              INLET
CO
Q
no
en
                                       N02, NH3
                                      CONVERTER
                            i SOLENOID
                               VALVE
                                    SIGNAL
                                    OUTPUT
                                   NOX,NH3
                          O
                   C/J-
                   V>-
                   <:.
                   o..
                   >•
                   en;
         NH3
      SCRUBBER
                                   J
 N02. NH3
CONVERTOR
                                                       AIR OR .
                                                      OXYGEN
                                                                   PHOTO
                                                                 MULTIPLIER
                                                                    TUBE
                                                                   PHOTO
                                                                 MULTIPLIER
                                                                    TUBE
                                                             VACUUM
                                                 OZONE
                                               GENERATOR
                  Figure 11. Diagram of ammonia measurement system using two separate detectors.

-------
                     MIXING BULB
                     MIXING BULB
ro
             f
           a
           LU
                                     NO,N02
 NO
CYL.
             OZONE
          GENERATOR
                        FLOW CONTROL (x)
                          DILUTION AIR
                                 SPLITTER
 SAMPLE
MANIFOLD
                                                                        VENT
                                                                  DEW POINT
                                                                 HYGROMETER
          Figure 12.  Diagram of modified NO, NO2 calibration system (gas phase titration).

-------
        PERMEATION
           TUBE
          CHAMBER
             I
             I
  MASS
  FLOW
CONTROL
  MASS
  FLOW
CONTROL
                      INSULATED BOX
                                                     PRESSURE
                                                       GAGE
 0	
CONTROL
 VALVE
                                CAPILLARY
                               RESTRICTION
              EXHAUST
           <
           Q.
                                                                O
                                                                cc
               SAMPLE
               OUTLET
                                                                    AIR
                                                                    INLET
   Figure 13. Diagram of RTI double dilution permeation system.
                                27

-------
                              THERMALLY INSULATED
                                    HOUSING
 AIR  ,
INLET
                              !   DILUTION
                       CARRIER
                        FLOW
                                  FLOW
CAPILLARY
          €3*
   PERMEATION

  DEVICE CAVITY
                                                     PRESSURE
                                                       GAGE
                            BEAD
                        THERMISTORS
                               THERMISTOR
                                 BRIDGE
                                 CIRCUIT
DIGITAL
PANEL METER
I

POWER SUPPLY
                                                                   SAMPLE
                                                                    FLOW
                                                                  CONTROL
                     Figure 14.  Function diagram, RTI permeation system.
                                      28

-------
                                  SECTION V
                                  RESULTS
     Initial calibration of the Aerochem monitor was accomplished using known
NO and N02 concentration from 2 to 50 ppb.  The response to both NO and N02
was determined to be linear with a noise equivalent to 0.25 ppb on a range of
50 ppb full scale.  Concentrations up to 50 ppb of N0« were generated and the
conversion efficiency of N02 to NO through the high and low temperature
converters was determined to be quantitative (>98%).
     Concentrations of NH3 in dry air were generated in the range of 1.5 to
50 ppb using the RTI double dilution permeation system.  The high temperature
converter quantitatively converted the NH3 concentrations while no conversion
was indicated in the low temperature converter.  A linear response for NFL
was obtained from 1.5 to 50 ppb.  Concentrations of NO and N02 were then
generated using dry air and were allowed to pass through the ammonia scrubber
where no decrease in response for NO or in conversion efficiency for N02 was
noted.  This was true for both channels and both high and low temperature N02
converters.
instrument alternately through the scrubber and by-pass.  An example of
output signal is shown in Figure 15, the 25 ppb concentration of NH3 was
quantitatively removed in the scrubber mode with a cycle time of 10 minutes.
     Humidified concentrations of NO and N02 from 2 to 50 ppb were then sent
through the scrubber and converters into the instrument.  The humidity
was varied from dry air (RH <5%) to a relative humidity of 70%.  Both NO and
                                     29

-------
NC^ passed quantitatively through the scrubber and conversion of NCL was
quantitative in each converter.   No losses of humidified NO or NOp were
observed through the scrubber.   Humidified concentrations of NHL from 2 to 50
ppb were generated with the modified permeation system.   Humidified samples
of NH3 passed quantitatively through the scrubber by-pass into the instrument
and were completely removed passing through the scrubber.
     Once the tests were completed using a high and low temperature con-
verter, two high temperature converters with the scrubber system in line with
one converter were subjected to  the same series of tests with humidity.  The
results are shown in Tables 1 and 2.
     The difference in signal output using either two high temperature con-
verters or a high and a low temperature converter with the scrubber system is
shown in Figure 16.  Part A of Figure 16 shows that when using two high
temperature converters, the magnitude of the signal is equal  in both channels
except when the sample passes through the scrubber to remove NH-j.  Part B of
Figure 16 shows a signal using a high and low temperature converter.  The
signals are normally different by the amount of NH3 present and are equal
when the sample passes through the scrubber and the NHo is removed.
     Once the NH3 measurement system calibration and performance tests were
completed, the system was used to sample ambient air inside the EPA labora-
tory and was also set up in a van in the parking lot of the EPA facility at
Research Triangle Park, N.C.  Figure 17 is a recorder trace of the ammonia
and NO  monitored in the parking lot.  The measurement system was set up with
      A
a high temperature converter in  each channel with the scrubber system in
channel one.  The outputs are equal except when the scrubber is in line and
                                     30

-------
the NHo is being removed  by  the  scrubber.   The NOV plus NI-U value  is approxi-
      O                                           X        -j


mately 20 ppb.  Five ppb  of  the  total  signal  is NM-.
                                      31

-------
                              — CHANNEL 1
                                LOW TEMPERATURE CONVERTOR
                              ...CHANNEL 2
                                HIGH TEMPERATURE CONVERTOR
             10
   20
TIME, minutes
30
                                                        40
Figure 15.  Ammonia response from optimized measurement system.
                             32

-------
o
a.
CO
UJ
CC
   PART A.
                                                    . CHANNEL 1
                                                     HIGH TEMPERATURE CONVERTOR
                                                    • CHANNEL 2
                                                     HIGH TEMPERATURE CONVERTOR
                                            I
  30
            25
20
    15

TIME, minutes
                                                      10
CO

o
&
CO
UJ
OC
   PART B.
NOX+NH3
                                               	CHANNEL 1
                                                     LOW TEMPERATURE CONVERTOR
                                               	CHANNEL 2
                                                     HIGH TEMPERATURE CONVERTOR
  30
            25
20
                  10
                                        15
                                 •^ TIME, minutes

Figure 16.  Ammonia response using different converters with the ammonia scrubber system.
                                            33

-------
                25
                20
                                                                       CHANNEL 1
                                                                       HIGH TEMPERATURE CONVERTOR
                                                                       CHANNEL 2
                                                                       HIGH TEMPERATURE CONVERTOR
              o
              <
                IS
CO
              LU
              O
              z
              o
              u
                10
                                                                                        AMMONIA
                                                                                        PRESENT
                                                                                    '   J
                                               W/SC RUBBER
                                                  NOX
                  30
25
20
     15
[TIME, minutes
10
                           Figure 17. Ambient measurement of ammonia with optimized system.

-------
TABLE 1.   SYSTEM RESPONSE TO HUMIDIFIED SAMPLES OF NITRIC OXIDE AND
           NITROGEN DIOXIDE HIGH TEMPERATURE CONVERTERS

% Relative Humidity


5%



50%



70%

Cone, ppb Response (% Chart) Converter Mode
22 (NO)
22 (NO)
22 (N0~)
22 (NOg)
25 (NO)
25 (NO)
25 (N09)
25 (NOg)
22 (NO)
22 (NO)
22 (N09)
22 (NOg)
45
45
45
45
50
50
50
50
44
44
44
44
HTC
HTC
HTC
HTC
HTC
HTC
HTC
HTC
HTC
HTC
HTC
HTC
By- Pass
Scrubber
By- Pass
Scrubber
By-Pass
Scrubber
By-Pass
Scrubber
By-Pass
Scrubber
By-Pass
Scrubber

TABLE 2.  SYSTEM RESPONSE TO HUMIDIFIED SAMPLES OF NITRIC OXIDE AND
 NITROGEN DIOXIDE HIGH (HTC) AND LOW (LTC) TEMPERATURE CONVERTERS

% Relative Humidity


20%



45%



70%

Cone, ppb Response (% Chart) Converter Mode
22 (NO)
22 (NO)
22 (NO,)
22 (NOg)
23 (NO)
23 (NO)
22 (N0«)
22 (NO^)
20 (NO)
20 (NO)
20 (N0~)
20 (NO^)
45
45
45
45
46
46
45
45
40
40
40
40
LTC
HTC
LTC
HTC
LTC
HTC
LTC
HTC
LTC
HTC
LTC
HTC
By-Pass
Scrubber
By-Pass
Scrubber
By-Pass
Scrubber
By- Pass
Scrubber
By-Pass
Scrubber
By-Pass
Scrubber

                                 35

-------
                                 REFERENCES


1.   Robinson, E., and R.C. Robbins.   Sources,  Abundance and Fate of
     Gaseous Atmospheric Pollutants.   Stanford  Research Institute, Menlo
     Park, California, 1978.  pp. 71-73.

2.   French, J.C., G. Lowrimore, W.C. Nelson, J.F.  Finklea,  T.  English,  and
     M. Hertz.  Arch. Environ.  Health, 27:129,  1973.

3.   Junge, C.E.  Air Chemistry and Radioactivity.   Academic Press, New  York,
     New York, 1963.  pp. 87-91.

4.   Pate, J.B., and J.P. Lodge.  Atmospheric Trace Constituents in the
     Humid Tropics IV. Environmental  Measurement in the Tropics.  Presented
     at the 9th Conference on Methods in  Air Pollution and Industrial
     Hygiene Studies, Pasadena, California, 1968-   pp. 7057-7064.

5.   Healy, T.V., H.A.C. McKay, and A. Pilbean.  Ammonia and Related Atmos-
     pheric Pollutants at Harwell.   AERE-R6231  AERE,  United  Kingdom Atomic
     Energy Authority Research  Report, Harwell  Bervshire, U.K., 1972.  pp.  1-20.

6.   Breading, R.J., J.P. Lodge, J.B. Pate, D.C. Sheesky, H.B.  Klonis,
     B. Folger, J.A. Anderson,  T.R. Englevt, P.L.  Haagensov, R.B. McBeth,
     A.L. Morris, R. Pogue, and A.S.  Wartburg.   J.  of Geophysical Research,
     78(30):7057-7063, 1973.

7.   Jacobs, M.B.  The Chemical Analysis  of Air Pollutants.   Interscience,
     New York, New York, 1960.   pp. 216-218.

8.   Morgan, G.B., C. Golden, and E.C. Tabor.  New and Improved Procedures
     for Gas Sampling and Analysis  in*the National  Air Sampling Network.
     J. Air Pollution Control Assoc., 17:300-304,  1967.

9.   Buck, M., and H.Z. Stratmann.   The Use of  the Impinger  Principle  for
     the Determination of Ammonia in the  Atmosphere.   Z. Anal.  Chemie,
     213:241,  1965.

10.  Stucliffe, R.A., and G.A.  Jones.  Research Into Some Methods of Analysis
     of Sewage.  Wat. Pollut. Control, 67:209-220,  1968.

11.  Tetlow, J.A., and A.L. Wilson.  An Absorptiometric Method  for Determining
     Ammonia in Boiler Feed Water.   Analyst, 80:453-465, 1964.

12.  Leithe, W., and G. Fetch.   Anal. Chem., 230:344-347, 1967.


                                     36

-------
13.   Kothny, E.L.  Health Lab.  Science,  10(2):115-118,  1973.

14.   Georgi, H.W., D.  Jost,  and W.J.  Muller.   Test of a Method  for
     Measuring Ammonium and  Ammonia Concentrations in the  PPB Range.
     Report No. 25. Institute of Meteorology  and Geophysics, University
     of Frankfurt, Frankfurt, Germany,  1973-   pp.  1-18.

15.   Okita, T., and S. Kamamori.   Atmospheric Environment,  5:621-627,
     1971.

16.   Kruse, J.M., and M.G. Mellon.  Colcrimetric Determination  of Ammonia
     and Cyanate.  Anal. Chem., 25(8):1188-1192, 1953.

17.   Vivameu, D.M., and J. Sech.   Anal.  Chem.,  44(2):395-398,  1972.

18.   Schiendrikar, A.D., and J.P. Lodge.   Atmospheric Environment,
     9:431-435, 1975.

19   David, D.J., M.C. Wilson,  and D.S.  Ruffin.   Analytical  Letters,
     9(4):389-404, 1976.

20.   Sawicki, E., J. Mulik,  and E. Wittenstein (Eds.).   Ion  Chromatograph
     Analysis of Environmental  Pollutants. Ann  Arbor Science Publishers,
     Ann Arbor, Michigan, 1978, pp. 41-51.

21.   Fontijn, A., A.J. Sabadell,  and R.  Ronco.  Anal. Chem., 42(6):575-
     579, 1970.

22.   Stedman, D.H., E.E. Darby, F. Stull, and H.J. Nikki.   Air  Pollution
     Control Assoc., 22:260, 1972.

23.   Breitenbach, L.P., and M.  Shelef.   Air Pollution Control Assoc.,  22(2):
     128-131, 1973.

24.   Sigsby, J.E., F.M. Black, T.A. Bellar, and  D.L.  Klosterman.   Environ.
     Science and Tech., 7:51-54, 1973.

25.   Hodgeson, J.E., K.A. Rehme, B.E. Martin, and R.K. Stevens. Measurements
     for Atmospheric Oxides of Nitrogen and Ammonia by Chemiluminescence.
     Paper No. 72-12.  Air Pollution Control  Assoc. Meeting, Miami, Florida,
     1972.

26.   Volltrauer, H.N.   Instructions Manual for Ultra Sensitive  NO/NO
     Monitor, Aero Chem.  Research Laboratories,  Inc., Princeton,  New Jersey,
     1976.

27.   Durham, J.C., W.E. Wilson, and E.B. Bailey.  Continuous Measurement of
     Sulfur  in Submicrometric Aerosols.  EPA 600/3-76-088.  U.S.  Environ-
     mental  Protection  Agency, RTP, North Carolina,  1975.  pp.  1-25.
                                     37

-------
28.   Rehme, K.A., B.E.  Martin,  and J.A.  Hodgeson.   Tentative Method for the
     Calibration of Nitric Oxide, Nitrogen Dioxide, and Ozone Analyzers by
     Gas Phase Titration.  EPA-R2-73-246.   U.S.  Environmental Protection
     Agency, RTP, North Carolina, 1974.

29.   White, H.,  and R.  Strong.   Double Dilution  Permeation System, Task No.
     2, Contract EPA 68-02-2291, Research  Triangle Institute, RTP, North
     Carolina,  1976.
                                    38

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
 REPORT NO.
 EPA-600/2-79-028
                             2.
                                                          3. RECIPIENT'S ACCESSION>NO.
 TITLE AND SUBTITLE
  OPTIMIZED CHEMILUMINESCENCE SYSTEM FOR MEASURING
  ATMOSPHERIC AMMONIA
            5. REPORT DATE
            february
1979
            6. PERFORMING ORGANIZATION CODE
 AUTHOR(S)
  Ralph E. Baumgardner,  W.A.  McClenny and R.K.  Stevens
                                                          8. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING ORGANIZATION NAME AND ADDRESS
  Environmental  Sciences Research Laboratory
  Office of Research  and Development
  U.S. Environmental  Protection Agency
  Research Triangle Park, N.C.  27711	
             10. PROGRAM ELEMENT NO.

              1AD712  BB-12  (FY-78)
             11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
  Environmental  Sciences Research Laboratory  --  RTP,  NC
  Office of Research  and Development
  U.S. Environmental  Protection Agency
  Research Triangle Park, N.C.  27711	
             13. TYPE OF REPORT AND PERIOD COVERED

               Tn-hnn<;p	
             14. SPONSORING AGENCY CODE
               EPA/600/09
 5. SUPPLEMENTARY NOTES
 6. ABSTRACT
       The  optimization and  testing of a  continuous measurement system for  analyzing
  atmospheric  ammonia  concentrations  (0  to  10 ppb)  is described.   The  measurement
  system combines an ultra-sensitive chemiluminescence nitric  oxide detector,  with a
  thermal  converter  for NH~ to  nitric oxide  (NO)  conversion,  and an acidic  scrubber
  for alternate  removal of NtU from the sample stream.

       The  chemiluminescence  monitor with a  high  temperature  converter  will  give a
  combined  measurement  of NO,  N02 and NH,.   Other nitrogen  containing compounds,  if
  present,  may also be converted.  By placing an acidic scrubber  in  the  sample line,
  the ammonia  will be  removed and  a  combined measurement of NO   (NO  and  NOp) will  be
  obtained.   By  taking the  signal difference when the scrubber  is  in  or out  of the
  sample line,  the concentration  of NH~ can be determined.

       Performance tests were conducted on the total measurement systems to verify its
  capability to  measure NO, NO^ and NH- under laboratory conditions as well as ambient
  monitoring conditions.   A  dynamic calibration system capable  of generating 1 ppb of
  NH,  was   used  for  the testing.    The  calibration  system  is  a  double  dilution
  permeation system.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                             COS AT I Field/Group
  *Air  pollution
  "'Ammonia
   Nitrogen oxides
  "Monitors
  *Chemi1umi nescence
   Evaluation
                               13B
                               078

                               07D
                               20F
18. DISTRIBUTION STATEMENT
   DPI (TACIT TO DIIRI TT
19. SECURITY CLASS (ThisReport)
     21. NO. OF PAGES
          45
                                              207S"E"CTjRIT? 'CLASS (This page)

                                                   ACCTCTCn	.
                           22. PRICE
EPA Form 2220-1 (9-73)
                                            39

-------