United States
 Environmental Protection
Environmental Sciences Research'
Laboratory                _
Research Triangle Park NC 2771
 Research and Development
EPA-600/S2-81-162  Jan. 1982
Project  Summary
A Water  Vapor  Monitor
Using   Differential
Infrared  Absorption
 D. E. Burch and D. S. Goodsell
  A water vapor monitor has been
developed with adequate sensitivity
and versatility for a variety of applica-
tions.  The  instrument  has  been
designed  for  the  continuous
monitoring of ambient air and the
measuring of the mass of H2O desorbed
from aerosol filters. The' sample gas
may be held static, or it may flow con-
tinuously through the 56 cm3 sample
cell, which is temperature controlled
at 45C.  Infrared energy from  a
tungsten-iodide bulb passes through a
rotating filter wheel  and the sample
cell to a PbS detector. The infrared
beam passes through the sample gas
twice to produce a total optical path of
40 cm. As the filter wheel rotates at
1800 rpm, the infrared beam passes
alternately through two semicircular
narrow  bandpass  filters;  one  is
centered in a spectral region of strong
absorption and the other is centered
nearby in a region of weak absorption.
Absorption by the water vapor in the
sample produces a 30-Hz modulation
of the detector signal that is propor-
tional to the HgO concentration. The
zero-setting of the monitor is stabi-
lized by controlling the temperatures
of the detector and  the filters. The
r.m.s. noise level corresponds to ~3
ppm of HaO. The maximum concen-
tration  that  can  be  measured
accurately is -5%; higher concentra-
tions could be measured by shortening
the sample cell.
  This Project Summary was develop-
ed by EPA's Environmental Sciences
Research Laboratory, Research Tri-
angle Park, NC, to announce key find-
ings of the research project that is fully
documented in a separate report of the
same title (see Project Report ordering
information at back).

  The  U.S. Environmental Protection
Agency (EPA)  would like  to obtain the
mass of H20 associated with ambient
air particulate matter that has  been
collected on Teflon filters. This meas-
urement is needed to estimate the
contribution of H2O to the deficit in mass
between the  total measured'aerosol
mass and the mass that  can be attri-
buted to currently measurable particu-
late components. This deficit is as much
as 30% in some cases.
  A positive and quantitative identifica-
tion of H20 poses an analytical problem
for several  reasons: (1) the amount of
H2O is  small  i.e., 30% of a typical
loading of 500 mg of aerosol per filter is
only 150 mg; (2) the  H20 associated
with the particulate matter is not acces-
sible; (3) volatilization of H2O from the
filter by heating also volatilizes  other
aerosol components so that a simple
measurement  of weight  loss is not
acceptable as  an indication of H20 on
the filter. However, volatilization of the
H20 coupled  with  a sensitive H2O
monitor capable of a real-time output
was conceived as a tenable approach
provided that the H20 vapor concentra-
tion was sufficiently high (parts-per-
thousand  range)  to  achieve  a

reasonable signal to noise as an output
  Previous research  for the EPA con-
cerned with the design and fabrication
of prototype instruments for the detec-
tion  of  CO and  CH4  at ambient air
concentrations   resulted   in   highly
successful  procedures  for  using
infrared  absorption  as the basis  for
sensitive  and  specific  ambient  air
monitors. Considerations based on this
analogous work indicated that the H20
could be detected if thermal desorption
over a period of one to two minutes
occurred into a carrier gas volume of 50-
100  cc. In other  work, a prototype unit
capable   of  effecting  the  thermal
desorption of volatile aerosols into  a
sulfur  detector  (for  H2SO4 detection)
was  developed. This prototype was to be
used for the  detection  of  aerosol-
associated H20 by replacing the sulfur
detector with an H2O monitor.

Materials and Methods
  The monitor consists of a radiation
source  (tungsten-iodide   lamp),  an
optical absorption cell,  a  PbS detector
and  associated  optical  and electronic
components. The unit is analogous in
many ways to the family of so-called gas
filter correlation  (GFC) monitors that
have  been  previously designed  and
fabricated  at  Ford  Aerospace  and
Communications Corp.
  The concentration of H2O vapor in the
sample cell of the monitor is determined
by electronically comparing the trans-
mittance of-the sample in two narrow
infrared spectral  intervals. One interval
occurs near 2.59 fjm where H2O vapor
absorbs much more  than in the other
interval   centered   near   2.51   /um.
Samples of gas  containing H2O vapor
may  be introduced into the 20-cm long
sample cell and  kept static during the
measurement, or they may flow contin-
uously through the cell. Good stability of
the zero-setting  and the sensitivity is
achieved  by controlling the tempera-
tures of the sample cell, spectral filters
and  detector. Water vapor  concentra-
tions as  high as 5% can  be measured
with  good  accuracy.  The  minimum
detectable concentration in air samples
at one atm total pressure is less than 10
ppm. Discrimination against interfer-
ence by other gases, particularly those
in the normal atmosphere, is excellent
due to the spectral position and width of
the spectral filters.
  A calibration curve is  provided as part
of the  main report text. Based on the
performance  of  the  monitor  during
calibration, the monitor appears to be an
excellent ambient air monitor, giving a
real-time  output proportional  to  the
ambient H2O content. Combined with
the temperature and pressure sensors
that are included in the instrument, the
absolute and relative humidities can be
obtained. This type of system may offer
significant   advantages  over   other
currently available H20 monitors.

  Variations of the monitor are recom-
mended  for  consideration for many
applications in which the concentration
of H20 vapor is to  be measured. The
required complexity of a monitor of this
type  depends   on   the   sensitivity,
stability,  accuracy and time response
involved. Therefore,  it is not practical to
design a  single  H20 monitor  for all
applications. Some  of the design fea-
tures  that should be considered for a
given  application are as follows:

  1.  The two spectral intervals (A =
     2.59 and 2.51  //m) passed by the
     two  filters  are appropriate  for
     many applications. If a gas species
     in the  samples absorbs strongly
     within  one  of these intervals, a
     different set of intervals should be
     selected. It  may be advisable to
     select another spectral interval to
     replace the 2.59jum interval if the
     absorption is too great. A different
     interval is recommended if pL is
     greater than ~2 atm cm (p is the
     H2O partial pressure and L is the
     optical path length through  the
     sample cell).

  2.  A PbS detector designed for long
     wavelength sensitivity is inexpen-
     sive,  durable,  convenient,  and
     adequate  for  most applications.
     However,  a cryogenically cooled
     PbS, PbSe, or InSb detector could
     provide   higher sensitivity  and
     possibly  better stability  when
     improved performance is required.
     Good stability of the zero-setting
     can  be  achieved  with   a  PbS
     detector  near  room temperature
     only if the detector is temperature

  3.  Controlling the temperature of the
     filters improves the stability of the
     zero-setting and the span calibra-
     tion. However, this is not required
     for many monitoring applications.
4.  The electronics that  process the
    detector signal should automat-
    ically account for small variations
    in   source  radiance,  detectoi
    sensitivity, window transmittance,

5.  The optical path in air outside  ol
    the sample  cell should be kepi
    short, and if very good stability is
    required, should be enclosed foi
    convenient purging or drying with

6.  Gas lines leading to the sample
    cell should be  heated to reduce
    adsorption or condensation of H2C
    on the walls.

7.  The same basic design may be
    used  to  monitor  gases withoui
    drawing them  into a sample cell
    For example, the monitoring bean-
    may  traverse  an  open  volume
    such as an atmospheric pathorar
    exhaust vent.

 D. E. Burch and D. S. Goodsell are with Ford Aerospace & Communications
   Corporation, Newport Beach, CA 92660.
 William A. McClenny is the EPA Project Officer (see below).
 The complete report, entitled"A Water Vapor Monitor Using Differential Infrared
   Absorption," (Order No. PB 82-114 422; Cost: $ 7.50, subject to change} will be
   available only from:
         National Technical Information Service
         5285  Port Royal Road
         Springfield, VA 22161
         Telephone: 703-487-4650
 The EPA Project Officer can be contacted at:
         Environmental Sciences Research Laboratory
         U.S. Environmental Protection Agency
         Research Triangle Park, NC 27711

United States
Environmental Protection
Center for Environmental Research
Cincinnati OH 45268
Postage and
Fees Paid
EPA 335
Official Business
Penalty for Private Use $300
       PS   0000329
       230  S  DEARBORN STREET
       CHICAGO  1L  60604