using molecular sieve to zero infrared gas analyzers for eddy...molecular sieve to remove co 2 and...
TRANSCRIPT
Using Molecular Sieve to Zero Infrared Gas Analyzers for Eddy
Covariance or Atmospheric Profile MeasurementsSteve Sargent, Campbell Scientific, Inc. Logan, Utah, USA
Poster 143
Acknowledgments• Kate Sargent: CAD support for proof-of-concept
design
• Paul Fluckiger: testing cylinders of ultra zero air
and general technical support
• Travis Kendall and Roy Sorenson: in-house
testing of the Zero Air Generator
• Karen G. Wolfe: poster layout and editing
ConclusionsMolecular sieve can safely, easily, and effectively remove CO2 and
H2O from air for zeroing an IRGA. It is available in a variety of
scrubber devices optimized for various applications:
• Remove traces of CO2 and H2O from cylinders of compressed gas
• On-line automated field zero; with or without pump
• Hand-held zero air generator with battery-powered pump
Molecular sieve scrubbers make checking and setting the zero of a
CO2/H2O analyzer easy and convenient, improving data quality and
confidence.
Closed-Path EddyCovarianceThe CPEC200 (Campbell Scientific, Inc.) is a closed-path EC flux
system used for long-term monitoring of atmospherebiosphere
exchanges of CO2, H2O, heat, and momentum. It includes an EC155
CO2/H2O analyzer, CSAT3A sonic anemometer, pump module, and
system enclosure that houses a CR3000 datalogger, and an optional
valve module for on-line automated zero/span.
The CPEC200 Scrub Module is
an accessory to the CPEC200. It
is used to provide a source of
zero air for zeroing the EC155
CO2/H2O analyzer. It uses a
small, low-power (2W)
diaphragm pump that is mounted
in a temperature-controlled and
weathertight fiberglass enclosure.
Ambient air is drawn through
molecular sieve to remove CO2
and water vapor, and push zero
air to the EC155 at
approximately 2 l·min-1 flow.
The pump and its heater are powered only when in use, to avoid any
significant effect on system power consumption.
Only an annual maintenance is required to replace molecular sieve in
one of the three stages.
-1.5
0
1.5
3
4.5
6
7.5
9
10.5
12
13.5
15
-50
0
50
100
150
200
250
300
350
400
450
500
6/4/15 0:57 6/4/15 0:59 6/4/15 1:00 6/4/15 1:01 6/4/15 1:03 6/4/15 1:04
H2O
(p
pt)
CO
2(p
pm
)
Mea
sure
zer
o
Rem
easu
re C
O2
sp
anA
fte
rZe
ro
Set
zero
Eddy Covariance Eddy Covariance
Zero/Span Sequence
Mea
sure
CO
2 s
pan
Bef
ore
Zer
o
Set
CO
2 s
pan
Molecular SieveType 13X molecular sieve is an effective desiccant and adsorber of
CO2. The CAS number for 13X molecular sieve is 6321-69-6. It is
available through Campbell Scientific, Inc. as pn 27450.
Scrubbing Compressed Air
Thirteen cylinders of Airgas Ultra Zero air were tested to verify specifications (CO2 < 1
ppm, H2O < 2 ppm) and to demonstrate the use of molecular sieve as a scrubber. A
cylinder of zero air was connected to two inlets of a CPEC200 zero/span module. One
inlet was connected directly, while the other inlet was connected through a BMT-4. The
datalogger program was modified to alternate four times between these two inlets, such
that the air stream spent 2 minutes on each. The CPEC200 measured the CO2 and H2O in
the air streams. The difference was calculated to quantify the CO2 and H2O in the zero
air before scrubbing by the molecular sieve.
All cylinders tested at CO2 < 0.3 ppm; below specifications. Water vapor was between 0
and 8 ppm, but still well below practical requirements for an IRGA intended for ambient
humidity. In all cases the zero air had equal or less CO2 and H2O after passing through
the molecular sieve.
0
1
2
3
4
5
6
<-0.02 0 0.04 0.08 0.12 0.16 0.2 0.24 0.28 0.32 0.36 >0.38
Nu
mb
er
of
Occ
urr
en
ces
CO2 (ppm)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Nu
mb
er
of
Occ
urr
en
ces
H2O (ppm)
AbstractInfrared gas analyzers (IRGAs) used to measure carbon dioxide (CO2) or water vapor for
eddy covariance (EC) flux or atmospheric profile applications can drift over time. It is
important to regularly monitor and set their zero (offset) and span (gain). The span of an
IRGA is generally more stable than its zero, but even an error in the zero can lead to
errors in EC flux and storage flux estimates.1 Most commonly, the zero and CO2 span are
checked using cylinders of compressed gasses, and water vapor span is checked with a
dewpoint generator.
Cylinders of compressed gases are difficult to transport to a field site because of their
bulk, weight, and safety issues, and dewpoint generators are generally not designed for
field use. Chemicals such as magnesium perchlorate and sodium hydroxide can be used
to scrub water vapor and carbon dioxide from ambient air to zero an IRGA, but these
chemicals are difficult to transport, use, and dispose of safely. Many IRGAs used in the
field are not checked regularly because of these difficulties.
Molecular sieve can provide a safer, more convenient alternative for zeroing an IRGA in
situ. Three example configurations are discussed: a field-rugged scrub module that
includes a 2 l·min-1 pump for automated on-line zeroing of a closed-path EC system, a
field-rugged scrub bottle with no pump for automated on-line zeroing of an atmospheric
profile system, and a light-weight (1.2 kg), hand-held zero air generator with a battery-
operated, 0.2 l·min-1 pump for manually zeroing an open-path IRGA. The molecular
sieve provided air with CO2 and H2O concentrations equal to or lower than ultra-zero
grade air, and required very little maintenance.
Molecular sieve is commonly used to scrub a tank of
compressed zero air of CO2 and H2O. Given a high-
pressure gas, safe operation is a necessary design
feature. Agilent Technologies’ BMT-4 contains 750
ml of 13X molecular sieve, and is made of heavy-
walled aluminum, making it a good choice to scrub
CO2 and H2O from tanks of compressed air.
Atmospheric ProfilesThe AP200 CO2 and H2O atmospheric profile system
(Campbell Scientific, Inc.) measures atmospheric CO2
and H2O at up to eight positions, usually spaced along
a tower to measure a vertical profile. The AP200 is
commonly used with an EC system to measure the
storage term to give a more complete measurement of
surface exchange.
The AP200 includes automatic zero and CO2 span
capability, with zero air provided either with a cylinder
of compressed zero air, or with the AP200 Scrub Bottle.
The AP200 Scrub Bottle uses molecular sieve to
remove CO2 and H2O from ambient air as it is pulled
through the scrub bottle by the AP200 sample pump.
The AP200 Scrub Bottle has two outlet connections:
one for normal operations; and one that allows the user
to assess the state of the molecular sieve. This test port
samples the scrubbed air from part way through the
column of molecular sieve to indicate when to replace
the sieve.
An AP200 Scrub Bottle was tested in
an accelerated lifetime test (February – October
2014) at the Campbell Scientific test site in
Logan, Utah. The bottle was connected to an
AP200 atmospheric profile system, configured
as follows:
• A step was added to the zero/span cycle
to measure the AP200 Scrub Bottle test port
• The zero/span cycle was configured to run
hourly, rather than daily
• The CO2 span setting was suppressed; a tank
of CO2 span gas was measured but not set-3
0
3
6
9
12
15
18
-100
0
100
200
300
400
500
600
8/24/2014 23:52 8/24/2014 23:56 8/25/2014 0:00 8/25/2014 0:03 8/25/2014 0:07 8/25/2014 0:10
H2O
(p
pt)
CO
2(p
pm
)
Mea
sure
Tes
t P
ort
Mea
sure
zer
o
Rem
easu
re C
O2
sp
an
Set
zero
Measure profile Measure profileZero/Span Sequence
Mea
sure
CO
2 s
pan
400
401
402
403
404
405
406
3/7/2014 0:00 4/26/2014 0:00 6/15/2014 0:00 8/4/2014 0:00 9/23/2014 0:00
CO
2(p
pm
)
CO2 Span Tank
The AP200 Scrub Bottle will easily last a year in normal
operation. The test port gives an indication it is time to
replace the molecular sieve with a large safety margin.
-0.01
0
0.01
0.02
0.03
0.04
0.05
3/7/2014 0:00 4/26/2014 0:00 6/15/2014 0:00 8/4/2014 0:00 9/23/2014 0:00
H2O
(p
pt)
H2O in AP200 Scrub Bottle Test Port
Portable Zero AirGenerator
An accelerated usage test was performed at Campbell Scientific, Inc., from
January to May, 2016. Open-path IRGAs (EC150 and IRGASON) were
zero and spanned using traditional compressed gases and dewpoint
generator, and then air from the Zero Air Generator was measured. The run
time for the pump was recorded for each test.
A total 54 IRGAs were tested, for a total run time of 15.5 hours. At the end
of the test there was no indication of CO2 or H2O in the scrubbed air stream,
and no indication of battery exhaustion. Other tests (data not shown)
showed the batteries generally run for approximately 40 hours before the
flow begins to decrease.
0
5
10
15
20
25
Nu
mb
er
of
Occ
urr
en
ces
CO2 (ppm)
CO2:mean: 0.00 ppmstd dev: 0.07 ppm
0
5
10
15
20
25
30
Nu
mb
er
of
Occ
urr
en
ces
H2O (ppt)
H2O:mean: -0.001 pptstd dev: 0.01 ppt
No H2O in test port over entire test (239 days, 2533 liters of air)
CO2 span tank was steady, confirming no CO2 in the air used to set the CO2 zero
Reference1 Fratini, G., McDermitt, D. K., and Papale, D. (2014) “Eddy-covariance flux errrors due to biases in gas
concentration measurements: origins, quantification and correction.” Biogeosciences, 11, pp 1037 – 1051.
-50
0
50
100
150
200
250
300
350
400
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
3/7/2014 0:00 4/26/2014 0:00 6/15/2014 0:00 8/4/2014 0:00 9/23/2014 0:00
CO
2(p
pm
)
CO
2(p
pm
)
CO2 in AP200 Scrub Bottle Test Port
CO2 appeared in test port after 132 days (3168 zero/span cycles, 1399 liters of air)
The Zero Air Generator (Campbell Scientific,
Inc. pn 31022) is a convenient tool for zeroing
an IRGA. The Zero Air Generator is very
similar to the AP200 Scrub Bottle but includes
a small diaphragm pump to circulate air
through the IRGA. The flow rate (0.2 l·min-1)
flushes the calibration shroud or sample cell of
the IRGA effectively without excess
pressurization. Power consumption is very low
and requires just one pair of AA batteries for
approximately 40 hours of use. The Zero Air
Generator weighs only 1.2 kg making it
extremely useful in zeroing IRGAs in remote
field sites.
AMS 32nd Conference on Agricultural and Forest Meteorology
CO2 remained below 300 ppm