U.S. Environmental Protection Agency

Office of Research and Development

EPAwww.epa.gov

Between June 2018 and July 2018, we ran the Aerodyne (for one week), Picarro, and Gasera

formaldehyde monitors in a research trailer at the Ambient Air Innovative Research Site (AIRS) in

Research Triangle Park, NC. We also ran collocated DNPH cartridges for several periods.

Evaluation of continuous formaldehyde measurements in ambient air1Andrew R. Whitehill, 1Lukas Valin,1David Williams, 1James Szykman, 1Russell Long, 1Surender Kaushik, 2Peter Furdyna, 2Dirk Felton1National Exposure Laboratory, Office of Research and Development, United States Environmental Protection Agency, 109 T.W. Alexander Drive, Research Triangle Park, NC 27709, United States of America2New York Department of Environmental Conservation, Albany, NY 12233, United States of America

Andrew R. Whitehill l whitehill.andrew@epa.gov l 919-541-4540

Formaldehyde is an important hazardous air pollutant (HAP) that is a leading driver for HAP-

related cancer risk in the United States. It is emitted directly by numerous anthropogenic and

natural sources, and formed as a secondary product from volatile organic compounds (VOCs)

photooxidation. Formaldehyde is a significant source of radicals in the atmosphere that result in

ozone and particulate matter (PM) formation. Routine measurements of formaldehyde in regulatory

networks rely on EPA Compendium Method TO-11A, which is based on HPLC analysis of

derivatized, time-averaged cartridge samples.

The U.S. EPA Office of Research and Development has been involved in the continuing evaluation,

analysis, and comparison of commercially available continuous formaldehyde measurements, as

well as the promotion of novel technologies for ambient air and source emissions monitoring

applications. We present results from our ongoing evaluation and intercomparison of newer

formaldehyde measurement technologies in both controlled laboratory experiments and ambient air.

Introduction Methods for Measuring Formaldehyde Instrument Performance Possible Influence of Water Vapor on Comparability

Instrument Evaluation Sites

GC-FID/MSHPLC-UV/MSSpectroscopic

Offline:

Collection/derivatization by

DNPH-coated silica gel,

followed by offline analysis

by HPLC. Standardized by

EPA Method TO-11A and

others.

Possible formaldehyde

collection efficiency issues

on some cartridges. Use of

HR-MS or MS-MS detectors

(versus UV/DAD) can

improve selectivity and

sensitivity.

Online (Auto-GC):

Collection on sorbent (e.g. MOF-

5), followed by thermal desorption

(TD) and analysis by GC-FID

(with oxidation/reduction catalyst,

e.g. Polyarc) or MS. Method is

not as well developed as other

methods.

Offline:

Collection/derivatization on

PFPH-coated Tenax, followed by

offline analysis by GC-MS.

Lower detection limits and higher

precisions than HPLC method

allows for quantification of lower

concentrations. Better separation

of GC vs HPLC reduces

interferences. Not well developed.

Online:

Direct analysis of underivatized

air at ambient concentrations

using infrared radiation with

various types of detection

(direct absorption, cavity

ringdown, photoacoustic

spectroscopy, etc.). Highest

time resolution and lowest

maintenance of these methods.

UV and millimeter wave

methods also possible, but not

commercially available.

Commercially Available Spectroscopic Instruments

Other Methods

Hantzsch Reaction

Aerodyne Research, Inc.

Direct absorption instrument

capable of high precision

(research grade) formaldehyde

and formic acid measurements,

large footprint.

Picarro

Cavity ringdown spectroscopy

measurements of formaldehyde

and methane, high stability.

Aeris Technologies, Inc.

Direct absorption

formaldehyde measurements,

lunchbox sized and portable,

automated zeros (every 30

seconds) using a DNPH

scrubber.

Gasera

Photoacoustic spectroscopy

measurements of formaldehyde.

0.001

0.01

0.1

1

10

1 10 100 1000 10000 100000

Alla

n D

evia

tion

σy(τ

), p

pb

Averaging Time, τ, seconds

AerodynePicarroPicarro_RTPGasera_RTPAeris

RTP

Flax Pond

Allen Deviation Analysis

Instrument precisions span approximately 2

orders of magnitude, with higher precision

instruments tending to be higher priced and

having a larger footprint.

y = 0.872xR² = 0.997

0

5

10

15

20

25

30

0 10 20 30Pic

arr

o F

orm

ald

eh

yd

e (

pp

b)

DNPH Formaldehyde (ppb)

5

6

7

8

9

10

11

12

01:16 06:04 10:52

CH

2O

(pp

b)

Time (hh:mm)

Picarro

PermTube

DNPH

y = 0.969x + 0.095R² = 0.999

-2

0

2

4

6

8

10

12

-2 0 2 4 6 8 10 12

Me

asu

red

CH

2O

(pp

b)

Target CH2O (ppb)

Aerodyne

(1-sec data)

y = 0.891x + 0.898R² = 0.999

0

2

4

6

8

10

12

-2 0 2 4 6 8 10 12

Me

asu

red

CH

2O

(pp

b)

Target CH2O (ppb)

Picarro

(5-min data)

Comparison to Reference Gas

Picarro and Aerodyne compare well to reference

formaldehyde from permeation tubes (top) and

gas cylinders (bottom).0

1

2

3

4

5

07-20 07-21 07-22 07-23 07-24

CH

2O

(pp

b)

Date and Time (EST, 2018)

Picarro

Gasera

DNPH

0

2

4

6

8

10

06-29 06-30 07-01 07-02 07-03

CH

2O

(pp

b)

Date and Time (EST, 2018)

Comparison to TO-11A (DNPH)

Note: Picarro data is as reported by the

instrument. Gasera data is corrected post-

collection using an empirical correction that

depends upon water vapor concentrations.

Instrument collocation Results

0

1

2

3

4

5

6

09-18 09-20 09-22 09-24 09-26 09-28 09-30 10-02

CH

2O

, p

pb

Date EST (MM-DD, 2018)

AerodynePicarroGasera

0

2

4

6

8

06-08 06-09 06-10 06-11 06-12 06-13

CH

2O

(pp

b) Aerodyne

Picarro

0

2

4

6

8

07-03 07-05 07-07 07-09

CH

2O

(pp

b) Picarro

Gasera

Collocation experiments show strong agreement between the Aerodyne and Picarro instruments,

even at high time resolutions (e.g. 5-minute averages). There are minor discrepancies between

the Picarro and Aerodyne instruments that correlate with water vapor/relative humidity.

Discrepancies between the Gasera and the Aerodyne/Picarro are larger and more variable.

Research Triangle Park, North Carolina (Ambient Air Innovative Research Site)

Flax Pond NY DEC (PAMS) Monitoring Site, Long Island, NY

Between September 2018 and October 2018, we ran the Aerodyne and Picarro instruments

collocated at the Flax Pond Marine Laboratory in Old Field, NY as part of the Long Island Sound

Tropospheric Ozone Study (LISTOS). Harvard University ran a fiber laser induced fluorescence

instrument and Aeris formaldehyde monitor, SUNY Stony Brook ran a PTR-MS, and the New York

DEC ran PAMS-style DNPH cartridges. The Gasera instrument was also run towards the end of the

study.

AIRS Research Trailer

Ambient

Inlet

NOy

Pandora

Spectrometer

Aerodyne

Harvard FLIF

(Keutsch)

Aeris

PTR-MS

(Mak)

Ambient

ManifoldZero / CAL

Manifold

Main

Inlet

Aerodyne

Inlet

Instrument

Rack

0

5

10

15

20

25

0 10 20

Gasera

Form

ald

ehyde (

ppb)

Picarro Formaldehyde (ppb)

Dry Set 1

Dry Set 2

1% Water Vapor

1:1 Line

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

0.0 1.0 2.0 3.0

Form

ald

ehyde D

iffe

rence

(A

ero

dyne -

Pic

arr

o,

pp

b)

Picarro Water Vapor (%)Comparison of reference formaldehyde

measurements show a strong correlation

between the Picarro and Gasera monitors,

but a difference in the correlation between

dry and wet air, suggesting a significant

water vapor interference in one (or both)

instruments.

Collocation experiments for the Aerodyne and

Picarro instruments (for both ambient air and

dry/humidified reference gas) suggest that the

minor discrepancies between the two instruments

may be correlated with water vapor.

All commercial infrared spectroscopic formaldehyde instruments will be influenced by water

vapor/relative humidity, as there are many water vapor lines in the spectral regions that absorb

strongly by formaldehyde. As a result, all the instruments will measure water vapor lines and

correct the formaldehyde for the water vapor concentration, either using multispecies spectral fitting

or an empirical/calibrated correction factor (or both).

Ongoing/Future Work

Formaldehyde concentrations in the southeastern United

States show strong diurnal and seasonal cycles and have a

large degree of variability, especially during the summer.

They also show the influence of meteorological factors

and other events (such as wildfire impacts). We will be

deploying the Aerodyne formaldehyde monitor long term

at the Duke Forest research forest in Durham, NC. In

2019, we plan a multi-month collocation study with the

Aeris formaldehyde monitor, and may collocate the

Picarro monitor as well, to continue to assess the

differences between instruments and to assess how the

instruments respond to environmental changes.

(Right) 4 month formaldehyde time series in Research

Triangle Park, North Carolina in 2016.

Acknowledgements

New York Department of Environmental Conservation; Picarro, Inc.; Gasera; J.J. Wilbur Company;

Aerodyne Research, Inc.; Aeris Technologies, Inc.; Frank Keutsch (Harvard University); Joshua

Shutter (Harvard University); David Shelow (EPA); Kevin Cavender (EPA); Ingrid George (EPA) DISCLAIMER: The views expressed in this poster are those of the author(s) and do not necessarily represent the views or policies of the

U.S. Environmental Protection Agency. Mention of or referral to commercial products does not imply EPA endorsement of those products.