Continuous Ammonia Monitoring

Jim Schwab, ASRC, University at Albany, SUNY

With contributions from many others!

Additional Contributers

• Eric Edgerton, ARA, Cary, NC• Minsuk Bae, Yongquan Li, Ken

Demerjian, John Spicer – ASRC• Sara Pryor – Indiana University• Bjarne Jensen - Danish National

Environmental Research Institute • Jian Hou, Xianliang Zhou – University at

Albany School of Public Health• Donna Kenski - LADCO

Continuous Methods• Potential Routine Methods

– Chemiluminescence• SEARCH TRN• API Model 201E or Thermo Model 17C (1 ppb LDL)

– Ion Mobility Spectrometer– Laser Acousto-Optical Detection– AiRRmonia – ECN, Netherlands, Mechatronics

Instruments, b.v.• Research Methods

– Tunable Diode Laser – TDLAS– Wet Effusive Diffusion Denuder (WEDD)– Wet Scrubbing Long Path Absorption Spectrometer –

WS-LOPAP– Wet Scrubbing IC analysis (URG AIM 9000)– MARGA – Steam Jet Collection – IC

SEARCH Continuous NH3 Method (TRN)• 2-channel oxidized AND reduced nitrogen

chemiluminescence analyzer• Channel 1 samples ambient air through a

sodium carbonate denuder (TN), then two heated converters

• Channel 2 samples ambient air through dual sodium carbonate and citric acid denuders (TN* ), then through a single moly converter

• NH3 operationally defined as TN-TN*

Total Reduced Nitrogen (TRN) Analyzer

650º C Pt Converter

350º C Moly Converters

Total Reactive Nitrogen after Na2CO3 denuder

Total Oxidized Nitrogen after Na2CO3 and citric acid denuders

0

15

30

45

60

75

5/4/05 0:00 5/4/05 3:00 5/4/05 6:00 5/4/05 9:00 5/4/05 12:00 5/4/05 15:00 5/4/05 18:00 5/4/05 21:00

SO

2 (p

pb

)

0

1

2

3

4

5

6

7

8

9

10

NH

3, N

H4,

NH

x (p

pb

)

SO2 NH4 NH3 NHx

NH3 and NH4+ during CFPP Plume

Event

NH4+ increases; NH3 disappears; NHx conserved

Method 2 - IMS• Candidate for routine monitoring• IMS – Ion Mobility Spectrometer (AirSentry-IMS

Ammonia Analyzer, Particle Measuring Systems, Boulder, CO)

• Ionization based time-of-flight technique performed at atmospheric pressure

• Instruments are targeted to a specific molecule in a specific concentration range

• Range for this instrument: 0–300 ppb (Lab Intercomparison); 0-100 ppb (Field Deployment)

• Cost (with OBC) ~ $26 K

Operational Details

• Requires up to 5 li/min of clean, dry instrument air (dewpoint ≤ -40º C)

• Internal sample pump option• Factory default “Smoothing Factor” (8)

should be increased for minute (or longer) averaged data

• Optional “On Board Calibration” - integral permeation oven set up (recommended)

• Nickel 63 sealed radioactive source (less than .005 microcuries) – ionization source

NH3 - PSP - Addison, NY; May 7-13, 2006

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

5/7/20060:00

5/7/200612:00

5/8/20060:00

5/8/200612:00

5/9/20060:00

5/9/200612:00

5/10/20060:00

5/10/200612:00

5/11/20060:00

5/11/200612:00

5/12/20060:00

5/12/200612:00

5/13/20060:00

5/13/200612:00

5/14/20060:00

NH

3 (

pp

b)

NH3 - PSP - Addison, NY; March 5-11, 2006

-1

0

1

2

3

4

5

6

7

3/5/20060:00

3/5/200612:00

3/6/20060:00

3/6/200612:00

3/7/20060:00

3/7/200612:00

3/8/20060:00

3/8/200612:00

3/9/20060:00

3/9/200612:00

3/10/20060:00

3/10/200612:00

3/11/20060:00

3/11/200612:00

3/12/20060:00

NH

3 (

pp

b)

Preliminary data: do not cite or quote!

IMS Field Data

Pranalytica Nitrolux NH3 Analyzer

Photoacoustic spectroscopyCO2 laserLow-noise microphoneIn-line particle filter (40 µm)LDL and cost depend on model

Model 1000 – LDL ~ 1 ppb; Cost ~$24KModel 200 – LDL ~ 0.2 ppb; Cost ~$38KModel 100 – LDL ~ 0.1 ppb; Cost ~$72K

Note: Instrument used was one of the first sold by Pranalytica and company claims to have made significant improvements to later (i.e. current) instruments

Laboratory Intercomparison

• September 12-30, 2005 at ASRC• Six measurement Methods – seven

instruments in all (Chemiluminescence method not included because instrument was not configured properly – my error)

• Heated glass manifold for uniformity of sampling environment

• Synthetic air and calibration system• “Ambient” sample for two extended periods

Method Characteristics Tested during Intercomparison

• Response to known [NH3] – accuracy and precision of calibration

• Time Response – to step changes and “spikes”

• Comparability during ambient sampling

• Potential interferences

9/28: Zero, 5, 10, 20 ppb

0

5

10

15

20

25

6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00

[NH

3]

- p

pb

IMS Nitrolux TDL LOPAP WEDD1 WEDD2

Preliminary data: do not cite or quote!

9/26: 40 ppb spike from zero

0

5

10

15

20

25

30

35

40

10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00

[NH

3]

- p

pb

IMS Nitrolux TDL LOPAP WEDD1 WEDD2

Preliminary data: do not cite or quote!

9/24: Ambient Spike

0

2

4

6

8

10

12

14

16

18

9/24/200518:00

9/24/200519:00

9/24/200520:00

9/24/200521:00

9/24/200522:00

9/24/200523:00

9/25/20050:00

9/25/20051:00

9/25/20052:00

9/25/20053:00

[NH

3]

- p

pb

IMS Nitrolux TDL LOPAP WEDD1

Preliminary data: do not cite or quote!

Sampling and response time

• This is an important issue for gaseous ammonia measurement – it clearly affects methods detecting gaseous ammonia directly and ammonium after scrubbing.

• The TDL is the system of choice for quick and accurate response – but it is not a “routine” method!

Spring 2006 Measurements – Pinnacle State Park, Addison, NY

• NH3 measured with AirSentry IMS

• HNO3 with denuder difference chemiluminescence method

NH3 HNO3

Mean 1.06 ppb 0.66 ppb

Standard Dev. 0.77 ppb 0.62 ppb

Median 0.87 ppb 0.48 ppb

Maximum 5.82 ppb 5.75 ppb

Est. MDL 0.3 ppb 0.2 ppb

Completeness 90.1% 85.4%

Ammonia at Pinnacle State Park, Addison, NY - 2006

Julian Day60 80 100 120 140

01

23

45

6

Nitric Acid at Pinnacle State Park, Addison, NY - 2006

Julian Day60 80 100 120 140

01

23

45

6

Co

nce

ntr

atio

n -

pp

b

88.0 88.5 89.0 89.5 90.0 90.5 91.0

Julian Day

0

1

2

3

4

5

6

NH

3 -

pp

b

0

1

2

3

4

5

6

HN

O3

- p

pb

NH3HNO3

NH3 and HNO3 - March 30 - April 1, 2006

145 146 147 148 149 150 151

Julian Day

0

1

2

3

4

NH

3 -

pp

b

0

1

2

3

4

HN

O3

- p

pb

NH3HNO3

NH3 and HNO3 - May 26-31, 2006

Response Time – Memory Effect• Note that on previous slide the HNO3

drops below 1 ppb each night

• However, NH3 does not drop very low between peaks

• We believe this is due – at least in part – to changing uptake and release of NH3 on the sample inlet, the switching valves, and the sample handling lines in the instrument

Candidate Methods for Monitoring – and their Major Shortcoming(s)

(Editorial Opinions!)

• ARA TRN (chemiluminescence)– Not ammonia specific– Three heated converters – difference method

• AirSentry IMS (Particle Measuring)– Response Time Problem

• Nitrolux 200 (or 100) (Pranalytica)– Response Time (worst of all tested

instruments, but should be better in current versions)

– Cost – Highest of “routine monitoring” candidates (for sufficiently low LDL)

• AIRRmonia, URG AIM, MARGA – need liquid reagents

Recommendation

• ARA TRN (chemiluminescence)– Custom implementation of proven

analyzer– Response time issue minimized

(eliminated?) since there is no “handling” of ammonia

– EPRI is considering commercialization of converter technology

Acknowledgements

• Funding from – EPA– NYSERDA– NYSDEC

• Although the research described in this article has been funded in part by the U.S. Environmental Protection Agency, it has not been subjected to the Agency’s required peer and policy review and therefore does not necessary reflect the views of the Agency and no official endorsement should be inferred should be inferred.

End of Presentation

Nitric Acid Measurement• Denuder difference in conjunction with NOy

• TECO 42S low-level chemiluminescence analyzer – operates with two measurement channels and one background (“prereactor zero”) channel

• Both measurement channels preceded by external heated molybdenum converters

• “Base” channel is NOy

• Second channel uses a NaCl coated denuder as inlet – this channel measures NOy – HNO3

• Difference of channels is HNO3

Schematic of HNO3 Method

SS NEMA box mounted outside (5 m agl)

Heated Mo converter

Heated Mo converter

Gas lines

NaCl coated denuder

PFA Teflon Tube

OUTSIDE INSIDE

Model 42S chemiluminescence analyzer

Mode switching valves

Power and signal wires

BYPASS PUMP

1-minute NH3 and WD at Yorkville

0

5

10

15

20

8/28/06 0:00 8/28/06 6:00 8/28/06 12:00 8/28/06 18:00

NH

3 (p

pb

)

0

45

90

135

180

225

270

315

360

WD

(d

eg.)

NH3 WD

Spikes between 90 and 180 degrees suggest nearbyNH3 sources.

Local Sources of NH3 at YRK

Poultry operationswithin several km line up with NH3 spikes.

2.6 km

1.5 km

3.3 km