Open Path Tunable Diode Laser

Document Sample
Open Path Tunable Diode Laser Powered By Docstoc
					          DRAFT VERSION NOT FOR DISTRIBUTION            JAN. 21, 2005

1    Measurement of Ammonia Emissions from Mechanically
2    Ventilated Poultry Houses using Multipath Tunable Diode
3    Laser Spectroscopy
 4   Paper # 542
 6   Eben D. Thoma, Richard C. Shores and D. Bruce Harris
 7   Office of Research and Development, National Risk Management Research Laboratory, Air
 8   Pollution Prevention and Control Division, US EPA, Research Triangle Park, NC
10   David F. Natschke and Ram A. Hashmonay
11   ARCADIS Inc., Research Triangle Park, NC
13   Kenneth D. Casey and Richard S. Gates
14   Biosystems and Agricultural Engineering, University of Kentucky, Lexington KY

16   Ammonia emissions from mechanically ventilated poultry operations are an important
17   environmental concern. Open Path Tunable Diode Laser Absorption Spectroscopy (OP-
18   TDLAS) has emerged as a robust real-time method for gas-phase measurement ammonia
19   concentrations in agricultural settings. This paper presents novel configurations of multipath
20   OP-TDLAS that allow near simultaneous measurement of poultry house inlet and exhaust
21   ammonia concentrations. The instrument and its deployment strategy are described and two
22   example applications are presented. In the first case example, an ammonia emission rate from
23   a 25,000-bird tunnel ventilated broiler grow-out house is discussed. From the several hour
24   observation period, an ammonia emission rate of 1.6 g/bird-day was determined for this house.
25   In the second example ammonia emissions from a 106,000-bird high-rise layer house were
26   investigated. This house type produced an ammonia emission rate of 1.3 g/bird-day over the
27   multi-hour observation period.

30   Emission of gas-phase ammonia (NH3) from concentrated poultry operations is the subject of
31   increased recent study1-3. Reliable estimates of NH3 emission rates are required to further
32   understanding of the agricultural industry’s potential impact on local and regional air quality.
33   Instrumental techniques capable of providing quality-assured, real-time NH3 concentration
34   data are needed for both monitoring and emission abatement strategy assessment. Near-
35   infrared Open-Path Tunable Diode Laser Absorption Spectroscopy (OP-TDLAS) has recently
36   emerged as a robust method for real-time NH3 gas concentration measurement in agricultural
37   settings. These systems utilize wavelength-scanning, narrow linewidth lasers to determine the
38   path-integrated concentration of a single pollutant species along user-configured optical paths.
39   Modern OP-TDLAS systems are simple to use, possess fast time response and allow for in-
40   field calibration and verification. The National Risk Management Research Laboratory
41   (NRMRL) of the U.S. Environmental Protection Agency (EPA) is developing OP-TDLAS-

          DRAFT VERSION NOT FOR DISTRIBUTION             JAN. 21, 2005

42   based techniques for whole-house NH3 emission assessments of both naturally and
43   mechanically ventilated poultry houses.
45   This paper describes the use of multipath OP-TDLAS for acquisition of NH3 concentration
46   data from mechanically ventilated poultry operations. The multipath OP-TDLAS used by
47   NRMRL is first discussed. Two example applications of the technique for poultry house
48   emission measurements are then described. Ammonia emissions data from a 25,000 bird
49   broiler house and a 106,000 layer hen high-rise house are presented.
51   Description of Multipath TDLAS Instrument:
52   The OP-TDLAS system used by NRMRL is a model CXL840 Multipath LasIR developed in
53   association with Unisearch Associates Inc. The instrument is capable of measuring four
54   separate analytes along as many as eight optical paths. The CXL840 consists of a hardware
55   controller, a laptop computer, and eight separate launch/receive telescopes. The hardware
56   controller contains four thermoelectrically cooled near-infrared tunable diode lasers, optical
57   control components, reference gas cells, and data acquisition systems. The controller
58   distributes the output optical signals from the lasers to the launch/receive telescopes via fiber
59   optic cables. By coaxial link, the controller receives return electronic signals from the
60   photodetectors housed in the telescopes. The laptop computer provides a software interface to
61   the hardware controller via an ethernet connection. A picture of the CXL840 with a single
62   telescope-retroreflector pair is contained in Figure 1. A block schematic of the CXL840 is
63   pictured in Figure 2.

66   The CXL840 sequentially coordinates the output of up to four lasers (L1-L4 in Figure 2) to
67   their respective reference gas cells and telescopes. Each laser has a dedicated analyte
68   reference cell for laser wavelength fine-tuning and correlation analysis. The outputs of the
69   lasers are first directed to beam splitters (BS1-BS4 in Figure 2) which split-off a small fraction
70   of the light for transmission through the lasers’ reference cells. The primary outputs of the
71   beam splitters are fed to a 4:1 optical switcher for time-division multiplexing of the laser light
72   through the calibration cell and then to the optical path. A 1:8 optical multiplexer is used to
73   select an individual launch telescope and is operated synchronously with an 8:1 electrical
74   demultiplexer that coordinates the return electronic signals. The acquisition system processes

           DRAFT VERSION NOT FOR DISTRIBUTION             JAN. 21, 2005

 75   detector signals from the launch/receive telescopes (field spectra) and from the reference gas
 76   cells (reference spectra).

 78   The CXL840 utilizes a fully automated direct-mode absorption spectroscopy scheme to
 79   quantify the target species concentration along the measurement path. For NH3 measurements,
 80   a single resolved rotational line from the first overtone band of the nitrogen-hydrogen
 81   stretching mode (approximately 1510 nm) is used. The calibration of the model CXL840 is
 82   set and verified through manual insertion of 10 cm long by 1 cm diameter cylindrical gas cells
 83   into the optical beam path during measurement. The gas cells are filled to atmospheric
 84   pressure using certified gas standards. The instrument provides the calibration cell insertion
 85   slot prior to the telescope multiplexing section of the hardware controller allowing for efficient
 86   calibration and quality assurance checks on all utilized optical paths. The model CXL840 has
 87   a manufacturer rated minimum detection sensitivity of 10 parts per billion for NH3 based on a
 88   500 m optical path and a 1 second signal integration time. Further information on the
 89   CXL840 OP-TDLAS system is contained in reference 4.

 91   Exhaust Fan Exit Plane Ammonia Concentration Measurements:
 92   Broiler Grow-out House
 93   The multipath CXL840 was used to help quantify NH3 emissions from a tunnel ventilated
 94   broiler house in central Kentucky. The measurement took place over a six-hour period in
 95   November of 2003. The four houses on this farm were each 12.2 by 152.5 m (40 x 500 ft),
 96   constructed in 2000 and housed approximately 25,000 birds. All houses had a 1.2 m (4 ft)
 97   curtain along the full length of both sidewalls for emergency ventilation. There are insulated,
 98   suspended ceilings in all houses. Each house has 8, 1220 mm fans [Choretime 38233-2 48”
 99   Turbo Fan (BD)] and 3, 915 mm fans (Choretime 38232-2 36” Turbo Fan). Box inlets are
100   located along both sidewalls and are automatically controlled based on static pressure

           DRAFT VERSION NOT FOR DISTRIBUTION               JAN. 21, 2005

101   difference. During tunnel ventilation mode, air is drawn into the house through two,
102   evaporative pad equipped 21.35 x 1.437 m (70 x 4.5 ft) openings. These openings are closed
103   off by a cable and winch operated curtain when the house not operating in tunnel ventilation
104   mode. The ventilation system at this site was managed by an electronic controller. A single
105   1220 mm (48 in) fan in a non-brood section of each house was used for minimum ventilation.
106   Operation of the end (915 mm) fans was prevented during the measurement period.
108   All houses reused litter with one annual cleanout and practiced half-house brooding. Litter in
109   U.S. broiler houses is typically reused for at least one year and is referred to as ‘built-up’ litter
110   in the industry. At the start of each annual cycle, the houses are completely cleaned out with
111   removal of all built-up litter and placement of fresh bedding (wood shavings). Caked litter is
112   removed at the end of each flock and a small layer of new wood shavings is added. This flock
113   was the 4th flock grown on the litter in this house.
115   This flock of birds was placed on 9/22/2003 and picked up on 11/14/2003 at 53 days of age.
116   At pickup, the birds would weigh approximately 3.2 kg each. The measurements reported in
117   this paper took place on 11/11/2003 with the birds being close to their maximum weight. A
118   schematic diagram of the barn along with OP-TDLAS optical paths and a fan bank photograph
119   is contained in Figures 3 and 4.



           DRAFT VERSION NOT FOR DISTRIBUTION               JAN. 21, 2005

122   Four primary optical paths were used for this measurement. Two identical optical path pairs were
123   configured along each side of the barn. The CXL840 telescopes were setup at the south end of
124   the barn. Retroreflector pairs were placed at the north end of the barn with the first
125   retroreflector positioned prior to the fan bank (short path) and the second placed after the fan
126   bank (long path). The short path represents the inlet background concentration measurement
127   for the configuration. The CXL840 produces a time dependent measure of the average NH3
128   concentration for each optical path in units of parts per million (ppm). For a particular optical
129   path, multiplication of the measured average concentration by the optical path length in meters
130   (m) yields the Path-Integrated Concentration (PIC) in units of ppm·m. To determine the
131   ammonia concentration at the output plane for a given set of operating fans, the PIC for the
132   short optical path is first subtracted from PIC of the long path. This result is then corrected to
133   account for the noncontributing space due to the nonoperating fans and the space between
134   fans. This correction assumes an average concentration of the background line or 10% of the
135   local concentration NH3, which ever is greater. The average analyte concentration is then
136   finally determined by division of the corrected PIC by the sum of operating fan diameters.

138   Figure 5 shows exhaust fan exit plane ammonia concentration data from the broiler grow-out
139   house for a 5.5 hour time period on 11/11/2003. Three fan operation conditions were
140   observed during the measurement time. The data sampling interval for each data point was
141   204 seconds. The two-fan configuration consisted of one fan operating on each side of the
142   barn. The three fan configuration added one additional operating fan on the east side of the
143   barn. The four-fan configuration consisted of two operating fans on each side of the barn. It
144   is observed that the exit plane ammonia concentration decreases with increasing ventilation
145   rate.

          DRAFT VERSION NOT FOR DISTRIBUTION            JAN. 21, 2005

147   High-Rise Layer House
148   In the fall of 2004, the CXL840 was used to measure NH3 emissions from a tunnel ventilated
149   high-rise layer house in eastern North Carolina. The measurements took place over multi-
150   hour periods on 9/1/2004, 10/13/2004 and 10/14/2004. The multilevel house had dimensions
151   of 16 m x 183 m and contained approximately 106,000 layer hens of 22 weeks in age (10/13
152   and 10/14). No birds were in the house on 9/1/2004, however, a partially dried manure load
153   was present and acted as an ammonia emission source. The interior of the house was dual-
154   level with birds on the upper level and no birds, but manure accumulation, on the lower level.
155   The manure in this house had been accumulating for approximately 1 year. The house is
156   ventilated by thirty-four 1.22 m (48 in) diameter fans located in two rows at each end of the
157   house. Box-shaped tunnel ventilation inlets (61 m x 2.5 m) with automated curtain control
158   were present on both the north and south sides of the house. Additional roof line inlets for
159   cross-ventilation were present but were closed during full fan operation conditions of this
160   study. A schematic diagram of the barn with OP-TDLAS optical path configurations is
161   pictured in Figure 6. A photograph of the fan banks and instrument components are contained
162   in Figure 7.



           DRAFT VERSION NOT FOR DISTRIBUTION            JAN. 21, 2005

166   Five primary OP-TDLAS optical paths were used for this measurement. The optical path
167   configurations were the same on each end of the barn. A single optical path traversed the
168   outputs of both the bottom and top fan banks on each end of the barn. A fifth optical line
169   measured the tunnel inlet ammonia concentration on the north side of the barn. On
170   10/13/2004 a sixth optical path measured the inlet concentration on the south side of the barn.
171   For the measurements presented, all 34 fans were operating.
173   Figure 8 presents exhaust fan exit plane NH3 concentration data for the high-rise layer hen
174   house. On 9/1/2004 no birds were present (manure only). On 10/13/2004 and 10/14/2004,
175   106,000 birds were present. Low levels (< 1 ppm avg.) of NH3 emissions from the manure
176   were observed on 9/1/2004. The manure had been drying for approximately 1 week prior to
177   the observation period. With birds present, NH3 concentrations from the output of the bottom
178   row of fans (manure level) is approximately three times higher than those from the output of
179   the top rows which are at bird level. This result is expected for this house design type since
180   the NH3 emissions are produced primarily from the manure. Table 1 summarizes the NH3
181   concentration data for this measurement.

183   Table 1: Summary of exhaust fan exit plane NH3 concentration data for a 106,000 bird
184   high-rise layer hen house with and without birds.
                                      Bottom Fan Sets    Top Fan Sets NH3           Inlet NH3
                       Number of
          Date                      NH3 Concentration Concentration ppm Concentration
                                      ppm (Std. Dev.)        (Std. Dev.)         ppm (Std. Dev)

         9/1/04            0             1.21 (0.26)             0.77 (0.41)           0.48 (0.27)

      10/13/04 and
                        106,000          11.29 (0.80)            3.73 (0.49)           0.30 (0.11)

           DRAFT VERSION NOT FOR DISTRIBUTION           JAN. 21, 2005


186   Ammonia Emission Rate Estimates:
187   Estimates of Building Ventilation Rates:
188   Ammonia emission rate estimates from animal feeding houses require both concentration data
189   and fan exhaust flow rate information. Determination of building ventilation rates for
190   concentrated poultry operations has have been the subject of recent research.1,2,5,6 For the
191   current study, volumetric flow rates are determined from building static pressure
192   measurements in conjunction with manufacturer fan calibration curves. In real-world
193   conditions, the deterioration of fan components coupled with accumulation of debris on the
194   fan blades results in significant reductions in fan performance. The fan performance curves
195   (flow rate vs static pressure) for the fans present on the 25,000 bird broiler house were
196   previously determined utilizing a Fan Assessment Numeration System (FANS)7 developed
197   and operated by the University of Kentucky.
199   Data from this study indicated an average performance reduction factor of 15% for the four
200   fans used in the current study. A preliminary measurement of fan flow rates for the 106,000
201   bird high-rise house was conducted using a 12 inch pitot tube grid and 11-point sampling
202   pattern across the fan face. An average 17% performance reduction factor was observed.
203   Based on these data, a standard fan performance reduction factor of 15 % with a 10%
204   uncertainty is assumed for the emission estimates contained in this work. Pertinent building
205   ventilation data is summarized in Table 2. Fan performance curve data was taken from the
206   Bess Lab test reports.8
208   Table 2: Summary of building ventilation data for the 25,000-bird broiler house and the
209   106,000-bird high-rise layer house from this study.
                                             Number of        Average Static
                          Fan Size /                                                 Volumetric
          House                              Operating           Pressure
                       Bess Lab Report                                             Flow Estimate*
                                               Fans         in. H2O (Std. Dev.)
                         48 in. / 96163           2            0.098 (0.009)            39,024

                         48 in. / 96163           3            0.112 (0.009)            56,706

                         48 in. / 96163           4            0.109 (0.007)            76,057

                         48 in. / 04331          34            0.089 (0.006)           647,360
210   (*) Includes –15% fan performance reduction factor.

           DRAFT VERSION NOT FOR DISTRIBUTION            JAN. 21, 2005

213   Estimates of Ammonia Emission Rates:
214   NH3 concentration data along with building ventilation rate information are combined to
215   produce ammonia emission rate estimates for poultry houses of this study. It is noted that the
216   following emission rate estimates are derived from a short observation time period and are not
217   representative of NH3 emission rates over extended periods of time.
219   Figure 9 shows NH3 emission rate estimates for the tunnel ventilated 25,000-bird broiler
220   grow-out house with accumulated litter and the 106,000 layer-hen high-rise house with
221   accumulated manure observed during this study. Emission rate estimates of Figure 9 are
222   presented in units of grams of NH3 per hour as a function of building ventilation rate.
223   Abscissa error bars indicate the assumed  10% building ventilation rate uncertainty.
224   Ordinate error bars combine concentration data variance ( 1 standard deviation) with the
225   building ventilation rate uncertainty to produce an estimated NH3 emission rate uncertainty.
226   Emission rates for the broiler house are similar for all three fan operation conditions at around
227   1650 g/hr. Ammonia emission rates for the high-rise house with 106,000 birds present
228   indicate an approximate 5700 g/hr emission rate. Emission rates for the no-bird condition
229   were not determined. Figure 10 presents the same NH3 emission rate estimate in units of
230   grams per bird-day. It is evident that the estimates NH3 emission rates on a per bird basis are
231   similar for both house types with the broiler house producing a 1.6 g/bird-day rate average and
232   the high-rise house producing a 1.3 g/bird-day rate.


           DRAFT VERSION NOT FOR DISTRIBUTION            JAN. 21, 2005

236   The emission rates determined in this study fall within the range of emission rates for broiler
237   house reported in recent U.S. studies. Emission rates from broiler houses have been shown to
238   vary from near zero to over 2 gm/bird-day depending on growth stage, ventilation rate and
239   season.1, 9 Average emission rates have been determined in the range 0.63 – 0.92 gm/bird-
240   day.9, 10, 11
243   A multipath near infrared OP-TDLAS system was used to measure the ammonia
244   concentrations at the exhaust fan exit planes of a 25,000-bird broiler grow-out house and a
245   106,000-bird high-rise layer house. Ammonia emission estimates were determined from
246   measured concentration values and building ventilation rates. This work shows the multipath
247   OP-TDLAS can a useful tool for real-time determination of ammonia concentrations emitted
248   from concentrated poultry operations. EPA NRMRL will continue its efforts to refine and
249   expand this emission measurement technique. Plans include 24 hour testing with automated
250   data logging of fan operation in addition to improvements in fan flow rate measurements.

254       1. Wheeler, E. F.; Casey, K. D.; Zajaczkowski, J. S.; Topper, P.A.; Gates R. S.; Xin, H;
255          Liang, Y., Tanaka, A., Ammonia Emissions from U.S. Poultry Houses: Part 3 –
256          Broiler Houses, Proceedings of the 3rd International Conference on Air Pollution from
257          Agricultural Operations, Research Triangle Park, NC, October 12-15, 2003.

          DRAFT VERSION NOT FOR DISTRIBUTION              JAN. 21, 2005

258      2. Liang, Y., H. Xin, E.F. Wheeler, R.S. Gates, H. Li, J.S. Zajaczkowski, P. Topper, K.D. Casey
259         and F.J. Zajaczkowski. 2004. Ammonia Emission for US Poultry Houses: Laying Hens. ASAE
260         Paper No. 044104. St. Joseph, Mich.: ASAE.
261      3. National Research Council. Air Emissions from Animal Feeding Operations: Current
262         Knowledge, Future Needs. The National Academies Press, Washington, DC, 2003.
263      4. Thoma, E. D.; Shores, R. C.; Thompson, E. L.; Harris, D. B.; Thorneloe, S. A.; Varma,
264         R. M.; Hashmonay, R. A.; Modrak, M. T.; Natschke, D. F.; Gamble, H.A., Open Path
265         Tunable Diode Laser Absorption Spectroscopy for Acquisition of Fugitive Emission
266         Flux Data, J. Air and Waste Manage Assoc., In Press.
267      5. Casey, K. D.; Gates R. S.; Wheeler, E. F.; Xin, H; Zajaczkowski, J. S.; Topper, P.A.;
268         Liang, Y., Ammonia Emissions from Kentucky Broiler Houses During Winter, Spring
269         and Summer. Proceedings of the Air and Waste Management Association 97th annual
270         Conference, Indianapolis Indiana, June 22-25, 2004.
271      6. Hong, L.; Xin, H.; Liang, Y.; Gates, R.S.; H.; Wheeler, E. F., Determination of
272         Ventilation Rates for Manure Belt Layer Hen House Using CO2 Balance, Paper
273         Number MC04-201, Proceedings of the 2002 ASAE International Meeting, Chicago
274         Illinois, July 28-July 31, 2002.
275      7. Gates, R.S., Casey, K.D., H. Xin, Wheeler, E.F., Simmons,J.D., Fan Assessment Numeration
276         System (FANS) Design and Calibration Specifications. Transaction of the ASAE. 47(5) 1709-
277         1715.
278      8. Bess Labs Online Agricultural Ventilation Fans Test Reports. University of Illinois,
279         Department of Agricultural and Biological Engineering, (2004), Web Site:
280 (accessed November 2004).
281      9. Siefert, R.L., Scudlark, J.R., Potter, A.G., Simonsen, K.A., Savidge, K.B.,
282          Characterization of Atmospheric Ammonia Emissions from a Commercial Chicken
283          House on the Delmarva Peninsula. Environ Sci Technol. 38() 2769-2778.
284      10. Burns, R.T., Armstrong. K.A., Walker, F.R., Richards, C.J., Raman, D.R., Ammonia
285          Emissions from a Broiler Production Facility in the United States. Proceedings of the
286         International Symposium on Gaseous and Odour Emissions from Animal Facilities, Horsens,
287         Denmark, June 1-4, 2003.
288      11. Lacey, R.E., Redwine, J.S., Parnell, C.B. Jr., Particulate Matter and Ammonia
289          Emission Factors for Tunnel-Ventilated Broiler Production Houses in the Southern
290          U.S.. Transactions of the ASAE. 46(4) 1203-1214.
293   TDLAS, Ammonia, Poultry, Fugitive, Emission