Development of an Acid Scrubber for Reducing Ammonia Emissions from Animal Rearing Facilities

Recent research has shown that over half of nitrogen excreted by chickens is lost into the atmosphere via ammonia volatilization before the litter is removed from poultry houses. Large quantities of particulate matter and volatile organic compounds (VOCs) are also emitted from animal rearing facilities. During the past decade we have developed and patented an acid scrubber for capturing ammonia, VOCs and dust from air exhausted from poultry and swine barns. The objectives of this project were; (1) to re-design the scrubber to improve the ammonia removal efficacy, (2) conduct full-scale testing of the scrubber under controlled conditions at various ventilation rates, (3) evaluate the cost, practicality and efficacy of various acids for scrubbing ammonia, and (4) install scrubbers on exhaust fans of poultry houses located in Virginia and Arkansas and measure the efficiency of ammonia removal from the exhaust air. The efficiency of ammonia removal by the scrubber varied from 55-95%, depending on the type of acid used, air flow rate, and the internal scrubber configuration. This technology could potentially result in the capture of a large fraction of the N lost from AFOs, while simultaneously reducing emissions of bacteria, dust, and odors, which would improve the social, economic, and environmental sustainability of poultry and swine production.

The objectives of this project were; (1) to re-design our ammonia scrubber to improve the ammonia removal efficacy, (2) conduct full-scale testing of the scrubber under controlled conditions at various ventilation rates, and (3) evaluate the cost, practicality and efficacy of various acids for scrubbing ammonia.

During the first year of this project the main task of our team was to re-design the ammonia scrubber developed and patented by Moore (2007). A full scale prototype was constructed of wood and a series of tests were conducted to evaluate various configurations on air flow and static pressure drop in tests conducted in a machine shop. The scrubber was connected to a 48” variable speed poultry fan. Air flow was measured using a fan assessment numeration system (FANS unit). Static pressure difference was measured using a Setra 2601MS1 differential pressure sensor. The effects of slat angle, number and arrangement of slats, and thickness of cool cell material were evaluated.

Following the initial testing a fiberglass mold was made and six scrubbers were constructed. One of these was used to evaluate the effectiveness of water, strong acids, acid salts, and a neutral salt on scrubbing ammonia. Anhydrous ammonia was metered out into a distribution system located within the fan at a sufficient rate to result in 25 ppm NH3 in the plenum between the fan and the dust scrubber. Evaluations of each acid were made with the variable speed fan set at 60 and 40 Hz, which corresponded to air flows of approximately 8,000 and 5,000 cfm, respectively. A stainless steel star sampler was used to take air samples from the plenum and from the air exhausted from the scrubber. Ammonia concentrations were measured using a photoaccustic multigas analyzer (Innova 1412). All personal involved in this testing wore respirators equipped with NH3 cartridges. Three 2-hour trials were conducted with solutions of the following acids at both 40 and 60 Hz: alum, aluminum chloride, ferric sulfate, ferric chloride, sodium bisulfate, sulfuric acid, hydrochloric acid, phosphoric acid, and nitric acid. The effects of water and calcium chloride were also evaluated. For these trials the amount of each acid added was equivalent to 2 liters of concentrated sulfuric acid.

In addition to measuring inflow and outflow ammonia levels, the mass accumulation of ammonia in both the dust and acid scrubber reservoirs was determined by analyzing the contents for ammonium using an auto-analyzer. Twenty ml aliquots of the scrubber solution were taken at times 0, 1 and 2 hours for ammonia and pH measurements. These data were used to validate that the difference in inlet and outlet ammonia were, in fact, due to accumulation of NH3 in the scrubber. Notes were also taken on each chemical’s ease of use and potential for problems. For example, some dry acids did not readily dissolve and some strong acids, like sulfuric acid, had very strong exothermic reactions. Salts of aluminum and iron become aluminum and iron hydroxides at high pH which have the potential to clog cool cell material.

Another performance issue that was monitored was the loss of fine droplets (mist) from the scrubber. When dealing with high air volumes and small droplet sizes, there is a potential for mist to exit the system, resulting in not only the loss of N, but of the acid used to scrub NH3. In order to measure mist loss, five 12.5 cm Whatman 42 filters were attached on a wire cage on the exhaust of the scrubber. These filters were placed in a 50 ml centrifuge tub at the end of each trial and shaken with 25 ml of DDI water, which was analyzed for ammonium, along with sulfate, chloride, nitrate, or phosphate, depending on the acid used.

Early on in this research we learned that two scrubbers (a dust scrubber and an acid scrubber) were needed rather than one. If the dust isn’t removed from the exhaust air of poultry houses, then a large amount of the acid will be wasted neutralizing the dust.

We found that the relationship between slat angle and pressure drop was exponential and the angle that would maximize particle collisions on a wet surface while minimizing pressure drop was 45o. We also found that as the number of rows of slats increased the effect on pressure drop was linear. The final configuration chosen was eight rows of slats in the dust scrubber and three rows of slats in the chemical scrubber, followed by one or two 6” thick layers of cool cell material. The pressure drop using this configuration was about 0.1” of water at 5,000 cfm and 0.3” of water at 8,000 cfm.

All of the acids scrubbed ammonium from air, whereas water and calcium chloride only worked for a very short period of time. The iron (Fe) and aluminum (Al) compounds tended to work a little better than the other acid salts or the strong acids. We believe this is due to Fe and Al compounds coating the cool cell material. Although no difference was observed in the static pressure during these short tests, we believe Al and Fe hydroxides would eventually form and may clog the cool cells. Due to the inherit danger in dealing with strong acids, we concluded that an acid salt that did not contain Al and Fe, such as sodium bisulfate, would be used for our research in the future. This product is sold under the tradename PLT for a poultry litter treatment and is readily available to poultry growers.

Four NH3 scrubbers will be attached to sidewall fans of a commercial broiler house located in Madison County, Arkansas. The efficacy of these scrubbers for reducing ammonia, volatile organic compounds (VOCs), and particulate matter will be evaluated. We will also measure the amount of sodium bisulfate, water and electricity used by the scrubbers, as well as the mass of nitrogen captured. A cost-benefit analysis will be performed based on this data. Data on the efficacy to scrub ammonia will also be conducted on farms in DE, VA, and PA.

This research was funding by USDA/ARS and by grants from USDA/NRCS and the National Wildlife Foundation.   The authors would like to thank the hard work and great ideas supplied by Scott Becton and Jerry Martin, without which this scrubber could not have been built.