The City of Houston has selected planning, engineering and program management firm Lockwood, Andrews & Newnam Inc. (LAN) to develop...
Online monitoring of ammonia-based on UV spectroscopy on gas sample
More wastewater treatment plants are being required to do more than remove biological oxygen demand. Increasingly, they are required to remove nutrients like nitrogen and phosphorous to reduce impact on the environment. Nitrogen removal requires additional tankage because the biological oxidation process requires more time and higher concentrations of microorganisms. Accordingly, it takes 4.5 mg of oxygen to oxidize 1 mg of ammonia.
Increased air demands and power consumption can also make operation more expensive. Grab samples or even flow paced composite samples do not provide the process engineer with information on the diurnal or weekly loading changes. Only online automated analysis can provide that information. If operators can optimize the use of existing facilities and delay expansion, there could be significant capital savings.
Stringent discharge limits to meet
The Durham Advanced Wastewater Treatment Plant in Tigard, Ore., is one of four treatment plants operated by Clean Water Services (CWS) in Washington County. The four plants together clean 72 mgd for nearly 450,000 customers. The Durham plant discharges into the Tualatin River and its 24 mgd discharge is a significant contribution to the river’s 100 mgd flow. Very low total maximum daily load has been allocated for ammonia and phosphorous to prevent dissolved oxygen depletion below the river standard.
The Durham facility has seasonal ammonia limits of less than 1 mg/L NH3-N and total PO4-P limits of 0.070 mg/L. To meet these stringent limits, the plant must closely follow the nitrification process as well as control the recycled dewatered nutrients. At the Durham WWTP, it was previously found that the centrate produced from the digester sludge dewatering process was 30% of the ammonia load entering the plant.
With a concentration of 800 mg/L of ammonia and 100 mg/L of phosphorous, it was decided to design and construct a centrate equalization basin to store this centrate during the daytime peak flow and peak ammonia loading period and pump it into the treatment process during the night when low flow and low ammonia loading conditions prevailed.
Durham closely monitors the nitrification process, especially in the spring when nitrification has to be restarted altogether. The limitations of a lab ammonia analysis based on a flow paced composite sample quickly became evident because no information as to the ammonia load over the diurnal loading cycle was available.
Review of online monitoring equipment
In 2003, CWS re-evaluated its use of the ion specific electrode technology for online monitoring of its primary effluent. Primary effluent has grease, surfactants, particles, fibers and plastics making it a very difficult sample to handle and analyze. Surfactants can affect the permeability of ion selective electrodes and the grease and fibers make filtration difficult and maintenance time-consuming. The turbidity of the primary effluent water makes measurement based on optical technologies—where a beam of light must go through the water sample itself—extremely difficult if not impossible.
CWS conducted a review of the sample handling and analytical method of 13 different ammonia analyzers in 2003. A UV absorption-based system known as the AX1000 and manufactured by AWA Instruments was found, and based on its unique feature where the ammonia measurement itself is performed in the gas phase, it had the potential to deal with the difficult primary effluent sample at the Durham WWTP.
The measurement method implemented with this analyzer is based on the following steps:
The analyzer looked promising, so a pilot was arranged at the Durham WWTP. The AX1000 was installed on the primary effluent in September and the test ran until the third week of October.
A temporary sample connection was made and the analyzer was easily installed within 15 minutes. A 10% caustic solution and an acid cleaner were the only chemicals needed. The plant laboratory installed a water sampler on the same line as the one going to the analyzer and it was programmed to take one discrete sample per hour.
An ISE-based analyzer was also installed at the same location. It was also easy to disconnect the sample pump suction and have the AX1000 draw in a standard for a quality control check or another water sample for a single grab type of analysis.
In AX1000 pilot testing at Durham WWTP, CWS’s objective was to check the reliability of this novel ammonia analytical method, as well as its accuracy, its ease of use and its cost
AWA data tracked the lab data very well, tightly following the diurnal curve and the ammonia concentration from the nighttime centrate release.
The AX1000’s unique analytical method based on measuring ammonia in the gas phase allowed its use on the most difficult wastewater sample streams. It earned positive reviews in the pilot test at Durham WWTP for ease of programming, reliability, accuracy, low reagent use and operating cost and human/machine interface.
CWS has since acquired an AX1000 ammonia analyzer for each of their Durham as well as their Rock Creek plant. The online ammonia data generated on the primary effluent will be used to fine tune the centrate ammonia release. The plants’ power rates are lower at night so oxidizing more ammonia in off peak hours will shift power use from the higher daytime rate to the lower nighttime rates, thereby reducing the electricity bill. The nitrifying bacteria also work better under more uniform loading. They cannot grow fast enough to metabolize sudden increases in loading which may result in unwanted ammonia bleeding through the secondary process and into the plant effluent.
Reducing the ammonia peaks increases the total load the plant can process, delaying plant expansion costs.