New European wastewater treatment technologies might help Florida, which is pursuing tighter discharge consents to sensitive surface waters, says Frank Rogalla of AqualiaA joint International Water Association and Water Environment Federation conference on Nutrient Management & Recovery, held in the US city of Miami in January coincided with new US Environmental Protection Agency (EPA) rules on Water Quality Standards (WQS) for the state of Florida. The rules define precise numeric criteria for total nitrogen (TN) and total phosphorus (TP) in five distinct watersheds - in accordance with the provision of the Florida Administrative Code that "in no case shall nutrient concentrations of water cause an imbalance in natural populations of aquatic flora or fauna".
From 6 March 2012, entities discharging nitrogen or phosphorus to lakes and Florida surface waters would be affected. The WQS are used in determining the discharge permit limits according to the National Pollutant Discharge Elimination System.
The EPA estimates that the annual cost of preventing discharges to impaired inland waters, as defined under this rule, is between US$135-206M/yr, including advanced treatment at wastewater treatment plants, and best management practice at agriculture operations. The regulator expects the additional annual cost to be between US$40-71 per household per year, or equal to 11-20 cents per day per household for clean water.
Other stakeholders, such as the Florida Water Environment Association, the local branch of WEF, released a study that projected costs to be upwards of US$8B, and was accused by the Florida Independent news service as "representing some of Florida's most notorious polluters", namely
water and wastewater utilities.
For southern Florida, apart from the five regions mentioned, the EPA proposes phosphorus limits of 10-50ppb in lakes, 107ppb in streams in southern Florida, and 42 ppb in canals. These limits are expected to be finalised in conjunction with criteria for Florida's estuarine and coastal waters by 14 November 2011. Current guidelines for the protected Everglades wetlands, adopted in 2003, set a 10ppb phosphorus limit. New technologies that would help in improving the sustainability of meeting tight discharge standards were widely discussed at the Miami conference. One particularly promising approach has now made the step to large scale applications - the anaerobic ammonia oxidation, or Anammox reaction: NH4 + + NO2 - àN2
This shortcut in the nitrogen metabolism pathway allows the conversion of ammonia to gaseous nitrogen without oxygen and carbon source. The reaction had been observed in the early 1990s when the Anammox process was developed and patented by Technical University of Delft.
The responsible bacteria were positively identified in 1999 as belonging to the bacterial phylum planctomycetes, their most striking feature is an extremely slow growth rate - the doubling time is nearly two weeks.
The major stumbling blocks to larger application of the process were that the wastewater had to be relatively warm and the ammonium level very high in order to work effectively. This problem has been solved by working with bacteria that have been tightly packed into grains, or fixed on a support media.
To apply ammonium removal from municipal and industrial wastewater with Anammox, a combination of aerobic nitrification and anaerobic conversion is required. The simplest version is a two-reactor process for the two steps of deammonification.
A first reactor, using partial nitrification (Sharon: 'Single reactor for high activity ammonia removal over nitrite') converts only 50% of the ammonium to nitrite: NH4+ + HCO3- + 0.75 O2 à0.5 NO2-+ 0,5 NH4+ + CO2 + 1.5 H2O. The conversion takes place in a single, completely mixed reactor without retention at average process temperatures between 30-40°C, and retention times of one to two days.
In a second reactor, the residual mixture of ammonia and nitrite is anaerobically transformed to elementary nitrogen gas with ammonium as electron donor. In order to prevent wash-out of the slow-growing Anammox bacteria, efficient biomass retention has to be achieved, which was obtained by the good granule formation capacity in gas-lift-loop reactors, ensuring simultaneously efficient mass transfer and mixing on a compact footprint.
The first full-scale application at Rotterdam-Dokhaven wastewater treatment plant (WwTP) in the Netherlands was for a 470,000 population equivalent (PE) site. It comprised these two separate steps in the treatment of reject water from anaerobic sludge digestion.
A Sharon reactor, designed by Grontmij, was installed in 1998 and in 2002 the Anammox reactor, designed by Paques, was started up. Sludge granulation to keep high concentration of microorganisms in the reactor was adapted from anaerobic granular sludge reactors to maintain 40-50g/l of suspended solids concentrations.
In contrast to future reactors, no seed sludge nor operational experience was available for the start-up of the first large reactor, and incidental loss of biomass or occasional toxicity (by high nitrite concentrations, methanol slip from the Sharon reactor and incidental discharges of chemical toilet waste to the digester) delayed the start-up over three and a half years, until the design-loading rate was attained in February 2006. When full dewatering capacity was available, the loading rate could be increased to a level of 750kg/d, yielding a volumetric rate over 10kg/m3/d and converting 90-95% of the nitrogen fed to the reactor.
This loading rate is approximately five to ten times higher than for conventional denitrification, with proportional reductions in land area requirement. The investment costs for the Sharon/Anammox installation with a capacity of 1,200kg NH4-N/d were estimated at €2M (2001), yielding operating costs of only €1/kg of N removed - about five times less than conventional operating cost for nitrogen removal.
The Anammox process can be applied to any stream with high concentration of ammonia or organic nitrogen (> 200mg/l), such as wastewaters from chemical industries, food industries, power plants, and from animal waste.
The first full-scale Anammox plant for an inorganic wastewater from a semiconductor factory was installed in Japan, retrofitting existing facilities of conventional nitrification and denitrification.
Further research showed that under oxygen limitation, the two reactions, of aerobic nitrification and anaerobic oxidation, can occur simultaneously, and three more variations of the Anammox process are commercially available:
1. DeAmmon is a process developed by Purac, Ruhrverband, and the University of Hanover, Germany. It uses mobile biofilm carriers to maintain the slow growing organisms. A full-scale plant in Hattingen, Germany, has been in operation since 2003. The plant is operated at 180kg N/d with a reduction above 80%.
A full scale DeAmmon reactor was taken into operation in 2007 at the Himmerfjärden plant near Stockholm, Sweden. Designed for a nitrogen load of 480kg d-1, the reactor is based on the moving bed biofilm reactor (MBBR), with about 32% of its volume filled with suspended carriers (Kaldnes K1H). An outer biofilm layer performs nitritation while an inner layer the Anammox reaction.
After a 10-month start up period, over 80% nitrogen removal efficiency was obtained and nitrogen removal rates reached almost 2g m-2 d at a temperature of 28°C. In 2007, a first order for China was announced for Dalian in the north-east.
2. DEMON was developed and patented by the University of Innsbruck, Austria, and has been in full-scale operation without interruption at Strass since mid-2004. Sludge return liquors in a single-sludge sequencing batch reactor (SBR) are deammonified to reach annual ammonia removal beyond 90%.
The specific energy demand of the sidestream process equals 1.16kWh per kg N removed, compared to about 6.5 kWh for mainstream treatment, contributing essentially to energy self-sufficiency of the plant. Biomass enrichment and start-up in Strass took two and a half years, whereas startup of the second system at Glarnerland WwTP in Switzerland was reduced to 50 days due to transfer of a 20m3 tanker of seed sludge.
DEMON has been installed at nine plants across Europe, with six more on the way. Apeldoorn in the Netherlands is the biggest so far, with a capacity of 1600kg N/d. The co-digestion of indigenous and industrial sludge is used to generate energy.
3. ANITA-Mox is a one-stage deammonification moving bed biofilm reactor (MBBR), developed for cost-effective autotrophic N-removal from N-rich effluents. Pilot testing for more than two years on two different sites in Denmark and Sweden to treat reject water from anaerobic sludge digesters achieved N-removal rates of 0.8 kgN/m3.d and NH4 removal rate of 0.9 kgN-NH4/m3.d with N removal and NH4 removal efficiency of 80% and 90% respectively.
The continuous aeration control strategy employed was successful in limiting the nitrate production by nitrite oxidising bacteria that is detrimental to the performance of any one-stage deammonification process. The specific seeding strategy greatly shortened the lengthy startup period that is generally required for most deammonification processes and can limit their commercial potential.
The impact of different carriers design (AnoxKaldnes K1, K3, BiofilmChip and two other prototypes called MiniChip and Chip K3) on the performance of the ANITA-Mox process was also investigated at laboratory and pilot scale. Chip type carriers always have much higher volumetric N-removal rate (kgN/m3 react.d) than 'K' type carriers due to their greater surface area (m2/m3).
The original developer of Anammox in the Netherlands, Professor Mark van Loosdrecht of TU Delft, is hopeful that the improvements in efficiency of the process will allow it to treat dilute and colder effluents, offering a brand new approach to municipal wastewater treatment. This new approach will consist of two steps:
Firstly, the organic material is concentrated, enabling it to be used for the production of biogas and therefore energy. Secondly, the Anammox process is applied to ensure the effective removal of ammonium.
This would allow to switch from a net energy consumption of around 16kWh/y per person in the treatment process to a net production of 9kWh/y per person. In terms of power consumption, this turnaround would make a difference of around 85MW in the Netherlands alone