Florida is the largest user of brackish water reverse osmosis (BRO) for groundwater treatment in the world in both capacity and number of plants (Mickley et al., 1993).  The capacity of BRO plants has increased exponentially along with the state’s increasing population (Figure 1.2).  Saltwater intrusion into the aquifer, degradation of surface water quality, along with a rapidly growing population and a potable water demand that should increase by 60% by the year 2020, mandate the development of alternative water resources for the coming decades (Vergara, 1998).
 

     Almost 200 MGD [760,000 m3/d] of potable water is produced by 123 desalination plants to serve 1.3 millions people out of the 15 millions of people living in Florida (Table 3-1).  Among those plants, only two use EDR (in Sarasota and in Venice), and ten use MS (in Lee, Broward and Palm Beach Counties); all of them have treatment capacities greater than 0.5 MGD [20,000 m3/d]. Thirty-one of these plants have a treatment capacity greater than 1 MGD [3,900 m3/d] and produce 171.4 MGD [670,000 m3/d] which represents around 90% of all the Membrane production in Florida (FDEP, 1998).  If 80% of the freshwater is recovered from the source water, concentrate production is estimated to be 40 to 60 MGD [156,000 to 234,000 m3/d].
     Florida is the largest producer of membrane water from groundwater in the world (Mickley et al., 1993).  Therefore, the sunshine state can be considered a pilot state for concentrate disposal investigation by industry, and legislative and regulatory agencies.  NPDES and UIC permits require monthly monitoring of concentrate quality.  The number of plants in Florida along with the years of monitoring have resulted in the largest database on the nature of concentrate discharges and their disposal in the country.  Most articles published in the literature on brackish desalination focus on Florida plants or projects whether they deal with cost considerations, plant operations, membrane upgrades or by-product disposal.
     Regardless of the difficulty in obtaining permits, FDEP favors RO and Membrane Technology for drinking water production in areas where adequate supplies of potentially potable water are not available.  In 1997, FDEP proposed changing the classification of membrane concentrate, removing the industrial waste designation.  The FDEP also proposed establishing a new classification for a blend of concentrate and domestic wastewater.  Changing this classification would also simplify the permitting process for smaller facilities (treatment capacity less than 1MGD [3,800 m3/d]).  This proposal also favors the establishment of mixing zones for RO discharges.  In addition, Class I municipal wells could be used for disposal of concentrate, avoiding the costs of upgrading to a Class V well.  The USEPA is presently considering this classification system.  In addition, legislation is pending in Congress which would implement this classification system (Bill HR 1106, 1999).
 

3.1 Monitoring

 3.1.1 Water Quality.

    Similar concerns are also true about gross alpha radiation and some heavy metals, which have been reported in concentrations higher that the surface water quality standard listed in the rule 62-302.400 FAC.  For example, in 1993, mercury concentrations of the concentrate from the Indian River County North Beach Reverse Osmosis Plant of Wabasso have been found around 0.34 µg/L, which exceed the 0.025 µg-Hg/L allowed in Class II waters (http://www.dep.state.fl.us/-labs/biol/Report_Smmrys/BSSY_Smmrys/-CEdist/9303_03CE.html).
    The toxicity of a concentrate is usually blamed on a specific toxic compound if it is detected in concentration violating water quality standards.  However, it is sometimes difficult to determine the cause of effluent toxicity if none of the common toxic species is found at suspect levels.  Thousands of different toxicants from heavy metals to pesticides are a potential menace for aquatic life.  Each of them can be detected with a specific method and is harmful at a specific level, and it is virtually impossible to check for all of them in a given concentrate.

3.1.2 Toxicity.

     A comparison between concentrate and source water at four Florida membrane plants is shown in Table 3-2.  These plants have been chosen as case studies, because they represent several treatment processes: one SWRO (Anclote River Pilot Plant), one low pressure BRO (Venice RO Plant), one medium pressure BRO (Jupiter RO Plant) and finally one NF (Ft Myers MS).
     The Concentration Factor (CF) is the ratio between the concentration of the concentrate over the source water concentration for a given ion.  It is dependent on the recovery rate of the plant.  The CF values for the four plants vary from 1.6 for the SWRO plant (50% recovery) to 4.6 for the NF plant (90% recovery).  The CFs of the different chemical species for the three RO plants seem to be consistent for each individual plant with relative standard deviations (RSD) between 40 and 110 ppt.  This means that the ions present in the source water are concentrated in the desalination by-product by the same amount.  On the other hand, the RSD of the NF plant of 1000 ppt is due to an impressive sulfate enrichment (CF=17) and a selective behavior of the NF membrane to double charged ions.  The sulfate probably comes from pre- (sulfuric acid for pH control) or post-treatments (high level of hydrogen sulfide transformed into sulfate by aeration).
     The calculated values are the theoretical values obtained from the concentrate’s chloride concentration using the standard seawater ions ratio (Berner and Berner, 1996).  Chloride has been chosen as it is the most abundant ions in seawater.  A deviation between the concentrate values and the calculated concentrations shows an ion imbalance.  It seems that such an imbalance may lead marine and estuarine aquatic life to be exposed to concentrate toxicity.  This toxicity may also prevent plants from discharging the concentrate to surface brackish water.
     In the case of the toxicity testing results from the Melbourne ROWTP presented in Appendix A, it seems that the undiluted concentrate from the facility was chronically toxic for both the island silverside fish (Menidia beryllina) and the mysid shrimp (Mysidopsis bahia) but only acutely toxic to the mysid shrimp.  However, a fourfold diluted effluent passes the toxicity screen in each cases.  A fourfold dilution can be obtained with an appropriate mixing zone.  In the case of the Melbourne plant, the mixing zone extends 1,500 feet after the point of discharge.  If a mixing zone is allowed, the water tested for toxicity should be collected at the end of the mixing zone and not at the point of discharge.
    In 1993, FDEP tried to develop a testing protocol that would indicate whether the concentrate represents toxicity to the receiving water aquatic life and whether this toxicity results from the presence of specific toxicants or from an imbalance in the major sea water ions ratio. Indeed, several concentrates from plants using fresh and brackish water as source water were tested during the elaboration of the protocol and clearly presented a major sea ions imbalance and were found toxic (FDEP, 1993).  However, the whole protocol relies on very accurate analytical chemistry and toxicology.  The concentrations of all the different major ions present in the concentrate have to be known with extreme accuracy.  Each steps of the toxicology testing increases the difficulty of obtaining reliable results. General consulting laboratories may not be able to perform such intricate method which will probably be reserved for exceptional situations..  The whole protocol still needs further examination and would benefits from peer review.
 

3.2 Proposed Options for Concentrate Disposal

3.2.1 Surface Discharge.

3.2.2 Blending with Domestic wastewater and POTW.

3.2.3 Deep Well Injection.

3.2.4 Land Application and Evaporation Pond.

     A combination of methods of disposal is most likely to offer the best answer to the question of concentrate disposal.  Surface discharge can be coupled with deep well injection or a sewer system; deep well injection may be associated with blending of wastewater (which is not considered Industrial Waste allowing the use of Class I Wells) where possible.  Domestic wastewater treatment plants receiving concentrate will probably need to increase their capacity to accommodate this increased flow.
 

back                        next

about the author
take a look at my resume.

drop me a lign or two with
your comments and advices.
[email protected]