© Marine Technology, Vol. 37, No. 2, Spring 2000, pp. 111-115

Technical Report: Novel Oil/Water Separator for Treatment of Oily Bilgewater J a s o n A. C a p l a n , 1 C h r i s N e w t o n , 2 a n d D o n a l d K e l e m e n 3

A novel device that combines physical separation methods with biotechnology to treat oily bilgewater is described. Laboratory and pilot-scale experiments were performed to examine the ability of this device, tradenamed PetroLiminatorTM, to both separate free oil and biodegrade the dissolved or emulsified oil from shipboard bilgewater. Laboratory experiments were conducted to isolate and enrich bilge oil-degrading microorganisms. These microbes were grown in specially formulated liquid nutrients containing several hundred parts per million (ppm) of bilge oil as the sole carbon source. These cultures were inoculated into a laboratory-scale aqueous fixed-film bioreactor for determination of the required flow rate (i.e., hydraulic retention time) to remove ->99% of the petroleum hydrocarbons in the bilgewater. This information was incorporated into the design and operation of a 500 gal pilot-scale bioreactor installed aboard the 700 ft Cape LobosMARAD motor vessel. The bioreactor was operated for 70 days processing more than 90 000 liters of petroleum hydrocarbon (PHC) contaminated bilgewater. The average PHC concentration in the untreated influent was 70 to 90 ppm. The TPH levels in all treated effluent samples analyzed were well below 15 ppm, the U.S. Coast Guard (USCG) limit for legal overboard discharge. In fact, the removal efficiencies for the system were greater than 99% with no operational or maintenance problems noted. A newer model was developed that incorporated a physical separation chamber (Stage 1) upstream of the bioreactor chamber (Stage 2) in order to minimize the oil load to the microbes. A series of tests was conducted that closely mimicked the USCG tests for oil/water separators (OWS). The results were dramatic. The PHC levels in the effluent were below 15 ppm in all samples analyzed for the specified flow rate. Based on these data, it is estimated that the subject system with a footprint of 6 x 5 x 5 ft (L x W x H) is able to treat up to 86 000 gal of oily bilgewater per month. This system was USCG and IMO approved in January 2000.

Description of problem BILGEWATER has traditionally been a challenge to effectively and consistently treat, given the Federal requirements to discharge water with an oil-in-water concentration of 15 ppm or less within the territorial waters of the U.S. Offshore restrictions are less stringent (100 ppm) but are still difficult to meet and promise to become more restrictive with IMO initiatives. OPNAVINST 5090.1B directs that Navy ships with a n oil/water separator (OWS) and oil content monitor (OCM) shall attempt to limit oil-in-water discharges to 15 ppm worldwide. Oil/water separation has recently become more difficult with a newer generation of ships with dryer bilges and thus higher concentrations of oily waste and detergents to remove. Recent performance tests of parallel plate oil/water separators and coalescing bead separators on MSC ships have shown effluent oil concentrations frequently exceeding 15 ppm. Oily waste transfer system pumps and bilge cleaners tend to mechanically and chemically emulsify the oil respectively m a k i n g conventional coalescing technology ineffective in m a n y cases. Emulsified oil droplets below 20 microns in diameter will not be separated by conventional OWS and rise time is exceedingly slow. Uniform National Discharge Standards (UNDS) will require marine pollution control devices (MPCDs) for 25 discharges including bilgewater which is al-

1 President and chief scientist, EnSolve Biosystems, Inc., Raleigh, North Carolina. 2 Manager of maritime applications, EnSolve Biosystems, Inc., Raleigh, North Carolina. 3 Senior scientist, EnSolve Biosystems, Inc., Raleigh, North Carolina. Manuscript received at SNAME headquarters January 2000. SPRING 2000

ready regulated for petroleum oil content and deck runoff which is not currently regulated. Additional contaminants may also be regulated including heavy metals, cleaners, glycol, solvents, phosphates and nitrates. A n u m b e r of OWS technologies have been tested by the U.S. Government and shipping i n d u s t r y with the goal of reliably polishing oil from the effluent of conventional parallelplate OWS to meet current and anticipated oil discharge standards. Sorbents had limited success with the liability of solid waste generation while ultrafiltration has proven to be very effective and reliable. Sorbents can be contaminated with gross oil, and if it becomes ineffective, must be replaced. Ultrafiltration has high capital and maintenance costs. U1trafiltration also adds significant mechanical complexity to the oil t r e a t m e n t and removal process. Membranes m u s t be replaced when fouled approximately a n n u a l l y at significant cost. The ultrafiltration process also adds water volume to the waste oil stream that must be disposed ashore or incinerated aboard ship due to its concentrate reduction factor of approximately 100 to 1. Measurement of oil-in-water concentrations as low as 15 ppm is very difficult due to the variation of oil constituents, other contaminants and use of emulsifying detergents by the ship's crew. Unreliable OCMs make it more likely to inadvertently discharge oil overboard. This further necessitates the need for more reliable and consistent oil/water separators that remove oil and contaminants plus detergents that have negative effects on the OCM accuracy.

Solution One approach to effectively t r e a t i n g the c o n t a m i n a t e d bilgewater is the process of bioremediation. Bioremediation refers to the t r e a t m e n t or remediation of contaminated soils

0025-3316/200013702-0111$00.35/0

MARINE TECHNOLOGY

111

or groundwater using biological means. The biochemical conversion of organic and inorganic chemicals by bacterial populations is a n a t u r a l process. As a general rule, mineralization of organic compounds (i.e., oil) is characteristic of growthlinked biodegradation in which the microorganisms convert the substrate to CO2, cell components, and products typical of the usual catabolic pathways. The overall goal of bioremediation is to degrade organic pollutants (i.e., bilge oil) to concentrations that are either undetectable or, if detectable, to concentrations below the limits established as safe or acceptable by regulatory agencies. The bioreactor design (i.e., PetroLiminator) evaluated in this study was based on previous research conducted by EnSolve Biosystems showing the ability of m a n y types of microorganisms to grow as a film attached to a synthetic support media. Stage 2 of the subject design, containing the support media, is filled with a water source (i.e., oily bilgewater) and inoculated with a specifed amount of n u t r i e n t s and hydrocarbon-degrading microorganisms. Aerators, located beneath the media, provide the required oxygen concentrations to support bacterial growth and oxidation of the targeted organic contaminants. Once the microorganisms are immobilized on the support matrix as a biofilm, c o n t a m i n a n t flow-through conditions can begin resulting in a relatively constant concentration of microorganisms to exist within the system. Overall, the biofilm in the bioreactor operates with a relatively constant total mass of microorganisms within it, even though individual portions of the biofilm are constantly undergoing cycles of growth and sloughing. The bioreactor offers a low-cost solution that is mechanically simple with a small installation footprint. Unlike most polishing technologies t h a t cannot handle bulk contaminants, the bioreactor has an oil separation chamber (Stage 1) upstream of the bioreactor chamber (Stage 2) that removes oil using conventional oil separation technology. The bioreactor is similar in effectiveness to ultrafiltration polishing with an oil-in-water effluent concentration of 5 ppm or less with detergents and other typical bilge or deck runoff cont a m i n a n t s present. Furthermore, it is effective at removing other organic pollutants that are of concern to the Navy, EPA, and Coast Guard such as glycols, solvents, jet fuel, detergents, nitrogen and phosphates. Waste oil and other organic pollutants are not a product of the bioreactor effluent because the bacteria consume them producing harmless carbon dioxide and water. The aerobic bioreactor process does not generate toxic or u n p l e a s a n t biological odors often associated with s t a g n a n t holding tanks and separators. Emulsified oil, which wreaks havoc with coalescing oil/ water separators, is actually preferred by the bacteria since small oil droplets are processed more quickly. In addition, the microorganisms also metabolize organic surfactants. All influent water is processed on a continuous basis 24 hours a day (if required) and discharged overboard after discrimination by an OCM. Rejected effluent with high oil content is reprocessed; however, this is a rare situation. Additional waste-oil storage is not required and no solid waste stream is produced. The bioreactor destroys the organic pollutants using n a t u r a l biological processes, so it has no waste oil stream to handle. Trace metals (copper, nickel, iron, etc.) in moderate concentrations and salt water to not hinder the process.

Technical approach The use of microbes to treat wastewater is not a new concept. Municipal sewage water t r e a t m e n t plants have used various forms of biological treatments to reduce biological oxygen demand (BOD) before discharge. Bioremediation of hydrocarbon and solvent c o n t a m i n a t e d groundwater has proven much more effective t h a n conventional pump removal processes. Bacteria-laden fertilizers have been used to speed up the removal of water contaminated oil from storage t a n k seepage and coastal oil spills. Oil-eating bacteria are naturally present in our environment; however they degrade the oil very slowly. If high concentrations of bacteria are mixed into a bilge or oily waste holding t a n k (OWHT) they metabolize some oil but then quickly die off due to lack of n u t r i e n t s or oxygen. The bacteria will eventually be pumped out of the t a n k or bilge with the effluent water. The authors' company has advanced this technology to efficiently apply it to shipboard bilgewater and deck runoff treatment. In 1998, a US Patent was awarded for a shipboard fixed-bed bioreactor system designed specifically to remediate petroleum hydrocarbons typically found in bilgewater. O i l - c o n s u m i n g b a c t e r i a n a t u r a l l y p r e s e n t in the waste stream are selectively bred using n a t u r a l means to achieve the most efficient colony of bacteria for the waste stream to be treated. Several species are used to digest a broad range and variation of organic pollutants that present themselves. In this way the most efficient aerobic bacteria are selected and bred using accelerated n a t u r a l selection processes. The bacteria are stored and shipped dry with an organic bran-like matrix with a shelf-life of several years. No refrigeration is required. They are in suspended animation in the dry organic matrix. There are approximately one to two billion bacterial cells per gram of matrix. When they are mixed in the bioreactor water, they activate and start eating organic pollutants and breed at an exponential rate until they reach equilibrium conditions. Approximately one pound of organic matrix is required for a 500 gal reactor. Once the bioreactor is activated, the bacteria continue to live and breed in the controlled environment not requiring reactivation with additional organic matrix unless the system is dormant for several weeks. Since there are not genetically altered microbes they are not regulated by the EPA. They are not toxic to h u m a n s and other animals and pose no waterborne or airborne health risks. The effluent water, however, is not suitable for drinking. The aerobic bacterial cells (i.e., Pseudomonas) consist of a n a t u r a l biosurfactant coating around the cell wall. Inside the cytoplasmic m e m b r a n e are multi-enzyme complexes that actually metabolize the organic pollutants. Bacteria b r e a k down the long hydrocarbon chain into smaller components. Some species are more effective at metabolizing different parts (compounds) of the molecule. The hydrocarbon molecule diffuses through the cell wall and is broken down by enzymes in the first stage intracellular metabolism. The rem a i n i n g components diffuse through the cytoplasmic merebrane and degradation is completed by a multi-enzyme complex. In the presence of oxygen and n u t r i e n t s the complex hydrocarbon is broken down into carbon dioxide and water by the bacteria.

D e v e l o p m e n t a n d field e v a l u a t i o n Objective The objective of the following laboratory and pilot study was to evaluate the ability of the subject system to reduce or eliminate hydrocarbon concentrations in contaminated bilgewater. 112

SPRING 2000

The subject bioreactor system (Fig. 1) is designed for the biological destruction of petroleum hydrocarbons found in bilgewater. The basic approach is to t r e a t the c o n t a m i n a t e d bilgewater using an interactive system comprising, in combination, a pumping system from the phase separator chamber (Stage 1) and a bioreactor (Stage 2), which biodegrades the MARINE TECHNOLOGY

gal of oily bilgewater on a daily basis with m i n i m a l maintenance requirements. The ship selected for the bioreactor pilot evaluation was the MARAD vessel Cape Lobos docked in Wilmington, NC. The 700 ft motor vessel generated 2000 to 4000 gal of oily bilgewater on a monthly basis. The bioreactor was installed in a 4 × 4 ft a r e a in the machine shop (see Fig. l(a)). The installation was completed in under two days. Once the bioreactor was installed, it was filled with approximately 500 gal of fresh w a t e r and inoculated with specially enriched microorganisms developed by the owners. The addition of bacterial nutrients and pH control chemicals was performed by the use of slow-release chemical addition tubes inserted into the lid of the bioreactor. Each tube lasted 3 to 4 days and was easily replaced with a new tube in j u s t a few minutes. The influent pump line was placed directly into the ship's bilge, completely bypassing the use of the ship's waste oil holding t a n k and OWS. P r e t r e a t m e n t (influent) and postt r e a t m e n t (effluent) samples were collected on a s t a n d a r d regime and analyzed as described below. Biomass development was monitored by density of observed growth, plate counts, pH titration curves and dissolved oxygen measurements. Once t h e s y s t e m was at s t e a d y s t a t e , t r i p l i c a t e s a m p l e s (40 mL each) were collected in volatile organic analysis (VOA) vials and analyzed by gas chromatography/ m a s s spectrometry (GC/MS) using EPA Method SW-846! 8270. All GC/MS analyses were performed by P a r a d i g m Analytical Laboratories (Wilmington, NC). Summary

b Fig. 1 Prototype biotreatment system installed in the machine shop aboard the MARAD CapeLobosvessel (a). Redesigned USCG and IMO approved PetroLiminator System (b)

dissolved constituents of the hydrocarbons in the bilgewater. The treated system water is transferred to a third chamber (Stage 3) for continuous monitoring by the OCM prior to overboard discharge. Preferably, the treated bilgewater is disc h a r g e d overboard if a p p r o p r i a t e cleanup s t a n d a r d s a r e achieved (i.e., ---15 ppm). The system was designed to remediate approximately 340

of results

The system was operated on board the Cape Lobos for over 70 days with no operational or maintenance problems noted. It is i m p o r t a n t to note t h a t the system was m a i n t a i n e d on a daily basis by the crew of the Cape Lobos and not company personnel. However, the company did provide technical support when requested. The proprietary slow-release chemical tubes m a i n t a i n e d the n u t r i e n t and pH levels within their desired range throughout the entire evaluation period. During this period, the entire cargo a r e a of the ship was washed down with liquid Tide detergent and this w a s h w a t e r eventually passed through the biotreatment system. No detrimental effects to the system were observed. The GC/MS d a t a clearly indicated the high efficiency of the system for degrading the petroleum hydrocarbon constituents found in the bilgewater. On average, the total petroleum hydrocarbon (TPH) concentration in the bilgewater influent was 70 to 90 ppm (Fig. 2).

90 80 70

TPH [ppm]

60 50 40 30

15 ppm USCG limit for discharge

20 10 0

,,,,,,,,,,,,,,,

Typical Influent

Week 1

Week 2

Week 8

Sampling time Fig. 2 GC/MS data indicating total petroleum hydrocarbon (TPH) concentrations in effluent (post-treated) samples from the bioreactor on board the Cape Lobos.

SPRING 2000

MARINE TECHNOLOGY

113

All effluent samples analyzed showed TPH concentrations of less t h a n the target 15 ppm concentration. In fact, typical effluent concentrations were less t h a n 3 ppm. Overall, the system performance was outstanding during the pilot evaluation.

Discussion The prototype bioreactor was constructed from an HDPE plastic t a n k t h a t is lightweight but not strong enough for stresses associated with rough seas. The l a t e s t version of the system is constructed from either 304 stainless or carbon steel (see Fig. l(b)). Another modification to the system was changing the shape of the bioreactor from cylindrical to rectangular. The rectangular shape is structurally stronger and increases the available surface area of the bacterial support matrix. The slow-release formulation of the EnCell bacterial nutrients has also been improved to provide higher concentrations of the required nutrients. In addition, the nutrients and pH a d j u s t m e n t chemicals are now available in compressed tablets, thereby eliminating the need for piping to house the powdered chemicals. The bioreactor processes w a s t e w a t e r most efficiently on a continuous basis 24 hours a day. This is a different operating philosophy t h a n batch processing currently employed aboard ships using conventional coalescing oil/water s e p a r a t i o n technology. The bioreactor, however, can lie dormant for extended periods without p u m p i n g effluent as long as aeration and nutrients are provided. The (hourly) processing r a t e is s o m e w h a t slower t h a n conventional oil/water s e p a r a t o r s typically used for several hours per day; however, the daily throughput of the "slow and steady" bioreactor is equivalent to the batch mode OWS. A good analogy is the story of the race between the tortoise and the hare. The slow and steady tortoise (bioreactor) wins the race while the fast h a r e (conventional OWS) loses. Conventional OWS technology loses because it can't consistently provide effluent with oil-in-

w a t e r concentrations of 15 ppm or less. It spends most of its time recirculating and reprocessing failed effluent and it cannot handle oil emulsions. Shipboard personnel have to carefully monitor the OWS to m a k e sure it operates properly and m u s t refrain from using bilge cleaners t h a t work. The bioreactor wins the race because it destroys oil, detergents and other organic wastes without the need for watching by the crew while providing oil-in-water effluent concentrations consistently below 15 ppm.

Commercial version The success of the pilot study led company engineers to develop a version for commercial applications (Fig. lb). This system is available in stainless or carbon steel (epoxy coated). On J a n u a r y 28, 2000, EnSolve received both U.S. Coast G u a r d and IMO approval of the PetroLiminator System. To date, this system is the only installed and operating USCG and IMO approved biomechanical oil/water separator. The system is r a t e d for a continuous flow rate of 2880 gallons of bilgewater per day. The results of the U.S. Coast Guard tests were dramatic. Stage 1 was successful in reducing #2 and #6 fuel oils from 60 lb/hr entering Stage 1 to 0.0004 and 0.026 lb/hr, respectively, exiting Stage 1 (Fig. 3). This represents reductions of 2300 to 145 000 times influent levels for #2 and #6 fuel oils, respectively. The Stage 2 biochamber was successful in reducing the total PHC concentrations to less t h a n 15 ppm for both fuel oil types (Fig. 4). These results were confirmed by both the OCM and IR analyses. The d a t a generated from the Turner OCM with very close (i.e., -+3 ppm) to those analyzed in the split samples sent to the lab.

System features Several key features of the subject system have been identified:

80-

70.

60

Hydrocarbon Loadin~ (lbs/hr)

50

40

30.

20.

10.

0.026 lb

0.0004 Ib

~ # 6 Fuel O i l

I

Stage 1

# 2 Fuel O i l

#6 Fuel Oil

.,dWP"J I

#'2 F u e l O i l

!

Stage 2

Fig. 3 Comparison of petroleum hydrocarbon loading for Stage 1 vs. Stage 2 in the PetroUminator bilgewater treatment system. Loading rates for Stage 1 based on directly added amounts of either bunker C (#6 fuel oil) or diesel (#2 fuel oil). Stage 2 loading rates based on water sample analysis using EPA Method 418.1 to determine hydrocarbon concentrations

114

SPRING 2000

MARINE TECHNOLOGY

45. 40. 35. 30.

TPH [ppm]

25.

20 "1

~

~

U

S Discharge C G Limit

15.,

_

o J

#6 Fuel Oil

4#2 Fuel Oil

Stage 2 Fig. 4

•"~°=~,~ , 7 Fuel C)il

# 6 Fuel Oil

Stage 3

Comparison data showing reductions of total petroleum hydrocarbon concentrations (TPH) from Stage 2 to Stage 3 in the biotreatment system for two different fuel types. Hydrocarbon concentrations in the water phases determined by EPA Method 418.1

• C o s t S a v i n g s - - S h i p o w n e r s c a n s a v e over $200,000 p e r y e a r for e a c h s h i p i n costs a s s o c i a t e d w i t h d i s p o s a l of bilgewater. • A u t o m a t i o n / L o w M a i n t e n a n c e - T h e s y s t e m is d e s i g n e d for c o n t i n u o u s , a u t o m a t e d o p e r a t i o n , t h e r e b y m i n i m i z i n g t h e n u m b e r of m a n - h o u r s r e q u i r e d to deal w i t h bilgew a t e r disposal. • P r o c e s s e s All O i l / W a t e r M i x t u r e s - - T h e s y s t e m is a t h r e e - s t a g e s y s t e m t h a t r e m o v e s p h a s e s e p a r a t e d oil as well as e m u l s i f i e d oil.

SPRING 2000

• Compatible with Strong Degreasers and Detergents-T h e s y s t e m a c t u a l l y p r e f e r s e m u l s i f i e d oil.

Metric conversion

factors

l f t = 0.3048m llb =0.45kg l g a l = 3.78L

MARINE TECHNOLOGY

115