Learn how government resources can help your business sell services internationally. David Josephson, managing direct of the Export-Import Bank of...
Goulston Technologies is a textile chemical and lubricant producer specializing in the development, formulation, manufacturing and application of spin finishes and other fiber lubricants and treatments. Goulston generates approximately 10,000 gallons per day of wastewater that is contaminated by about 3 percent concentration of various oils, lubricants, synthetic organics, emulsifiers, dispersants, silicone fluids, viscosity modifiers, antistats and detergents. The emulsion generated is difficult to treat and varies sufficiently to make conventional separation and flocculation techniques inadequate or undependable.
Rather than haul the wastewater away at a cost of $0.28 or more per gallon, Goulston has practiced evaporator techniques over four years to separate the water from the oil and surfactant mixture. An early model vapor compression evaporator was installed in January 1994 that had nominal capacity of 5,000 gpd of water throughput. The efficiency and throughput of this unit was not sufficient to support either the volume increases since then or Goulston’s corporate goal of eventually attaining zero process water discharge and materials recovery.
A joint development effort was pursued with a local company, Recovery Technologies Corporation (RTC), Charlotte, N.C., and its technology took the evaporation process one step further in terms of dewatering, and also toward active product recovery and potential recycle. Goulston’s strategic goals include the desire to become independent of third party waste treatment vendors, water recycle and eventual oil and surfactants recovery for future use or value extraction.
The current two-step process incorporates a primary evaporator with vapor compression and heat recovery, and a secondary evaporator with vapor compression and heat recovery. The vapor compressed from the second stage is theoretically ported to the primary evaporator for supplemental heat, or its heat can be extracted to the waste stream coming into the primary evaporator (current orientation).
The primary evaporator takes 3 percent wastewater and concentrates it to 65 percent, and the second stage takes the 65 percent and concentrates it to 95 percent. Total dewatering (100 percent concentration) is possible but not practical with the mixture that is yielded. A theoretical third stage (TF2) is being piloted that will take the 95 percent stream, dewater it and separate it into two distinct phases–lights and heavies. Different potential uses are envisioned for the lights and heavies. The water recovered in the process of condensing the compressed vapor is currently discharged to POTW, but is intended in the near future to be recycled for plant use, either for plant washing purposes or for finished product makeup. A simple schematic of the Goulston implementation is shown in Figure 1.
In the past, inorganic additive methods for destabilizing and separating the wastewater mixture were used. However, significant inorganic materials remained in the water phase and the separation was neither good enough to satisfy the effluent discharge permit nor consistent and dependable enough for continued use.
Therefore, pilot work was done with ultrafiltration and reverse osmosis technology. The initial tests looked promising, so a much larger system was installed for study. In the long run, the approach proved to be inappropriate for this type of wastewater. Over a year’s time experiments with different membrane compositions, recycle rates, backwash frequencies, etc., were conducted. These trials could not stabilize a steady state operation. Membrane deterioration, blinding and plugging were constant problems, and the maximum concentration that could be achieved without binding the system was about 8 percent. At this level, the system generated approximately 1/3 more wastewater due to backwashing than the plant generated. The system was abandoned and dismantled as a result.
Biological systems were investigated, but concerns about waste stream variability and the effect of surfactants and detergents on bacteria resulted in a decision to embrace more fundamental technology that was not subject to biological inconsistency or frailty. Laboratory work ensued, focusing on distillation. Laboratory bench scale studies were performed that indicated distillation was feasible. Foaming and overhead carryover with the wastewater appeared not to be a problem.
A larger pilot study was run comparing a vacuum distillation unit and vapor recompression unit. Both units performed reasonably well, but the vacuum unit was estimated to be more costly to purchase, operate and maintain. The vapor recompression unit performed consistently and appeared to be a simpler process with superior heat recovery potential.
Goulston installed its first vapor recompression unit in January, 1994. This was a shell and tube designed kettle, with a small de-entrainment section and vapor compressor (blower) assembly. The unit was sized for nominal 5,000 gpd water distillation capacity, and yielded about 3,000 gpd of actual wastewater handling capacity. The output was about 15 percent concentrate. These limitations caused Goulston to collaborate with RTC. The product was a system that used a different heat exchange and recovery design. More heat recovery was built into the system, with a two-stage approach to achieve essentially complete dewatering if desired.
A third stage is envisioned that will allow the system to partially refine the oil and surfactant mixture that is recovered from the two stage process. Bench studies have indicated possible success, and the pilot work to validate this approach currently is under way.
The goals of collaboration with RTC for the improved process were greater throughput, higher concentration, greater heat recovery, process variable remote control and reporting, and more accessible design for repair and cleaning purposes. These goals were largely met with the current design. It has a nominal water handling capacity of greater than 15,000 gpd, and wastewater handling capacity of about 12,000 gpd. The first stage normally achieves 65 percent concentration, which is more than a 20 fold reduction in wastewater volume (four times more than the old evaporator), and the second stage completes the dewatering process.
Approximately one third of an operator’s time is required for adequate management of this system. This includes incidental duties such as changing bag filters, taking samples, running pH, TOC measurements, observing system operation and recording any specific data required for study or management. The operator should be a certified wastewater systems operator or chemical operator. Generally this will require a high school or associate degree in a mechanical, chemical or engineering with a certain degree of mechanical and electrical process control experience.
Project justification was based on a comparison of the cost of hauling all the raw waste to an outside waste processing vendor and also on the then current partial evaporation situation. Figure 2 describes the general statistics used to estimate payback period assuming all the wastewater generated is disposed of by hauling (no pretreatment). Assumptions were that the wastewater volume would increase 12 percent annually (conservative); disposal costs were scheduled to increase 5 percent per year. No other benefits from future project goals such as refining and recovering raw materials were considered. Figure 3 compares the cost of the new installation with the cost of disposing of Goulston’s current portion of wastewater. This is a smaller volume because the present evaporator system concentrates the wastewater to about 15 percent. Higher concentrations cannot be achieved with the older evaporator system because of design and capacity limitations.
Figure 4 represents the current operating costs of the system. Ten percent of these costs currently come from the disposal of high concentrate to outside vendors. This practice will stop when the pilot for TF2 is complete and equipment is installed to refine the O/SM. Forty percent of the cost is a depreciation estimate, 20 percent is maintenance and human interaction with the system and 30 percent is the approximate fraction due to utilities (primarily electricity) required to operate. Looking at pure operating cost when the refining unit is in or when an agreement with a vendor to take the high concentrate for free (i.e., for incineration and BTU value) is reached, one could add the utilities and maintenance to come up with a price of $.025 per gallon under ideal, uninterrupted operation. Occasional maintenance downtime for cleanouts or mechanical repairs would realistically get this cost up to about 3 to 3.5 cents per gallon. This is still a very practical pre-treatment cost.
These cost estimates do not take into account any future value of water or O/SM recycle. Taking these factors into account, it is expected that the whole process will eventually be a profit center. The essence of good design incorporates recovery of value. In the process of dealing with a difficult problem, a company may actually realize an opportunity for cost reduction in utilities and raw materials. In this case, pre-treatment actually pays off.
Recovered Wastewater and Quality
Figure 5 shows the quality of combined distillate coming off the main evaporator and TF1 (linear regression of data). Since the TF1 operates at higher temperatures, it contributes a higher concentration. However, this is low flow compared to the main evaporator. These values are well below the company’s lbs/day BOD limitation. Statistically, the BOD runs about 1/2 to 2/3 of TOC. Since the effluent is a combination of distillate and RO reject water from out demineralized water make-up system, these are the best figures currently available for the individual process output. This level of performance is adequate for the permit application. Bear in mind that performance will vary upon the application and the presence of higher volatility components in the waste stream. Applications where water forms an azeotrope with a constituent in high concentrations or where components in the waste stream are highly volatile are probably not suitable candidates for this approach. These waste streams would demand a different separation process. The distillate currently is discharged to the local POTW, but the intention is to distribute the distillate back into the plant for washdown purposes. Distillate, being essentially demineralized, is highly aggressive to plant floors and certain other materials. However, combining it with the distillate from the RO effluent, which is hard water, should form a slightly soft industrial water stream for general plant washing use.
The recovered byproduct currently is being analyzed. The analytical laboratory did a split of this material under vacuum at various temperatures and obtained some surprising results. Experiments ranged in temperature from 130° to 220° C after dewatering the sample to full dryness. The fractions were called distillate and residue. With increasing temperature,
• the amount of collected distillate increases as the amount of residue decreases,
• the color of distillate gets darker, and
• the viscosity of distillate increases.
Interestingly, the composition did not change much with increasing temperature. The distillate was mostly composed of lubricating esters (palmitate/stearate types). The distillate is primarily non-ethoxylated types, though at 220° C some alkoxylates become visible. The distillate also is more unsaturated than the residue. Acid values indicate that some cleavage of free acid may be taking place at these temperatures. The residue is composed mostly of fatty acids C8—C18, at rather evenly distributed proportions.
From these experiments it is clear that there are potential values to be reclaimed in the waste oil mixture and that a minimum of refinement will allow a split of a liquid mixture from a semi-solid grease. It is estimated that 60 to 70 percent of these materials can be successfully reused in some way. If so, and assuming half the market value for such materials, Goulston stands to recover in excess of $500,000 in materials. This would result in a payback for the TF2 and support equipment somewhere in the range of six to nine months. When folded into the costs for operating the entire system, the company expects to materialize a profit of well over $350,000 per year.
This process takes fundamental engineering concepts, laws and processes and applies them in a novel way. The basic technology used in all stages is proven in a variety of applications from water de-salination to caustic soda manufacturing, to foods and pharmaceuticals processing. With the application of any technology to a new use, whether proven or not, there always are surprises and pitfalls. There has been a learning curve about proper heat exchanger selection and design, recycle rates, limits on concentration, venting, temperature settings in various process stages, etc. Further developments are expected to improve thermal efficiency and improve capacity. New exchanger designs are being investigated and pilot work is scheduled.
Fortunately, the length of time to acquaint ourselves with the various facets of this technology has left us with a renewed commitment that this process technology works, works well and is robust in the face of variable waste composition. Having a quality supplier and collaborator (Recovery Technologies Corporation) was important in that this company provided support for the project from the beginning through today with total commitment.
Another key factor is having an open-minded and intelligent engineering and operating team that is committed equally to the process of making the process work. The teamwork and commitment in both organizations has to be complete. Patience during the shakedown and adaptation period was very useful. There were times when we thought we had hit a wall, and through patient review of principles and mechanics we were all able to not only work through some serious problems, but actually improve the process.