|
BIOFILTRATION PILOT UNIT STUDY TO CONTROL
FUGITIVE EMISSIONS
FROM A SOLVENT MIXING AREA AT
IMATION ENTERPRISE'S WHITE CITY, OREGON FACILITY
by
Ray Willingham/Scot Standefer - PPC Biofilter
Janice Tacconi/Bruce Bullough - Imation Enterprises (Borne of 3M Innovations)
Introduction
PPC Biofilter, in conjunction with Imation Enterprises conducted a six month
pilot study to determine if biofiltration was a viable option for controlling
fugitive VOC emissions from a solvent mixing area. Imation Enterprises manufactures
medical imaging film at their White City, Oregon facility. This process uses
a variety of solvents including methyl ethyl ketone (MEK), acetone, methanol,
ethanol, toluene, heptane, and ethyl acetate, with gas stream concentrations
ranging from 0 to 100 ppm (as solvent).
PPC provided a commercial pilot unit containing two cubic meters of biofilter
media, from which operating data and background information on biofiltration
was gathered to evaluate the effectiveness of the technology for controlling
VOC emissions from a highly variable source. Although the VOC concentrations
were dilute, they represented a significant fraction of the uncontrolled emissions
from the facility. This study indicates that biofiltration can be an effective
and economical control option for dealing with highly variable VOC emissions
sources.
Equipment
The pilot unit provided by PPC is a skid mounted unit, consisting of a pretreatment
humidification chamber, followed by a totally enclosed insulated vessel containing
two separate biofilter beds operating in series. Each bed contained one cubic
meter of filter media. The pilot unit is fully instrumented and all pertinent
operating parameters are logged and controlled via a computer control system.
Air flows through the pilot unit are measured using orifice plates and differential
pressure transmitters. The flows were set using manual gate valves. Differential
pressure transmitters were also used to measure the pressure drop across the
humidifier and both filter beds.
The moisture content of the filter media was maintained via the computer
control system. To monitor the moisture content of the media, load cells are
located in the reactor vessels to monitor the weight of the filter media.
This weight is related directly to the moisture content of the filter bed.
Initially, the control system was calibrated for the load cells to activate
solenoid valves which control the addition of water to the filter media. During
the course of the study, the control logic was modified so that water was
added to the media on a timed basis and the load cells were used to monitor
the system.
Pilot Unit Operation
Gas Residence Time
During the course of the study, the pilot unit was operated at a number of
different flow rates. Initially, the flow rate through the biofilter was set
to provide an empty bed gas residence time (EBRT) of approximately one minute.
Two months into the study the flow rate was adjusted to provide a 15-20 second
EBRT. During the third phase of testing the flow rate was set for a 30-35
second EBRT..
Media Moisture Content
The performance of a biofilter is directly related to the moisture content
in the filter media. Throughout the course of this study, the moisture content
of the filter material was measured periodically, and the control system was
adjusted to maintain a uniform moisture content. Initially the pilot unit
was set to maintain and approximate moisture content in the filter media of
60-65%. During the course of the study, it became evident that the load cell
control system was not capable of adequately responding to moisture needs
of the filter material and a timer system was incorporated in the control
system. The load cells were, however, reliable as monitoring tools for determining
stable moisture contents and system operation. During the final phase of testing,
the media moisture content was maintained at 65-70%.
Gas Temperature
The temperature of the gas stream entering the biofilter was monitored continuously
and controlled as much as possible during the course of the pilot study. Inlet
gas temperatures of 4.4°C (40°F), after the humidifier, were observed
along with low removal efficiencies. Experience has shown biofilters to be
much more effective at temperatures above 21°C (70°F). Attempts were
made to control the inlet gas stream temperature using the humidifier sump
heater. At the low flow rates (~60 second EBRT) the sump heater was able to
maintain gas temperatures above 32°C (90°F). When the flow rate was
increased to 460 m3/hr (15 second EBRT), the inlet temperature dropped to
21°C (70°F). At a 38 second EBRT, the inlet gas temperature fluctuated
between 21°C (70°F) and 37°C (98°F), depending on ambient
conditions.
VOC Monitoring
Throughout the pilot study, VOC concentrations were measured using both a
flame ionization detector (FID) and by gas chromatography with a FID (GC/FID)
The FID was used in an effort to obtain continuous inlet and outlet VOC concentration
readings. The FID proved to be an unreliable measurement tool because moisture
in the sample lines, with a high organic content, fouled the detectors. The
FID results can only be used to note trends in the data, although the correlation
between the FID and GC/FID data was fairly close.
Ambient temperature also affected sample delivery conditions to the FID.
Thorough and rigorous design of the sample delivery system with an analyzer
intended for continuous duty, could provide the destruction and process efficiency
monitoring needed. However, once a stable ecosystem has been established within
the biofilter vessel, monitoring the process variables such as inlet temperature
and media moisture content, provides reliable information regarding system
performance, in much the same way that operating temperatures in a thermal
oxidizer are related to system performance.
Waste water drainage from the biofilter had no detectable solvent content
and required no pH adjustment before discharge. There was no blowdown of the
recirculation water during this study, and at the end of the trial, solvent
concentrations were below discharge limits.
System Performance and Results
The results of this pilot study indicate the biofiltration system works
well in removing VOCs from the tested gas stream. The biofilter was able to
handle the variability in the VOC concentration and composition and still
provide consistent, reliable performance. The overall performance of the biofilter
throughout the course of the study was 83% removal of the VOCs. During the
final stage of the study, the biofilter consistently demonstrated removal
efficiencies greater than 90%.
Of particular interest is the resiliency of the system. During the course
of the study, the biofilter media dried out and this was reflected in the
performance of the system. When the moisture content was brought back in line,
the system performance rebounded, in a relatively short period of time, to
previous performance levels.
Gas Residence Time
As previously mentioned the gas residence time in the bed was varied between
15 and 60 seconds. Good removal efficiencies occurred at all residence times,
but the most consistent removal was obtained at 30-35 second EBRT.
VOC Loading
The wide variability in the concentration and composition of the VOCs in the
process gas stream caused some initial concern regarding system performance.
Due to this variability, removal efficiencies were not stable and did vary
over time, however the system was able to respond to changes and maintain
an average control efficiency of greater than 90%.
The availability of regular GC analysis allowed for the evaluation of the
biofilter performance on specific compounds. As expected, low molecular weight,
highly biodegradable compounds were removed most readily. Alcohols present
in the gas stream were removed consistently, with average removal efficiencies
of 97-98%. Average acetone removal efficiencies were 73% at a 15 second EBRT,
91% at a 30 second EBRT, and 93% at a 60 second EBRT. Average MEK removal
efficiencies were 82% at a 15 second EBRT, 93% at a 30 second EBRT and 90%
at a 60 second EBRT. It is interesting to note that the increase in EBRT from
30 to 60 seconds did not provide a significant improvement in VOC removal
efficiency.
Of particular interest was the affect of alcohols in the gas stream on acetone
removal. When alcohols were present in the gas stream, acetone removal was
adversely impacted and removal efficiencies dropped by as much as 10%. Although
there was a direct correlation between alcohol concentration and acetone removal,
there was not enough data to develop a relationship. Although the preferential
degradation of compounds within a biofilter is accepted, more work needs to
be done to explore the relationship between specific compounds in a particular
gas stream and the removal efficiencies of the specific compounds.
Media Moisture Content
The media moisture content is the single most important variable affecting
biofilter performance. During the most consistent period of operation, the
moisture content of the filter media was maintained at 65-70%. Unfortunately,
gas flow changed and temperature fluctuated during this period and it is difficult
to attribute the excellent system performance to any single parameter. However,
during the course of the study, there were a number of times the system performance
dropped off significantly. During each of these periods of reduced removal
efficiency, it was observed that the moisture content of the media had dropped
significantly. When the moisture content of the media was returned to previous
levels, high removal efficiencies returned.
Conclusions
The biofilter effectively dealt with the emissions from the solvent mixing
room. It was able to adapt to the variable organic loading that was inherent
in the system and provide consistent removal efficiencies of greater than
90%. Of particular importance, the biofilter was able to respond quickly and
effectively to variations in gas stream composition and concentration. Based
on the results of this study, the biofilter is a technically and economically
viable alternative to more traditional VOC control technologies such as thermal
oxidation.
This study also indicates the attention needs to be paid to the delivery
of the treated gas stream to the biofilter. To provide the most stable operation,
minimizing fluctuations in the inlet gas conditions at the biofilter may be
necessary. Of primary importance is the temperature of the gas stream after
humidification. To maintain a stable inlet temperature, the inlet duct work
may have to be insulated to minimize the impact of ambient conditions. Additionally,
if the process gas stream has an inlet temperature, after humidification,
below 21°C (70°F) it will probably be necessary to increase the gas
stream temperature prior to humidification.
Pilot studies such as this one give a clear indication of the applicability
of biofiltration for controlling VOC emissions. Biofiltration will not meet
all VOC control needs, but as more commercial installations come online, biofiltration
will become an accepted alternative to more costly control technologies in
situations were it does apply.
For more information, contact Scot Standefer, president of PPC Biofilter,
3000 E. Marshall Avenue, Longview, TX 75601; phone (903) 758-3395; fax (903)
758-6487.
Back
to Menu
|
