|
The
management of volatile organic compounds in bakery air
emissions was identified as a high priority problem
by the American Bakers Association in conjunction with
their member companies. Biotrickling filter technology
was evaluated among other approaches and was selected
for a pilot study to control ethanol emissions from
a bakery oven exhaust gas stream. The pilot study was
conducted by PRD Tech Inc. in conjunction with other
partners.
Biofilters
and biotrickling filters use microbial populations in
biofilms that grow on support media to degrade or transform
contaminants in the air. A biofilter uses natural support
media, and relies on indigenous microorganisms and nutrients
present in the media. A biotrickling filter uses synthetic
support media and employs a water/nutrient recycle stream
to distribute nutrients (phosphorous, nitrogen, etc.)
and to keep the media moist.
The
biotrickling filter employed in the pilot study used
PRD Tech's proprietary support media. PRD Tech has developed
a unique trickling biofilter that has been demonstrated
at pilot-scale to treat odors and emission of volatile
organics. The biofilter design is based on research
funded by U.S. EPA., and conducted at the University
of Cincinnati. Figure 1 shows a schematic of the biotrickling
filter system.
FIGURE
1
Air
contaminated with odors and organics flows through a
bed of proprietary support media, which has active microrganisms
immobilized on its surface. Nutrients, specially formulated
for a biofilter system, are sprayed into the bed to
keep the support media moist and supply chemicals to
enhance biological activity in the biofilter bed. Biomass
growth in the biofilter bed does not cause plugging
and the media requires no periodic cleaning to maintain
low gas-phase pressure drop.
A
pilot study was conducted from September 1997 to February
1998, at the Wonder Bread Bakery of Interstate Brands
Corporation in Columbus, Ohio. The objective was to
assess the performance of the biotrickling filter using
PRD Tech's proprietary support media, in treating organic
emissions from bakery ovens. Pilot studies were performed
at stack gas flowrates in the range of 1.8-3.0 standard
cubic meters per minute (65-104 standard cubic feet
per minute).
The
daily average inlet concentration of ethanol versus
time (Figure 2) showed that the concentration varied
from 0 ppmv to 3,500 ppmv. Variations in inlet ethanol
concentration were due to changes in the type of bakery
product made, production capacity, and weekend shutdowns.
FIGURE 2
The
performance of the biotrickling filter was quantified
by the Percent Removal Efficiency, which was defined
as follows:
Percent
Removal Efficiency = [(Inlet Conc. - Outlet Conc.)/(Inlet
Conc.)] x 100
The
system achieved at least 80% ethanol removal efficiency
for 99.6% of the operating time. The system achieved
an overall average removal efficiency of 91%.
One
of the major concerns with the use of biofiltration
technology is its robustness, i.e., ability to respond
to changes in inlet ethanol concentration or stack gas
flow rate (ethanol loading) and the viability of the
microorganisms after no ethanol had been supplied to
the biofilter for some time. These issues were addressed
in this study by measuring the biofilter performance
after severe changes had occurred in ethanol loading
and after the gas blower had been shutdown for a few
days.
The
dynamics of the biotrickling filter (Figures 3A and
3B), showed that after the inlet ethanol concentration
had changed abruptly from an inlet value of 200 ppmv
(Loading of 50 g/m3/h) to 3,000 ppmv (Loading of 750
g/m3/h) over a period of 3 hours, the biotrickling filter
maintained a removal efficiency exceeding 90%. This
was consistent with the earlier observation that the
pilot-scale biotrickling filter was capable of handling
higher ethanol loadings. Further, the active microorganisms
were able to quickly adapt to increased ethanol concentrations,
since ethanol is an easily biodegradable chemical.
FIGURE 3
The
start-up response of the biotrickling filter (Figure
4) was obtained after the blower had been shut down
for various time periods ( 3 - 9 days). During the shutdown
period, the blower was not operating, and the liquid
nutrients were being recirculated through the two biofilter
sections. The only ethanol load available to the biofilter
microorganisms was the ethanol dissolved in the liquid
nutrients, and this ethanol load was expected to decrease
over the shut down time period, due to biodegradation
of the ethanol. The experimental data for the percent
ethanol removal efficiency of the biofilter after the
various shut down periods shows that the microorganisms
decay rate is very slow and the biofilter is able to
achieve over 95% removal efficiency in a few hours.
FIGURE 4
COST
ANALYSIS
Economic data presented in this brochure are intended
to provide a reasonable first-cut estimate of biotrickling
filter costs as applied for treating ethanol emissions
from typical bakeries. The capital and operating costs
have been calculated for only the biotrickling filter.
The costs of connecting the biotrickling filter to a
bakery oven have not been included, since such costs
will be site specific.
Figure
5 shows the variations in Total Capital Cost and Total
Annual Cost for various stack gas flow rates treated
in the biofilter system. Both costs increase in a non-linear
fashion with stack gas flow rate. Further, the Total
Capital Cost per scfm ranges from $50/scfm to $160/cfm.
FIGURE 5
Figure
6 shows the Total Annual Cost per ton of ethanol treated
($/ton), as a function of stack gas flow rate (scfm).
FIGURE 6
^
TOP
|
