Figure
1: Flow Diagram of Process Technologies Inc. Photolytic Destruction Unit
Process Technologies Inc. (PTI) used a technology called photolytic destruction of volatile organic compounds (VOCs) to treat the emissions from a soil vapor extraction (SVE) unit operated at Naval Air Station (NAS) North Station, Site 9 in southern San Diego County. The SVE waste stream had several contaminants of concern (COCs), including:
Ø
Tetrachloroethene
(PCE)
Ø
Trichloroethene
(TCE),
Ø
Cis-1,
2-dichloroethene, toluene
Ø
Octane. (NFESC, 1998)
The concentrations of
the COCs in the SVE vapor are shown in table 1 below.
|
Table 1: SVE Vapor Concentrations
of COCs at NAS North Island |
|||||
|
|
|
|
Concentration in SVE Vapor |
||
|
Chemical of Concern |
|
(ppmv) |
|||
|
Octane |
|
|
|
96.44 |
|
|
Tetrachloroethene
(PCE) |
|
31.40 |
|
||
|
Trichloroethene
(TCE) |
|
|
27.60 |
|
|
|
cis-1,2-Dichloroethene |
|
|
22.20 |
|
|
|
Toluene |
|
|
|
14.20 |
|
This photolytic destruction unit, pictured in Figure 1 consists of two primary components, a fluidized bed concentration unit and a photolytic destruction unit. A brief description of these two processes is included below.
The fluidized bed concentration unit uses three interrelated processes to concentrate organic chemicals found in the SVE vapor waste stream:
Ø An Adsorber, which uses adsorbent “beads” to collect VOCs from the SVE waste stream
Ø A Desorber, which uses a heat exchanger operated at atmospheric pressure to vaporize adsorbed VOCs, and moves (or “sweeps”) these contaminants to the condenser.
Ø A Condenser, which then uses chilled water to preferentially remove water vapor and non-halogenated organic compounds in the concentrated “sweep” vapor. A portion of the halogenated vapors is also removed in the condenser. The liquid condensate, which results from this Fluidized Bed process, is collected and shipped off-site for proper disposal. The regenerated adsorbent is pneumatically recycled to the top of the adsorber for reuse.
The non-condensable vapors are mixed with ambient air, then sent
to the PDU for further processing. The
photolytic reactors use ultraviolet (UV) lamps to break down the VOCs directly
(photolysis) as well as forming free radicals (highly reactive chemical species
such as hydroxyl radical) that chemically oxidize the compounds, converting
them to products such as carbon dioxide and water and hydrochloric acid. Adsorbent panels tie up the hydrochloric
acid to prevent its release. A small
amount of unreacted chemical and reaction by-products were observed in the exhaust
gases from the PDU, and these are passed back to the adsorbent bed. Periodic replacement of the reagent panels
is necessary.
During the process demonstration, two sets of adsorbent
panels were used. When the
demonstration was completed, the panels were tested using the EPA Toxicity
Characteristic Leaching Procedure (TCLP) to verify that the panels could be
disposed as sanitary waste, rather than hazardous, waste. The treated gas from the photolytic reactors
flows through a caustic scrubber system to remove any remaining trace amounts
of hydrogen chloride, or other acidic by-products. The clean, scrubbed gas flows back to the inlet of the fluidized
bed concentration unit, so that portion of the process does not discharge
directly to the air. (NFESC, 1998)
Figure
1: Flow Diagram of Process Technologies Inc. Photolytic Destruction Unit
The PTI System used for this demonstration was designed
to treat 500 scfm of SVE vapor, and to remove a minimum of 3.6 pounds per hour
(lbs/hr) of VOCs, but was only able to treat a maximum flow rate of 440 scfm at
an average composition of the SVE vapor of 192 ppmv of. Thus, the actual capacity of the system was
approximately one-third the projected VOC removal capacity during this
demonstration.
Table
2 below shows the average, by compound, and total DRE for the PDU.
|
Table 2: PDU Average Total and
Individual VOC Removal Efficiencies |
||||||
|
|
|
Inlet |
Mass |
Outlet |
Mass |
Average |
|
|
|
Concentration |
Rate |
Concentration |
Rate |
DRE |
|
Compound Name |
(ppmv) |
(lbs/hr) |
(ppmv) |
(lbs/hr) |
(%) |
|
|
cis-1,2-Dichloroethene |
742.86 |
0.0623 |
8.11 |
0.0007 |
98.85 |
|
|
1,1,1-Trichloroethane |
12 |
0.0013 |
0.08 |
0 |
99.27 |
|
|
Trichloroethene |
688.57 |
0.0799 |
17.7 |
0.0022 |
97.29 |
|
|
Toluene |
|
205.86 |
0.0172 |
11.62 |
0.001 |
94.18 |
|
Tetrachloroethene |
334.29 |
0.0501 |
11.79 |
0.0018 |
96.36 |
|
|
Ethylbenzene |
2.8 |
0.0003 |
0.1 |
0 |
96.21 |
|
|
Xylenes (total) |
11.6 |
0.0012 |
0.44 |
0 |
95.89 |
|
|
1,2,4-Trimethylbenzene |
4.5 |
0.0005 |
ND |
0 |
>92.22 |
|
|
Totals |
|
2,002.47 |
0.2128 |
49.82 |
0.0058 |
>97.27 |
|
Reprinted with permission. |
||||||
The destruction and removal efficiencies or
individual compounds ranged from 94 to 99%, which is not particularly high, though
because the exhaust stream is returned to the adsorber they do not represent
air emissions.
Table
3 below shows the major byproducts of the system and their concentrations.
|
Table
3: Major byproducts measured in the
PDU outlet. |
||
|
|
|
Max Allowable |
|
Contaminant |
Concentration |
Emission |
|
Carbon
Monoxide |
5.9 ppmv |
none |
|
Chlorine |
0.04 ppbv |
no standard |
|
Hydrochloric
Acid |
0.18 ppbv |
<10 ppmv |
|
Phosgene |
23.8 ppbv |
no standard |
No dioxin tests were
performed during the demonstration. The
report states that during past demonstrations of the PTI system dioxins were
not observed when PCBs were not present.
The system was designed to run 89% of the time; however, it is not clear how many hours a day the system would be operating.
(However, UCD was unable to
confirm the actual time the system was operating during the demonstration.)
Requires
a technical service person necessary for approximately 2080 hours per year.
Maintenance
of this system includes replacement of the reagent panels, UV lamps and caustic
solution. Additionally, the
concentrator needs periodic maintenance.
This
configuration of the PTI’s design had a 218 kWh total power load.
The
total PDU system footprint was approximately 20 feet by 40 feet, which included
work areas for the technical support personnel. The concentrator unit was approximately 25 feet long by 8 feet
wide.
Worker health and safety issues include the potential for release of the concentrated VOC stream or a spill of the liquid condensate stream.
A catastrophic failure in this system could be the direct emission of a more concentrated solvent stream or spill of the liquid condensate stream from the desorption process. Spill containment and prevention of accidental rupture or venting from the ductwork would mitigate against such releases. Another possible operational failure would be a UV lamp malfunction. A lamp sensor could be used to detect a lamp failure and the system could be automatically shut down
Concentrated solvent spills could affect workers in close proximity to the spill. Worker exposure would be by inhalation of vapors. Depending upon the location of the spill, ground water contamination could also occur. .
Secondary waste streams from the PTI system included
spent reagent panels, scrubber blowdown and liquid condensate from the
condenser. After TCLP analysis to
determine that the reagent panels were non hazardous waste, they were disposed
of in a landfill.
Several
pumps were located throughout the system.
These pumps could be considered potential sources for noise
pollution. The PDU system was located
within a building that was 9 feet tall and 40 feet long, and the concentrator
was located on a trailer adjacent to the PDU building. Other equipment located outside the PDU
building included a solvent storage tank, various pipes and wiring for the PTI
system.
The PDU is most cost-effective when treating high
concentration vapors containing chlorinated hydrocarbons. The Navy determined that a 45 % cost savings
was provided by the PTI system over the traditional activated carbon or
alternative flameless thermal oxidation systems.
What
are the minimal and maximum flow rates, generally and by process component?
What
concentration levels can this technology accommodate, generally and by process
component?
Do
any relationships exist between different flow rates, concentration levels and
destruction efficiencies?
How
was this technology monitored? (QA/QC)
At
what other sites has this technology been field-tested?
Has
this technology been subjected to third party analysis?
What
is the adequacy of the database for dioxin measurements in previous tests
referred to in the report when partially chlorinated oxidation products are
observed?
Was
the composition of the “clean” air from the adsorber column measured?
Process
Technologies Incorporated
1160
Exchange Street
Boise,
Idaho 83716-5762
(Vendor is out of
business)
EPA
Project Manager:
Paul
de Percin
US
EPA
National
Risk Management Research Laboratory
26
West Martin Luther King Drive
Cincinnati,
OH 45268
513-569-7797
Technology
User Contact:
Kevin
Wong
SM-ALC/EMR
5050
Dudley Boulevard Suite 3
McClellan
AFB, CA 95652-1389
916-643-0830
NFESC,
1998, Photolytic Destruction Technology Demonstration-Final Report, NAS North
Island Site 9, Naval Facilities Engineering Service Center, CR 98.015-ENV,
August.
SITE, 1999, Superfund Innovative Technology Exchange Technology Profiles, Demonstration Program.