7

TiO₂ Photocatalytic Destruction—(Matrix Photocatlytic)

Current Status of Technology

Matrix photocatalytic is currently developing and supplying photocatalytic treatment systems for aqueous and gas environmental remediation and control of process emissions.

Description of Technology

The Matrix Photocatalytic TiO₂ system basic component is the photocatalytic cell which can be linked in a serial mode or parallel mode depending on the concentration of the contaminants, the required throughput, and the amount of organic destruction desired. The system works by passing ambient temperature contaminated air through a fixed TiO₂ catalyst bed activated by UV light. The Matrix system is an advanced reduction/oxidation process developed by Matrix Photocatalytic to remediate vapors contaminated with chlorinated compounds in both the vapor and aqueous phases in fractions of a second. Regenerative blowers draw vapors laden with chlorinated compounds through the PCO treatment system. The treatment system itself is composed of a series of reactor cells, called wafers, each of which is comprised of six reactor cells. Each reactor cell comprises an outer stainless steel jacket which contains an internal photocatalytic matrix wrapped around a quartz sleeve and a UV lamp. Each lamp is mounted coaxially within the quartz sleeve and emits low intensity UV light. The UV light in the presence of oxidants such as O₃ and hydrogen peroxide, and a semiconductor will produce hydroxyl radicals, (OH^-). The matrix system offers an effective method for the mass production of hydroxyl radicals, which are capable of destroying chlorinated compounds in the vapor phase.

The seminconductor, TiO₂, catalyst is activated by the light to produce two effects:

· The momentary shifting of an electron into a much higher orbital shell where it can react with an electron acceptor (such as ozone) to create oxidizing hydroxyl (OH^-) radicals

· The "hole" left open momentarily by the shifting of the electron to a higher orbital shell exhibits a reduction effect where it may participate in reactions to create oxidizing species or directly with contaminants.

As contaminated air flows through the reactor it passes through the catalyst matrix where the organic contaminates undergo both oxidation and reduction processes.

Semiconductors studied for commercial photocatalytic process include TiO₂, strontium titanium trioxide, and zinc oxide. TiO₂ is generally the preferred semiconductor since it has a high level of photoconductivity, it is readily available, it has a low toxicity, and a low cost. One problem associated with semiconductor photoconductivity is the electron-hole reversal process where photons or heat are generated in place of hydroxyl radicals. This electron-hole reversal process results in a significant decrease in the photocatalytic activity of a semiconductor. This problem can be sidesteped and the photocatalytic activity of the semiconductor can be increased by adding O₃ to the vapor to be treated. The addition of the O₃ inhibits the electron hole reversal, prolongs the lifetime of the photo-generated hole and it generates additional OH^·.

Organic compounds can be destroyed by a variety of reactions with OH^· such as hydrogen abstraction, electron transfer, and radical-radical combination. If insufficient OH^· is generated, and contaminants are not completely oxidized to CO₂ and H₂O, stable intermediates may be formed. Types and concentration of intermediates formed depend on the composition of the initial vapor stream to be treated.

The PCO unit instrumentation and control consisted of:

Averaging pitot tubes to calculate flow rate.
Bi-metal thermometers for temperature readings at the PCO inlet and outlet.
Start-up and shutdown switches to control the regenerative blowers.
Individual circuit breakers to control the bank of UV lamps.
Interconnection with the air stripper supplying the vapor stream to shut the system down in the event of an air stripper shutdown.

The Matrix PCO system is modular in design and its size is based on flow rate, contaminant concentrations, and target reduction objectives. Systems of 100 ft³/min have been tested on both dry and moist air from vapor extraction operations, air stripper emissions, steam from desorption processes and VOC emissions from manufacturing facilities. Systems of up to 1,000 ft³/min can be cost competitive with thermal destruction systems. Matrix Photocatalytic, Inc. TiO₂ photocatalytic systems will treat air streams of 1-5,000 ft³/min and water streams of 1-1,000 gal/min. The systems are configured based on the number of wafers required to treat the contaminants, and this system can function over a broad range of pressures and pH. The benefits TiO₂ PCO systems offer over other technologies in the treatment of vapor contaminated with organic compounds include:

On-site treatment of organic pollutants.
Ambient temperature process (no combustion source).
Photocatalytic air treatment systems generate minimal amounts of NOx and therefore do not raise the concerns associated with obtaining permits for thermal (high temperature) treatment systems.
Vapor phase systems are not adversely affected by humidity.

The monitoring system for the PCO system consisted of pitot tubes to calculate flow and bi-metal thermometers for temperature readings upstream of the PCO inlet and downstream of the PCO outlet. Sampling occurred at the inlet and exhaust of the PCO unit. Additionally, sampling for chlorine and phosgene was conducted at the caustic scrubber exhaust to monitor the stack emission for protection of workers and the general public.

The Matrix Photocatalytic, Inc. system can also be used to treat CVOCs in ground and wastewater. Further information is available in the SITE Innovative Technology Evaluation Report EPA-540-R-97-503.

Site and Contaminants Description

The National Renewable Energy Laboratory (NREL) demonstrated the Matrix Photocatalytic titanium dioxide (TiO₂) based photocatalytic oxidation (PCO) system at McClellan AFB. The objective of the NREL project was to demonstrate the feasibility of using a TiO₂ based PCO system to treat SVE vapor phase mixtures of chlorinated hydrocarbons, mainly solvents such as TCE, PCE and 1,2-DCE. The demonstration was divided into two phases, system optimization and operation. Each of these phases took approximately 2 weeks to complete. During the optimization phase three parameters were varied, flow rate, ozone (O₃) concentration and lamp type.

The demonstration was originally to be conducted at operable unit (OU) C1 at McClellan AFB. However, the catalyst was fouled by elevated concentrations of long chain saturated petroleum hydrocarbons (LCSPHs) in the vapor stream resulting in low destruction removal efficiencies (DREs). Additionally lamp performance suffered due to an increase in the temperature of the influent stream to above 120 ^oF by the regenerative blowers from the SVE. This higher temperature is 10 ^oF above the optimum operating temperature, which caused the power output of the lamps to drop significantly. The combination of the LCSPHs and the low lamp performance resulted in DREs in the 10 to 40 percent range, which a high degree of scatter in the data. Therefore the demonstration was moved to another location, IC 29, which contained a richer chlorinated gas stream containing trichloroethene, chloroform, carbon tetrachloride, and cis-1,2-dichloroethene. IC 29 contains a large VOC groundwater plume that was treated by Radian International with an air stripper to transfer the chlorinated solvents from groundwater wells into a vapor stream.

Major contaminants at the McClellan IC 29 site and their concentrations are presented in table I below.

Table I. Major Contaminants at McClellan site IC 29 and their concentrations.
Contaminant	Concentration (ppbv)
1,1-Dichlorethene	270
Methylene chloride	88
cis-1,2-Dichloroethene	460
Chloroform	950
Carbon Tetrachloride	1500
1,2-Dichloroethane	86
Benzene	810
Trichloroethene (TCE)	22,000
Tetrachloroethene (PCE)	60
Total Target List VOCs	22,224
ppbv = parts per billion by volume
Results obtained from modified Method E18 (GC/Mass Spectroscopy (MS)).

Performance of Technology and DRE

Optimization Phase

During the optimization phase the following system variables were examined, flow rate, concentration of O₃ and lamp type. During the optimization the flow rate was varied from 45 to 50 standard cubic feet per minute (scfm), while the concentration of ozone was held at 0 or 10 L/min, and the UV wavelength was either 254 nanometer (nm) or 185 and 254 nm simultaneously. The analysis of the system optimization showed that the maximum DREs resulted when O₃ was present, the flow rate was at its minimal level and when a hybrid lamp system was used (both 184 and 254 nm wavelength light were present). These optimal conditions were used during the demonstration phase of the TiO₂ based PCO technology.

Demonstration Phase

During the operation phase, the Matrix PCO technology demonstrated mixed removal results. The overall DREs for the major contaminants of concern at the IC 29 site are presented in table II below. Further details on each compound are presented in the field performance data section.

Table II. Range of DRE for major contaminants of concern at the IC 29 site by the PCO technology.
Contaminant	Lowest DRE observed (percent)	Highest DRE observed (percent)	Average DRE observed (percent)
TCE	97.8	98.3	98.1
PCE	86.0	87.4	86.7
1,2-DCA	52.6	64.6	58.6
cis-1,2-DCE	98.4	98.6	98.5
Benzene	98.0	98.1	98.1
Chloroform	2.2	57.7	29.9
Carbon tetrachloride	18.1	34.7	26.4
Total VOC’s	89.1	92.3	90.7
DRE Ranges were calculated from EPA Method TO-14 except for PCE, 1,2-DCA and CTCL.

On average the Matrix PCO system satisfactorily removed TCE and cis-l,2-DCE from the vapor stream, however it did not meet removal requirements for PCE, chloroform, carbon tetrachloride, and 1,2-DCA. DREs for total VOCs exceeded 90 percent, but did not meet the Sacramento Metropolitan Air Quality Management District “best available control technology” requirement of 95 percent removal.

To successfully implement the Matrix Photocatalytic system at a site, the off-gas stream needs to be completely characterized to ensure compatibility between contaminants and catalyst. This technology should not be applied to sites contaminated with high concentrations of LCSPHs. To meet the applicable DRE requirements, residence time of a system should be conservatively designed. It is recommended that the system be allowed to continue to operate/maintain "steady state" by passing ambient air through it via a three way valve, to remove the effect of any periodic downtimes.

The results seen at McClellan were supported by a field-test of the Matrix TiO₂/UV system at the U.S. DOE Savannah River Superfund Site in Aiken, South Carolina. TCE and PCE concentrations in the SVE off-gas ranged from 110 to 190 ppmv and 700 to 1200 ppmv respectively, and flow rates treated ranged from 0.0028 to 2.8 scmm. The SVE –off-gas passed through a cyclone separator and a filter to remove moisture and particulate matter prior to entering the Matrix system. Maximum TCE removal was 98.1 percent, and maximum PCE removal was 95.2% The Matrix system demonstrated at Savannah River did not remove 1,1,1-TCA, an byproducts produced included small quantities of carbon tetrachloride, chloroform, dichloroacetyl chloride (DCAC), hexachloroethane, methylchoroformate, pentachloroethane, phosgene and trichloroacetyl chloride.

Field Performance Data

Performance results from the PCO demonstration at McClellan on 6 different days and by EPA Method TO-14 and Method 8021 are presented in table III below.