Study on Photocatalytic Oxidation (PCO) Raises Questions About Formaldehyde as a Byproduct in Indoor Air

Research at Lawrence Berkeley National Laboratory finds that Photocatalytic Oxidation (PCO) reduces indoor VOC's but could produce formaldehyde as a byproduct.

A very promising new technology for the reduction of Volatile Organic Compounds (VOC's) in indoor environments is Photocatalytic Oxidation (PCO). This process exposes ultraviolet light to a catalyst such as titanium dioxide to produce primarily hydroxyl radicals (OH). These hydroxyl radicals are extremely reactive and can oxidize or "break down" typical VOC's in indoor environments. The objective of this study conducted by A.T. Hodgson, D.P. Sullivan and W.J. Fisk entitled "Evaluation of ultra-violet photocatalytic oxidation (UVPCO) for indoor air applications: conversion of volatile organic compounds at low part-per-billion concentrations" (LBNL-58936) was to determine if this process could be used to reduce indoor VOC's to the extent that "acceptable indoor air quality in office buildings could be achieved with less energy by combining effective air cleaning systems for VOC's with particle filtration than by relying solely on ventilation."

The researchers point out that most of the studies of this technology have been conducted in laboratory settings. The large majority of these investigations have employed relatively large concentrations of just a few VOC's primarily to better understand the PCO process. This study was designed to simulate low VOC concentrations that would be found in real indoor environments.

Theoretically all VOC's will be broken down into carbon dioxide and water. However, in many cases the reactions to receive this end state have numerous stages, can be complex and can produce relatively stable intermediary byproducts. The question is whether or not the Photocatalytic Oxidation process can react quickly enough and completely enough with VOC's to neutralize them and not create harmful VOC's as unintended byproducts.

To test this the researchers created three challenge VOC mixtures. One was a combination of 27 VOC's commonly found in office buildings. The second was a mixture of three commonly used cleaning products - a pine-oil based cleaner, a cleaner using 2-butoxyethanol, and an orange-oil (ie. d-limonene) based cleaner. The third consisted of a mixture of VOC's commonly emitted from building products such as painted wallboard, composite woods, carpeting and vinyl flooring. Air flow speeds and VOC concentrations were varied with each mixture to create a total of nine experiments. Measurements of the intake VOC's and single pass outflow VOC's were taken. Other experiments were also conducted with just a formaldehyde and acetaldehyde mixture and the PCO device.

Generally, the efficiencies of the conversions of the challenge VOC's varied by the type of VOC and the speed of the airflow. Interestingly, the concentration of the challenge VOC's did not have much of an effect. Despite increasing concentrations by two or three times the Clean Air Delivery Rate (CADR) remained about the same. For the cleaning product VOC's the reaction efficiencies varied between 20% and 80%. For the building product VOC mixture the reaction efficiencies varied between non-significant and up to 80%. The aldehyde mixture conversion efficiencies ranged between 18% and 49%. Generally the efficiency of the conversions broke down in the following order with the most effective being alcohols and glycol ethers; then aldehydes, ketones and terpene hydrocarbons; then aromatic and alkane hydrocarbons; and finally halogenated aliphatic hydrocarbons. In general, the conversion rates were determined to be very encouraging and the authors of the study point out that this was achieved at a very low pressure drop thus supporting the proposition that PCO's could lead to energy conservation.

However, there was a negative coming out of these experiments. The researchers found that because of incomplete decomposition of the VOC's in the inlet air stream there was a net production of formaldehyde, acetaldehyde, formic acid and acetic acid. Of particular concern was that the outlet concentrations of formaldehyde and acetaldehyde were 3.4 and 4.6 times the inlet concentrations, respectively. Both formaldehyde and acetaldehyde are recognized as important indoor toxicants. Formaldehyde is classified as a human carcinogen. Governmental guidelines suggest keeping indoor concentrations of formaldehyde and acetaldehyde at very low levels.

While the VOC exposure to PCO devices creates formaldehyde and acetaldehyde, the PCO device also decomposes these compounds. The question then becomes whether or not this results in a net increase of these compounds in an indoor environment. Using modeling based on the results of the study the authors conclude that there would be roughly a three-fold increase in indoor formaldehyde and acetaldehyde concentrations with a PCO operating in an office buiding (depending on concentrations and types of VOC's).

In conclusion, the researchers state that while the VOC conversion efficiencies with the PCO device may be beneficial for the large-scale treatment of air in occupied buildings, the increases in formaldehyde and acetaldehyde need to be researched futher and better quantified. Work needs to be done to either reduce the production of the formaldehyde and acetaldehyde or to combine the technology with some sort of scrubber to pull out the toxic byproducts before they are brought back into the occupied space.

This reseach continues as can be seen by the minutes of the February 7, 2007 meeting of the Federal Interagency Committee on Indoor Air Quality. The Department of Energy (who is the biggest sponsor of this research) representative summarized the above results and stated that experiments are being conducted using several types of sorbent media scrubbers downstream of the PCO device. Initial results show that a sodium permanganate chemisorbent has considerable potential.

Another approach is to improve the productivity of the reactions of the VOC's and the hydroxyl radicals and other ROS. The difficulty with this is that it is unlikely that the reactions will ever be total and produce no byproducts. In those same CIAQ minutes the point was made that all 10 of the VOC's tested produced formaldehyde. Another issue is the air speed and exposure time near the PCO. The tests that were conducted at LBNL were done at two speeds. Significant decreases were seen in the percentages of VOC's that were broken down as the the speed was increased. This stands to reason since the VOC's would be in the presence of the ROS's for a shorter period of time. What makes this troublesome was that the "high" speed was only 340 cfm. Most residential systems produce at least 1,000 cfm while commercial systems are generally rated at 2,000 cfm. At these higher speeds one would have to assume even lower percentages of reactions and higher levels of byproducts though further research would have to be done to confirm this.

Further work also needs to be done on the use of Photocatalytic Oxidation (PCO) in areas where you have smokers. The reduction in discernable odors for houses with smokers or places like bars and casinos is very appealing. However, cigarette smoke has over 1,000 different chemicals. There is a lack of good research to determine what is coming from the reactions with these 1,000 plus chemicals and the hydroxyl radicals and other reactive oxygen species (ROS) from the PCO devices. Given results we have seen with cigarette smoke and ozone (another ROS) and the results of the above detailed study, it is a pretty safe assumption to make that formaldehyde is one of the byproducts. What other byproducts, the levels of those byproducts, and the possible production of ultrafine particles are all unanswered questions.

What this illustrates to me is the complexity of indoor air and the dangers of making assumptions about the outcomes of chemical reactions. What you want is often what you do not get. While the PCO technology is very promising, in my mind, the "jury is still out" on whether or not it should be universally recommended for indoor occupied spaces.