All commercially available ozone monitors based on UV absorbance exhibit a water vapor interference.  Although not widely recognized, most scientists who work with UV ozone instruments are aware that rapid changes in humidity, as may occur during a frontal passage, will cause a large zero shift that may last for several hours.  This interference is especially troublesome in balloon and aircraft measurements where the instrument may rapidly pass through alternating dry and humid air parcels.  It also is a problem when one calibrates an ozone instrument using extremely dry tank air.

Dr. Kevin Wilson, a graduate student of Prof. John Birks at the University of Colorado, Boulder, recently completed a Ph.D. thesis in which he elucidated the mechanism of the water vapor interference and discovered a simple way to completely eliminate it (Wilson, K.L., Ph.D. Thesis, University of Colorado, Boulder, 2005).  He discovered that a Nafion® tube can be used to equilibrate humidity during I and Io measurement cycles and thereby completely eliminate this interference -- even for step changes between 0 and 90% RH.  Nafion® tubes of the appropriate length and internal diameter and modified to easily attach to 2B Tech detection cells are available under our trademark name DewLine™.  All new instruments are provided with DewLines™.  We recommend installation of a DewLine™ on all instruments returned for service if they do not have one.  Ozone monitors manufactured by other companies benefit from the use of the DewLine™ as well.

The following is the abstract of a peer-reviewed article published on this subject:

Mechanism and Elimination of a Water Vapor Interference in the Measurement of Ozone by UV Absorbance

Environmental Science and Technology (2006) 40, 6361-6367

Kevin L. Wilson and John W. Birks

A water vapor interference in ozone measurements by UV absorption was investigated using four different ozone monitors (TEI Models 49 and 49C, Dasibi Model 1003-AH and a 2B Technologies Model 202 prototype).  In the extreme case of step changes between 0 and 90% relative humidity (RH), a large interference in the range of tens to hundreds of ppbv was found for all instruments tested, with the magnitude and sign depending on the manufacturer and model.  Considering that water vapor does not absorb at the wavelength of the Hg lamp (253.7 nm) used in these instruments, another explanation is required.  Based on experimental evidence and theoretical considerations, we conclude that the water vapor interference is caused by humidity effects on the transmission of uncollimated UV light through the detection cell.  The ozone scrubber acts as a water reservoir, either adding or removing water from the air sample, thereby modulating the detector signal and producing a positive or negative offset.  It was found for the 2B Technologies Ozone Monitor that use of a 1-m length of Nafion® tubing just prior to the entrance to the detection cell reduces the water vapor interference to negligible levels (±2 ppbv for step changes between 0 and 90% RH) while quantitatively passing ozone.

Download a free reprint of the full paper at the ACS publication website as a pdf file (requires filling out a form) or email John Birks at johnb@twobtech.com for a copy.