A calibration equation for oxygen optodes based on physical properties of the sensing foil. (14th March 2014)

• Record Type: Journal Article
• Authors: McNeil, Craig L.; D'Asaro, Eric A.
• Is Part Of: Limnology and oceanography, methods. Volume 12:Number 3(2014:Mar.)
• Page Start: 139
• Page End: 154
• Publication Date: 2014-03-14
• Subjects: Limnology -- Methodology -- Periodicals; Oceanography -- Methodology -- Periodicals; 551.48
• DOI: 10.4319/lom.2014.12.139
• Languages: English
• ISSNs: 1541-5856

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>We present a new physically based calibration equation for Aanderaa Inc. oxygen sensing optodes. We use the two site fluorescence quenching model of Demas et al. (1995) to describe the nonlinear Stern‐Volmer response of the optode foil to oxygen partial pressure. Seven (minimally six) coefficients quantify foil response to oxygen and temperature; another quantifies response to hydrostatic pressure. These eight coefficients are related, theoretically, to basic physical properties of the foil. The equation provides a framework to study causes of variability and drift in optodes and to develop better quality control and handling procedures. We tested the equation using factory calibrations of 24 optode foils. When accurate multi‐point calibration data are unavailable, two additional coefficients empirically correct the usually large differences observed between factory foil calibrations and post‐factory laboratory/field calibrations; we cannot eliminate this major cause of uncertainty in optode calibrations. Excluding two potentially anomalous foils, the calibration equation fits 13 similarly calibrated foils, totaling 455 calibration points over 3 − 40°C to −0.57 ± 1.48 mbar. Analysis of the resulting best fit coefficients reveals an underlying variability in optodes associated with variability in site 2 and site 1 Stern‐Volmer coefficients of 32% and 20%, respectively. The fraction of unquenched fluorophores responding with the more accessible site 1 quenching characteristics varies by only 3%. The equation fits multi‐point data for two optodes within manufacturer's specifications, the greater of ± 2.5 µmol.kg<sup>−1</sup> and ± 1.5%. Detailed measurements of calibration changes over time will be required to understand the causes of optode drift.</p> </abstract>