GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 7-8
Presentation Time: 10:10 AM

A NOVEL METHOD FOR CREATING CO2 DISSOLVED-GAS STANDARDS USING COMMERCIAL COLA BOTTLES AS ACID REACTORS


VESPER, Dorothy J. and MITCHELL, Jonney, Dept. of Geology & Geography, West Virginia University, Morgantown, WV 26506

In this study, a novel method was developed to create standard solutions for dissolved CO2 in used, sealed, commercial cola bottles as reaction vessels. This method was designed to overcome problems associated with creating dissolved gas standards. Each standard is created in used and cleaned cola bottles using calcium carbonate and deionized water to create solutions of dissolved inorganic carbon (DIC) with known concentrations. Small hand-blown glass bulbs containing 2-4 mL of concentrated HCl are added to each solution-filled bottle along with 4 steel ball bearings. The bottles are shaken causing the glass bulbs to break, allowing the acid to mix in solution and convert the DIC into CO2. The bottles are placed in an Anton Paar pressurized device that punctures the cap and transfers the solution without degassing to an Anton Paar CarboQC, a device that measures the dissolved activity of CO2. A bypass valve redirects the excess solution into a flow-through cell where the pH and temperature of the final mixture are measured. CO2 was measured in replicates for concentrations from 1, 4, 6, 8, 12 and 16 mM DIC. Final pH values for the mixtures were all less than 3.3 indicating that the DIC was fully converted to CO2. The accuracy of the method was determined by calculating the DIC from the resulting CO2, pH and temperature data and comparing it to the original solution chemistry. At 1 mM, the accuracy was ca. 10% but that is the approximate detection limit. Standards with 4-8 mM CO2 had measured values close to the expected concentration (<5% error). Standards with > 12 mM CO2 had greater error (10-15%) but these CO2 concentrations are higher than what is typically measured in natural water systems and thus field data is unlikely to have similar error. Based on the tested standards, this method is very accurate at concentrations between 1 and 6 mM making it useful for natural waters with elevated CO2 concentrations.