Optimization of the Sampling Technique for the Determination of Dissolved Hydrogen in Groundwater

Abstract

In this study a field‐sampling technique for dissolved hydrogen (H2) in groundwater will be presented which allows the transport of gaseous samples into the laboratory for further analysis. The method consists of transferring the headspace trapped in a gas‐sampling bulb which is continuously purged by groundwater into previously evacuated vials using a gas‐tight syringe. Three transfer steps with preceding evacuation of the vial led to a H2‐recovery of 100 % in laboratory experiments. The method has been applied to determine H2 concentrations in an aquifer contaminated with chlorinated solvents. Tests concerning the effect of different pumping techniques on H2 concentrations revealed that most reliable values were obtained with a bladder pump, while an electrically driven submersible pump generated considerable amounts of hydrogen due to electrochemical interactions with the sampled water. Concentrations of dissolved hydrogen in field and laboratory samples were about two orders of magnitude higher when sampling was performed with the electrically driven submersible pump compared to sampling with the bladder pump and a peristaltic pump. Lab experiments with a Plexiglas reservoir to produce H2‐enriched water were used to study the effect of two tubing materials (PVC, polyamide) on H2 losses. PVC tubing turned out to allow transfer of H2‐enriched water over 25 m without significant losses, while PA‐tubing was not suitable for sampling of H2.

Purification and Characterization of the [NiFe]-Hydrogenase of Shewanella oneidensis MR-1

“…The ultrapure H 2 was used as a standard to determine the amount of H 2 in the headspace of serum bottles. Based on Henry's law, the H 2 concentration in the Shewanella liquid cultures was then calculated (18).…”

The quantification of hydrogen and methane in contaminated groundwater: validation of robust procedures for sampling and quantification

“…It was reported that pump sampling using a similar submersible pump led to obvious overestimation of hydrogen contents in groundwater aquifers. 25 Hydrogen levels were about two orders of magnitude higher than in the case of the use of a bladder pump. This effect cannot be ruled out here as reason for the greater hydrogen values after the bubble stripping.…”

“…The headspace in the gas sampling bulb was not transferred into evacuated vials for transportation to the laboratory as recommended elsewhere. 25 The withdrawal of a gas aliquot from that vial for direct injection into the GC causes underpressure in the vial which easily leads to infiltration of the injection syringe with air. Then, oxygen and nitrogen are detected in higher concentrations and a diffusion loss of hydrogen is likely as well.…”

“…It should also be taken into consideration that the contact time between steel surfaces and groundwater in the case of bubble stripping was about 1 hour in contrast to 7 days in the case of passive sampling (see above). The alternative to the MP1 type of pump would be a bladder pump as recommended by Heimann et al 25 Neither type is readily available completely without ironcontaining surfaces. It should be noted that the stripping is a feasible method for methane (see above) and even in the case of hydrogen the whole procedure (stripping, immediate transport of the bulbs to the laboratory and analysis) is regarded as feasible if artefacts by the employed pump can be avoided.…”

A Simple Pore Water Hydrogen Diffusion Syringe Sampler

“…The syringe sampler is deployed in the sampling zone (soil or water), and the internal gas is allowed to come to equilibrium with the in situ H 2 . The recovered gas sample can be injected directly into a gas chromatograph if analysis can be completed within 2 h. If prompt analysis is not an option, the gas can be transferred to a sealed borosilicate glass vial for storage (Heimann et al 2003;Microseeps Inc. 2003). Samples stored in 20-mL serum vials containing high-purity (99.999% purity) inert gas are stable for at least 1 month (P. McLouglin, 2005, personal communication).…”