These errors were caused by transferring hot or cold solutions in a manner similar to procedures used in quality control (QC) laboratories when working with compounds of limited solubility or temperature sensitive products.
The expeditions have been designed to highlight the impact of environmental factors such as temperature, pressure and humidity on the accuracy of liquid volumes dispensed from pipettes.
Inaccurate pipetting can lead to erroneous data generation, expensive QC issues and problems with regulatory bodies if the errors remain undetected until after products have been released.
In an attempt to increase laboratory accuracy, Artel has developed instruments that allow liquid handling devices to be calibrated in their native environments, rather than in conditions preset by the manufacturer.
These instruments use a ratiometric photometry technique and include the PCS (personal calibration system) for calibrating pipettes and the MVS (multichannel verification system) for calibrating automated high-throughput liquid handling devices.
The severity of the errors measured with the PCS during the expedition to Yellowstone surprised even the expedition's organisers with errors as great as 37 per cent observed when delivering 0.2µl of cold liquid at 3ºC using a 2µl syringe.
Errors as large as 24 per cent were observed when delivering warm liquids at 37ºC with the same pipette setup.
"We are about to embark on further studies, but since we saw a greater issue here than we expected so we are facing a greater amount of work to develop guidelines for pipette users," said Artel's president, Kirby Pilcher.
These errors were reduced if the pipette tips were set to deliver their maximum volume (2µl) with only a one per cent over-delivery error being observed for cold liquids and a 7 per cent over delivery observed for warm liquids.
According to Dr George Rodrigues, Artel's senior scientific manager, these 'thermal disequilibrium' errors can be explained by the temperatures causing the volume of air above the liquid to expand when pipetting a warm liquid, which affects the air displacement operation of the pipette.
When a pipette tip is inserted into a warm liquid the air in the tip is at ambient temperature.
During the aspiration the pipette tip heats up and causes the air to expand and push liquid out of the tip.
This results in less liquid being delivered than expected, with the opposite effect happening for cold liquids.
"Equilibrating the liquids to ambient temperature would remove this problem, but that's not possible for all assays - there are many assays that involve handling liquids at extreme temperatures," said Dr Rodrigues.
"These could include assays using restriction enzymes that are stored on ice and decompose at ambient temperatures, or assays that use DNA bead extraction techniques where you need to heat the beads up to 75 º C to release the DNA from the bead."
These errors can also be observed when handling mammalian cell cultures at 37ºC and when conducting PCR (polymerase chain reactions) at temperatures that can exceed 60ºC
He continued by recommending that one way to reduce these errors would be to mimic the process using the PCS instrument and measure the errors involved.
The process would need to be copied as accurately as possible as the depth at which the pipette is placed into a hot or cold container and the time the liquid is resident in the tip can both play a part in causing errors.
"We saw that using a pipette close to its maximum setting reduces the dead space above the liquid and this gives more accurate results than if the pipette is dialled to its minimum delivery setting," said Dr Rodrigues.
The dead space in a pipette varies from model to model so judicious pipette selection could play a part in minimising the errors.
Dr Rodrigues was careful to note that so far he had not seen any papers that have addressed this issue and that Artel will aim to develop guidelines after they have conducted more extensive experiments themselves.
"Our initial advice would be to pick a pipette and tip combination and pipette setting that involves the lowest volume of dead air," said Pilcher.
"You should also minimise the immersion of the tip into the liquid as well as the time and extent to which the pipette tip is exposed to the microenvironment."
The company is currently planning further expeditions to test other factors that could introduce errors into pipetted volumes.
"One of the suggestions has been to go to the bottom of a salt mine to study the effects of high pressure, much the same as would be observed in a glovebox," said Pilcher.
"We haven't as yet found anyone that wants to go on such an expedition because it's a decidedly uncomfortable and unphotogenic environment - having said that the drive of science may cause us to go."