Miniaturizing nutrient sensors for research platforms

I recently visited the NOAA Sand Point campus in Seattle to install a new type of nitrate sensor on an oceanographic buoy as part of the ITAE project.  I work as a researcher and technology developer at the National Oceanography Centre (NOC) in Southampton, UK, where, as part of Dr Matt Mowlem’s Ocean Technology and Engineering Group, I work towards developing the next generation of chemical sensors and analysers for ocean science.  Through a scientific collaboration, we are providing one of our nutrient sensors for installation on the ITAE buoy to be deployed in the Arctic.

Scientists have been measuring nutrients (nitrate, phosphate etc.) using colorimetric techniques (mixing seawater with various specifically chosen chemicals and carefully measuring the resulting colour change) for years, and most chemical oceanography labs now have high performance machines that can automate this process by analysing tens of  seawater samples per hour.  However, rather than taking samples of seawater and analyzing them back at the laboratory, it would be much more effective if we could leave an instrument out in the ocean (deploy it on a buoy or put it on a vehicle such as the Saildrone) in order to analyse seawater automatically over long time periods.  This way we wouldn’t have to keep visiting the same sites to collect samples and we would obtain much more detailed datasets.

The problem with this approach (taking in situ measurements) has always been that chemical analysers require a lot of power and consume large amounts of liquid chemical reagents.  This means that they tend to be large, bulky, and limited in how many measurements they can perform.  At NOC, we wanted to use new technologies to make chemical nutrient analysers smaller and more efficient in terms of power and fluid consumption, and this is what led us to the world of microfluidics.

Microfluidics (sometimes referred to as lab-on-chip technology) involves manipulating very small amounts of fluid using tiny fluidic channels and miniaturized valves and pumps.  The fluidic channels in our lab-on-chip sensors are just 150 micro-meters wide (a typical size for a human hair), and each measurement requires just 250 micro-litres (one quarter of a millilitre) of seawater.  The research field of microfluidics has traditionally found most of its applications in the biomedical sciences (aiming to miniaturize and speed-up medical tests, for example), but we were one of the first groups to apply this technology to the environmental sciences.

Using microfluidics means we really cut down on the amount of fluid used by the analysers (both in terms of liquid reagent and seawater sample required), and on the power consumed (it takes less energy to pump smaller volumes of fluid).  This means that we can deploy the analyzers for longer time periods in remote locations and obtain long time series measurements.  Another advantage of microfluidics is that the equipment generally tends to be smaller, meaning that we can deploy the analyzers in situations where space is limited, such as on autonomous underwater vehicles or ocean gliders.

At NOC Southampton we have been developing miniaturised chemical analyzers using lab-on-chip technology since about 2007, although in the last couple of years we have made significant gains in terms of reliability and performance.  Just this year we have deployed LOC sensors in rivers, estuaries, on the seafloor, in glacial meltwater streams, in the Arctic Ocean (Fram straight), as well as on the ITAE buoy in the Arctic.

The LOC system that we have deployed on the ITAE buoy measures nitrate, although we are also developing LOC sensors to measure phosphate, silicate, dissolved iron and pH.  An understanding of all of these parameters is often essential in order to assess the state of the ocean ecosystem and understand the underlying chemical and biological processes.  We engineered our LOC systems so that they are all based on the same common platform, so that, apart from a slightly different chemical method, there are very few differences between each of the sensors.  This really helps when scaling up production, and also when developing a LOC sensor for a new parameter.  In future we hope to expand our range of parameters by developing sensors for, among other things, dissolved inorganic carbon (DIC), total alkalinity and organic nutrients. 

Just this year we have deployed LOC sensors in rivers, estuaries, on the seafloor, in glacial meltwater streams, as well as on the ITAE buoy in the US Arctic. - Alex Beaton

Producing microfluidic systems capable of being deployed in the ocean requires specialist manufacturing facilities and techniques that we have developed in house.  The “chip” part of lab-on-chip refers to the plastic substrate that contains the microfluidic channels.  We have specialist micro-mills that can precisely cut these channels and features, and specially developed bonding techniques for sealing the microfluidic channels once they have been cut.   We are often asked why, if the channels are so small, they don’t get blocked by particles in the seawater.  We place a filter on the inlet of the sensor that blocks anything larger than 0.45 micro-meters from entering the channels.  Because the analysers only suck in small amounts of fluid, the filter lasts quite a long time before it itself becomes completely blocked.

In the last year we have produced over 30 copies of our LOC nitrate sensor, each with its own serial number, and these have been deployed in various locations around the world.  We are excited to be part of the ITAE project, and hope to work together again in future to combine more of our innovative technologies to advance the exploration of science the Arctic.

Written by A. Beaton