Tuesday, 12 July 2016

DIY fluorometer for detecting oil pollution in water

The citizen scientist community, which in the United States has its roots in the gulf oil crisis, has been interested in measuring the amount of oil in water to track pollution like oil spills. One test kit that has been developed by the organisation Public Lab uses a laser and a webcam spectrometer to measure the fluorescence of different oils in order to identify the type of oil (https://publiclab.org/wiki/oil-testing-kit).
This is useful for identifying different types of oils but quantifying the amount of oil in a sample is also needed to track its spread and quantify the degree of pollution. However, detecting oil in water is difficult as the two separate. The standard method in the lab is to use a solvent (tetrachloroethylene ASTM D7066-4) to extract the oil from the water. Once extracted, either fluorescence (where the oil is illuminated with UV light and glows green/red) or infrared absorption of the C-H bond can be used to measure the amount of oil in the sample.

Figure 1 - Infrared transmission spectra of solvent without (left) and with (right) oil added. The band at 3000 cm-1 corresponds to vibrations in the C-H bonds which the solvent doesn't have and is used to quantify the amount of oil in the sample. (source: patent US4164653-2)

Figure 2 - Fluorescence spectrum from different oils excited with UV light (308 nm). (source)

The problem with these methods is that the equipment needed to perform the measurement costs a considerable amount, 500-1000 USD for a second-hand Fourier transform infrared spectrometer or a second-hand fluorometer on Ebay. I had done some work on making low-cost colourimeters using LEDs before and so I thought building my own fluorometer would be a good place to start.

I started by taking my multimeter colourimeter and measuring the voltage of a green LED, using it as a photodiode. There was a signal for highly fluorescent compounds such as vitamin B12, but it was very weak and would not be sensitive to small amounts of oil. One option was to construct an amplifier circuit, however this incurs a high assembly cost. I stumbled across this instructable on how to construct a very sensitive light sensor from a TSL237 light to frequency converter, which has a photodiode along with a built-in amplifier and voltage to frequency converter: (http://www.instructables.com/id/Highly-sensitive-Arduino-light-sensor/). This light sensor is very sensitive but what is more helpful is the conversion of the light intensity into frequency - a microcontroller like the Arduino is much better at measuring a frequency than a voltage. This means you can plug the sensor directly into an Arduino and take very accurate readings by converting the frequency to an illuminance value. The sensor only costs about 3.33 USD and is probably the most accurate sensor you can buy for that price.

After finding an accurate sensor, the other problem is the use of solvents. These are hard to source, require fume cupboards and are not safe for use at home. I found a solvent-free method online to extract oil from water, from the company Turner Designs, which makes fluorometers. The procedure consists of adding a solution to the oil and water mixture, heating it until it became cloudy and then cooling it down until it became clear and the oil is dispersed throughout. Certain detergents will drop out of solution when their temperature is raised. This only works for detergents that are not charged; heating an uncharged detergent interrupts the detergent-water interactions and causes them to no longer be soluble in water, the opposite of what you would expect. We had some Triton X-100 in the lab which is a non-ionic surfactant (detergent) with a cloud point of ~65 degrees C. Triton X-100 is safe to use and very cheap to buy. I prepared a 0.1% solution of Triton X-100 and added a small amount of olive oil, which has a good light emission at the wavelength where crude oil emits light. As such, this oil could be used as a standard for crude oil measurements.

Figure 3 - Drops of olive oil floating on 0.1% Triton X-100 in water (left). Heating above the cloud point (70 degrees) (right) it becomes opaque; upon cooling it becomes completely clear.

I bought some TSL237 light-to-frequency converters from RS and used some Veraboard to solder one to some wires to attach to the Arduino. I also attached a 405 nm LED with a resistor (220 Ohms) between 0 and 5V of the Arduino. I used four layers of green cellophane which allow through ~500 nm, the peak of the emission, and block out the UV light from the LED. I found that four layers were necessary to block the UV LED to the point where it was not picked up and without any sample added the reading was the same whether the LED was on or off.
Figure 4 - Veraboard circuit which connected the light sensor and LED to wires, which plugged into the Arduino. It is attached to a laser cut colourimeter, and to my laptop.

I made use of the great Freqperiod library to measure the frequency, a software library for Arduino that is designed to measure frequencies below 20 kHz. It measures frequency in Hertz to six decimal places. There are other libraries to measure above 20kHz that means that by changing the software you can measure a huge dynamic range of light intensities. (http://interface.khm.de/index.php/lab/interfaces-advanced/frequency-measurement-library/) Most of the readings were in the hundreds of Hz. Full room lights were 20kHz while the maximum reading the sensor can do is 1000kHz (see Figure 5).
Figure 5 - Frequency vs. light irradiance (left) and response curve (right) taken from the datasheet for the TSL237 light-to-frequency converter. 

Below are the readings for different samples. The filter attenuated a lot of the light. The next improvement is to buy a better filter. In fluorescence measurements large lenses are often used to collect as much light as possible. So integrating a lens could also help, however all of these additions increase the cost of the unit. 

SampleFrequency (Hz)
Triton X-10045.0±0.2
100 ppm olive oil55.1±0.2
1000 ppm olive oil62.3±0.2
Pure olive oil5985
Room lights20,000

Figure 5 - Olive oil excited with 405 nm LED emitting red light. The decay of the red emission into the olive oil indicates it is saturating and absorbing all of the light before it can reach the other side of the cuvette. This leads to incorrect readings; for an accurate reading the olive oil would need to be diluted. 

Measuring the stability of the reading over time gives a fairly narrow band of frequencies with a deviation of only ~0.2 Hz. The readings are stable enough to compare samples ranging from 100 - 1000 ppm. With a better filter, there is potential to detect samples of under 100 ppm. 
Figure 6 - Measurement over a few minutes (inset top) with the histogram of frequency readings (top) and fitting to a normal distribution (perhaps log-normal would've been better) giving a width of 0.2 Hz.

The next step is to integrate it into an Arduino shield. The cost of something like this would come down to the price of the Arduino but because you only really need one pin to measure the frequency you could use something like the Adafruit trinket (6.95 USD) and make the whole thing for under $20. Any interest in helping to do this would be greatly appreciated as I have left all of my soldering supplies in New Zealand (I now live in Cambridge, UK). 

Figure 7 - Diagram of a shield for an Arduino Micro, with an added RGB LED for optical absorption measurements on the same device. In order to take a reading, a box would be needed to put over the device, or a 3D printed holder for the cuvette could be designed.

Another interesting application is in water testing. Turbidity is an important measure of water health and my biology friends tell me it is very important as fish can't see their prey if the water is cloudy. This accurate light sensor could be used to measure turbidity. A student and I have also been working on using milk as a standard for preparing turbidity standards to calibrate the measurement. Anyway, there is more to come on this. If anyone is interested in helping out or if you have any questions, please let me know in the comments. 

Thursday, 16 June 2016

Detecting 70 molecules in a billion by converting a blu-ray disc into a chemical sensor

At the Photon Factory (the lab I used to work at) Dr Michel Neuwoudt our expert Raman spectroscopist had been hard at work detecting contaminants in milk using a technique called Raman spectroscopy which she has recently published (doi:10.3168/jds.2015-10342). Prof. David Williams, Dr Cather Simpson and Michel started this project after the melamine scandal in New Zealand where melamine was put into baby milk formula to make it appear as if the protein content of the milk was higher than it was. This fooled people because the test for protein naively digests everything in the product and measures the nitrogen content of the mixture. Melamine has a huge number of nitrogen molecules so it appeared as if the milk had more protein than it did. I had been playing around with using Blu-ray discs colour sensors. We thought about whether the nanostructures on the Blu-ray disc could be able to enhance the weak singles you measure using Raman spectroscopy and so along with other in the lab Reece, Nina, Andy (Xindi) and Jenny Malström we started the project.
Molecule of melamine showing large number of nitrogens Link

The paper we recently published first as a conference article and then as a journal article in Analytical and Bioanalytical Chemistry converts a Blu-ray disc into a sensor that can detect accurately and reproducibly down to 70 parts per billion of melamine.

Front cover of journal showing the gold nanoparticles on top of the Blu-ray disc. You can also see the pattern of the nanostructure of the Blu-ray grating the gold was deposited on.

The technique we used is called Raman spectroscopy which measures the vibration of molecules using a laser (see this video for more details). This effect is very weak and so cannot be used for low concentrations of molecules. To enhance the signal and amplify it we use a surface which has small antennas that are tuned to the light we are using. The electrons on the surface of these metal nanoparticles start to oscillate with the electric field of light and so you get regions on the metal where the electric field is much higher.

Plasmon resonance as the electrons move in the electric field of light which is an electromagnetic wave Link

The effect is not linear so when you double the intensity you actually get out four times the amount of signal. As the nanoantenna produces very intense hot spots near the particles you can enhance the signal by >1000000000 times. Two things are then needed to make surface enhanced Raman spectroscopy work; the molecules need to be able to adsorb onto the particle so they are in the high electric field and the surface of the nanoparticle must be the right size to resonate with the light. Actually, it is not that hard to prepare SERS substrates that meets these criteria and gives large enhancements but they are notoriously unreproducible this is a huge problem for the field as analytical measurement demands that you can measure things accurately. It turns out preparing reproducible nanoparticles of gold on the nanometre scale is really challenging and is why commerical single use SERS substrates usually retail for about $NZD100 each.

We managed to produce a SERS substrate that is cheap and easy to make and most importantly gave reproducible results. We did this by using a sputter coater to deposit nanogold onto a Blu-ray disc. Sputter coaters are used in almost every lab where there is microfabrication and it deposits metal using a plasma. They are relatively cheap to buy I have seen a few on eBay. There is a nice video about how to make your own sputter coater. The amount of gold is actually quite small so it would only be a few dollars of gold on each sample. Most  You could deposit a SERS substrate on any non-conductive material however the trick for making a reproducible SERS substrate is that the Blu-ray disc provides a very uniform clean sample. The Blu-ray disc is made up of layers of plastic and metal with the grating hot embossed into the larger plastic disc.

Diagram of layers found in Blu-ray disc with the hard coat the polymer being easily peeled off to reveal the grating.

By peeling off the plastic top coat with tweezers you reveal a perfect surface for depositing the gold.

Using tweezers the top plastic coat can be removed to reveal the grating.

In order to tune the nanoparticles to the wavelength of light we used (785 nm) we changed the sputtering time. This produced smaller nanoparticles and led to the substrate resonating at the right wavelength. Putting a drop of melamine doped water and measuring the spectrum using a Raman microscope we were able to measure down to 68 molecules out of a billion water molecules. The amounts deemed to be dangerous by the world health organisation are 1 ppm in infant formula so we are well under the limit of detection.

Above the Raman spectrum of melamine at different concentrations. Below the melamine vibration we are measuring is the ring breathing mode. Link to animation 
The measurement was also very reproducible with only 12% variations which is very low when considering other SERS substrates with you can have over 100% variations between areas of the same sample. 

The Raman spectrometer we used was quite a large unit (table top) but there are now smaller handheld Raman spectrometers and these will only be getting cheaper. We think the applications for a low cost testing kit could be developed making use of a portable Raman spectrometer.

Portable Raman spectometer Link
An interesting application we would like to explore is using these low-cost SERS substrates to identify different bacterial strains in a hospital setting. Below is a presentation I gave on this at Advanced Materials and Nanotechnology conference in Auckland.

Sunday, 17 April 2016

Sifting for carbon dioxide

Some colleagues and I have just had a paper accepted for publication and with a title like "Enhancement of Chain Rigidity and Gas Transport Performance for Polymers of Intrinsic Microporosity via Intramolecular Locking of the Spiro-carbon", I thought I would explain it in more simple language.

In short

  • If we could filter out $CO_{2}$ from the emissions of powerplants we would be able to collect and economically pump it into underground storage without having to compress all of the other gases emitted from the powerplant, such as nitrogen (which can be up to 60%).
  • New plastics have been recently made called 'polymers of instrinic microporosity' (PIM). They can let through $CO_{2}$ selectively, but more importantly they are very permeable which means they can separate out $CO_{2}$ on an industrial scale.
  • Jianyong Jin (University of Auckland) and his team found a way to make the best PIM so far by introducing a locking mechanism between the molecules.
  • To show how this locking mechanism improves the polymer, I used computer simulations to show that the lock increased the rigidity of the polymer and also produced the optimal geometry for the polymer. This made the pores just the right size for the $CO_{2}$ to pass through, and the large rigidity made it selective and very permeable.
  • Here is a video of a similar material and the sort of separations that this polymer enables.

What is the most interesting detail if you are a scientist?

The most interesting detail for scientists is the idea of locking polymer molecules using an intermolecular locking mechanism. Many different polymer systems rely on the interplay between entropy and enthalpy. By locking the polymer backbone you can play with how the polymer packs and distributes its vibrational energy. One really interesting application we comment on in the paper is locking protein molecules using intramolecular locks. We are very interested in exploring this idea with other groups.

From a materials point of view, the idea of engineering the size of the pore by changing the geometry of the linkage is an interesting concept. You can go straight from a chemical structure to a material property.

Overview of the paper with some further details

The polymer monomer that was locked is called a spirobisindane, SBI for short, formed by adding bromine onto the 6-membered rings and then performing a silver oxidation which forms an ester bridge. This forms an 8-membered ring which bridges the weakly bonded spiro centre. Here is what it looks like with a 2D drawing.


It looks quite ridiculous in a two dimensional drawing. How can those two atoms be linked all the way across by an oxygen atom? Well, looking at the 3D drawing it makes a lot more sense, the 6-membered rings are actually a lot closer together than you might think and the bridge is only slightly strained. I have coloured the different rings so the 5-membered rings are coloured blue, the 6-membered rings are coloured pink and the 8-membered ring is coloured yellow (this also helps when comparing with the 2D representation shown earlier).

The image above is actually a 3D model you can rotate in the web page so click and hold on the picture and you can rotate you can also zoom in with the mouse wheel. Thanks to molview.org for the plugin.

The polymer, made up of thousands of repeats of the monomer, forms a very rigid structure which packs very poorly leaving lots of space (pores) for the gases to permeate through. With PIMs you want to optimise the pore size for the molecule you want to sort and also it is important this pore doesn't change size by much. Thermal energy from the polymer being at room temperature causes the pores to change size as the polymers move around. Changing the pore size reduces the selectivity for $CO_{2}$ over other gases so one method that has been used in the past is to rigidify the polymer. By replacing the spirobisindane centre with something more rigid you can make it more selective. Many of these attempts, however, changed the pore geometry so it was no longer big enough for $CO_{2}$. So what we have done was to rigidify a particular PIM, PIM-1, which we know has a good geometry for $CO_{2}$ separation, while keeping the large pore size. 

To show how this intramolecular lock improves the rigidity of the polymer I simulated the movement of the polymer in the computer. This heats the polymer, causing it to vibrate and the rigidity is tracked by looking at the distance between the atoms at the end of the chain (end-to-end distance). This fluctuates over time performing a repetitive motion, almost like a snake. The polymer that is more rigid will have the less movement over time. There is a plot of the end-to-end distance over time (a) and also the bar graph showing the frequency of the end-to-end distances being at a certain distance.


What we found was that the unlocked polymer, labelled SBI, was much more mobile and the end-to-end distance varied a lot more than for the locked polymer.

Using a more advanced technique (taking into account the quantum mechanical description of all of the electrons in the molecule) we compared how much more rigid this locked PIM was, compared with other methods used to rigidify the polymer. The plots you see below show how the potential energy increases, from the baseline value, as I forcefully twist the polymer in the computer. Think of it like a spring - as it is twisted past its natural equilibrium, it will want to spring back. I actually used the equation for a spring to describe how rigid each polymer is. When describing a spring there is something called the spring constant - the larger this value, the more the energy increases with the deflection. The equation for the spring's potential energy is a parabola:


Fitting the spring model to the potential energy plots I calculated for the different polymers allowed us to compare with other linkages. We showed that there was a 230% increase in the rigidity for the locked SBI compared with the unlocked SBI. It also showed there are other linkages that are actually more rigid. However the locked SBI has both the correct geometry to get the right pores and has reasonably large rigidity. 

Excellent gas separation experiments were performed by Tim Merkel and Sylvie Thomas at the Membrane Technology and Research, Inc. in the US. These  are called Robeson plots and have selectivity on the y axis and permeability on the x axis. It was good to have a selective polymer but for industrial scale separations the real winner is having a permeable polymer. A permeable polymer allows for industrial scale separations.


All of the black open circles are common polymers. They have a trend that Robeson saw - as you increase their permeabilty you decrease their selectivity. He set an upper bound, the black line, that was not surpassed until these rigid polymers of instrinic microporosity were developed. You can see that for these two gas pairs $O_{2}/N_{2}$ and $CO_{2}/CH_{4}$ PIM-C1 is the most permeable polymer above the line. The first gas pair $O_{2}/N_{2}$ is useful for separating oxygen from the air which is mainly used to generate oxygen in medical applications. The second gas pair $CO_{2}/CH_{4}$ is important for removing carbon dioxide from natural gas which means you only need to transport the methane not the carbon dioxide as well; this is called natural gas sweetening. However, the important application is removing $CO_{2}$ from factory emissions. This is the $CO_{2}/N_{2}$ pair, which we didn't show in the paper, so I plotted it below.

I plotted the data with a linear scale and a log-log scale (which is most often used) as it might be more intuitive to see the data on a linear scale. You can see you get some good selectivity and a huge permeability for $CO_{2}$ which is exactly what we wanted for greenhouse gas capture. So the take home message is that careful design of the molecules leads to a huge change in the material properties and from a chemist's point of view, this was very exciting.