…or why adding chloroform to coffee can be a good idea.
Perhaps you remember a story from a few weeks ago that up to one in ten bags of “100% arabica ground coffee” on sale contained “substantial” amounts of robusta coffee?
The story suggested that, perhaps unsurprisingly, fraud is quite common in the coffee industry with cheaper robusta being substituted for the more coveted arabica in a substantial number of packs of pre-ground coffee. But how did the authors of the paper measure this and why did speciality coffee play an important role in the study?
The study used ¹H NMR (nuclear magnetic resonance) spectroscopy to measure the concentration of a particular “finger print” chemical known as 16-O-methylcafestol or 16OMC for short. This aspect of the study was not new. The compound 16OMC was known to be found in robusta (Coffea canephora) while it had not been previously found in arabica (Coffea arabica) beans. It had therefore been considered an excellent marker chemical as to whether a sample of arabica beans had been contaminated with a cheaper robusta.
Previous studies had also used NMR to check for 16OMC but in those studies, they had used a conventional NMR machine and the data collection and analysis had taken a long time. It was also expensive, which meant that it had shortcomings as a technique for quickly investigating fraud within the industry.
The difference in this new report was that firstly, the scientific team investigating the coffees used an NMR machine that fits on a lab table-top: portable, commercially available, and so a possible tool to quickly detect fraud. But secondly, the authors ‘double brewed’ the coffee using chloroform. They first dissolved the ground coffee in chloroform which was filtered using filter paper and then dried and re-dissolved in fresh chloroform to produce a super-concentrated coffee-chloroform brew. This super-concentrated chloroform coffee enabled the authors to obtain a much better signal to noise ratio on the data and so improve the reliability of the detection of any rogue robusta.
But why could this group use a portable NMR machine whereas previous studies required far more expensive and bulky pieces of kit? NMR works because, just like electrons, atomic nuclei (protons, neutrons) have a property called spin. This spin gives rise to a magnetic moment which means that when you apply an external magnetic field to the sample, some nuclear magnetic moments are parallel, some perpendicular and some antiparallel to the applied field. Consequently, the different moments have different energies which, being on the atomic scale, are quantised meaning that they form discrete levels. This difference in discrete energy levels means that the nuclei will emit/absorb energy (i.e. light) at specific frequencies, which we can calculate. Moreover, the frequency is directly proportional to the applied magnetic field (because the larger the field, the bigger the energy difference between the levels): increase the field applied and you increase the resonance frequency of the nuclei.
But there is one more detail. The nuclei do not exist in isolation, they are affected by the chemical environment that they are in. So a proton in 16OMC will respond slightly differently to an applied magnetic field than a proton in say, water. Rather than be at the resonance frequency we have calculated, the frequency will shift as a consequence of the chemical environment surrounding the proton. As you may expect, this shift is small, but it is significant. It is partly because of this effect that NMR is such a fantastic tool for chemical analysis¹. Typically, the shift is of the order of parts-per-million from the non-shifted resonance frequency. So, in the coffee study discussed here, the interesting “fingerprint” peak is at 3.16ppm. Given that the machine was operating at 60MHz, this means that the scientists were looking at shifts of 189.6 Hz to the non-shifted resonance signal.
It seems sensible that the bigger the shift, the easier it would be to resolve these chemical fingerprints. To get the larger shift requires using a higher operating frequency which is exactly what more traditional NMR spectrometers used. However, given what we know about nuclear energy levels (above), a large energy level split (i.e. high operating frequency) requires a large magnetic field, and large magnetic fields require expensive and bulky pieces of kit. To put this all in perspective, the magnetic field of the Earth at its surface is (variable but around) 0.00005T. A fridge magnet has a field about 0.01T. Commercially available, small rare earth magnets can have fields about 0.3T. A 60 MHz NMR spectrometer looking at ¹H nuclei would need 1.5 T, higher frequency NMR spectroscopy would require still higher fields. The sort of magnetic fields that would be needed for the more traditional NMR technique therefore require large superconducting magnets which are bulky and require expensive cooling. Being able to use a lower frequency for such sensitive measurements is a significant engineering, as well as scientific, achievement.
So where does the speciality coffee come in? Well, it turns out that by measuring speciality coffee the team uncovered a surprising result: 16OMC was present in arabica beans too.
In order to calibrate the technique, the study had obtained traceable coffees known to be purely arabica or purely robusta. Some of these coffees were sourced from Ethiopia and were grown far away from any possible robusta hybridisation. They were speciality coffee. When the team measured these samples with their concentrated coffee extraction technique, they found that these too contained a small peak at 3.16ppm. Previous studies had missed this because it is such a small quantity. So, as well as determining a technique to quickly establish whether a given coffee on sale is fraudulently being marketed as 100% arabica, this new technique enabled the scientific community to learn something new about arabica. The coffee is more chemically rich than was realised.
If you would like to read more about the study, the authors have summarised it here as well as publishing the paper as open access (so you can download it for free) here. A summary of the results by the company that made the spectrometer can be found here. You can learn more about NMR spectroscopy online, or by obtaining a book from the library such as:
¹ Nuclear Magnetic Resonance Spectroscopy, Robin K Harris, Longman, (1983,1986)