The Colourful World of Chemical Sensors
Illuminating a 60-Year Misconception
Betaine 30 (left) and the new Extended Betaine 30 created by the Wong group. The vials on the right show the range of colours of the extended dye in solution.
While synthesising a highly sensitive molecular sensor based on a famous colour-changing dye, a research group has rectified two mistaken reports about a 60-year-old dye and showing that a previously misunderstood design motif has potential to make more sensitive sensors for chemical analysis and biological imaging.
You may know that water and oil don’t mix: but why is this the case? The answer lies in solvent polarity. Broadly speaking, polar solvents are those like water while non-polar solvents are like oil. Solvent polarity, despite its influence on reaction rates, equilibria and mechanisms, is often difficult to quantify due to the complex interactions occurring in solution. In 1963, Christian Reichardt was studying solvatochromism: molecules that can change colour as solvent polarity changes. One dye he made, Betaine 30, was extremely sensitive to polarity, with the molecule absorbing shorter wavelength light as the solvent polarity decreased. It became the basis of the ET(30) polarity scale and found uses in chemical analysis and biological imaging. A recent application was to identify fake or poor-quality sanitisers during the COVID-19 pandemic.
The past two years, Stephen Franzese, a researcher in the Wong group, has been designing and investigating new variants of Betaine 30. Betaine 30 is a zwitterion: it has both a negative and a positive charge. This makes it stable in polar solvents and progressively less stable as polarity decreases. If the distance between these charges were increased by making a dye with a longer bridge, it was predicted that this change in stability would be even more pronounced. The new extended Betaine 30 was found to be more sensitive than the original dye, although it degraded in low polarity solvents, turning permanently yellow. Because it is more unstable in low polarity solvents than Betaine 30, it shows greater changes in colour but is also more reactive. Additionally, the bridge between the two ends is more open to attack than the shorter Betaine 30.
Reviewing the literature, Franzese noticed two other groups had previously synthesised a similar extended dye, Betaine 21. One group was Reichardt himself who had found it to be non-solvatochromic but had only tested it in two solvents, one of which had degraded Franzese’s dye. The second report was published 60 years after Reichardt’s and claimed that the dye showed ‘solvatochromic inversion’: unlike Betaine 30, it would absorb longerwavelengths of light as solvent polarity increased and then, once the polarity was large enough, would start absorbing at shorter wavelengths instead.
Perplexed by these conflicting results, Franzese made Betaine 21 himself and tested it in a wider range of solvents. He found that the dye was in fact strongly solvatochromic, but that it broke down more easily than any other dye he had made, also turning permanently yellow. In addition, by comparing his methods with those of the second group, he discovered that the inverted solvatochromism was actually caused by methanol, a highly polar solvent, being present in solution and interfering with the results.
In revealing these mistakes, Franzese has proved that dyes with larger separation between their negative and positive sides can still function well as solvent polarity sensors. This opens a pathway to make new, more sensitive variants of already-existing dyes, leading to even more potent chemical sensors. His work also provides a salient reminder of the importance of care and rigour in the design of experiments: a simple oversight can cause huge misunderstandings