The Positive Side of Proton Transfer
For the past year, I’ve been studying the mechanisms of proton transfer in Professor Kana Takematsu’s physical chemistry lab at Bowdoin College. Physical chemistry is a branch of chemistry that is based largely on mathematical properties, and as such is often feared by even chemistry majors. Breaking down chemistry from the macroscopic to the microscopic and studying molecules and atoms at their fundamental level is hard to grasp. But this fundamental nature of physical chemistry is precisely why I love it. Underlying all chemical reactions are interactions between positively charged protons and negatively charged electrons and if the interaction is favorable, a reaction will occur. Studying proton transfer, occurring on the nanosecond timescale, is thus hard to imagine, but super important to understanding chemistry between all sorts of more complex molecules, especially in biological systems.
To give a bit more context, proton transfer (PT) describes the movement of a positively charged hydrogen (interchangeable with the term ‘proton’, and shown as ‘H’ in the figures below). The most well known example of this phenomenon is when protons are transferred across a gradient to generate energy (ATP) in cells. PT in molecules occurs when the proton is transferred from one molecule to another (intermolecular PT) or from one part of a molecule to another (intramolecular PT). Intramolecular PT is more likely to occur when there are multiple proton-binding sites that are close to one another. The less distance between the sites, the less interference from the environment around the molecular there is! Intramolecular PT is what I have been primarily studying. I have been examining aminonaphthols, molecules with 2 proton-binding sites. The aminonaphthols that I have been studying are 3-amino-2-naphthol (3N2OH), and 1-amino-2-naphthol (1N2OH). As you can see from the figures below, the molecules look similar: there are two proton binding sites that are right next to each other. The only difference is the placement of the second amino (NH2) site. (NOTE: The figures below are drawn in line-bond structure where each corner represents a carbon, one connecting line is a single bond, and two connecting lines is a double bond).
So then what is the difference between these two molecules? Why do I care about studying both 1N2OH AND 3N2OH if they look so similar? Well, just by moving that NH2 site, the molecule actually significantly shifts its electron density. This essentially means that the negative charge is localized on a different part of the molecule. The localization of the negative charge actually plays a huge role in the chemistry of the molecule, including how it participates in proton transfer. My research is focused on understanding the mechanisms of proton transfer, and just how electron density influences it.
As I mentioned, a lot of physical chemistry is based in a molecule’s mathematical properties. These properties are determined by the molecule’s quantum mechanics. Quantum mechanics is far too complicated of a subject to explain even at the simplest level (and, if I’m being honest, I still don’t completely understand it), but suffice to say that all molecules can be described by what is called their wave function. A wave function is a mathematical representation of molecular behavior, and how an atom or molecule interacts with other atoms or molecules, and how it behaves by itself can be described by this wave function. This makes it sound like we have found the solution to all of science! Just solve for every molecule’s wave function, and we can know everything we need to about all kinds of scientific questions. Unfortunately, the one caveat is that we can’t exactly solve for this wave function. This is because we can’t precisely describe the interaction between electrons for a whole host of reasons that I won’t get into because people since Einstein have been trying unsuccessfully to do so.
So even though it seems I’ve introduced this mystical wave function for no reason because we can’t exactly solve for it, it turns out that we can do the next best thing: approximate it. However, if humans tried to approximate the solution to wave functions, it would take years (if we’re lucky). Instead, we employ the much faster computer technologies to approximate solutions to a wave equation. This is a branch of physical chemistry called computational chemistry, and has exploded over the past two decades, as computers get faster and smarter. Thanks to computational chemistry giants like Hohenberg and Kohn1 and Kohn and Sham2 the most popular computational chemistry program, called Gaussian, does a fairly good job of approximating the wave function of individual molecules.
As I’ve said, these computational methods are good at approximating a molecule’s wave function. But this approximation is due to purely mathematical calculations. How a molecule actually behaves is often times influenced by a variety of factors that the computational program doesn’t necessarily do the best job of taking into account. One such environmental factor is the solvent of a molecule. Though computational programs can introduce a blanket solvation, this often times doesn’t pick up on the specific interactions between the solute and solvent. This is why it is important to combine both computational and experimental data to see where they agree, and where they don’t.
For the past year, I’ve been examining the excited state properties of 1N2OH and 3N2OH by performing computational and experimental investigations. The excited state, where an electron is promoted to an upper level, is where a lot of excited state PT occurs. Because aminonaphthols are photoacids, this means that they become more acidic, or more likely to lose a proton, when they become excited. Experimentally, this means shooting an intense beam of light at the molecule at different wavelengths in a process called spectroscopy. My research primarily involves UV/Visible absorption spectroscopy, and steady-state fluorescence spectroscopy. UV/Visible spectroscopy allows us to measure the wavelength (and, by extension, energy) of light that causes the electron to be excited. Fluorescence spectroscopy measures the wavelength (and energy) that is emitted from the electron relaxing back down to the ground state. These processes can also be mimicked in computational calculations.
My primary interest is studying the proton transfer (PT) mechanisms of aminonaphthols. How does spectroscopy allow us to do this? Well, at different pH values, aminonaphthols can exist as four different species in the excited state: the cation (where the NH3+ and OH are protonated), zwitterion (where the OH is deprotonated but the NH3+ is protonated) neutral (where the OH is protonated and the NH3+ is deprotonated), and anion (where both the NH3+ and OH are deprotonated).3
Each of these species has a distinct fluorescence spectra, so we can tell at what pH a new species is existing. This can also help us understand how the proton is being transferred. For example, if we change the solvent environment and one species is no longer present, we know the proton was most likely transferred along a solvent chain. Fluorescence spectra can also tell us the stability of a particular species: if the light fluoresced is at a large wavelength, that corresponds to a low energy, and that means the species is stable in the excited state. We can also compare our experimental fluorescence peaks and our predicted fluorescence peaks calculated computationally to try and determine what in-lab factors are not being accounted for by quantum mechanics.
Some results
One frustrating thing about physical chemistry is that the majority of the work is understanding the theory behind the experiments. Knowing what steps to take involves a lot of reading papers and having hypotheses based off of previous work. From the past year, I would say 70% of my time has been spent reading or trying to understand concepts, and only 30% has been producing actual data. But what I have so far is exciting! Starting with 3N2OH, my experimental and computational results have correlated very well! This indicates that computational predictions are fairly accurately accounting for experimental results.
Interestingly, computational and experimental results suggest that the zwitterion (Figure 2 top right) is preferentially formed at lower pHs to the neutral (figure 2 bottom left). This indicates that there is a stabilization of the zwitterion due to the interaction of the NH3+ and O– groups that we hadn’t expected! This is exciting for us because it is another piece of the puzzle to determining just how the proton is being transferred from the cation (figure 2 top left) to form the zwitterion.
Excited by the success of 3N2OH, I moved on to looking at 1N2OH. As is usual with science, I was not getting the results I expected. Computational analysis showed that the fluorescence spectra were much more stabilized than experimental spectra was showing. I couldn’t accurately match my spectra to determine which species I was seeing fluoresce, as I had been able to with 3N2OH. An additional frustration was that 1N2OH breaks down at very basic pHs. For a compound so similar to 3N2OH, it certainly was giving me a lot of problems! But why? A likely explanation is that the shift in electron density (as I mentioned above) is causing some interesting/different chemistry to occur. But exactly what affects the electron density is having, I still don’t quite know.
As you can tell, a lot of my experiments have been giving me more questions than answers. This is what I love about science: it is never finished, and there is always a next step. Something that seems so simple, the transfer of a positively charged hydrogen, is actually very complicated and intricate. After a year of experiments, we still cannot definitively say what is going on! But we press ahead because curiosity drives us to. And that is what science is all about.
Author Bio: Holly is a senior at Bowdoin College majoring in chemistry and physics. She has done extensive research in physical chemistry, and has worked in Professor Kana Takematsu’s lab for the past year. Her primary research focuses on studying the mechanisms of proton transfer using computational and experimental techniques. Holly also participates as a member of the Bowdoin College varsity swim team and loves being a volunteer at a local elementary school as a math tutor. The research described in this article was funded by the Maine Space Grant Fellowship and the James Stacy Coles Fellowship.
References
- Hohenberg, P.; Kohn, W., Inhomogeneous Electron Gas. Physical Review 1964, 136 (3B), B864-B871.
- Kohn, W.; Sham, L. J., Self-Consistent Equations Including Exchange and Correlation Effects. Physical Review 1965, 140 (4A), A1133-A1138.
- Ellis, D. W.; Rogers, L. B., The measurement of ionization constants of electronically excited species from aminonaphthols. Spectrochimica Acta 1964, 20 (11), 1709-1720.