62nd Annual Report on Research 2017

Chemically fueled dissipative (energy-expending) assembly is one of the hallmarks of biochemical systems, yet its development as a tool for abiotic systems chemistry is in its infancy. The goal of this project is to develop carbodiimides as useful chemical fuels for the assembly of carboxylic anhydrides, in water, from the corresponding carboxylic acids. Since anhydrides are hydrolytically unstable, they should spontaneously decompose back to the starting acids. This basic functionality, the transient formation of an unstable covalent bond, is one of the fundamental building blocks for functional systems such as molecular motors, nonequilibrium gels, etc.

As outlined in the original proposal, the project has three Specific Aims: (1) The development of anhydrides as dissipative covalent bonds. (2) The construction of macrocycles using this chemistry. (3) The generation of out-of-equilibrium molecular geometry changes using this chemistry. Over the past year, we have made substantial progress on Specific Aims 1 and 2 and have begun work on Specific Aim 3.

Our first key results were to establish conditions for the transient assembly of simple commercially available carboxylic acids MEA-Ac and KSB-Ac fueled by the common carbodiimide reagent EDC, as shown in Figure 1a (Specific Aim 1). After some optimization, suitable conditions were identified and kinetic runs performed for both systems. Substantial characterization was done to show that anhydride is indeed formed in these systems. The kinetics of both systems are readily monitored by NMR spectroscopy (Figure 1b,c), and the concentrations of all species could be fit to a simple kinetic model (solid lines in the plots). From the fits we were able to extract further information about the systems, particularly the yields (net anhydride formed per EDC added).

These results demonstrate the basic feasibility of the system. To show that this chemistry could be used to assemble more complex architectures, we explored the formation of crown-ether-like macrocycles from simple diacids, as shown in Figure 2a (Specific Aim 2). The behavior of these systems is comparable to the monoacids, with both cyclic (TEG-Cy) and acyclic anhydrides observed (Figure 2b). Interestingly, the yields of the macrocycles were found to be sensitive to the presence of alkali metal cations, as shown in Figure 2c, with the “matched” cation acting to decrease the yield of the macrocycle. This effect, a sort of “negative templation” by the species’ own guest, is a counterintuitive example of out-of-equilibrium effects on assembly. We have published a paper describing these results in the Journal of the American Chemical Society.

We are presently continuing our efforts toward Specific Aims 1 and 2 by taking a careful look at structure–property effects in a systematic series of monoacids and by designing new cage-like hosts that should exhibit much stronger cation binding constants. Materials for these experiments are currently being synthesized. We have also made significant progress toward Specific Aim 3. We have shown that this same chemistry works on diphenic acids to give diphenic anhydrides (which results in a large decrease in the twist about the biaryl bond), as shown in Figure 3. We are currently synthesizing substituted diphenic acids to probe structure–property effects in this system.