Unsolved Problems in Chemistry

Originally posted 2011-10-23


(Update from 2019: This list has been cleaned up a bit and extra links added, but nothing has fundamentally changed about the list. One emerging area I’m very excited about in 2019 is neural network approximations to predicting chemical properties.)

A google search for “unsolved problems in chemistry” returns some meager results. The wikipedia article is uninspiring, and there’s a lot of speculation from professors and graduate students on how their personal research is an important unsolved question in chemistry. If I were a young student, these lists would do a fairly poor job of convincing me to go into chemistry.

Here’s my own take on what I think this list should look like.

Unsolved Problems in Chemistry:

  1. The Origin of Life / Abiogenesis. ’Nuff said.
    • Homochirality: Why are L-amino acids and D-sugars so dominant in the biological world, and how did it happen? There’s been much more headway made on this problem than on other abiogenesis-related problems (See Viedma and Blackmond’s research on chiral amplification). The gist is that once a small enantiomeric bias is generated, most likely by statistical fluctuations, then there exist processes that can amplify this initial bias into homochirality.
    • Prebiotic sugar chemistry: How did sugars come about in the prebiotic world? Since RNA molecules contain sugars, An answer to this question is a prerequisite for the RNA world hypothesis. The formose reaction, which takes formaldehyde and outputs sugars, seems to be our leading candidate here, although the resulting complex sugar mix contains just about every sugar you could possibly imagine, and then some.
    • Prebiotic lipid chemistry: How did the first compartments come about? What is the chemical nature of the first micelles/lipid bilayers?
    • Molecular paleontology: The investigation of biochemical renmants of the earliest forms of life. Traditionally, people think of physical remains, but I think it’s more interesting to dig into hints of how biochemical systems came about. For example, the amino acid code is often redundant in the third base: e.g. TC* codes for serine, regardless of the third base’s identity. That’s probably a hint at something. And there appear to be RNA fragments in all sorts of core biochemical processes; ATP is literally an RNA nucleotide. The abundance of RNA in core processes leads to the RNA World Hypothesis: the idea that modern life descends from an “RNA world” in which RNA served multiple roles now occupied by proteins and DNA.
  2. Chemistry on a medium scale. We understand small molecules (\(10^{-10} - 10^{-9}\) meters) very well - quantum mechanics and molecular orbital theory do a pretty good job of explaining many phenomena. We understand bulk materials (\(10^{-5} - 10^5\) meters) very well - statistical mechanics, continuum mechanics, and crystallography can explain the behavior of most inorganic materials. However, there’s an unexplored gap in between these two length scales. We don’t understand the behavior of biological macromolecules, nor do we understand the behavior of nanomaterials such as inorganic macromolecules.
    • What are the chemical/electrical/mechanical properties of buckyballs, nanotubes, and graphene? (\(10^{-9} - 10^{-8}\) meters). How can we bring silicon chip manufacturing down to a scale where bulk silicon properties no longer hold?
    • What are the rules governing biomolecular processes? The cell is commonly envisioned as a bag of electrolytes and proteins and organelles, but that’s almost certainly missing important details. The nascent field of cell biophysics tries to explore the \(10^{-7} - 10^{-6}\) meter range.
  3. Better approximations to the Schrodinger/Dirac equation. All of chemistry essentially arises from quantum mechanics. If we could solve these equations faster and more accurately, then we could scale up the set of chemical simulations we can run on a computer. Density Functional Theory was a breakthrough that allowed significantly faster computations. But today, most practicing chemists ignore computationally derived properties, because the properties they compute are either uninteresting or fail to take important effects into consideration, like solvent molecules.
    • Protein folding is a busy area of research now.
    • We still have no way to accurately model large numbers of solvent molecules, and most of our computations are done in unrealistic gas-phase conditions. (Gas-phase is a nice way to say, “Let’s pretend that this molecule doesn’t interact with anything”.)
  4. Energy, sustainable living. Resources in general. Can we rid ourselves of all dependence of finite resources, and rework our civilization to operate in a steady-state manner, with only solar energy as an energy input? Phosphates, helium, oil - these are all finite resources that are going to become very scarce within our lifetimes. Other longer-term finite resources exist.
    • Cheap, efficient solar photovoltaics.
    • Energy storage technology to smooth out variations in energy production and demand.
    • Efficient recovery of salts from seawater (the eventual solution to phosphate and lithium shortages).