As the recent work done by IBM to use polymer chemistry to produce smaller data storage domains, or the recent assembly of molecular FETs by TIIT using a combination of biochemical methods, or the assembly of a single electron tunneling transistor in part with chemical methods at the Chalmers Univ. of Technology, or the commercial work being done by Calmec and Zettacore there is no shortage of problems for creative chemists to work on.

Four problems I can think of without devoting much energy to it would be:

• Self-assembly of molecular CPUs, starting with migrating from a molecular diodes or transistors to a molecular flip-flops to a molecular adder (see for example Figure 15 (pg 41) of Mitre's Pink Book) [Simple description in EETimes].
• Guided assembly of molecular wires — perhaps made from nanotubes.
• Development of self-assembly methods for molecules that can store data and replace the Level 2 or perhaps provide a Level-3 cache for future microprocessors. The Level-2 caches are starting to take up a significant fraction of microprocessor chip real estate and generate much of the heat because they use static RAM transistors. Modifications for the self-assembly of molecular memories that can take the heat on current chips or perhaps a completely separate chip assembled using chemical methods that can be mated directly to existing chips (moving us into real 3D assembly) that could put the Level-3 cache much closer than DRAM or MRAM main memory would be very desirable.
• Working out more of the chemistry to optimize various reactions and assembly methods described in Chapter 8 of Nanosystems. (And communicate it to many more chemists so they will become convinced it really *will* work!)

All of these are going to involve chemists. So while yes, some aspects of solution phase organic chemistry will diminish in importance, that is going to happen anyway whether or not Drexler is right! This will be due to the simple fact that designer enzymes and/or evolved enzymes (e.g. work being done by Maxygen) are going to be more efficient at many, perhaps most, reactions that have traditionally been handled by solution phase organic chemistry. If that were not enough, the development of retrosynthetic programs to work out very complex synthesis paths is making at least a couple of years of education that the typical organic chemist receives rather obsolete.

So the need for people to reeducate themselves is going to become a continual and accelerating process (due to the singularity). The chemists are going to simply have to catch up with people such a physicians who have had things like requirements for attending CME seminars and conferences for a long time in order to retain their medical licenses.