One of the hazards of having a degree in chemistry is the appealing idea that you can explain everything and predict everything on the basis of textbook notions on solubility, electronegativity, pKa’s, or molecular orbitals. These are important things to be sure. But in the field, the recall of knowledge isn’t always enough. More often than not you have to collect data and generate new knowledge.
Rationale of a result on the basis of hand waving and a few reference points can seem compelling in a meeting or brainstorming with a colleague to understand a problem. But in the end, nothing can top having solid data from well conceived experiments.
My chemical “intuition” have proven wrong enough times now that I am deeply skeptical of it. After prolonged periods of absence from the lab I find myself resorting to a few cherished rules of thumb in trying to predict the outcome or explain the off-normal result of a process.
In chemical process development there is no substitute for running experiments under well controlled conditions and capturing solid results from trustworthy analytical methods. It is hard work. You may have to prepare calibration standards for chromatographic methods rather than the preferred single-transient nmr spectrum in deuterochloroform.
We’re all tempted to do the convincing quick and dirty single experiment to finesse the endpoint. Certainly time constraints in the manufacturing environnment produce an inexorable tilt towards shortcuts. But in the end, depth of knowledge is only had by hard work and lots of struggle in the lab. The most important part of science seems to be to frame the most insightful questions.The best questions lead to the best experimental results.
Lamentations on Chemistry 
Nice thoughts…. It is especially tempting to show how smart you are by explaining a one-off experiment. Often, though, the explaination is just plain wrong.
Pharma began as many chemists diddling around with natural products’ hints. That evolved to structure-activity stuff and fitting a putative target – with a smaller R&D staff. When that ran dry – combinatorial chemistry and massively parallel assays! Hardly needs any folk at all. A single failed reaction is a setback. A million failed reactions are a combinatorial library.
Pharma now has two divergent paths. 1) Finding profits in therapies for butt itches. If you don’t mind risking crushing depression, projectile vomiting, icthyosis, and a modest incidence of idiosyncratic death… your right second toe need not itch between 2114 and 2233 hrs each evening. 2) Desperately seeking a managerial paradigm for the next “breakthrough.”
(2) Is coming along. So far it contains terminating US domestic research and substituting outsourced efforts using the same preferentially hired personnel but remaining in their native lands. US education specializes in disgorging deeply indebted domestic unemployables. “Diversity” does not carry its own costs.
My Dear Uncle Al.
Let us not forget the over-reliance on in silico design, fragment based discovery and all the other so-far-its-not-as-useful-as-we-thought molecular modeling to design the next lipitor. If all the fancy toys are really that good, why are they droppin’ deuces at the thought of expiring IP?
“One of the hazards of having a degree in chemistry is the appealing idea that you can explain everything and predict everything on the basis of textbook notions on solubility, electronegativity, pKa’s, or molecular orbitals. “
I have better luck with predicting lottery numbers than I do predicting experimental results. Forunately, the number of predictions I make is inversely correlated with my age.
In grad school we joked that you can answer any mechanistic question on exams with one of the following four:
1. Steric effects
2. Secondary orbital interactions.
3. HOMO-LUMO interactions
4. P pi- d pi interactions
Halogen bonding! Compare the crystal structures of tetrakis(4-bromophenyl)methane with that of co-crystallized Ph4C plus CBr4, Can. J. Chem 80(10) 1351 (2002). LCAO has been very, very good to Uncle Al over the decades, including Framework Molecular Models. KISS.
Brimonidine is a sexy alpha-blocker as glaucoma eye drops. Brimonidine is photosensitive and ships in an opaque dispensor bottle. Brimonidine was typically preserved with chlorine dioxide. Would you put brimonidine in your eyes then walk outside into sunlight? Betoptic-S, beta-blocker, does not adversely interact with Cialis, etc. This is a major selling point to “senior” villages. Men die off after 65. A surfeit of women are on the hunt with casseroles in their hands and quicker picker uppers in the purses. Know your market.
HyperChem-Lite ver. 3 is quite wonderful. it costs substantially less than an organic textbook. Howver, it is my observation that if you are stupid without a computer you are still stupid with a computer. God save us from the congenitally inconsequential and ACS Project SEED. I’ve never met a synthesis that was socially advocative. You fight for every percent of yield – and none of that microchemistry crap. TLC spots don’t pay commissions.
Never confuse product with process. Process like birth is an unwelcome painful mess after the main player gets… screwed.