It’s funny how you can deceive yourself into thinking that you understand a reaction. Then you do that last experiment and get a result that shows unexpected sensitivity to one thing or other. Depth of knowledge comes from doing a lot of experiments, not hand waving. It is important to try to learn something from every experiment. If the rxn went south, what happened? Can you do a mass balance? What happens to the mass that doesn’t convert to product? Consider every “poor” result as an opportunity to extend your understanding of the reaction.
If you want to claim true expertise in a process, you have to know what the reaction system is broadly sensitive to and what it may be insensitive to. In short, you need to know what affects the velocity of the reaction or what steers it to side product formation. Exactly what are the boundaries of “normal”?
Running the same reaction a hundred times successfully by carefully following the directions confers proficiency, not expertise. It is fine for most workers to have proficiency. But someone should take the trouble to acquire broad expertise for the inevitable off-normal event somewhere down the timeline.
An upset condition can stem from an off-normal engineering input or from some reactivity issue. How many watts per liter will your reaction generate? How will your reaction mass behave if there is a solvent boil-off? Does the solvent boiling point fall below the maximum temperature of the reaction in an off-normal condition? In other words, can the reaction mass self-heat in a manner leading to a runaway condition? If so, what layers of protection are in place to prevent this kind of event?
The ability to push electrons in a mechanism or facility with named reactions is not enough skill for process scale-up. A chemist has to walk over the entire acreage to thoroughly map out the hills and valleys of the process. The people who operate the big pots and pans, and their families, are depending on your thorough knowledge to keep them safe.
Acquiring expertise is going to annoy people. It necessarily slows things down. It will make you a colossal boor at parties. But never confuse motion for progress or data for knowledge. Over time, people will come to you for advice on things. Be patient.
Lamentations on Chemistry 
Funny, I try to tell my postdocs this same thing seemingly every day…
Ain’t that the truth! I’m sure the folks at the T2 Lab that went up in 2007 would agree.
If we were mathematicians wouldn’t we say that it is the corollary to the idea (often attributed to Corey) that every reaction can be tweaked to a 100% yield once you start getting product.
You just got to figure out how to read the map!
Yes – I use to annoy my direcors a great deal by attempting to understand what goes on when a reaction goes astray. I had a bromination reaction once where you used NBS. New bottles failed old bottles worked fine. Turns out a reaction that normally isn’t a radical reaction, benifits from a free radical intiator. Add a little Benzoyl peroxde to the reaction using new NBS and it worked out fine if not better than use of a old bottle of NBS. Took a while to figure that one out.
Radiation grafting of solution N-vinyl pyrrolidinone to PMMA surfaces. Rxn mix was thick goo, device saline contact angles were all over the map, $pinning band distillation of the monomer did nothing. Engineers intensively studied the problem to no avail. Chemist Uncle Al had once distilled dimethylaniline.
5-liter flask and its heating mantle, long thick Vigreaux, head, condenser, vac line. Blasted the VP up, over, and down at 0.005 torr full suck into a LN2-cooled flask. The crowd laughed. Rxn mix was thin syrup, device contact angles where uniform 19 degrees vs. spec’d maximum of 25 degrees.
More deserving others tried it and got the SOP result. Nobody noticed that Uncle Al broke to argon and never let the monomer see air from the lab to the polymerization carrier. A charge-transfer complex with oxygen rapidly forms, and that mucks the polymerization. Read JACS, learn stuff. The project was abandoned.
“Running the same reaction a hundred times successfully by carefully following the directions confers proficiency, not expertise.”
BING! Kind of like this (as also mentioned above).
Your reaction is probably talking to you, but if you don’t listen, it’s just noise. If you don’t listen long enough or don’t care what it says, well, then it can be a lot of noise.