Stored alongside the welcome packages detailed previously was a lecture that David Collum apparently gives to new students each year. Titled “Mother Liquors”, it’s a concentrated collection of good laboratory practices, progress expectations and reaction shortcuts. I was already planning on putting together something similar, and so used his outline as a framework for this post.
Graduate school has been an incredible experience, filled with some of my best–and worst–memories. I was fortunate enough to work under a supportive supervisor who allowed me to find my own path, and through guided trial and error gained confidence in my techniques and my instincts. Along the way I learned what worked for me and what didn’t, building a set of guidelines that helped me be both productive and content. These guidelines are detailed below. I hope you find them useful during your own career.
Progress and Time Management
Set up at least one reaction per day, completing more than ten reactions each week. Complete in this case ranges from full characterization (NMR, MS, MP, etc.) for new, pure compounds to clean-up and lab book annotation for reactions which lead only to decomposition.
Never sit idle, especially when you are just starting research. If you have a surplus of free time while your reactions are running clean dirty glassware, characterize purified compounds, catch up on literature reading or set up another reaction.
At the end of each day make a list of experiments and tasks planned for the next day. At the beginning of each day start working through the list.
Always work on at least two projects at once, but never more than five. Switch between projects as conditions demand (ie. when a needed reagent must be ordered or frustration mounts). If necessary tape a to-do to your desk, to remind you what needs doing on a weekly or monthly basis.
Before starting a project, take the time to thoroughly examine the literature surrounding the planned synthetic scheme or reaction. Look for alternative approaches to each intermediate (if any are published) and similar substrates which have undergone your proposed transformations. Be on the lookout for potential side reactions and difficult to separate byproducts. Two days of planning can eliminate a month of frustration.
When testing a hypothesis try to use experiments that give data for both positive and negative results. Remember: absence of evidence is not evidence of absence.
Planning an Experiment
Stagger your reactions, such that when one is ready to purify another has just begun. When preparing a series of compounds run the reactions and workup in parallel, while purifying compounds sequentially (unless you have an automated purification system) Try not to tweak reactions or purification conditions from sample to sample.
Build from published procedures, modifying conditions as appropriate. Even adapting a procedure designed for a completely unrelated substrate is better than building the synthesis from whole cloth. Order reagents for 2-4 steps at once, to cut down on time wasted waiting for chemical orders to arrive.
Begin with a trial reaction, on a convenient scale (ie. 200 mg). When scaling up never increase beyond a factor of 5, and never change the reaction conditions.Ex. Trial reaction: 200 mg. Second reaction: 1g. Third reaction: 5g.
When working at the “front” of the synthesis, use the smallest appropriate scale. 75 mg of material is far easier to workup and purify than five grams, and will consume your precious intermediates slower. When preparing said intermediates work in bulk, to reduce the amount of non-productive time (repeating your work is not making progress).
Use appropriately sized glassware:
Reactions less than 750 mg: 4 mL screw cap vial, with a “flea” stir bar (8.5 mm x 1.5 mm).
Reactions less than 3 g: 20 mL screw cap vial.
Reactions greater than 3 g: Round bottomed flasks.
Never fill a reaction vessel more than 2/3 full with solvent. Extract reaction mixtures weighing less than 3 g in the reaction vessel or another vial; use separatory funnels for larger quantities of material.
When ordering chemicals consider the cost of your labour, in both money and time. In general, spending an extra $800 is worthwhile if it saves a week spent preparing a literature compound or reagent (~$400 in wages, and equal part in reagents and solvents). At the postdoc level the cost of wages almost doubles. Within reason time is more valuable than money.
Most of the reagents stored in the lab since time immemorial are still of acceptable purity . Unless you have reason to be suspicious don’t routinely check purity before a first reaction (do check if the reaction fails). If a reaction does fail, always double-check your technique and reactants/reagents before heaping scorn on the published procedure.
Running a Reaction
Always monitor your reactions. New reactions should be observed via TLC at a standard set of timepoints (ex. 5min/15 min/30 min/1 hr/4 hr/overnight), regardless of the published reaction time. Run a new TLC after the reaction has been quenched and/or immediately before purification.
Clean glassware, set up flash columns, or otherwise occupy your time while TLC plates are running. Visualize TLCs with a general purpose stain like vanillin/anisaldehyde/PMA/CAM, even if your compound is UV active . Draw scale models of key TLC plates in your notebook, or photograph/scan the plates.
Anoxic and anhydrous conditions range from essential to overkill to counter-productive, depending on the reaction. The first time you run a reaction use your discretion, but if low yields occur switch to strict anhydrous conditions (use a glovebox or glove bag, if necessary).
Working up the Reaction
Deviating from a published workup may increase the yield of a reaction, but for a new student should not be the first impulse. Begin by replicating the author’s procedure, and deviate on the second attempt if the workup fails or gives low yields.
When devising a new procedure, these general rules may be helpful:
First remove your reaction from any sources of heat and allow it to cool to room temperature. If the reaction is chilled quench it before allowing the flask to warm.
Strong bases can be quenched with a small quantity of saturated ammonium chloride in water or acetic acid, while acidic reactions are amenable to sodium bicarbonate or triethylamine. Radical reactions can be quenched by sodium thiosulfate, and biphasic reactions will stop when stirring ceases.
If possible always add a quenching agent. Simply removing the solvent will quench many neutral reactions, but during concentration side reactions may occur, decreasing yields.
If possible, use an extraction to remove otherwise difficult to separate impurities. A basic fractionation between water and ethyl acetate will remove any salts from the crude, but by using acids (NH4Cl, 0.1 M HCl), or bases (NaHCO3, 0.1 M NaOH) amines, carboxylic acids, phenols, and other functional groups can be removed as well. Saturated lithium chloride will scavenge DMF from the organic layer, while a brine wash is essential when ethyl acetate is used as the extracting solvent.
Always keep the size of glassware to a minimum. Phases separate faster in 20 mL vials than in separatory funnels, and give higher yields as well. Weigh your compounds both before and after the workup, to ensure that large quantities are not lost in the water layers. Do not take crude NMR spectra unless you have reason to suspect compound degradation during workup or have not observed your expected product in the TLCs.
Once the crude has been worked up either purify it immediately or store it in the dark at reduced temperatures. Otherwise stable compounds can rapidly degrade when impure.
Carefully consider the impurities in your sample. While flash chromatography is the most general purpose purification method it is not always the most suitable, especially for gram-scale quantities. When working on an early step in a long synthesis take time to explore recrystallization, distillation, tandem extractions and other less common techniques.
Obtain all necessary data the first time you purify a given compound, not the fifth. This limits subsequent compound characterization to a simple 1H NMR, and ensures that during paper/thesis writing all spectral data is accounted for.
At a minimum collect 1H NMR, 13C NMR, COESY, HSQC and LRMS. High resolution mass spec data and/or elemental analysis should be obtained after the compound’s purity and structure are established. Additional tests (optical rotation, melting point, IR, HPLC retention time, etc.) will be at the discretion of your supervisor and committee.
Key peaks in the NMR spectra can be used as a preliminary way to determine if the transformation was successful, for quick turn-around time between reactions. The remainder of the spectra should be assigned weekly, and written up into a publishable format by the end of each month. This limits the amount that has to be written immediately before a paper is published, and speeds up thesis writing.
Pure samples should be stored at reduced temperature, out of the light. Most will likely be stable for months or more at room temperature, but one bad sample can ruin a week or more of work.
Back up all digital data on a regular basis, on a cloud server if possible (encrypt the data if necessary). Lab books should be written in indelible (gel) ink and scanned once complete. Publish your work as soon as possible, or failing that write up all necessary information for new compounds (if the spectra is recorded losing the sample in a far has less impact). Be aware of the damage that a crashed hard drive or lab fire can have, and limit your exposure as much as possible.
Treat graduate school like any other job you care about. Arrive at approximately the same time each day (I prefer to start between 8 AM and 9AM), take short lunches, and only the occasional break. Work hard and efficiently, keeping time spent surfing the web or chatting to a bare minimum.
At the same time, do not allow graduate school to consume your life. Long hours in the lab are not generally productive, and quickly degrade your mental health. Leave the lab at a reasonable hour and go play sports, knit, photograph the scenery, and just generally unwind. Above all, limit the time you work to 40-50 hours each week. Productivity drops off rather sharply after a few weeks of 40+ hours, and the quality of the work suffers even more.
Beware burn-out. Take time off when you need it, including at least a few weeks of vacation each year. Graduate school is a marathon, and solid, steady work will always win out.
 I once came across a half-empty bottle of anisaldehyde from 1969. It was >99% pure by NMR.
 The universal stain will show impurities that may not have delocalized bonds.