Air Free Reaction Setup

Materials

• Schlenk line (vacuum + inert gas manifold) or a N₂/Ar tank with bubbler + balloon
• Oven dried Schlenk flask/tube (or septum capped round bottom), stir bar
• Rubber septa, needles (various gauges), gas tight syringes
• Liquid N₂ (or dry ice/acetone) bath for freeze pump thaw
• Glovebox, if your catalyst is extremely O₂ sensitive (e.g. some Cu(I) complexes)

Step by step

1. Dry the glassware. Oven dry the flask overnight at 120 to 150 °C, or flame/heat gun dry under vacuum on the Schlenk line, then cool under inert gas. Any residual water will hydrolyze catalysts or terminate chains.
2. Purge the empty vessel. Cap with a septum, then do 3 evacuate/backfill cycles on the Schlenk line (vacuum until it holds, then backfill with N₂ or Ar). This renders the flask's headspace inert before anything goes in.
3. Add solids under counter flow. With inert gas flowing out through the septum (needle vent, or briefly removing the septum under positive pressure), quickly add air stable solids: initiator, ligand, monomer if solid, then reseal immediately. Anything pyrophoric or very O₂ sensitive (e.g. neat CuBr) goes in inside a glovebox instead.
4. Add solvent and monomer via syringe.
5. Degas: freeze pump thaw (FPT), the gold standard. See the dedicated Freeze Pump Thaw Method section below for the full procedure. Repeat 3 to 4 cycles total; backfill with inert gas after the last one.

   Simpler alternative: sparge with N₂/Ar through a long needle to the bottom of the flask (with a second needle as vent) for 20 to 30+ min. Less rigorous than FPT and can strip volatile monomer, but adequate for RAFT/FRP or ARGET type ATRP.

6. Add the catalyst last, under inert atmosphere. For ATRP, the Cu(I)/ligand complex is usually the most O₂ sensitive component. Prepare it as a degassed stock solution and add it by syringe (or cannula) right before heating, since this is effectively what starts the clock. Same idea for ROMP: add the Grubbs/Hoveyda catalyst as a stock solution last.
7. Heat with positive pressure maintained. Submerge in a preheated oil bath or heating block; keep a slight positive N₂/Ar flow (bubbler or balloon) throughout so air can't diffuse back in through the septum.
8. Sampling (if tracking conversion). Withdraw aliquots by syringe/needle quickly, ideally with gas still flowing out of a second needle to prevent air being drawn in. Use an internal standard added to the bulk reaction mixture up front so the conversion you calculate from each aliquot is trustworthy regardless of aliquot size.
9. Quench. Open to air (this terminates ATRP by oxidizing Cu(I)→Cu(II)) or add a radical inhibitor/scavenger, then cool and work up (precipitation, dialysis, etc.).

Freeze Pump Thaw Method

Freeze pump thaw (FPT) removes dissolved gas (mostly O₂ and N₂) from a liquid by alternately freezing it solid, pulling vacuum on the headspace, and letting it thaw. Each cycle releases gas that was trapped in the liquid. It's the most thorough degassing method and the standard for ATRP.

What you need

• Schlenk flask/tube with a sidearm stopcock (or a septum plus a needle to a Schlenk line), rated for vacuum
• Vacuum source with a cold trap in line (liquid N₂ or dry ice/acetone) to protect the pump from condensing solvent/monomer vapor
• Liquid N₂ in a wide mouth Dewar, deep enough to submerge the flask's contents
• Insulated gloves, cryogenic rated face/eye protection

Procedure (one cycle)

1. Freeze. Lower the flask into the liquid N₂ bath (stopcock still open to the inert gas line, not to vacuum yet) and swirl briefly until the contents are fully frozen solid. A partially frozen liquid can still splash or bump when vacuum is applied.
2. Evacuate. With the contents frozen (so nothing can boil over into the vacuum line), open the flask to vacuum. Evacuate until the vacuum gauge reads close to the line's static base pressure and holds steady. This pulls the headspace gas out through the solid.
3. Isolate. Close the stopcock to seal the flask under vacuum, then remove it from the liquid N₂ bath.
4. Thaw. Let the flask warm to room temperature undisturbed (a room temperature water bath speeds this up). As it melts, dissolved gas that migrated to the surface during freezing bubbles out into the now evacuated headspace. Let it fully liquefy and sit a minute so the gas has time to escape.
5. Repeat. Refreeze and repeat the evacuate → isolate → thaw sequence 3 to 4 times total. Each cycle should show visibly less bubbling on thaw than the last; once you barely see bubbles on the final thaw, degassing is essentially complete.
6. Backfill. On the last cycle, instead of opening to vacuum, backfill the flask with inert gas (N₂/Ar) while still cold, then let it warm under a positive inert atmosphere.

Technique specific notes

• ATRP: most O₂ sensitive. Do the full 3 to 4x FPT. ARGET/ICAR/SARA variants continuously regenerate Cu(I) and tolerate trace O₂ better, so sparging alone is sometimes acceptable there. See oxygen tolerant methods below for how far that can be pushed.
• RAFT: no O₂ sensitive metal, but O₂ scavenges initial radicals and causes an induction period. FPT (3x) or thorough sparging both work; less rigor needed than ATRP. PET-RAFT with an added tertiary amine can skip degassing entirely.
• ROMP: Grubbs type catalysts are fairly bench stable as solids but degrade in solution with O₂/moisture. Degas the monomer solution and add catalyst last as a fresh stock solution.
• FRP: most O₂ tolerant, but degassing still shortens the induction period and improves reproducibility.

When you can skip degassing: oxygen tolerant methods

Everything above assumes you are removing oxygen before you start. A newer body of work instead makes the chemistry consume oxygen while it runs, to the point that some systems polymerize in an open vial on the bench. This is worth knowing about because degassing is usually the slowest, most error prone part of a controlled polymerization, and because the reason each method works tells you exactly how far you can push it.

Why ATRP is more oxygen tolerant than it looks

Oxygen attacks an ATRP two ways: it quenches propagating radicals, and it oxidizes the Cu(I) activator to Cu(II). Both reactions are essentially diffusion controlled, so which one dominates comes down to concentration, and at the ATRP equilibrium there is far more Cu(I) around than there are radicals. Oxygen therefore hits the catalyst first. That is genuinely useful: the catalyst acts as a sacrificial oxygen scavenger and your chains survive. In classical ATRP the damage is permanent, because nothing puts the Cu(II) back. In the regeneration variants (ARGET, ICAR, SARA, eATRP, photoATRP, mechanoATRP) a reducing agent or an external stimulus continuously converts Cu(II) back to Cu(I), so the catalyst keeps scavenging and the polymerization survives a real oxygen load.

The limit is worth stating plainly, because it is where people get burned: this buys you tolerance in a sealed vessel with a finite amount of oxygen, usually at the cost of an inhibition period while the catalyst clears it. It does not generally buy you an open flask. Regeneration has to outrun oxygen diffusing in through the surface, and typically it cannot. Cu(0) powder systems show no conversion at all in an open vessel, and copper wire systems have been run open for 48 hours with no polymer. Minimizing headspace is the cheap practical win here.

Enzymatic deoxygenation: genuinely open flask

To actually work in an open vessel you need something that destroys oxygen faster than it arrives. Glucose oxidase (GOx) with glucose does exactly that, and it turns ATRP into an open vessel reaction. There is a catch worth understanding rather than memorizing: GOx makes hydrogen peroxide, and H₂O₂ reacts with Cu(I) in a Fenton type reaction to give hydroxyl radicals, which start new chains. The symptom is a molecular weight far below target (roughly five times low when reaching for DP > 200) rather than an obvious failure. Adding sodium pyruvate fixes it by scavenging the peroxide to carbon dioxide, acetate, and water, restoring the expected M_n and dispersity.

Combined with photoATRP, the enzymatic system is not just tolerant but faster: 93% conversion of OEOMA in two hours versus more than five hours for the same reaction without the enzyme, while keeping full light on/off temporal control. It is mild enough that the enzyme's own secondary structure survives, which is why this route is used for grafting from proteins and DNA. A related discovery is that sodium pyruvate under UV, with no enzyme at all, gives an oxygen proof photoATRP that works in both water and organic solvent, avoiding the problem of separating enzyme from product.

Oxygen tolerant RAFT

PET-RAFT has its own route: run the photocatalyst through a reductive quenching cycle by adding a tertiary amine. Eosin Y with triethylamine polymerizes fully open to air under ordinary visible LEDs, blue (~483 nm) or green (~532 nm), at intensities of a few mW/cm². This is what makes RAFT based 3D printing possible on a standard digital light processing printer with no inert atmosphere, and because the trithiocarbonate survives inside the printed network, the object stays "living": a printed part can be reactivated later to insert more monomer into its outer layers.

Solvent choice can also do the work. A deep eutectic solvent made from tetrabutylammonium chloride and ethylene glycol supports open to air photoiniferter and PET-RAFT polymerization with no additives and no degassing at all, while increasing rate: even dithiobenzoate, a poor photoiniferter, ran about 4.5 times faster than in DMSO, and the medium improved chain end survival. Eosin Y works in it too, to the point that the polymerization runs in natural sunlight. It is a nonvolatile, inexpensive, green medium, which makes it an appealing option when your monomer tolerates it.

Choosing an approach

• Still degas properly for classical ATRP, for anything where you need reproducible kinetics from the first minute, or when you are chasing high molecular weight, where an inhibition period and a few dead chains matter most.
• Sealed vessel, minimal headspace, regeneration variant (ARGET/ICAR/SARA/photoATRP) is the low effort middle ground: accept a short inhibition period instead of a Schlenk line.
• Enzymatic (GOx/glucose/sodium pyruvate) when you truly need an open vessel or biological compatibility. Do not omit the pyruvate.
• Amine assisted PET-RAFT, or a deep eutectic solvent for open-to-air RAFT, photo-patterning, and 3D printing.

Sources: Szczepaniak, Fu, Jafari, Kapil, and Matyjaszewski, Acc. Chem. Res. 2021, 54, 1779 to 1790 (oxygen tolerance in ATRP); Bagheri et al., ACS Appl. Polym. Mater. 2020, 2, 782 to 790 (open-to-air eosin Y / triethylamine PET-RAFT 3D printing); Li and Yu, Macromolecules 2021, 54, 9825 to 9836 (open-to-air RAFT in deep eutectic solvents). See also the PET-RAFT mechanism and the RAFT calculator.