A glove box that held sub-1 ppm oxygen for years can read 200 ppm within a day of arriving at a new site. Moving a box disturbs door gaskets, glove ports, flange fittings, and the purge manifold, and the purifier usually sits idle long enough to take on moisture. A glove box relocation leak test is therefore the first job on site, not the last. Leak integrity has to be proven before you spend gas and time on purification.
This procedure covers the mechanical checks, the leak test itself, and the purification startup sequence for a standard nitrogen or argon box with a copper-catalyst or molecular sieve purifier.
Pre-Checks and the Glove Box Relocation Leak Test
Start with the mechanical side. Inspect the main door O-ring, antechamber door seals, glove port gaskets, and any KF or CF flange gaskets; replace anything cut, flattened, or visibly twisted, and never reuse a deformed O-ring. Level the frame and confirm it is not twisted, because a warped frame keeps the main door from seating evenly no matter how hard you clamp it.
Check the utility side next. Confirm that feedthroughs, sensor ports, and vacuum lines are tight, that the pump oil is clean and at level, and that the inert gas regulator holds a stable 2-5 mbar positive pressure on the box. Leave the box under slight positive pressure overnight before testing so gross leaks show up early.
For the pressure test, use the rate-of-rise method on the box volume. Evacuate through the pump to below 1 mbar, close the isolation valve, and log pressure every five minutes for at least 30 minutes while recording ambient temperature. A rise of roughly 0.05 mbar per minute or less is a reasonable pass line for a 100 L box, but the OEM specification always wins. Where the rise is too high to localise a leak, follow up with a helium mass spectrometer on suspect joints, starting with the glove ports and the antechamber inner door.
Gloves deserve their own test. Slightly pressurise the box and brush soap solution on the outside of each glove, or pressurise a single glove and listen for a hiss. Pinholes picked up in transit are the most common cause of a failed startup.
Purification Startup: Regeneration, Purge, and Circulation
Do not start circulation before the glove box relocation leak test passes and the purifier has been regenerated. A copper catalyst bed is normally regenerated at 200-250 C under a 5-10% hydrogen in nitrogen or argon flow for eight to twelve hours, while a molecular sieve column is regenerated under vacuum or dry gas at 250-300 C. Follow the manufacturer’s ramp and never open the regenerated bed to the box while it is hot.
Purge the gas lines and the antechamber while the purifier cools. Use three fill-and-evacuate cycles, taking the antechamber to at least 0.5 bar of working gas and back to rough vacuum each time, which removes the air and moisture that entered during the move. Then fill the antechamber, open the inner door, and let the box equalise.
Cool the purifier under continuous gas flow, then start the circulation blower. Run the box empty for 12-24 hours with the analysers live, and log O2 and H2O rather than watching them continuously; moisture held in the walls, gloves, and window keeps readings high for the first several hours. A heated antechamber or an internal bake-out at 60-80 C shortens that tail considerably.
Calibrate the analysers before you trust any number. Span the O2 sensor with certified gas, follow the maker’s procedure for the H2O sensor, then wipe the gloves and interior with lint-free, solvent-free wipes.
Acceptance Criteria and Common Faults
A standard box should settle below 1 ppm O2 and H2O, and a research-grade system below 0.1 ppm. With all valves closed and the box idle, readings should rise by less than about 0.05 vol%/h, which is your combined leak-and-outgassing benchmark.
Read the trend, not only the number. A box that plateaus at 20-50 ppm usually has a small leak or a partly regenerated purifier, while a value that climbs steadily points to a real leak or heavy outgassing from a new component. Re-check the antechamber inner door gasket, the glove port clamps, and the analyser calibration in that order.
If the readings will not fall, do not compensate by running the purifier hotter or longer; find the leak first. Finish the glove box relocation leak test with a documented rate-of-rise record, and the purification startup will reach specification on the first pass.
