A power outage at a biotech lab in Watertown last August started at 2:47 in the afternoon. The PI was at a conference in San Francisco. The grad student who found out did the right thing. She didn't open the freezer. She also called her supplier, who told her "you've got about four hours, probably." Four hours later, the freezer was at -62°C. The cell line was fine. The reference standards in the door were not.
That's the conversation. Not whether to keep dry ice on hand for outages (you should), but how much, in what form, and for what timeline. If you run a lab anywhere in the Northeast, the question is going to come up. Probably this summer.
The Physics of the First Two Hours
A -80°C ULT freezer, in the moment the compressor stops, is a well-insulated box of cold air and cold samples. The samples are the thermal mass. The freezer holds because they hold.
Published numbers for a closed, undisturbed ULT vary by manufacturer and load. The working range:
Empty, door closed. Roughly 4 to 5°C of warming per hour at the start. It accelerates to 6 to 8°C per hour as the unit climbs past -60°C.
Fully racked, door closed. Published Stirling and Thermo data put new units at about 5 hours from -80°C to -60°C, roughly 4°C per hour averaged. The first hour is closer to 2.
One door opening. Anywhere from 5 to 15°C of immediate rise in the affected zone. Recovered slowly.
The FDA and USP ultra-low envelope for most biologics is at or below -70°C. That's the line you're protecting. Not -80°C exact. The 10°C of cushion is your runway.
For a typical 23 cubic foot lab ULT loaded with racks and samples, that runway is 1.5 to 3 hours if nobody opens the door. That's the entire margin for any decision you're going to make.
The Fork You Choose in the First Hour
Once the power's out, there are two strategies. They don't combine well. Pick one in the first thirty minutes.
Fork 1: ride it out in the freezer. Door stays closed. You add dry ice on top of the load (not directly on samples; see below) to slow the warming curve. Reasonable for outages you expect to clear inside 12 hours.
Fork 2: transfer to dry ice coolers. You move racks into pre-staged insulated transport coolers packed with dry ice. You hold them there until power returns or you can get them to a backup freezer. The right call for outages over 12 hours, or anything where utility ETA is unknown.
The mistake we see most often. Teams try Fork 1 for six hours, realize it's not enough, then start transferring at hour seven with the freezer already at -55°C. The samples are already compromised. If you're going to transfer, decide in the first hour.
Fork 1: Ride It Out
Standard short-outage protocol at most academic and biotech ULTs. Layout matters.
Remove the bottom rack. Place dry ice blocks or slabs on the floor of the freezer. Not pellets; they sublimate too fast and pack too dense. Most ULT cabinets fit 25 to 40 lb (11 to 18 kg) of slab dry ice in the bottom zone without obstructing rack rails. Close the door. Don't open it again until you have to.
The energy math is simple. Latent heat of sublimation of CO₂ is roughly 571 kJ/kg. To absorb the heat ingress through a typical 23 ft³ ULT cabinet at the temperature gradient you're trying to maintain, you're looking at a heat load somewhere in the 80 to 150 watt range. One kilogram of dry ice absorbs about 0.16 kWh of energy as it sublimes. Run the arithmetic. Eleven kg (about 24 lb) of slab dry ice buys you roughly 12 to 16 hours of additional hold time below -70°C in a full unit. 8 to 12 hours in an empty one.
A few practical notes the textbooks don't emphasize.
Never seal the door. The CO₂ has to vent. ULT door gaskets aren't airtight. If you've taped the door, untape it. A pressurized cabinet is worse than a slightly leaky one.
Slab beats pellet for this job. Pellets give up about 12 to 18 percent of their mass per hour in a -80°C cabinet because of surface area. Slab gives up 4 to 6 percent. Same total cooling, longer runway.
Don't put dry ice on or against vials. Direct contact thermal-shocks glass and stresses cryo tubes. Use the bottom shelf or a cardboard buffer layer.
One open is one open. Every door opening is heat ingress you don't get back. Use the monitoring system.
Fork 2: Move to Coolers
For anything longer than 12 hours, or when the utility company won't give you an ETA, transfer is the right call. The math is different. The dry ice load is higher.
A standard 95-quart EPS or VIP transport cooler holds about 2 to 3 ULT racks. To keep that payload below -70°C for 24 hours at room ambient:
EPS cooler, walls 2 inches thick. 40 to 50 lb (18 to 23 kg) of slab dry ice per 24 hours.
VIP-paneled cooler. 18 to 25 lb (8 to 11 kg) of slab dry ice per 24 hours.
Pellet in the same containers. Add 30 to 50 percent.
For a fully loaded 23 ft³ ULT freezer (10 racks, 400 boxes), plan on 4 EPS coolers or 3 VIP coolers, and a dry ice charge of 160 to 200 lb (72 to 91 kg) for the first 24 hours. Then half that again for each additional day of holding.
That's the number to write on the inside of your freezer door right now. Two hundred pounds of slab dry ice for one day of full-freezer holding. Argue with it if you want. Write it down.
What the Prepared Labs Keep on the Shelf
The labs that handle outages well in the Northeast do roughly the same five things.
They keep a minimum 50 lb standing reserve of slab dry ice on site during the May-through-October storm season. Enough for a Fork 1 ride-out of any reasonable summer outage.
They have a transfer plan written down, not in someone's head. Who calls the supplier. Who pulls the coolers off the loading dock. Who decides Fork 1 vs Fork 2.
They have a supplier on speed dial who answers the phone after hours. Delivers inside four hours during business hours. Inside eight hours overnight. A regional supplier matters. National couriers don't run emergency dry ice.
They have the coolers staged. Right number of pre-cut cardboard separators. Gloves. A CO₂ monitor for the transfer room.
They have wireless temperature monitoring with text alerts that fire at -72°C, not -65°C. The early-warning threshold gives you the full decision window.
Common Mistakes (We See All of These)
Confusing chest freezer rules with ULT freezer rules. The 2.5 lb per cubic foot rule of thumb comes from -20°C household freezers holding food. A -80°C ULT freezer with biologics inside operates in a different regime. The dry ice load is roughly double per cubic foot per day.
Using pellet because that's what's in the standing order. Pellet is right for shipping. Slab is right for emergency hold inside a stopped freezer. If your standing order is pellet only, get the supplier to add a 50 lb slab reserve to the next delivery.
Opening the door to "check." Don't. Read the external display, read the external probe, read the wireless logger. Do not open the door. Every open costs you 30 to 90 minutes of runway.
Treating the building generator as protection. Most academic and biotech buildings in Greater Boston, Cambridge, and Worcester have generators rated for emergency lighting and life-safety loads. Not freezer rooms. Confirm with facilities. Don't assume.
Underestimating duration. National Grid and Eversource publish post-storm restoration data for the New England territory. The long-tail of multi-day restorations in summer thunderstorm season and winter nor'easters is real. If a town goes dark at 4 PM on a Tuesday in a thunderstorm, plan for Wednesday morning at the earliest.
The Northeast Outage Calendar
We've watched this rhythm for several years. The shape of the year, for lab planning purposes.
May through September. Thunderstorm season. Most outages are 2 to 6 hours. The Cambridge/Watertown/Waltham transformer-fire cluster runs longer. Plan Fork 1 with a Fork 2 standby.
October through November. Nor'easter shoulder. Outages start trending longer. Pre-position extra dry ice.
December through February. Ice storms and wind events. Worst-case outage durations of the year. Multi-day events in CT, RI, and Worcester County aren't unusual. Fork 2 with deep reserves.
March through April. Spring storms. Fewer events but harder to predict. Standing reserve of 50 lb still applies.
A regional supplier can flex an emergency delivery during these periods, but the lead time stretches. The closer to a storm the call comes in, the longer the queue. Standing orders with a guaranteed emergency response clause are the way to handle this. So is keeping the 50 lb reserve.
Questions We Get Asked
Can I put dry ice directly into a powered ULT freezer to "help" it cool? No. Dry ice in a powered cabinet stresses the compressor. It creates a CO₂ buildup the unit isn't designed to manage. The dry ice play is for when the compressor is off, not as supplemental cooling during normal operation.
Will my samples be okay if the freezer reaches -65°C briefly? Depends on the samples. For most clinical research material, mRNA, DNA, protein reference standards, and bulk drug substance, a short excursion to -65°C is recoverable but should be logged. For viable cells in DMSO, no. Those belong in liquid nitrogen. Dry ice isn't the right tool. Always check your sample-specific stability data.
How much dry ice for a 24-hour outage in a typical Cambridge or Boston academic lab? For a single 23 ft³ ULT freezer ridden out closed: 25 to 40 lb of slab dry ice, loaded in the bottom at the start. For a multi-freezer lab transferring to coolers: roughly 200 lb of slab per full freezer per 24 hours.
What form of dry ice is right for outage use? Slab. 4.5 kg blocks or equivalent. Slabs sublimate slower per unit mass than pellets. That's exactly what you want for hold duration. Order it that way. Store it that way.
Do you deliver emergency dry ice in the Northeast? Yes. Cryo Life Solutions runs same-day and after-hours emergency delivery across MA, NH, RI, CT, NY, NJ, and PA for biotech and lab clients on standing orders. After-hours response for active outage events.
How do I set up a standing reserve without burning through it? Most labs we work with rotate the reserve into normal use on a 30-day cycle. The dry ice never sits long enough to fully sublimate. The supplier replenishes on the same delivery as the standing pellet order. Costs less than people expect.
The Summary, on One Line
Two hundred pounds of slab dry ice covers one day of full-ULT holding. Fifty pounds on the shelf covers most of a 12-hour ride-out. Decide Fork 1 versus Fork 2 in the first hour. Don't open the door.
Cryo Life Solutions supplies pelletized and slab dry ice for biotech, pharma, and academic research labs across the Northeast US. We operate under documented GMP-aligned procedures and are working toward full GMP certification. Standing orders, emergency response, and outage planning support are available across MA, NH, RI, CT, NY, NJ, and PA. Call 603-802-6650 or email info@cryolifesolutions.com.
