Our lab, like many others, relies on freezers in our daily work activities. Freezers are often taken for granted but are one of the most useful devices in any biological lab. They give us the capacity to extend the life of expensive reagents, store materials for later use, and greatly prolong the storage of biological samples. Within our various freezer stocks, we even have samples from the 1980s that were used in the PhD of our advisor when organisms like Sulfolobus were starting to be explored and utilized for their genetic potential. At any given time, we have a mixture of both -20°C and -80°C freezers in use, and while we use them every day, we rarely think about the large amount of energy it takes to keep things cold.
Keeping things cold is no easy task, especially at freezing temperatures. It typically requires the use of potentially toxic chemicals combined with compressors to force a heat exchange in a direction it would not normally take. These compressors require energy to do this, and the colder the freezer, the more energy it takes. This led our group to ask the question, “How cold is cold enough?”. In today’s world, it seems that most biological labs utilize a -80°C freezer to store particularly important or sensitive samples such as cells collected from volcanoes or the bottom of the ocean. We have not only one, but three -80°C freezers that we are constantly running to store reagents and biological samples that span decades of scientific work. We are well aware that these samples are important, but we began to question if -80°C was really necessary.
After some research into the subject, we found that we were not the only ones thinking about this issue. As it turns out, many universities in the U.S. are already raising the temperatures of their -80°C freezers [1,2]. In some cases there are even a competitions to see how high it was possible to go without compromising the biological samples [3,4]. The entire concept has also given way to long term studies about the effects that freezer temperatures have on different types of samples [5,6]. Some of these studies, however, are still in progress. From an energy savings perspective it also seemed to make a lot of sense. Raising the temperature by just 10°C could lead to savings of approximately 33% (19 kW-hr/day to 12.6 kW-hr/day, based on company specifications) and could even extend the life of the previous compressors doing all of the work. After discovering this, we thought “Why not? Why can’t we also raise our -80°C freezers to -70°C?”.
Of course, this process is not nearly as simple as it sounds. Our -80°C freezers contain a large variety of samples that belong to researchers spanning at least 4-5 different working groups. Simply changing the temperature without coordinating with the people involved and going through the proper channels would likely be a recipe for disaster. Nevertheless, we set about initiating the process by first communicating our intentions with all of the working groups involved. At the same time, we began to assemble a list of everything that was present in our -80°C freezers. We reasoned that if we know what is there, it would be much easier to make an informed decision about the risk of raising the temperature.
After assembling our list (containing everything from RNA, DNA, soil samples, enzymes, and rumen fluid) we took a closer look and decided that nothing on this list would be any worse for wear at -70°C. Naturally, we arrived at this decision with the consensus of the group leaders and the keepers of the individual samples. Once we were in agreement, and had the final approval from the group leaders, we were finally able to raise the temperature to -70°C.
I am happy to say that so far (about 5 months) there have been no ill effects of this decision. And, presumably, we are saving up to 6.4 kWh-day based on the manufacturer’s analysis. This change is further complemented by good freezer etiquette. In our lab, we try to keep an up to date and organized list of all the contents of our -70°C freezers. This way, people can minimize the amount of time searching for what they need (which results in an increase in temperature). This is also helped by organizing all of our samples with the use of freezer racks (which we would highly recommend to anyone using a low-temperature freezer). This allows content to be easily removed and minimizes the time that a freezer door is open to mere seconds.
One of the final concerns that we, and others, have had is what happens if the freezer fails? If it is at a higher temperature, won’t it warm faster and ruin the samples? As it turns out, we unintentionally tested this question due to a universal power outage within our building. To make a long story short, everything was fine. In today’s world, most ultra freezers come equipped with an internet connection that will automatically alert the owner of the freezer if there is an issue (of course, if the power is out this won’t help, but labs are typically alerted of power outages by building management). This alert system, either from the freezer or from facility management, should give ample time to rescue anything from the freezer if necessary. In our case, the freezer remained closed during the outage (also important) and the temperature was only increased to -55°C before the issue was corrected. In an even more impressive scenario from the University of California-Santa Cruz , an ultra freezer was even able to maintain the integrity of samples although the building was on fire. This undoubtedly supports the idea the ultra freezers are well equipped to keep things cold with their insulation alone if they must.
Altogether, we have so far had a positive experience in adjusting the temperature of our freezers. At the end of the day, we concluded that a freezer can be just as effective without pushing it to the furthest possible limits.
- Wandro S, Carmody L, Gallagher T, LiPuma JJ, Whiteson K. Making It Last: Storage Time and Temperature Have Differential Impacts on Metabolite Profiles of Airway Samples from Cystic Fibrosis Patients. mSystems. 2017;2(6):e00100-17. Published 2017 Nov 28. doi:10.1128/mSystems.00100-17