Several innovations, currently pending at CryoSpan, should reduce the liquid nitrogen boiloff rate of patient dewars considerably. Although the main ideas are mine, I wish to thank Mark Connaughton, Greg Fahy, Steve Harris, and Brian Wowk for valuable discussions while the development of these ideas was proceeding.
While patents may have been very helpful to the progress and development of modern technology, it is also clear that they are not optimal and have often done a great deal of harm. I believe that the spontaneous order of a truly free society would not include patenting as it is currently implemented. Therefore, I have decided to proceed without the use of a patent. In making this disclosure, I am declaring that the innovations herein described are my intellectual property, and that by this publication I am not giving away the right to their use. I wish to make this information public so that others may use and/or extend these innovations more easily. If this technology is effective in reducing someone's costs of operation, then I expect and require that some reasonable value--a percentage of that cost reduction, a one time payment, or an appropriate technology exchange with CryoSpan--should be forthcoming from that party.
Although I have known the innovations disclosed herein for over two years, I have not disclosed them earlier because they have not yet been fully implemented on CryoSpan's patient storage system and, therefore, their full potential has not yet been demonstrated. However, the basic idea and part of the potential have been proven for some time, and it has become clear to me recently that for various reasons (mainly the extremely slow rate of CryoSpan's cryonics patient expansion) it will still be some time before the dewar system will be completed to a stage where the full potential of the boiloff reduction method will be known.
LEFT: Cross-section of a dewar in its underground silo at CryoSpan, showing current and future innovations to minimize boiloff of liquid nitrogen.
TOP: Detail from the drawing below, showing vapor (and heat) flow.
After procuring CryoSpan's first whole-body dewar (based on the Alcor "bigfoot" design) at the beginning of 1994, I proceeded to analyze the heat flow into the dewar by means of theoretical calculations. I also studied previous analyses by others and placed thermocouples at various points about the dewar. I concluded that the heat flow into such a high-vacuum, superinsulation-wrapped dewar is approximately as follows:
Therefore, it was clear that while using high vacuum and superinsulation to reduce heat flow through the largest part of the surface area is very efficient, and even the thick foam top does a reasonable insulating job (partly because it interfaces only with warmer vapor), the major source of heat inflow and subsequent boiloff is the continuous metal path between the liquid nitrogen and the room air effected by the inner stainless steel dewar wall. (This is one of the reasons why fiberglass and perlite/foam cryostats are used at the Cryonics Institute.)
After many thoughts of possible ways to reduce the heat flow through this path, I decided that the only feasible method was to lengthen the path from LN2 to room air and to keep the temperature gradient along this path as small and as uniform as possible. This is being implemented in the following ways:
1) CryoSpan's new dewar is 10 inches taller than the original Alcor design. This additional height was used because the sheets of stainless steel from which the dewar was built allowed that amount if they were used uncut.
2) The thickness of foam in the dewar neck will be increased by ten inches. Both theory and measurements showed that it is very important to have the escaping nitrogen gas generated by the boiloff "hug" the dewar walls and be warmed as much as possible by the inflowing heat. By doing so the heat removed to warm the gas is carried away back to the outside by the escaping gas. This effect is of major importance to the overall efficiency of the dewar. The ideal arrangement, therefore, would be some kind of neck plug which would expand downward to fill all the space above the surface of liquid nitrogen as it falls due to boiloff, and yet would be solid in the center, only allowing the gas to escape up the inner wall. Although I have some ideas, we have not yet developed a practical design for such a neck plug.
Even if the total thickness of foam in the neck remains constant, it is important to split the convection volume between plug and liquid nitrogen surface into several small parts. This can be accomplished by placing several foam pieces (solid circles tight to the inner dewar wall which we call "baffles") below the fixed neck plug, so that each will drop by only a certain amount as the liquid nitrogen surface falls.
In addition, we have found that it is particularly beneficial to have a baffle riding on the surface of the liquid nitrogen. This last appears to greatly reduce the convection cooling of the walls, and to keep all escaping nitrogen vapor close to the walls.
We have been able to effect this last innovation to some extent with our original dewar, even though it was the standard Alcor height, because the containers which we inherited with our three whole-body patients from Trans Time were shorter than our standard patient pods. In addition, they did not have the vertical "ears" of the Alcor designed pods which would greatly hinder the effect of a floating baffle when the level of liquid nitrogen is low. With a 12-inch thick foam lid, a 3.5-inch floating foam baffle, and perhaps some natural reduction in the ambient silo temperature, we have been getting a boiloff rate of 11.5 liters per day from a dewar which boiled off 14 liters per day with the standard 14-inch foam top. Unfortunately, it will not be possible to say for some time what boiloff advantage has been obtained with our new taller dewar because it is currently being used as a reservoir with only neuros in the bottom. Consequently, its level varies widely.
3) The escaping nitrogen gas flow will be directed down the outer wall of the dewar. This is possibly my most innovative idea. Others had tried thicker tops, evacuated tops, or even "mushroom" type caps partly extending down the outer wall of the dewar, all with little success. However, the thickness of the dewar's outer wall at 3/16" or 187.5 mils is over 5 times that of the inner wall (35 mils). Therefore, the heat conduction of a given length of outer wall is over 5 times that of the inner wall. Thus, to gain much by making the temperature gradient extend down the outer wall by insulating it from the external ambient heat source, the insulation must be extended down the full length of the outer walls. (Actually, I found out, after I made this discovery, that the idea--and the actual implementation--of inverting another, slightly larger, dewar upside down over the top of a given dewar had been tried before.) Therefore, my final proposed design (already partly implemented) of the containment of the dewar is as follows:
The outer wall is surrounded by corrugated fiberglass roofing material (Home Depot) with corrugation running vertically to allow space next to the outer wall down which the escaping nitrogen gas can flow. This is covered by 5 inches of half-inch-thick foam layers, bent to fit the curvature, with interleaved joints and sealed with plastic wrapping. A 6-inch foam external top is planned which will be sealed down tight onto the top of the side foam and corrugated fiberglass, but not against the top lip of the dewar. Thus, the escaping nitrogen gas, which warms as it escapes and carries much of the inward flowing heat back to the outside, after creeping up the sides of the inner wall will be directed down the outer wall of the dewar to its bottom. Finally, the gas will flow up the outside of the foam wrapping the dewar, next to the 5-inch-thick concrete silo walls and earth insulation.
Once this arrangement is complete (we still don't have our sealed tops because our automated filling system is not yet in place), I expect that there should be a considerable temperature gradient down the silo wall and up the dewar outer wall. The effect of this should be that the ambient temperature at the very top of the dewar inside wall (which is normally at room temperature) will be well below freezing. Therefore, the gradient up the inner wall and the consequent boiloff rate should also be reduced considerably.
An additional advantage of this design will be that the entire silo will be filled with cold, slowly exhausting nitrogen gas. No water ice should be able to form anywhere around the lid, and various types of data media can be stored safely around the dewar at the bottom of the silo.
Finally, the 5 inches of foam surrounding the dewar, with its outer surface only 1 inch from the silo wall, should also provide a substantial cushioning effect from any buffeting which the dewar might incur during an earthquake.
My calculations and my expectations are that when the final system is complete, the boiloff rate of a dewar which was originally 14 liters per day will be reduced to, at most, 9 liters per day.