Home Energy Nuclear Electricity Climate Change Lighting Control Contacts Links



By Charles Rhodes, P. Eng., Ph.D.

An important aspect of radio isotope container design is the sealing methodology. A container suitable for long term radio isotope storage actually consists of a metal container inside a ceramic container. Both containers need airtight, water tight seals that will readily come apart after 300 years in storage.

The seal of the inner metal container can readily be made using a soft metal corrosion resistant gasket formed from copper or gold between bolted machined stainless steel flanges. Such seals are commonly used today in high vacuum equipment. However, the seal of the outer porcelain container must be more sophisticated.

The porcelain container seal must make an air tight, water tight, radiation resistant seal between porcelain surfaces that will easily come apart after 300 years in storage. Precise dimension control of the porcelain surfaces is more difficult than for stainless steel. Porcelain can readily be ground flat but more complex precision shaping is expensive.

After over 1000 years in storage the sealant can start to become hard but the sealant must not significantly shrink or evaporate away while it remains a grease or liquid. Over its working life, even with a lead shield protecting the sealant, the sealant may be exposed to about 20 X 10^6 rads of gamma radiation.

To realize the required seal with the required working life while exposed to radiation the porcelain container seal actually consists of two seals. The inner seal is a grease seal realized using a Dupont supplied fluorocarbon grease known as Krytox 143AD. The outer seal is a ground porcelain seal realized by grinding flat the upper and lower facing porcelain surfaces at the outer rim of the porcelain assembly. The purpose of the outer seal is to reduce the evaporation rate of the Krytox 143AD by a factor of 100. The Krytox 143AD forms the main inner seal.

If there was no gamma radiation present a suitable sealing material might be tar, bitumen or tree resin. Tar, asphalt and similar long chain hydrocarbon substances are widely used in the modern construction industry for external water sealing of concrete structures. Absent gamma radiation such seals will likely function reliably for several centuries.

Over very long periods of time, in the absence of gamma radiation, certain types of tree resin form amber, which in effect becomes rock.

However, a problem common to all hydrocarbon based sealing materials is that they consist of long hydrocarbon molecular chains that involve the relatively weak hydrogen-carbon chemical bond. This bond is easily broken by absorption of gamma radiation. The result is that the hydrocarbon reforms into a combination of low molecular weight and high molecular weight components. The low molecular weight components such as methane (CH4) rapidly evaporate and the remaining high molecular weight components form solids, which defeats the objective of allowing the seal to come apart after 300 years in storage.

Another possible sealing material which was considered is fine unfired clay. However, the problem with all clays is that for sealing they rely on the presence of water of hydration. As the water of hydration is lost the clay becomes porous. The porisity of fired china is countered by glazing, which is a surface coating of a low melting point glass. However, glazing is not a practical method for sealing a nuclear waste container.

In esence what is required is a radiation resistant grease that will not significantly evaporate or otherwise change in physical properties over 300 years while exposed to 20 X 10^6 rads of gamma radiation, and that over thousands of years will gradually form a solid.

Due to the weak hydrogen carbon bond of hydrocarbon materials it is necessary to instead use a grease formed from silicon-fluorine molecular chains or from carbon-fluorine molecular chains. Silicon-oxygen based sealing materials are readily commercially available but lack the stability of fluorocarbon materials.

The requirement for negligible evaporation over a 300 year period means that the vapor pressure of the sealing material must be very low. A low vapor pressure is achieved by having a combined high molecular weight and a high molecular binding energy.

There are many practical problems with forming high molecular weight fluorosilicon compounds. Hence for now we will focus on high molecular weight fluorocarbon compounds. During the late 1980s Dupont developed a family of fluorocompounds intended for mechanical lubrication in space craft. The trade name of this family of fluorocarbon compounds is Krytox. Krytox compounds have been successfully used for lubrication and sealing purposes in various applications involving substantial exposure to ionizing radiation.

Krytox compounds are available as both liquids and greases. The process for making Krytox produces a mixture with various molecular weights. This mixture is then separated by molecular weight into different products for use in different applications, in a manner analogous to refining of petroleum into its different molecular weight hydrocarbon components. Prior to exposure to radiation the molecular weights of some of the Krytox products are:


The different Krytox products exhibit a wide range of vapor pressures. The vapor pressure is a function of the compound's surface binding energy and its molecular weight. Representative vapor pressures at 20 degrees C are:
143AZ1 X 10^-4
143AA4 X 10^-5
143AY1 X 10^-5
143AB1 X 10^-6
143AX2 X 10^-7
143AC2 X 10^-8
143AD1 X 10^-9


A few of the Krytox products have been tested at large radiation exposures. The data for Krytox 143AB Fluorinated Oil and Krytox 240AB Fluorinated Grease exposed to 20 X 10^6 rad is as follows:


From the above data it is possible to calculate the Krytox evaporation rate for a unirradiated Krytox seal alone if used in a nuclear waste container located in vented storage space at a temperature of 20 degrees C. Note that in unvented storage space the Krytox will come to an equilibrium between the seal material and the surrounding atmosphere so there is no net loss of Krytox from the seal. However, the problem with unvented subterranean storage is that there is no heat removal except by conduction through the surrounding ground, which conduction is very slow and if relied upon will lead to an undesired temperature rise in the storage space.

Consider the circumstances of a Krytox vapor pressure measurement in a sealed vacuum chamber. The vapor comes into equilibrium with the liquid such that the number of krytox molecules leaving the liquid surface per unit time equals the number of krytox molecules captured by the liquid surface per unit time.

We can use properties of a gas to calculate the number of krytox molecules captured by the liquid surface per unit time. At equilibrium in the vacuum chamber that number equals the Krytox evaporation rate. When the Krytox is used as a sealing material in a vented storage space at the same temperature that evaporation rate will be unchanged.

At 0 degrees C and at a pressure of 1 bar one mole of a gas occupies 22.4 litres. At 20 degrees C one mole of a gas occuies:
22.4 lit X (273.16 + 20) / 273.16 = 24.04 lit

One mole of a gas contains 6.023 X 10^23 molecules and hence at 20 degrees C and a pressure of 1 bar has a molecular density of:
(6.023 X 10^23 molecules / 24.04 lit) X (1000 lit / m^3) = 0.2505 X 10^26 molecules / m^3.

Consider use of Krytox 143AD sealant which has a vapor pressure of 10^-9 / 760 mm Hg at 20 degrees C.

At a pressure of 10^-9 mm of Hg the molecular density of a gas is:
(10^-9 mm / 760 mm) X (0.2505 X 10^26 molecules / m^3 = 3.296 X 10^13 molecules / m^3.

Let V be the average molecular velocity.
V^2 = Vx^2 + Vy^2 + Vz^2 = 3 Vx^2
Vx = V / 3^0.5

Let M be the molecular mass

Consider a rectangular container 1 m X 1 m X 1 m

Let N = number of molecules in the container.

The internal pressure force on a container wall is:
(N 2 M Vx) / (2 m / Vx) = N M Vx^2 / 1 m

Pressure on container wall = N M Vx^2 / (1 m X 1 m^2)
= [N / 1 m^3] [M Vx^2]
= [3.296 X 10^13 molecules / m^3] [M Vx^2] = [10^-9 / 760] X 101,000 Pa
M Vx^2 = [10^-9 / 760] X 101,000 Pa / [3.296 X 10^13 molecules / m^3]
= [10^-9 / 760] X 101,000 Newton m / [3.296 X 10^13 molecules]
= [10^-9 / 760] X 101,000 J / [3.296 X 10^13 molecules]
= 10^-22 X 1.01 X 10^5 J/ (760 X 3.296 molecules)
= (40.32 X 10-22 J / molecule) X (1 ev / 1.602 X 10^-19 J) = 25.16 X 10^-3 eV / molecule

For Krytox 143AD the molecular weight M is given by:
M = 7480 X 1.67 X 10^-27 Kg molecule

Vx^2 = M Vx^2 / M
= (40.32 X 10-22 J / molecule) / (7480 X 1.67 X 10^-27 Kg / molecule)
= .00322777 X 10^5 m^2 / s^2
Vx = 17.966 m / s

The number of molecules incident upon a 1 m^2 wall per unit time is:
N / (2 m / Vx)
and the number of molecules incident on a wall per unit area per unit time is:
N / [(2 m / Vx) (1 m^2)] = N Vx / 2 m^3
= molecular capture rate
= evaporation rate in molecules per unit area per unit time

The evaporation rate in mass per unit area per unit time is:
(N / 1 m^3)(M Vx / 2)
= [3.296 X 10^13 molecules / m^3][7480 X 1.67 X 10^-27 Kg / molecule][ 17.966 m /2 s]
= 739702 x 10^-14 kg / m^2 s
= 7.397 X 10^-9 kg / m^2 s X 10^3 g / kg X 1 m^2 / 10^4 cm^2
= 7.397 X 10^-10 g / cm^2 s
= 7.397 X 10^-10 g / cm^2 s X 3600 s / hr X 8766 hr / year
= 233.43 X 10^-4 g / cm^2 year
= 0.022243 g / cm^2 year

This evaporation rate is tolerable for 10 years but it is not tolerable for 1000 years. Hence the Krytox 143AD evaporation rate must be further reduced by at least a factor of 100 by use of an external ground porcelain seal. The open area of the ground porcelain seal must be less than 1% of the Krytox exposed area at the internal seal's Krytox-air interface. For example, if the porcelain is ground flat to within +/- 5 um the width of the exposed Krytox at the inner seal must be at least 1 mm. This exposed Krytox width must be maintained as the Krytox evaporates. Hence the seal must have a depth of at least 1 cm. Hence the Krytox seal has a minimum width and a minimum depth which together with the seal's linear length define the required volume of Krytox in each seal.

If the porcelain grinding tolerance is increased to +/- 10 um the required minimum width of the exposed Krytox increases to 2 mm causing the required volume of Krytox to double. Thus there is a tradeoff between the cost of porcelain grinding and the cost of Dupont Krytox sealant. The seal width is also affected by the porcelain fabrication tolerance. This author suspects that the least expensive route may be to totally avoid porcelain grinding and simply use more Krytox. For example, if the Krytox width is 0.25 inch, the porcelain rims must be flat to within +/- 0.00125 inch, which may be attainable with little or no porcelain grinding.

This web page last updated December 15, 2014.

Home Energy Nuclear Electricity Climate Change Lighting Control Contacts Links