Award Date


Degree Type


Degree Name

Doctor of Philosophy in Mechanical Engineering


Mechanical Engineering

Advisor 1

William Culbreth, Committee Chair

First Committee Member

Brendan O’Toole

Second Committee Member

Robert Boehm

Third Committee Member

Woosoon Yim

Fourth Committee Member

Yi-Tung Chen

Graduate Faculty Representative

Laxmi Gewali

Number of Pages



The storage of nuclear waste in underground storage facilities presents numerous engineering challenges and risks. Experimental verification of engineered underground storage is impractical or prohibitively expensive, leaving scientists with few options. A 1995 report by Bowman and Venneri of the Los Alamos National Laboratory generated considerable controversy by hypothesizing that wastes composed of fissionable plutonium leached from underground storage containers could pose a nuclear criticality hazard. They proposed cases where plutonium collected in underground fractures could lead to sustained nuclear fission. In overmoderated cases, they argued that the resulting release of energy from fission could result in steam explosions, or even an underground nuclear explosion (autocriticality). Their hypothesis had severe implications for the feasibility of long-term nuclear waste storage in geologic repositories.

The Bowman and Venneri hypothesis led to the need for a study of conditions that could lead to a critical event in a geologic repository due to releases of uranium or plutonium. Information about the likely consequences of a critical event is also important in repository design. To accomplish this study a numerical simulation code, GEOCRIT, was written to model radionuclide transport from the repository into a fracture below the repository. Once sufficient material has accumulated in the fracture and rock matrix, the neutronics portion of the code is started to simulate heat generation and fluid flow. The thermohydraulics portion of the code calculates heat generation from fission, stream functions, velocity, and pressure of the fluid in the fracture and rock matrix. The transport portion of the code incorporates numerous parameters that can be varied to simulate different radionuclide buildup in the fracture and rock matrix.

Variation of the solubility, diffusion, deposition coefficients in the program yield different accumulations of radionuclides in the fracture and rock matrix. In each case the majority of the deposition occurs in the fracture and adjacent rock matrix. The variation in accumulated radionuclides led to different neutron distributions within the fracture and rock matrix. In each case the largest neutron flux is in the fracture leading to the highest temperature also being within the fracture. The increase in temperature leads to a transient bifurcating flow within the rock matrix.

One potential risk for the geological repository is the buildup of radionuclide in the saturated rock with subsequent drying out of the rock matrix. This case is known as overmoderation, whereby the neutron flux increases as the water dries out. It has been hypothesized that certain configurations of radionuclides may be overmoderated and drying out of the water in the rock may lead to a critical event. Three cases of overmoderation were simulated with the code.

Three different levels of accumulation of radionuclides within the fracture and rock matrix, with subsequent drying out of the crack, lead to an increase in neutron flux. This increase in neutron flux indicates that the accumulated radionuclide in the saturated fracture and rock matrix is overmoderated. In one of the overmoderated cases, the reactor reached a steady state with transient neutron behavior due to water in the surrounding rock changing from liquid to steam and back. One case shows a significant increase in neutron flux.


Actinides; Computer simulations; Criticality; Geologic repositories; Nuclear waste; Overmoderation; Plutonium-239; Water


Mechanical Engineering | Nuclear Engineering