Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

Legacy Department

Mechanical Engineering


Grujicic, Mica

Committee Member

Joseph , Paul F

Committee Member

Singh , Rajendra

Committee Member

Li , Gang


Development of new transparent armor systems is essential for the protection of the current and future US armed forces, especially in light of the recent military operations The Operation Iraqi Freedom in Iraq and The Operation Enduring Freedom in Afghanistan. These conflicts have introduced a new military theater without a well-defined battle front and new types of threats (e.g. improvised explosive devices, IEDs). Development and modeling of new transparent armor systems for use in numerous applications from vehicle windows to face shields is a current area of thrust aimed at addressing the shortcomings of existing systems in order to better protect US soldiers and align with the military's goal of becoming more mobile, deployable, and sustainable.
This dissertation is focused predominately on the computational modeling of transparent armor materials and structures. Glass remains the dominant constituent in many modern transparent armor systems for a number of performance and manufacturing related reasons and thus is the material of focus in the present work. The present work is concerned with the development and further enhancement of a continuum-level, physically-based, high strain-rate, large-strain, high-pressure mechanical material model for soda-lime (and borosilicate) glass. The model is being developed in attempt to capture the complex stochastic, pre-existing flaw-controlled damage nature of glass under blast and impact conditions and do so in a computationally efficient manner. Numerous finite element simulations were carried out using the computational code ABAQUS/Explicit to assess the utility of the model under physically realistic ballistic loading conditions, including multi-hit impact scenarios. Further enhancements of the glass material model are made with the inclusion of the following: (i) differentiation of the mechanical properties of the so-called air-side and tin-side of glass plates manufactured using the float glass process; and (ii) a damage tensor to produce an orthotropic macro-cracked material. In addition a multi-length scale modeling approach for glass is taken to elucidate phenomena at different length scales (e.g. glass irreversible densification, shock response, etc.) with the ultimate objective of enhancing the efficacy of the current continuum-level material model. The irreversible densification of glass under ballistic (shock) loading conditions is investigated at multiple length scales (atomistic-level and continuum-level) in order to understand its effect on the ballistic penetration resistance of glass. The findings related to the material shock response and irreversible densification of glass were subsequently included in the continuum-level glass material model equation of state to further increase its efficacy.
The results from the various test scenarios and modifications to the continuum-level glass material models reveal that: (a) transient non-linear dynamics computational analyses, when utilizing the glass material model, have demonstrated to be a useful tool in understanding the multi-hit ballistic-protection performance of laminated glass/polycarbonate transparent armor systems. The loss of the ballistic-protection performance of the armor caused by a sequence of closely spaced bullet impacts has been observed and the results of these analyses are validated against their experimental counterparts; (b) while it was expected (based on quasi-static mechanical testing result) that orienting the borofloat tin-side as a three-layer laminate strike face would enhance its ballistic protection performance, experimental findings did not support this conjecture. Computational simulations of the laminate impact established the capability of the borosilicate glass material model to capture the prominent experimentally observed damage modes and the measured V50, reconfirming the experimental findings; and (c) a 2-4% (shock strength-dependent) irreversible density increase in glass is capture computationally at multiple lengths scales. Subsequent modifications of the continuum-level material model for glass to include the effect of irreversible-densification resulted in minor improvements in the ballistic-penetration resistance of glass and only for high projectile initial velocities.