Using Shock Wave Simulation to Optimize Body Armor

Advanced computational modeling helps to improve protective gear designs

Tomas Friend, Ph.D.

American war fighters in Iraq and Afghanistan face a high risk of serious injury or death from mines and improvised explosive devices (IEDs). Since the end of conventional warfare in Iraq, blasts have killed more than 1,000 Americans and many more Iraqis. A typical explosion creates a shock wave traveling at three to five times the speed of sound. Temperature across the shock wave can increase over 1000 degrees Celsius near the explosion and pressure can increase abruptly to more than 20 atmospheres. Closed spaces such as rooms and street canyons cause reflections and shock wave interference patterns that can amplify pressure changes. A large percentage of fatal blast injuries result from primary blast injuries (PBIs), in which the shock wave directly damages the lungs through violent, localized pressure changes. Most body armor is designed to prevent penetrating wounds and not blast wave injuries. The working environment of today's war fighters requires body armor that protects against blast waves as well as shrapnel and other projectiles.

Because a shock wave travels faster in liquids and solids than air, organs like the lungs, intestines, and inner ear are damaged by shear stresses when a shock wave reaches tissue-gas interfaces. The lungs are most vulnerable to PBIs because they contain large surface areas of fragile alveoli where oxygen is exchanged for carbon dioxide. Differences in wave propagation mechanics at the interface between air and blood in alveolar membranes cause large deformations of lung tissue that collapse alveolar sacs, tear alveolar membranes and rupture blood vessels. The extent of lung injury is a decisive parameter for mortality in victims surviving an explosion. Because the symptoms of primary blast injury are often delayed, victims may not receive timely treatment. Improvements in body armor design can reduce the incidence of fatal injuries resulting from PBI to the lungs.

CFD Research Corp. (CFDRC) is working on a project sponsored by the Defense Advanced Research Projects Agency (DARPA) to mathematically model blast wave interactions with the human body and help design body armor that protects war fighters against PBIs. This modeling effort requires a combination of diverse disciplines including gas and human body dynamics, tissue biomechanics, and the pathophysiology of organ damage.

Not for dummies (or pigs)
Several approaches have been used to study blast injuries and design body armor. Dummies containing pressure sensors and synthetic approximations to human organs have been used to study bomb blasts and car crashes. Large animals such pigs also have been equipped with pressure sensors exposed to blast waves. These methods can be used to relate shock wave pressures to the degree of injury and compare body armor prototypes but they are not well suited for the efficient optimization of body armor design.

Computational modeling can be used to optimize the design of body armor by combining the techniques of computational fluid dynamics (CFD) and computational structural dynamics (CSD). CFD is used to simulate airflows and fluid flows while CSD is used to model stresses and strains in solids. A combined model that describes fluid flows impacting a deformable solid by direct linking of a CFD code and a CSD code is a challenging task. Fluid dynamics codes typically solve Navier-Stokes equations and use fixed coordinate systems, whereas structural dynamics codes are formulated in a Lagrangian frame, where the coordinate system moves with the structure. As a result, once a structure deforms, it will generate a hole or a void in the CFD mesh. CFDRC has developed fluid structure interaction (FSI) software to study the fluid-dynamics of blast waves as they impact and pass through a human chest. The fluid dynamics module solves Navier-Stokes equations in a hybrid Lagrangian-Eulerian frame.

CFDRC has developed an FSI model for modeling blast wave interactions with a flexible human body and for pressure wave propagation in the human thorax. FSI uses a time-dependent geometry, a moving deforming mesh and overset surface body motion

The computational simulation of traveling shock waves and wave diffraction on walls requires high resolution in both space and time, which in turn requires a high mesh density and high numerical accuracy, especially in shockwave interaction regions. Since these regions change over time with the solution, CFDRC has developed an advanced computational modeling technique based upon local adaptive mesh refinement. The hierarchical adaptive mesh refinement (HAMR) code uses a unique computational approach based upon hierarchical data structures that results in an efficient means of performing solution adaptive mesh refinement for gas dynamic applications.

Blast wave modeling
FSI and HAMR have been used to simulate the effects of a bomb blast on an "instrumented soldier" in an enclosed space (Figure 1). The blast is created by an explosive charge detonated at a predetermined position in a room. The simulation was performed in an overnight run on a single-processor PC with high accuracy. Without using the HAMR grid adaptation, the run time would have been weeks instead of overnight. The results of numerous simulations of this type are in good agreement with physical data obtained from blast wave experiments.

Figure 1: Example of a single time point from a blast wave simulation in an enclosed space with color-coded blast wave pressures. Complete simulations will show how the body is thrown by the blast as well as the location and severity of blast lung injury. Many different scenarios can be simulated and the results used to plan emergency responses to attacks against military and civilian targets.
The room scale blast simulation was used to provide boundary conditions for modeling blast wave interactions with the human thorax (chest). Figure 2 shows four time points from the simulation of a blast wave passing through a human chest. The pressure wave propagates from the front of the body to the back. When the shock wave reaches the chest, the chest wall is compressed, sending a pressure wave into the lungs. Since the speed of sound in the air of the lungs is slower than in the surrounding tissue, the induced wave front inside the lungs falls behind the blast wave front. While the air space in the lung is still experiencing high pressure, the chest wall deformation reaches a maximum and starts to rebound. This rebound locally expands the lungs, sending a low-pressure wave into the lung airspace. At the same time, the original blast wave is exiting through the back. The wave propagation becomes complex as the wave also propagates along the thorax wall, sending a pressure pulse into the airspace. The complex patterns of pressure waves create shear stresses that damage the lungs.

Figure 2: Four time points from a 2-D simulation of a blast wave impact on the front of a human chest. The geometry of the human thorax was meshed from an image taken from the Visible Human Databank (
The patterns of shear stress calculated by the blast wave model are used as input data for a 3-D, physiology-based lung model that simulates the physiological consequences the blast. The geometry used in the lung model is generated using a medical image of the upper respiratory tract and a mathematical algorithm that generates airway branches and terminal alveolar "equivalents" (Figure 3). Based on the predicted location and degree of damage, the model estimates the resulting gas exchange in damaged lungs and blood oxygenation.

Figure 3: Example lung geometry generated during CFDRC modeling of blast lung injury. A lung geometry is created from medical images of segmented lung lobes and the first four branches of the airways. Additional branches are created using a physiology-based tree generation algorithm. Each airway terminates with an "equivalent" alveolus that represents thousands of alveoli. Image courtesy of Dr. J. Reinhardt, University of Iowa
A wide range of personal protective vests and suites are currently deployed in the battlefield and for de-mining operations. These devices incorporate composite materials to protect the thoracic region from penetrating wounds, but are not as effective for blast protection. In some cases, vest materials may even exacerbate blast lung injury.

CFDRC assessed the relative effectiveness of several different laminations for blast protection including variations of soft materials layers (modeled as air layers), rigid layers, and blast attenuating foam (viscoelastic layers). The results of simulations show that body armor made of eight low density rigid layers separated by air pockets actually increases blast lung injury. As the blast wave reaches the outer rigid layer of the body armor, the rigid layer accelerates and compresses the adjacent air pocket. The blast wave is traveling at a supersonic speed but the inertia of each of the subsequent rigid layers slows the wave front while increasing the pressure in subsequent air pockets. This compression wave propagates to the chest wall, causing more lung damage than no armor. The aim of body armor for blast protection is to spread out the pressure spike over time, lowering peak pressure and reducing shear stresses in the lungs. The use of air and rigid layer has just the opposite effect.

The results of simulations with alternating layers of various soft and rigid materials, including a high density outer shell were more encouraging. The best design combines an enhanced, high-density outer shell and alternating layers of blast attenuating foam and light rigid plates. The inherent damping of the wave passing through the blast attenuating foam reduces the rate at which pressure rises and spreads out the overpressure loading over a longer duration. The high density of the outer shell has more inertia than the subsequent light plates reducing or eliminating the compression effects.

Conclusions and future direction
CFD and CSD modeling have been used to optimize parameters for body armor design. CFD and CSD coupled to a physiology-based model of blast lung injury also have been used to predict the degree of lung injury and the treatment victims will likely require before symptoms become life threatening. In the future, similar models will be applied to develop means of preventing damage to other organs including the brain and colon. The same computational techniques can be applied to transportation safety and protective gear to prevent blunt trauma injuries.

Tomas Friend is a senior scientist at CFDRC. He may be contacted at