Background of The Laser Spallation Technique

U.S. Patent 5, 438, 402; Aug, 1995

Schematic of The Laser Spallation Technique.

A method for characterizing interfacial adhesion is through laser driven shocks of thin films. This is a fundamental technique that quantifies interfacial adhesion. In the laser spallation experiment (shown in figure to the right) developed to measure the tensile strength of a thin film interface, a 2.5 nanosecond (ns) long Nd:YAG laser pulse is impinged over a 3 mm-dia area on a 0.5 µm thick Al film sandwiched between the back surface of a substrate disc (25 mm diameter and 1 mm thick) and a 50 to 100 µm thick layer of SiO2. The melting- induced expansion of Al under confinement generates a compressive stress wave (with 1 ns rise time) directed towards the test coating, which is deposited on the substrate's front surface. The compression stress pulse reflects into a tensile wave from the coating's free surface and leads to its spallation (complete removal) at a sufficiently high amplitude. During this film separation process, the transient free surface velocity of the coating is continuously recorded using a state-of-the art optical interferometer, a schematic of which is shown in the figure to the right. Optical fringes corresponding to stress pulses of 1-2 ns rise-time and 16-20 ns total duration are recorded in a single shot mode, with a resolution of 5 ps! In addition, the patented design of the interferometer can handle reflection from optically rough surfaces.

Applications of the Laser Spallation Technique.

Next, the recorded fringe information is reduced to transient free surface velocity profile v(t), and is used to calculate the normal interfacial tensile stress in the following manner. For a coating of density ρ and thickness h, the interface stress σ is calculable from the measured transient velocity v(t).Because of high strain rate loading conditions, all inelastic effects are largely suppressed during interface decohesion, and thus, the measured strengths are intrinsic and related directly to the atomic microstructure and chemistry of the interfacial region. This was exemplified by measuring the tensile strengths of Nb/sapphire interfaces whose structure and chemistry were characterized using high resolution transmission electron microscopy and modified by a combination of heat treatment and deposition of 5 to 40 Å thick interlayers of Cr and Sb. Thus, atomic-scale tailoring of the interfacial region for either maximizing or reducing adhesion can be accomplished. Using this technology, interfacial strengths in the range of 0.1 GPa (14.5 ksi) to 2.5 GPa (362 ksi) have been measured in a variety of engineering systems (paints, multilayer electronic devices, engines, tribology) involving metal, ceramic and polymeric coatings deposited on metal, semiconductor and ceramic substrates.

  1. Measurement of Thin Film Adhesion

    A spalled spot of a SiC coating from a Si substrate via the Laser Spallation Technique.

    A laser spallation experiment has been developed to measure the strength of planar interfaces between a substrate and a thin coating (in the thickness range of 0.3-3 µm). In this technique a laser pulse of a high enough energy and a pre-determined duration is converted into a pressure pulse of a critical amplitude and width that is sent through the substrate toward the free surface with the coating. The reflected tensile wave from the free surface of the coating pries-off the coating. The critical stress amplitude that accomplishes the removal of the coating is determined from a computer simulation of the process. The simulation itself is verified by means of an interferometry process that is capable of mapping out the profile of the stress pulse generated by the laser pulse. Interface strength values ranging from 0.1 to 2.5 GPa have been determined for various thin film systems system.

  2. Measurement of Joint Strength

    The interface strength of an E-glass/polymer composite as a function of exposure time at 50°C and 90% RH.

    The dynamic tensile strengths of various joints have been analyzed. One study included E-glass composite/polymer and polymer/steel interfaces within the E-glass composite/polymer/AL-6XN stainless steel joint were measured using a laser spallation technique. Values were obtained for the polymer/composite interface while a much higher value was obtained for the steel/polymer interface. Because of the transient nature of the stress pulse, the strain rate changes continuously as the interface stress builds up. A peak strain rate of 5 x 105 s-1 was estimated. The effect of moisture on the tensile strength of the E-glass/polymer interface was also examined. The effect was found to be minimal, with the tensile strength stabilizing after 30 days of exposure to a 90%RH, 50°C environment. When comparing the strengths of corresponding interfaces in an epoxy-bonded joint from a previous study, it was concluded that our polymer results in a much stronger and durable joint.

Polymers for Impact Applications

A micrograph showing the failure at the steel/polymer interface. Similar to the fracture energy measurements, the effect of moisture in the polymer-bonded samples was minimal with the interface tensile strength stabilizing at 320 ± 20 MPa after 30 days of exposure at 50°C and 90% RH environment.

Our research began by introducing a polymer as an adhesive for joining E-glass and stainless steel substrates. A recipe was discovered that resulted in a joint stronger than one of its adhered. Experiments to characterize the intrinsic and total fracture energies of the joints under both static and dynamic loads have been studied. The effect of hygrothermal loads is also discussed and shows that joints with this polymer remain virtually undamaged by such loads. Next, ultrahigh strain rate constitutive behavior of our polymer is presented at peak strain rates of 107 s-1 under pure tension and combined pressure-shear and tension-shear loads by subjecting thin polymer layers to laser generated stress waves of several nanoseconds in duration. We also studied the application of the polymer in multilayer armor design by showing two unusual effects. First, a sample flipping effect, in which polymer/foam bilayer specimens show almost a 40% reduction in their impact resistance depending upon which face bears the impact, and a second, resonance effect, when thin polymer layers are sandwiched between acrylic and polycarbonate plates.

  1. Dynamic Strength and Fracture Toughness of Polymer Joints

    Comparison of the fracture energy values for the E-glass/Hysol epoxy and E-glass/polymer interfaces as a function of sample exposure times at 90% RH and 50ºC.

    Our research on this polymer started nine years ago. In the beginning, the research was motivated by the desire to find a durable adhesive to join AL-6XN stainless steel plates and E-glass composite sections for lightweight ship construction that could withstand static and dynamic loads during ship operation. To that challenge, a polymer chemistry was designed that resulted in failures within the E-glass composite section, away from the joint, when the bonded specimens were separated using a double cantilever beam (DCB) experiment. Additionally it was shown that hygrothermal loads did not degrade the mechanical integrity of the steel/polymer/E-glass joints as their fracture energies were reduced from 1232 ± 15 J/m2 to 1070 ± 35 J/m2 when subjected to an accelerated 90% RH, 38°C environment for 30 days. In contrast, the fracture energy of the same joints constructed using the Hysol 9394 epoxy which is widely used in the aerospace industry underwent a 90% reduction from 950 J/m2 to 88 J/m2 when subjected to the same conditioning loads. To separate the size-related inelastic component of the fracture energy, DCB experiments were also carried out at the cryogenic temperature of 77 K to determine the intrinsic fracture energy value which can be used as a local failure criterion in large scale structural simulation of polymer joints.

  2. Characterization of Polymer Impact Conditions

    Since these joints are expected to perform under blast loads, their ability of resist dynamic loads was also studied by measuring their tensile strength and fracture energy at extremely high rates of loading. This was accomplished by separating the joints using well-characterized laser-generated stress waves of several nanoseconds in duration. Because of the sharp rise-time of the stress pulse, failures in both experiments were accomplished at a peak strain rate approaching 106 s-1. Like the intrinsic fracture energy, these values can be used to set the local failure criterion in design simulations of polymer joints subjected to dynamic loads. To further aid the designer in determining the durability of these joints, these parameters were also measured in conditioned samples. Together with the chemical recipe, the measurements of both the intrinsic and total fracture energies under static as well as dynamic loads provide a complete set of parameters for designing and constructing durable polymer joints for large scale ship building and other structural applications. Following studies were also extended to measure the dynamic tensile strength and fracture energy of the polymer by essentially using the same procedures except that the polymer layer on the steel substrate was chosen thick enough such that the returning tensile wave caused cohesive failure within the polymer volume. An average tensile strength of 93.1 ± 5 MPa at a peak strain rate of 1.67 x 107 s-1 and fracture energy of 6.75 J/m2 at a peak strain rate of 1.2 x 107 s-1 were obtained.

  3. Characterization of Polymer under High Strain Rate Stress Conditions

    Comparison of the calculated (by TTS method) and experimentally derived stress-strain response for the polymer.

    After de­vel­op­ing and char­ac­ter­iz­ing poly­mer joints that were stronger than one of the sub­strates (E-glass com­pos­ites), the pro­ject focus shifted to un­der­stand­ing the con­sti­tu­tive and frac­ture be­hav­ior of poly­mer at ul­tra­high strain rates, ex­ceed­ing 105 s-1. This was mo­ti­vated by the need for de­vel­op­ing poly­mer-based ar­mors against weapon in mod­ern war­fare, such as shaped charges and ex­plo­sively formed pro­jec­tiles, which can at­tain speeds in the range of 2,500 to 9,000 m/s and there­fore cause dy­namic load­ing within 50 ns upon im­pact. Since be­hav­ior at such high rates of load­ing can­not be stud­ied even with the most so­phis­ti­cated plate-im­pact and split-Hop­kin­son bar se­tups, our goal was to first de­velop ex­per­i­men­tal pro­ce­dures sim­i­lar to those uti­lized in these tests and then study the poly­mer be­hav­ior at such rates under pure ten­sion, com­bined ten­sion-shear, and com­bined pres­sure-shear load­ings. Two major ef­fects were un­cov­ered. First, it was demon­strated that the Time Tem­per­a­ture Su­per­po­si­tion prin­ci­ple holds for the poly­mer even at strain rates as high as 105 s-1. This strain rate is two orders of magnitude higher than those reported by Zhao et al. (2008) for the similar phenomenon. This important finding suggests that blast simulations of large-scale structures and armors involving the polymer can be based on constitutive data gathered under quasi-static conditions and translated to higher rates using the TTS principle.

    Interferometrically-reduced stress wave profiles measured in a (a) bare Al (input) plate and an Al/Polymer/Al sample (output) at an energy fluence of 311 mJ/mm2;(b) bare Acrylic (input) plate and an Acrylic/Polymer/Acrylic sample (output) at an energy fluence of 270 mJ/mm2; (c) bare Polycarbonate (input) plate and a Polycarbonate/Polymer/Polycarbonate sample (output) at an energy fluence of 310 mJ/mm2; and (d) bare Stainless steel (input) plate and a Steel/Polymer/Steel sample (output) at an energy fluence of 310 mJ/mm2.

    To study this, thin polymer layers were sandwiched between acrylic, glass, steel and Al plates and loaded with stress waves of 50 ns duration. The stress waves exiting the sections were compared with those entering the sections by the use of interferometry. An unusual resonance phenomenon was uncovered when the polymer layers were sandwiched between the acrylic and polycarbonate plates suggesting a substantial increase in the transmitted stress wave amplitudes contrary to a reduction that was expected. The expected results were however obtained when the polymer was sandwiched between Al and steel plates, each showing a dramatic reduction in the transmitted stress wave amplitude. Beside this ultrahigh rate testing, the effectiveness of the polymer in managing more traditional impacts in the low strain rate regime (103 s-1, or impulse loading over 1-50 ms range) were also studied by carrying out tests in a standard drop tower facility. To start, dense polymer sheets of 1-2 mm thickness were combined with commercially available foams to prepare bilayer specimens. A sample flipping effect was discovered which showed that the impact force was lower by up to 40% when the dense polymer layer was away from the striking impacter head. This observation, along with the discovered resonance effect, is very important to consider when designing polymer-based multilayer armors. Details pertaining to this research can also be found in Youssef & Gupta (2012) and Youssef & Gupta (2011).

  4. Testing for polymer use in helmets

    Professor Gupta's interview regarding use of a UCLA developed polymer in football helmets

    Video of helmet (left) and polymer (right) impact testing at the Dynamic Research Institute (DRI)

    The mechanism in which a trauma to the head may cause concussions or other brain injuries is described in the slides below.

Copper Nano-structures Loaded by Shockwaves

The deformation of metals subjected to ultra-high strain rates, in excess of 107 s-1, is associated with a variety of unique physical phenomena, not normally observed in metals deformed at lower strain rates. Additionally, nano-structured materials are now recognized to exhibit extreme strength, reversible plasticity and dislocation-free twin boundary motion. In this study, a combination of laser-generated shockwave experiments, TEM observations, and Molecular Dynamics (MD) simulation are used to study the dynamic response of several nano-structured fcc metals when loaded at extremely high strain rates. MD simulation conditions are remarkably close to those achieved experimentally, allowing for direct interpretation of collective atomic motion and associated defect structures. The combination of nano-geometry and nanosecond rise-time stress waves is able to deform the material at stress (50 GPa) and strain rate (in excess of 107 s-1) levels never accomplished before. Two different loading setups are shown below:

Set-up for Copper Nanopillars via Longitudinal Loading

Set-up for Copper Nanopillars via Transverse Loading

  1. Chemical Activation of Copper Nanopillars under Longitudinal Loading

    Scanning Electron Microscope image of copper nano-pillar 36 hours after shock loading.

    Scanning Electron Microscope image of reference copper nano-pillar.

    An apparatus and test procedure for fabrication and loading of single crystal metal nanopillars under extremely high pressures (>1 GPa) and strain rates (>107 s−1), using laser-generated stress waves, are presented. Single-crystalline Cu pillars (∼1.20 µm in tall and ∼0.45 µm in diameter) prepared via focused ion beam milling of Cu(001) substrates are shock-loaded using this approach with the dilatational stress waves propagating along the [001] axis of the pillars. Transmission electron microscopy observations of shock-loaded pillars show that dislocation density decreases and that their orientation changes with increasing stress wave amplitude, indicative of dislocation motion. The shock-loaded pillars exhibit enhanced chemical reactivity when submerged in oil and isopropyl alcohol solutions, due likely to the exposure of clean surfaces via surface spallation and formation of surface steps and nanoscale facets through dislocation motion to the surface of the pillars, resulting in growth of thin oxide films on the surfaces of the pillars.

  2. Bending of Copper Pillars via Transverse Loading

    TEM bright-field images of the nanopillar before (a) and after laser shot at 320 mJ (b) and 450 mJ (c). Insets show corresponding diffraction patterns taken from the tip of the pillar. Scale bars in all images are 200 nm.

    An experimental study to bend FIB-prepared cantilevered single crystal Cu nanopillars of several hundred nanometers in diameter and length at ultrahigh strain rate is presented. The deformation is induced by laser-generated stress waves, resulting in local strain rates exceeding 107 s−1. Loading of nano-scale Cu structures at these extremely short loading times shows unique deformation characteristics. At a nominal stress value of 297 MPa, TEM examination along with selected area electron diffraction characterization revealed that twins within the unshocked Cu pillars interacted with dislocations that nucleated from free surfaces of the pillars to form new subgrain boundaries. MD simulation results were found to be consistent with the very low values of the stress required for dislocation activation and nucleation because of the extremely high surface area to volume ratio of the nanopillars. Specifically, simulations show that the stress required to nucleate dislocations at these ultrahigh strain rates is about one order of magnitude smaller than typical values required for homogeneous nucleation of dislocation loops in bulk copper single crystals under quasi-static conditions.

Biomechanical Research Topics

  1. Biofilm Removal in Grossly Infected Wounds

    Schematic of proposed method of how to use laser generated shockwaves to delaminate and destroy biofilms.

    Bacterial biofilm-related infections are a burden on the healthcare industry. The effect of laser generated shockwaves through polycarbonate, a flexible polymer, is explored for its ability to generate high peak stresses, and also for its ability to conform to complex wound surfaces. Shockwave pulses in Al coated polycarbonate substrates and a resulting peak stress of greater than 60 MPa was measured which should provide sufficient pressure to kill bacteria.

  2. Bio­me­chan­ics of a Cow Eye

    Re­ported me­chan­i­cal prop­er­ties of or­bital con­nec­tive tis­sue and fat have been too sparse to model strain–stress re­la­tion­ships un­der­ly­ing bio­me­chan­i­cal in­ter­ac­tions in stra­bis­mus. We have per­formed rhe­o­log­i­cal tests to de­velop a multi-mode upper con­vect­ed­Maxwell (UCM) model of these tis­sues under shear load­ing. From 20 fresh bovine or­bits, 30 sam­ples of con­nec­tive tis­sue were taken from rec­tus pul­ley re­gions and 30 sam­ples of fatty tis­sues from the pos­te­rior orbit. Ad­di­tional sam­ples were de­fat­ted to de­ter­mine con­nec­tive tis­sue weight pro­por­tion, which was ver­i­fied his­to­log­i­cally. Me­chan­i­cal test­ing in shear em­ployed a tri­borheome­ter to per­form: strain sweeps at 0.5–2.0 Hz; shear stress re­lax­ation with 1% strain; vis­com­e­try at 0.01−0.5 s−1 strain rate; and shear os­cil­la­tion at 1% strain. Av­er­age con­nec­tive tis­sue weight pro­por­tion was 98% for pre­dom­i­nantly con­nec­tive tis­sue and 76% for fatty tis­sue. Con­nec­tive tis­sue spec­i­mens reached a long-term re­lax­ation mod­u­lus of 668 Pa after 1,500 s, while cor­re­spond­ing val­ues for fatty tis­sue spec­i­mens were 290 Pa and 1,100 s. Shear stress mag­ni­tude for con­nec­tive tis­sue ex­ceeded that of fatty tis­sue by five-fold. Based on this data, a mul­ti­mode UCM model was developed with vari­able vis­cosi­ties and time con­stants, and a damped hy­per­e­las­tic re­sponse that ac­cu­rately de­scribed mea­sured prop­er­ties of both con­nec­tive and fatty tis­sues. Model pa­ra­me­ters dif­fered sig­nif­i­cantly be­tween the two tis­sues. Vis­coelas­tic prop­er­ties of pre­dom­i­nantly con­nec­tive or­bital tis­sues under shear load­ing dif­fer markedly from prop­er­ties of or­bital fat, but both are ac­cu­rately re­flected using UCM mod­els. These vis­coelas­tic mod­els now fa­cil­i­tate re­al­is­tic global mod­el­ing of EOM be­hav­ior in binoc­u­lar align­ment and stra­bis­mus.

  3. High Speed video analysis for Biomechanics of a Baseball Swing

    Professor Gupta and a few of his graduate students took to using the lab's high speed video camera in order study the biomechanical effect of the axe-handle. Professor Gupta's report can be found here.

    High-speed video of baseball impact utilizing traditional handle bat

    High-speed video of baseball impact utilizing Axe handle bat

Group News
October 2014 Graduate student Brian Ramirez presents a poster at the SACNAS National Conference in Los Angeles. Brian won the graduate student poster presentation award. Congrats Brian!
Brian and his poster!
Brian announced the winner!
September 2014 USA Today releases article referencing Professor Gupta's work with the Axe-handle bat
Find article here.
April 2014 Professor Gupta releases a biomechanical study for a new baseball bat.
Find report here.
September 2013 Local radio station KCRW interviews Professor Gupta regarding his research to make safer helmets for football players
Find article here.
September 2013 Professor Gupta featured in LA Weekly article regarding his research to make a safer football helmet to reduce concussions
Find article here.
September 2013 Gupta Group research on polymer found online on Gizmag
Find article here.
September 2013 Gupta Group research on polymer liner included in UCLA's Daily Bruin
Find article here.
Aug. 2013 Professor Gupta is featured on the UCLA Mechanical and Aerospace Engineering website for research to make football helmets safer.
Find article here.