Shock treatment

17 Jun 2002

Advances in repetition rates and pulse energies of high-power lasers mean that they can now be used for strengthening materials such as steel. Rob van den Berg finds out more.

From Opto & Laser Europe December 2001

Turbine blade

Although bombarding a metal component with intense laser pulses may not be the most obvious way to strengthen it, laser peening - or laser shock processing - is becoming an increasingly accepted technique. It aids the performance of materials by improving their resistance to cracking, stress corrosion and fatigue.

The first experiments in laser peening took place more than 30 years ago at the Battelle Institute in Ohio, US. However, it is only today - with the advent of higher-power lasers at higher repetition rates - that the method is beginning to compete with traditional shot-peening methods, which employ small metal particles to deliver the impact.

High power and pulse energy

A laser system appropriate for peening at an industrial level requires an average power in the hundred watt to kilowatt range, a pulse energy of around 100 J per pulse and a pulse duration of 10-30 ns. Five years ago, when just such a laser became available to Lloyd Hackel of the Lawrence Livermore National Laboratory (LLNL), US, Hackel began to look for an industrial partner who would be interested in developing a laser peening system. He eventually teamed up with the Cooperative Research and Development Agreement (CRADA) and the Metal Improvement Company (MIC) of New Jersey, US.

Laser peening employs high-energy laser pulses from a solid-state laser to create intense shock waves in a material. The area to be treated is locally covered with two different layers: one that absorbs the laser light and one that is made of a transparent substance, usually water. The laser energy that is absorbed by the surface layer rapidly vapourizes the layer, but the layer of water suppresses the outward expansion of the vapour. This causes a shock wave to propagate through the material, generating peak pressures of several gigapascals.

The plastic deformation caused by the shock wave gives rise to compressive residual stresses that can extend to more than 1 mm in depth, making the material more resistant to fatigue and stress corrosion cracking. Hackel said: "[The shock wave] acts like a compressive force, keeping small cracks pinched together and making them grow much more slowly."

Laser peening is already used in the automotive, medical, and aerospace industries. Spin-off technologies like laser peenmarking, which provides an easy way to identify parts, and peenforming, which enables complex contouring of problematic thick metal components, such as parts of large aircraft wings, demonstrate that pulsed lasers are making an impact.Generating high-energy pulses from a solid-state laser is a well-developed technique, but generating high average powers at high pulse energies is more problematic.

Hackel's Nd:glass laser consists of a single master oscillator combined with one or more power amplifiers. The amplifier gain medium is Nd-doped phosphate glass configured in a 1 cm thick slab. As the laser beam propagates through the gain medium it follows a zigzag path, bouncing off the slab faces due to total internal reflection.

The slab is water-cooled and pumped by two flashlamps on each side. It has a high heat-transfer efficiency, which increases the speed capability of the laser. A specially shaped reflector surrounds the flashlamps and provides uniform optical pumping.

The zigzag path that the light follows averages thermally induced pathlength differences. After four passes through the amplifier, however, the wavefront is highly distorted. Hackel said: "To correct this, the pulses are reflected off a phase conjugate mirror. The distorted beam thus corrects itself while retracing the same path. In this way we can achieve 25-50 J per pulse with a perfect beam quality."

It is the mechanical strength of the glass that determines how fast the laser can fire. "With our current repetition rate of 6 Hz - which is 25 times as high as was formerly possible - we have been able to transform laser peening from a laboratory curiosity into a technique that is seriously considered in an industrial environment," said Hackel. "We can treat a surface of 1 m2 in less than an hour, which is comparable to the performance of shot peening."

When a new type of glass that does not fracture so easily is employed, the repetition rate may be increased to 12 Hz. "And using diode pumping, by which you can tune the excitation to the absorption band of the neodymium ion, we could increase the rate by two to four times," said Hackel.

One of the first applications of laser peening was in the aerospace industry, where it was used for the treatment of turbine blades. Peening can also be applied to other areas, for instance prosthetic devices such as knee and hip implants. In collaboration with the biomechanics department of UCLA, MIC has been studying the use of laser peening to treat knee implants.

James Daly, MIC's senior marketing vice-president, says that the technique can also be used on transmission gears for helicopters and oil drilling components, but added: "We are not yet at the stage where we can offer our system for the treatment of automotive components. The speed of the assembly line does not allow that."

Hackel, however, says that the automotive industry has shown considerable interest in the technique for the treatment of car frames. "This might make it possible to manufacture lighter frames, with obvious reductions in fuel use," he said.Laser peening has other applications in peenforming and peenmarking. Hackel explained: "By applying a negative residual stress on one side of a metal part, you can make it bend in a highly controlled way. Using laser peenforming you could potentially make the 40 m outer skin of an aeroplane wing fit precisely onto its frame with just one treatment. There would be no need to go back and forth to make slight adjustments, as is the case in classical shot peenforming. And we can easily treat plates with a thickness of up to 1 inch, which is twice the capability of shot peenforming."

In laser peenmarking, a tiny identification mark is applied to stress-critical parts using a laser. Hackel said: "Conventional marking techniques reduce the fatigue strength of a part. With peenmarking, it is safe to mark them anywhere."

MIC hopes to bring its first production system to market in the spring of next year, but faces strong competition from LSP Technologies (LSPT) of Ohio, US.

David Lahrman, head of marketing at LSPT, said: "Apart from the exclusive rights to Battelle's original technology, we have 21 laser peening patents with 25 pending." He plays down the importance of high repetition rates in reducing costs and increasing throughput: "There is often no advantage in treating the whole part. It is much more important to use lasers that are not temperamental and can be relied upon on the factory floor."

Although most commercial activity is concentrated in the US, a great deal of fundamental research is carried out in Europe. In the last two years Thomas Schmidt-Uhlig and Peter Karlitschek from the Laser Laboratory in Göttingen, Germany, have developed a fibre-delivery system for laser shock processing in co-operation with Japanese firm Toshiba.

Schmidt-Uhlig said: "If you want to transmit high-energy laser pulses through a fibre, you are limited by the damage threshold of the fibre material and by dielectric breakdown of the air in front of the fibre. We developed an optical homogenizer - an array of small lenses - that separates the incoming laser light into several partial beams that are then combined again at the fibre entrance. These partial beams have their own individual focal spots, so breakdown of the air is prevented and the total energy is distributed over a very smooth, flat profile which is coupled into the fibre."

In this way 125 mJ pulses from a frequency-doubled Nd:YAG laser could be delivered via a standard multimode optical fibre. Surface analysis of a treated stainless-steel sample showed that sufficiently large compressive stresses could be induced. Schmidt-Uhlig said: "The energy per pulse is not very high, so several thousand pulses are necessary to process 1 cm². This is compensated for, however, by the high repetition rates of commercially available laser sources. Our approach means that no protective coating has to be applied to the metal surface prior to processing."In the Laboratoire pour l'Application des Lasers de Puissance in Arcueil, France, other fundamental aspects of laser shock processing are being studied. Rémy Fabbro and colleagues have been able to characterize the induced shock waves at the surface of the material using Doppler velocimetry. In this way, they have studied the effect of changes in laser wavelength and energy flux on the efficiency of the process.

Fabbro said: "At shorter wavelengths - from an XeCl excimer laser, for instance - and at power densities of 10 GW/cm², the shielding effect of the water layer becomes dominant. Owing to multiphoton ionization in the water, the plasma that is formed begins absorbing the laser light before it can reach the metal interface. The induced pressure is therefore less effective."

These results strongly favour the second harmonic of Nd:YAG at 532 nm, which is hardly absorbed by water. "You can even have the whole system - both optics and target - immersed in water. Toshiba has developed a peening robot that goes down into the cooling water of a nuclear reactor to treat welds," said Fabbro.