A growing number of applications involve the transmission of high-intensity laser pulses through optical fibers.
Previously, our particular interests led to a series of studies on single-fiber transmission of Q-switched, 1064 nm pulses
from multimode Nd:YAG lasers through step-index, multimode, fused silica fibers. The maximum pulse energy that
could be transmitted through a given fiber was limited by the onset of laser-induced breakdown or damage. Breakdown
at the fiber entrance face was often the first limiting process encountered, but other mechanisms were observed that
could result in catastrophic damage at either fiber face, within the initial "entry" segment of the fiber, and at other
internal sites along the fiber path. These studies examined system elements that can govern the relative importance of
different damage mechanisms, including laser characteristics, the design and alignment of laser-to-fiber injection optics,
fiber end-face preparation, and fiber routing. In particular, criteria were established for injection optics in order to
maximize margins between transmission requirements and thresholds for laser-induced damage. Recent interests have
led us to examine laser injection into multiple fibers. Effective methods for generating multiple beams are available, but
the resulting beam geometry can lead to challenges in applying the criteria for optimum injection optics. To illustrate
these issues, we have examined a three-fiber injection system consisting of a beam-shaping element, a primary injection
lens, and a grating beamsplitter. Damage threshold characteristics were established by testing fibers using the injection
geometry imposed by this system design.
A promising new tool in shock wave physics is the generation of shock waves in test materials through the impact of small, laser-accelerated discs ('flyers'). In order to achieve the necessary one-dimensional condition of uniaxial strain in the shock-loaded material, it is vital that flyers maintain a nearly planar geometry during the acceleration and impact processes. The geometry of the flyer is significantly influenced by the spatial intensity profile of the driving laser beam. With the goal of achieving a nearly uniform drive intensity for this application, we have evaluated a diffractive, microlens-array beam shaper for use with a high-energy, Nd:Glass laser driver. Based on the near-field spatial profile of this multimode laser, a 30-mm-diameter array containing multiple hexagonal diffractive lenslets was designed and fabricated. In combination with a primary integrator lens of 76.2-mm focal length, this optical element was intended to produce a uniform intensity distribution over a 2-mm-diameter spot at the focal plane of the primary lens. Beam profiling studies were performed to determine the performance of this optical assembly. At the focal plane of the primary lens, the beam shaping optics generated a reasonably uniform profile over a large portion of the focused beam area. However, a small amount of undiffracted light resulted in a high-intensity, on-axis spike. A beam profile approaching the desired 'top hat' geometry could be obtained by moving the flyer launch plane a few mm inside or outside of the focal plane. The planarity of flyers generated using this optical assembly was evaluated using a line-imaging, optically recording velocity interferometer system (ORVIS). Results of these measurements demonstrate the deleterious effect of the on-axis spike on flyer planarity. Acceptable conditions for useful flyer impact experiments can be obtained by operating at a position that provides a near-top-hat profile.
An increasing number of applications are requiring fiber transmission of high-intensity laser pulses. Our particular interest have led us to examine carefully the fiber transmission of Q-switched pulses from multimode Nd:YAG lasers at their fundamental wavelength. The maximum pulse energy that can be transmitted through a particular fiber is limited by the onset of laser-induced breakdown and damage mechanisms. Laser breakdown at the fiber entrance face is often the first limiting process to be encountered, but other mechanisms can results in catastrophic damage at either fiber face, within the initial entry segment of the fiber, and at other internal sites along the fiber path. In the course of our studies we have examined a number of factors that govern the relative importance of different mechanisms, including laser characteristics, the design and alignment of injection optics, fiber end-face preparation, and fiber routing. The present study emphasizes the important criteria for injection optics in high-intensity fiber transmission, and illustrates the opportunities that now exist for innovative designs of optics to meet these criteria. Our consideration of diffractive optics to achieve desired began in 1993, and we have evaluated a progression of designs since that time. In the present study, two recent designs for injection optics are compared by testing a sufficient number of fibers with each design to establish statistics for the onset of laser-induced breakdown and damage. In this testing we attempted to hold constant other factors that can influence damage statistics. Both designs performed well, although one was less successful in meeting all injection criteria and consequently shows a susceptibility to a particular damage process.
Laser-induced damage mechanisms that can occur during high- intensity fiber transmission have been under study for a number of years. Our particular interest in laser initiation of explosives has led us to examine damage processes associated with the transmission of Q-switched, Nd:YAG pulses at 1.06 micrometers through step-index, multimode, fused silica fiber. Laser breakdown at the fiber entrance face is often the first process to limit fiber transmission, but catastrophic damage can also occur at either fiber end face, within the initial 'entry' segment of the fiber, and at other internal sites along the fiber path. Past studies have examined how these various damage mechanisms depend upon fiber end-face preparation, fiber fixturing and routing, laser characteristics, and laser-to-fiber injection optics. In some applications of interest, however, a fiber transmission system may spend years in storage before it is used. Consequently, an important additional issue for these applications is whether or not there are aging processes that can result in lower damage thresholds over time. Fiber end-face contamination would certainly lower breakdown and damage thresholds at these surfaces, but careful design of hermetic seals in connectors and other end-face fixtures can minimize this possibility. A more subtle possibility would be a process for the slow growth of internal defects that could lead to lower thresholds for internal damage. In the current study, two approaches to stimulating the growth of internal defects were used in an attempt to produce observable changes in internal damage threshold. In the first approach, test fibers were subjected to a very high tensile stress for a time sufficient for some fraction to fail from static fatigue. In the second approach, test fibers were subjected to a combination of high tensile stress and large, cyclic temperature variations. Both of these approaches were rather arbitrary due to the lack of an established growth mechanism for internal defects. Damage characteristics obtained from fibers subjected to each of these aging environments were compared to results from fresh fibers tested under identical conditions. A surprising result was that internal damage was not observed in any of the tested fibers. Only breakdown at the fiber entrance face and catastrophic damage at both end faces were observed. Fiber end faces were not sealed during the accelerated aging environments, and thresholds at these faces were significantly lower in the aged fibers. However, most fibers transmitted relatively high pulse energies before damaging, and a large fraction never damaged before we reached the limits of our test laser. The absence of any observable affect on internal damage thresholds is encouraging, but the current results do not rule out the possibility that some other approach to accelerated aging could reveal a growth mechanism for internal defects.
High laser intensities are being transmitted through optical fibers in a growing number of applications. Our interest in laser initiation of explosives has led us to examine the transmission of Q-switched, Nd:YAG laser pulses through step-index, multimode, fused-silica fibers for a number of years. A common limiting process is a plasma-forming breakdown occurring at the fiber entrance face. The breakdown threshold at this face depends on the surface characteristics that result from the particular method of end-face preparation. In previous studies we examined entrance-face breakdown thresholds for several different mechanical polishing schedules, and also for several schedules of CO<SUB>2</SUB>-laser conditioning following mechanical polishing. In the present study we examined three end-face preparation methods that were base on the recent availability of exceptionally good cleaved surfaces for our fibers of interest. Using test procedures similar to those in past studies, we examined the cleaved fibers directly, fibers with cleaved surfaces that were subsequently flame polished, and fibers with cleaved surfaces that were subsequently conditioned with a CO<SUB>2</SUB> laser. All of these preparation methods resulted in fibers that showed a broader range of entrance-face breakdown conditions than found in past studies, together with a susceptibility to subsurface exit-face damage. By introducing additional cleaning steps with the cleaved surfaces, we were able to reduce the variability in breakdown thresholds observed after subsequent CO<SUB>2</SUB>- laser conditioning. A consistent location of exit-face damage sites indicates that subsurface fracturing occurs during the cleaving process, and that the subsequent end-face processing steps were not effective in mitigating damage at these sites. Threshold energies for entrance- face breakdown are also affected by the relation between incident laser energy and the resulting peak local fluence at this surface. Laser characteristics and the design of the laser-to- fiber injection optics determine this relation. The present study utilized an improved injection system consisting of a custom diffractive optical element combined with a lens having a very short focal length. This system produced the lowest value for the ratio of peak-to-average fluence at the entrance face that we have observed, and was very successful in inhibiting internal damage mechanisms along the fiber path.
A system has been built and is now being tested that can inject a very high power. pulsed yttrium aluminum garnet (YAG) laser beam into a multimode fiber without damaging the fiber. This design is quite tolerant to changes in the laser beam quality and alignment errors, thereby making the system largely maintenance free. This beam-injection architecture is expected to be useful for medical and industrial applications.
Interest in the transmission of high intensities through optical fibers is being motivated by an increasing number of applications. Using different laser types and fiber materials, various studies are encountering transmission limitations due to laser-induced damage processes. For a number of years we have been investigating these limiting processes during the transmission of Q-switched, multimode, Nd:YAG laser pulses through step-index, multimode, fused- silica fiber. We have found that fiber transmission is often limited by a plasma-forming breakdown occurring at the fiber entrance face. This breakdown can result in subtle surface modifications that leave the entrance face more resistant to further breakdown or damage events. Catastrophic fiber damage can also occur as a result of a variety of mechanisms, with damage appearing at fiber entrance and exit faces, within the initial 'entry' segment of the fiber path, and at other internal sites due to fiber fixturing and routing effects. System attributes that will affect breakdown and damage thresholds include laser characteristics, the design and alignment of laser-to-fiber injection optics, and fiber end-face preparation. In the present work we have combined insights gained in past studies in order to establish what thresholds can be achieved if all system attributes can be optimized to some degree. Our multimode laser utilized past modifications that produced a relatively smooth, quasi-Gaussian profile. The laser-to-fiber injection system achieved a relatively low value for the ratio of peak-to-average fluences at the fiber entrance face, incorporated a mode scrambler to generate a broad mode power distribution within the initial segment of the fiber path, and had improved fixturing to insure that the fiber axis was collinear with the incident laser beam. Test fibers were from a particular production lot for which initial-strength characteristics were established and a high-stress proof test was performed. Fiber end faces were prepared by a careful mechanical polishing schedule followed by surface conditioning using a CO<SUB>2</SUB> laser. In combination, these factors resulted in higher thresholds for breakdown and damage than we had achieved previously in studies that utilized a simple lens injection system. Probability distribution functions were fitted to the threshold data, providing a means for estimating the probability for transmission failure at lower laser energies.
For a number of years we have been investigating laser-induced damage mechanisms that can occur during the transmission of Q-switched, Nd/YAG laser pulses through fused silica fibers. We have found that fiber end-face characteristics, laser characteristics, and aspects of the laser-to-fiber injection typically determine dominant damage mechanisms. However, an additional damage process has been observed occasionally at internal sites where fibers were experiencing significant local stresses due to fixturing or to bends in the fiber path. A transmission reduction prior to damage was typically not measureable at these sites. Damage would not always occur during initial testing, but sometimes occurred later in time at laser levels that previously had been transmitted without damage. In these cases the time at stress appeared to be more important than the number of transmitted shots prior to damage. A possible relation between internal damage thresholds at stressed sites and the total time under stress is suggested by the fact that silica fibers experience static fatigue processes. These processes involve the slow growth of local defects under tensile stress at rates that depend upon environmental conditions. Defects reaching sufficient size and having appropriate location could be sites for reduced laser-induced damage thresholds. This possibility could have important implications for high-power fiber transmission systems that must satisfy extended lifetime requirements. The needs of the telecommunications industry have motivated extensive studies into initial fiber defect characteristics and their likely growth mechanisms. The present work used the understanding developed in these studies to guide a preliminary experimental investigation into the possibility that static fatigue processes can affect damage thresholds. The experiments used a laser injection and fiber routing configuration that produced significantly elevated fluences within fiber core regions under tensile stress. In one set of experiments, internal damage thresholds were determined in available fiber samples that had been assembled in stress-imposing fixtures for periods up to 24 months. A decline in mean thresholds with time was observed, although measured values showed significant scatter. In order to establish initial strength and fatigue properties for these fibers, a number of additional samples were used to generate time-to-failure data at various stress levels. Based on these results, other fiber samples were subjected to conditions that greatly accelerated fatigue processes. Internal damage thresholds were then measured in these fibers and compared to thresholds measured in fresh fibers. Conclusive comparisons were frustrated by sample-to-sample and lot-to-lot variations in fiber defects.
For several years we have been investigating laser-induced damage mechanisms encountered when transmitting Q-switched Nd/YAG laser pulses through step-index, multimode, fused silica fibers. Previous studies primarily addressed end-face breakdown and damage processes and how corresponding thresholds could be affected by different preparation techniques. However, we frequently encountered two internal mechanisms for damage that influence test procedures and results. An `entry' mechanism is related to the laser mode structure and to the geometry of laser injection into the fiber entrance face. Internal damage is also observed at sites where a fiber is experiencing significant local stresses, either due to fixturing or to severe bends in the fiber path. The present study continued to address these fundamental issues, and began to address additional concerns that may arise in the design of practical high-power fiber transmission systems. End-face preparation issues were examined through a comparison between purely cleaved faces and mechanically polished faces.
A previous investigation of laser-induced damage mechanisms and corresponding thresholds in step-index, multimode fibers was motivated by an interest in optical systems for firing explosives. In the initial study, the output from a compact, multimode Nd/YAG laser was coupled into fiber cores of pure fused silica. End-face polishing steps were varied between successive fiber lots to produce improved finishes, and each fiber was subjected to a sequence of progressively increasing energy densities up to a value of more than 80 J/cm<sup>2</sup>. Essentially all of the tested fibers experienced a 'laser conditioning' process at the front fiber face, in which a visible plasma was generated for one or more laser shots. Rather than produce progressive damage at the front surface, however, this process would eventually cease and leave the surface with improved damage resistance. Once past this conditioning process, the majority of fibers damaged at the rear end face. Other modes of damage were observed either at locations of fixturing stresses or at a location of high static tensile stress resulting from bends introduced to the fiber. Although the previous results were encouraging in terms of achieving useful damage thresholds, a number of areas for further study were indicated. In the present study, a similar experimental procedure was used to address these areas. The relative permanence of front-surface laser conditioning was examined by re-testing fibers that had experienced this process at least a year previously. End-face mechanical polishing was again examined by testing fibers prepared using a refined polishing schedule. Attempts to use a single fixture to hold an entire lot of fibers throughout end-face polishing and damage testing met with mixed results, with fiber positions subjected to fixturing stresses likely sites for initial damage. In an effort to prepare fiber faces with the improved damage resistance observed with front faces following 'laser conditioning,' two schedules for CO<sub>2</sub>-laser polishing of end faces were developed and evaluated. Finally, to improve resistance to damage at sites with significant static stresses, fiber samples which passed a much higher tensile proof test during manufacturing were tested. The current experiments were conducted with a new laser having a shorter pulsewidth and a significantly different mode structure. The beam was injected into the fiber using a geometry that had been successful in the previous study in minimizing a damage mechanism which can occur at the core/cladding interface with the first few hundred fiber diameters. However, the different mode structure of the new laser apparently resulted in this mechanism dominating the current results.