Previous research showed that mid-infrared free-electron lasers could reproducibly ablate soft tissue with little collateral damage. The potential for surgical applications motivated searches for alternative tabletop lasers providing thermally confined pulses in the 6- to-7-μm wavelength range with sufficient pulse energy, stability, and reliability. Here, we evaluate a prototype Raman-shifted alexandrite laser. We measure ablation thresholds, etch rates, and collateral damage in gelatin and cornea as a function of laser wavelength (6.09, 6.27, or 6.43 μm), pulse energy (up to 3 mJ/pulse), and spot diameter (100 to 600 μm). We find modest wavelength dependence for ablation thresholds and collateral damage, with the lowest thresholds and least damage for 6.09 μm. We find a strong spot-size dependence for all metrics. When the beam is tightly focused (∼100-μm diameter), ablation requires more energy, is highly variable and less efficient, and can yield large zones of mechanical damage (for pulse energies >1 mJ). When the beam is softly focused (∼300-μm diameter), ablation proceeded at surgically relevant etch rates, with reasonable reproducibility (5% to 12% within a single sample), and little collateral damage. With improvements in pulse-energy stability, this prototype laser may have significant potential for soft-tissue surgical applications.
Purpose: It is hypothesized that 6.1 μm produced by a portable laser would be useful for incising tissue layers such as performing a retinectomy in detached retina with extensive anterior proliferative vitreoretinopathy.
Methods: An alexandrite laser system, which provides a high-intensity Q-switched pulse (780 nm, 50-100 ns duration, 10 Hz), is
wavelength-shifted by a two-stage stimulated Raman conversion process into the 6-7 μm range (Light Age, Inc.). Fresh cadaver porcine retinas were lased with 6.1 μm using a 200 μm diameter spot at 0.6 mJ after removal of the vitreous. Specimens were examined grossly and prepared for histological examination.
Results: The Raman-shifted alexandrite laser produced a smooth Gaussian profile. A narrow spectrum was produced at 6.1 μm. A full-thickness retinal incision with minimal thermal damage was obtained at a low energy level of 0.6 mJ in the retinas. However, the depth of the incision did vary from an incomplete incision to a full-thickness incision involving the underlying choroidal layer in
attached retinas. Conclusions: The 6.1 μm mid-infrared energy produced by a portable laser is capable of incising
detached retinas with minimal thermal damage.
The Mark-III Free Electron Laser (FEL), tuned to λ=6.45 μm has been demonstrated to provide for efficient ablation in ocular and neural tissues with minimal collateral damage. To date, the role of the FEL pulse structure on the mechanism of ablation has not been determined. In an effort to study the role of the FEL micropulse on the ablation of corneal tissue, the native pulse structure of the FEL, a 2.85 gigahertz repetition of picosecond pulses within a five microsecond macropulse envelope, was changed using a a pulse stretcher. This device changes the duration of the micropulse from 1 picosecond to 30-200 picoseconds in length, thus reducing the peak intensity of the micropulse by as much as 200x the original intensity, while the macropulse energy remains unchanged.
Two basic metrics were studied: the ablation threshold on water and the ablation crater depth on gelatin. These metrics were employed at λ=6.45 and 6.1 μm for 1, 100, and 200 picoseconds in micropulse duration. The results showed a very slight difference between the 1, 100, and 200 picosecond micropulse duration, given a 200 fold decrease in peak energy for both the threshold and crater depth measurements. Brightfield imaging was also performed to probe the ablation dynamics and showed no difference between the 1 and 200 ps micropulses.
The effect of changing the micropulse duration was studied on the ablation of canine cornea. Craters (500 micron diameter) were created with 25 pulses at three times the ablation threshold as determined for water on freshly enucleated corneas within 12 hours of removal. Three rows of seven craters were created on the center of each cornea. The native one picosecond micropulse and 200 picosecond stretched micropulse were compared at λ=6.1 and 6.45 μm. Histological data shows that less thermal damage is present at 6.1 μm compared with 6.45 μm; however, there is no significant difference between the native and stretched pulses with respect to thermal damage.
The Mark-III Free Electron Laser (FEL), tuned to 6.45 microns in wavelength has been demonstrated to provide for efficient ablation in ocular, neural, and dermal tissues with minimal collateral damage. To date, the role of the unique pulse structure of the FEL on the ablation mechanism has not been determined. In this study, the native pulse structure of the FEL, a 2.85 gigahertz repetition of picosecond pulses within a five microsecond macropulse envelope, was changed using a pulse stretcher. This device changes the duration of the micropulse from its native one picosecond to 30-200 picoseconds in length, thus reducing the peak intensity of the micropulse down to 1/200th of the original intensity, while the macropulse energy remains unchanged.
Two basic metrics were studied: the ablation threshold on water and mouse dermis and the ablation crater depth on gelatin and mouse dermis. These metrics were employed at 6.45 and 6.1 microns in wavelength for 1, 100, and 200 picoseconds in micropulse duration. In addition, bright-field imaging was used to compare the ablation dynamic between 1 ps and 200 ps micropulses on water at 6.1 and 6.45 microns. The effect of changing the micropulse duration was also studied on the ablation of mouse dermis for histological analysis. Craters (500 micron diameter) were created with 25 pulses at three times the ablation threshold as determined for mouse dermis within 8 hours of removal. Three rows of twenty craters were created on each piece of mouse dermis for a given parameter set. The native one picosecond micropulse and 200 picosecond stretched micropulse were compared at 6.1 and 6.45 microns in wavelength. There was no difference seen between the native 1 ps micropulse and the stretched micropulse durations with respect to the ablation threshold, efficiency, dynamics, and thermal damage.
Pulsed mid-infrared (6.45 μm) radiation has been shown to cut soft tissue with minimal collateral damage (<40 mm); however, the mechanism of ablation has not been elucidated to date. The goal of this research was to examine the role of the unique pulse structure of the Vanderbilt Mark-III FEL and its role in the efficient ablation of soft tissue with minimal collateral damage. The pulse structure consists of a 2.865 GHz train of one picosecond micropulses within a 4-5 μs macropulse envelope operated between 2 and 30 Hz. The effect of the picosecond micropulses was examined by running the native FEL pulse structure through a pulse stretcher in order to increase the micropulse length from 1 picosecond up to 100 picoseconds. This allowed us to determine whether or not the picosecond train of micropulses played any role in the ablation process. The pulse stretcher was varied between 1, 30, 60, and 100 picoseconds. The ablation threshold was determined for water and mouse dermis for each micropulse length using PROBIT analysis of 100 individual observations of the macropulse. The results of the analysis showed no statistical difference between 1 and 100 picoseconds.
The ablation efficiency was also measured on 90% w/w gelatin and mouse dermis for the different micropulse lengths. Multiple ablation craters were made by varying the number of pulses delivered between 5 and 500. The ablated crater depth was measured using OCT. No significant difference was observed between 1 and 60 picoseconds; however, the 100 picosecond micropulse did show a reduction in the efficiency of ablation. We have shown that the effect of micropulse duration of the FEL on the ablation process is negligible between 1 and 100 picoseconds. Further analysis is needed beyond 100 picoseconds.
The pulse train from a Mark III FEL tuned to a wavelength of 6.45 microns has been shown to be efficient at ablating soft tissue with minimal collateral damage. This laser has a unique pulse structure consisting of a train of 1ps micropulses spaced 350ps apart, which is maintained for 4-5 microseconds (the macropulse) and is repeated at 1-30Hz. We are investigating the role of the pulse structure in the ablation mechanism. In order to determine the importance of non-linear effects potentially induced by the high peak power of the micropulses, we are using a grating pulse stretcher optimized for 6.45 microns to vary the micropulse duration while maintaining the macropulse duration and micropulse frequency. The technique allows use of the same pulse energy and average power with widely variable peak power. Ablation thresholds were measured using PROB-IT analysis and crater depths were measured using OCT imaging. In water, gelatin, and mouse dermis, we have found no statistically significant difference in the ablation threshold of pulses having widths of 1, 30, 60, and 100ps. The measured ablation efficiency of mouse dermis also showed no significant difference over the same range of pulse widths. This data suggests that the ablation characteristics obtained with the FEL at 6.45 microns are independent of the micropulse duration and do not rely on the high peak power of the FEL pulse train.
A fast electron energy spectrometer has been built using a photodiode array measuring the backward optical transition radiation from a thin film of aluminum. The resolution of the electron energy spectrometer is about 0.2% with a time resolution of 50 ns. The maximum energy spread that can be measured is 6.4%. We present the measurements of the time-resolved electron beam energy spectrum on the Mark III linear accelerator at Vanderbilt University, while lasing at different wavelengths and while not lasing. We also discuss the effects of different parameters, such as cathode heating, alpha magnet strength and RF phase, on the electron energy spectrum and optical spectrum. The diagnostics of time-resolved electron energy spectrum and time-resolved laser spectrum provide the technology to understand the physical process of the FEL interaction. Based on these diagnostics, the FEL facility can realize some special modes of operation, such as macropulse chirping and macropulse two color lasing.
The W.M. Keck-Vanderbilt Free-electron Laser Center operates a reliable free-electron laser (FEL) that is used in human surgical trials, as well as in basic and applied sciences. The wavelength of the FEL is tunable from 2.1 micrometers to 9.6 micrometers , delivering above 50 mJ per macropulse with a repetition rate of 30 Hz. For soft tissue surgery, especially neurosurgery and surgery on the optic nerve, a wavelength of 6.45 micrometers has been found to ablate with little collateral damage. The free-electron laser beam is delivered to experiments approximately 2000 hours each year. The Center also supports several other tools useful for biomedical experiments: an optical parametric generator laser system with tunable wavelength similar to the free- electron laser except it has much lower average power; a Fourier transform infrared spectrometer to characterize samples; several devices for in vivo imaging including an optical coherence tomography setup, a two-photon fluorescent confocal microscope, and a cooled, integrating camera capable of imaging luciferin-luciferase reactions within the body of a mouse. The Center also houses a tunable, monochromatic x-ray source based on Compton backscattering of a laser off of a relativistic electron beam.
The Vanderbilt Mark III FEL is a tunable source of coherent mid-infrared radiation occurring as a train of high- intensity (picosecond) micropulses with a repetition rate of 3GHz that continues for 3-5 microseconds (the macropulse). We have measured the spectral output of the Vanderbilt FEL as a function of time during the macropulse with ~10nm resolution in wavelength and ~20ns resolution in time. The measurement takes about one minute and gives a representation of the micropulse spectral width average over many macropulses. Data collected thus far indicates a surprising amount of structure produced by overlapping periods of growth, saturation, and decay within the macropulse. It is found that the central wavelength of the FEL slips over the course of the macropulse, and that the instantaneous output typically has a much smaller spectral bandwidth than the macropulse bandwidth. Thus, a user slicing portion fo the macropulse with a Pockel's Cell can obtain different central wavelengths by slicing at different times during the macropulse. The evolution of the macropulse spectrum as a function of cavity de-tuning and electron beam parameters is studied with the goal of improving the stability and spectral density of the FEL output.
We have investigated the experimental consequences of two picosecond infrared lasers, both tuned to 6.45micrometers and focused on ocular tissue. The exposure conditions were comparable, other than pulse repetition rate, where an optical parametric oscillator/amplifier laser (OPA) system operates at a kilohertz and the Mark-III FEL at 3 gigahertz. In both cases, the peak intensity was near 2x10<SUP>14</SUP> W/m<SUP>2</SUP> and the total delivered energy was approximately 125 mJ. The Mark-III consistently ablates tissue, while the OPA fails to ablate or to damage corneal tissue. In particular, there is no experimental evidence for protein denaturation due to OPA irradiation. WE account for these observations in terms of a theoretical model based on thermal diffusion and threshold conditions for superheating and chemical kinetics. We comment on the relevance of tissue geometry.
The Vanderbilt Mark III FEL is a tunable source of high- intensity coherent mid-infrared radiation occurring as a train of picosecond pulses spaced 350ps apart. The laser beam is transported to each laboratory under vacuum, but is typically transmitted through some distance of atmosphere before reaching the target. Losses due to absorption by water vapor and CO<SUB>2</SUB> can be large, and since the bandwidth of the FEL is several percent of the wavelength, the spectrum can be altered by atmospheric absorptions. In order to provide an accurate representation of the laser spectrum delivered to the target, and to investigate any non-linear effects associated with transport of the FEL beam, we have recorded the spectrum of the FEL output using a vacuum spectrometer positioned after measured lengths of atmosphere. The spectrometer is equipped with a linear pyroelectric array which provides the laser spectrum for each pulse. Absorption coefficients are being measured for laboratory air, averaged over the bandwidth of the FEL. The high peak powers of this Fel have induced damage in common infrared-transparent materials; we are also measuring damage thresholds for several materials at various wavelengths.
Tunable, pulsed radiation sources in the ultraviolet, visible, and infrared wavelength ranges offer novel opportunities for investigating laser-induced biomedical effects. Free-electron lasers (FELs) deliver continuously tunable, pulsed radiation in the infrared, providing the capability to selectively target radiation into the vibrational modes of water or other biopolymers. Experimental techniques for measuring the absorption spectra of biological samples are described. These spectra indicate wavelengths that potentially serve as the basis for laser-induced biomedical effects. Some practical considerations for infrared, visible, and UV spectroscopy of biological samples are summarized, and the connection between biomedical research and more fundamental investigations of vibrational energy transfer are emphasized.
Free-electron lasers (FELs) provide tunable, pulsed radiation in the infrared. Using the FEL as a pump beam, we are investigating the mechanisms for energy transfer between localized vibrational modes and between vibrational modes and lattice or phonon modes. Either a laser-Raman system or a Fourier transform infrared (FTIR) spectrometer will serve as the probe beam, with the attribute of placing the burden of detection on two conventional spectroscopic techniques that circumvent the limited response of infrared detectors. More specifically, the Raman effect inelastically shifts an exciting laser line, typically a visible frequency, by the energy of the vibrational mode; however, the shifted Raman lines also lie in the visible, allowing for detection with highly efficient visible detectors. With regards to FTIR spectroscopy, the multiplex advantage yields a distinct benefit for infrared detector response.
Our group is investigating intramolecular and intermolecular energy transfer processes in both biopolymers and more traditional materials. For example, alkali halides contain a number of defect types that effectively transfer energy in an intermolecular process. Similarly, the functioning of biopolymers depends on efficient intramolecular energy transfer. Understanding these mechanisms will enhance our ability to modify biopolymers and materials with applications to biology, medecine, and materials science.