Lasers in Medical Science

, Volume 19, Issue 3, pp 155–160

Transmission of Q-switched erbium:YSGG (λ=2.79 μm) and erbium:YAG (λ=2.94 μm) laser radiation through germanium oxide and sapphire optical fibres at high pulse energies

Authors

    • Department of UrologyJohns Hopkins Medical Institutions
  • Yubing Yang
    • Department of UrologyJohns Hopkins Medical Institutions
  • Charles A. Chaney
    • Department of UrologyJohns Hopkins Medical Institutions
  • Daniel Fried
    • Department of Restorative DentistryUniversity of California
Original Article

DOI: 10.1007/s10103-004-0316-8

Cite this article as:
Fried, N.M., Yang, Y., Chaney, C.A. et al. Lasers Med Sci (2004) 19: 155. doi:10.1007/s10103-004-0316-8

Abstract

The erbium:YSGG and erbium:YAG lasers are used for tissue ablation in dermatology, dentistry and ophthalmology. The purpose of this study was to compare germanium oxide and sapphire optical fibres for transmission of sufficient Q-switched erbium laser pulse energies for potential use in both soft and hard tissue ablation applications. Fibre transmission studies were conducted with Q-switched (500 ns) Er:YSGG (λ=2.79 μm) and Er:YAG (λ=2.94 μm) laser pulses delivered at 3 Hz through 1-m-long, 450-μm germanium oxide and 425-μm sapphire optical fibres. Transmission of free-running (300 μs) Er:YSGG and Er:YAG laser pulses was also conducted for comparison. Each set of measurements was carried out on seven different sapphire or germanium fibres, and the data were then averaged. Fibre attenuation of Q-switched Er:YSGG laser energy measured 1.3±0.1 dB/m and 1.0±0.2 dB/m for the germanium and sapphire fibres, respectively. Attenuation of Q-switched Er:YAG laser energy measured 0.9±0.3 dB/m and 0.6±0.2 dB/m, respectively. A maximum Q-switched Er:YSGG pulse energy of 42 mJ (26–30 J/cm2) was transmitted through the fibres. However, fibre tip damage was observed at energies exceeding 25 mJ (n=2). Both germanium oxide and sapphire optical fibres transmitted sufficient Q-switched Er:YSGG and Er:YAG laser radiation for use in both soft and hard tissue ablation. This is the first report of germanium and sapphire fibre optic transmission of Q-switched erbium laser energies of 25–42 mJ per pulse.

Keywords

ErbiumEr:YAGEr:YSGGGermaniumQ-switchedSapphire

Introduction

The Q-switched erbium:YAG (λ=2.94 μm) and Er:YSGG (λ=2.79 μm) lasers have been studied experimentally for precision tissue ablation in several medical specialities, including ophthalmology [16] and dentistry [710]. Peripheral thermal damage caused in tissue during Er:YAG and Er:YSGG laser ablation may be reduced from 10 μm to 50 μm in the long-pulse, free-running mode to only 5–10 μm in the short-pulse, Q-switched mode [11]. The laser fluence necessary for rapid and efficient ablation of a variety of hard and soft tissues has been previously reported for the Q-switched Er:YAG and Er:YSGG lasers. For soft tissues, such as skin, cornea, and aorta, a laser fluence of 1–5 J/cm2 is sufficient [11]. However, for hard tissues, such as bone, dentin, and enamel, a higher laser fluence of 20–40 J/cm2 may be necessary, and studies have demonstrated that shorter laser pulses provide more efficient ablation with fewer thermal effects [8, 12].

The replacement of a large, rigid articulated arm with a small, flexible optical fibre could prove beneficial in medical applications, where operating space for access of surgical instruments is limited (e.g. ophthalmology and dentistry). Several mid-infrared optical fibre delivery systems are commercially available, with the most promising for erbium laser delivery being sapphire, germanium oxide, and hollow waveguides [13]. The sapphire [1422] and germanium oxide [2327] fibres have been studied extensively for transmission of free-running erbium laser pulses, but little work has been performed that demonstrates short-pulse, Q-switched transmission. Previous studies that demonstrated fibre transmission of Q-switched erbium laser radiation have been conducted at only relatively low pulse energies of less than 10 mJ [2327]. This pulse energy may be insufficient for certain medical applications, especially those involving hard tissue ablation (e.g. dentistry).

Preliminary studies in our laboratories have demonstrated increased fibre optic transmission of Q-switched erbium:YSGG laser radiation using germanium oxide and sapphire optical fibres [28, 29]. The purpose of this study is to compare germanium oxide and sapphire optical fibres of similar size and length for transmission of both free-running and Q-switched Er:YSGG and Er:YAG mid-infrared laser radiation at high pulse energies sufficient for both soft and hard tissue ablation.

Materials and methods

A Schwartz Electro-Optics SEO 1-2-3 laser, operated with either an erbium:YSGG (λ=2.79 μm) or an erbium:YAG (λ=2.94 μm) laser rod, was used for the experiments. The laser was operated in free-running mode with a pulse length of 300 μs and in short-pulse mode with a rotating mirror Q-switch (Shiva Laser, Los Angeles, Calif., USA) producing 500 ns pulse lengths (Fig. 1). The Q-switched laser produced Er:YSGG energies up to 60 mJ per pulse and Er:YAG energies up to 25 mJ per pulse, at a repetition rate of 3 Hz. A single-mode Gaussian laser spatial beam profile was used in all experiments. The laser pulse–pulse stability was within ±5% for the free-running mode and ±10% for the Q-switched mode operation. The laser pulse energy was measured with a pyroelectric detector (Molectron, Model EPM 1000, Portland, Ore., USA) and the temporal pulse length was measured with a photovoltaic infrared detector (PD-10.6, Boston Electronics, Brookline, Mass., USA).
Fig. 1

Temporal pulse profile for the a free-running and b Q-switched Er:YSGG laser

The laser radiation was focused with a 50-mm focal length calcium fluoride lens into either 450-μm-core germanium oxide optical fibres (Infrared Fiber Systems, Silver Spring, Md., USA) or 425-μm-core sapphire fibres (Photran, Amherst, N.H., USA). The input end of the fibres was placed at the focal point of the lens with a laser spot diameter of 340 μm. All the sapphire and germanium fibres tested in this study measured 1 m in length. Light microscopy was used to examine the input ends of the germanium and sapphire fibre tips for evidence of damage after high power Q-switched erbium:YSGG laser transmission studies.

Fibre attenuation comparisons were made either between sapphire and germanium fibres, for a given laser and pulse length, or between the Er:YAG and Er:YSGG laser, for a given fibre and pulse length. A linear fit to the data points provided the average percent transmission of the fibres over the entire input energy range (including the losses due to Fresnel reflection). This percent transmission measurement was then corrected by subtraction of the Fresnel losses, and the attenuation was calculated in decibels per metre. Statistical analyses were performed with the paired Student’s t-test. Differences were considered significant at a level of P<0.05. The data points for each graph represent the average of seven independent measurements from seven different fibres, and the error bars represent the minimum and maximum values recorded.

Results

Erbium:YSGG laser

The transmission of free-running and Q-switched Er:YSGG laser radiation through sapphire and germanium oxide optical fibres is shown in Fig. 2. Fibre optic transmission of free-running Er:YSGG laser radiation averaged 76±2% and 88±6% for the germanium and sapphire fibres (n=7), respectively (P<0.05). It should be noted that much of the loss was due not to transmission but, rather, to reflection at the fibre ends. Both the germanium and sapphire fibres have high indices of refraction. For sapphire the index of refraction is 1.71, which results in reflection losses of 6.9% at each end. The germanium fibre has an index of refraction of 1.84, which results in Fresnel losses of 8.7%, in comparison with only 4% reflection losses from silica fibres (ni=1.4) [13]. After correction for those Fresnel losses, attenuation through the germanium and sapphire fibres measured 0.3±0.1 dB/m and 0.2±0.1 dB/m, respectively.
Fig. 2

Transmission of a free-running, 300-μs and b Q-switched, 500-ns Er:YSGG laser pulses through 1-m-long samples of 425-μm sapphire and 450-μm germanium oxide fibres at 3 Hz (seven fibres each)

Fibre optic transmission of Q-switched Er:YSGG laser radiation averaged 57±1% and 65±3% for the germanium and sapphire fibres (n=7), respectively (P>0.05). This corresponded to an attenuation of 1.3±0.1 dB/m and 1.0±0.2 dB/m, respectively. The Q-switched transmission was lower than free-running transmission, overall, for both the fibres, and the data exhibited more variability than the free-running data. This is most likely to be due to two factors. First, operation of the Er:YSGG laser in Q-switched mode resulted in greater pulse-to-pulse energy instability. Second, damage at the input end of several of the germanium (n=2) and sapphire (n=2) fibres was observed at Q-switched pulse energies above 40 mJ (input end) and 25 mJ (output end), respectively. Thus, at the higher energy levels (>40 mJ input), permanent degradation and damage to the input end of the fibre might have occurred, affecting the transmission results. In at least one case a germanium oxide fibre also experienced catastrophic failure, as evidenced by a burning and snapping of the fibre jacket when the laser input energy was increased above 40 mJ. For both the free-running and Q-switched Er:YSGG lasers, the sapphire optical fibre had lower attenuation than the germanium fibre.

Erbium:YAG laser

The transmission of free-running and Q-switched Er:YAG laser radiation through sapphire and germanium oxide optical fibres is shown in Fig. 3. Fibre optic transmission of free-running Er:YAG laser radiation averaged 68±2% and 77±5% for the germanium and sapphire fibres (n=7), respectively (P<0.05). This corresponded to an attenuation of 0.7±0.1 dB/m and 0.4±0.2 dB/m, respectively. It is not a surprising result that the transmission of free-running Er:YSGG laser radiation is higher than that of the free-running Er:YAG laser. For example, the attenuation through germanium fibres is documented to be 0.7 dB/m for the Er:YAG and 0.3 dB/m for the Er:YSGG laser wavelengths [28]. This can be explained by the OH component in both the germanium and sapphire fibres, which is less absorbing at 2.79 μm than at 2.94 μm [19, 21].
Fig. 3

Transmission of a free-running, 300-μs and b Q-switched, 500-ns Er:YAG laser pulses through 1-m-long samples of 425-μm sapphire and 450-μm germanium oxide fibres at 3 Hz (seven fibres each)

Fibre optic transmission of Q-switched Er:YAG laser radiation averaged 64±6% and 74±4% for the germanium and sapphire fibres (n=7), respectively (P<0.05). This corresponded to an attenuation of 0.9±0.3 dB/m and 0.6±0.2 dB/m, respectively. There was no evidence of fibre tip damage during these studies. However, it should be noted that the Q-switched Er:YAG laser output energy was limited to less than half that of the Q-switched Er:YSGG laser. The Q-switched transmission was lower than in free-running mode. However, it is unclear why the fibre attenuation of the Q-switched Er:YAG is lower than that of the Q-switched Er:YSGG. However, if the variability observed in the data at high Q-switched Er:YSGG laser energies is considered, there was no significant difference between the fibre optic transmission of Q-switched Er:YAG and Er:YSGG laser radiation (P>0.05). A summary of the Q-switched erbium laser results for the fibre transmission studies is provided in Table 1. Previously reported free-running erbium laser pulse energies are also provided for comparison [22, 28].
Table 1

Transmission of Er:YSGG and Er:YAG laser radiation through 1-m-length germanium oxide and sapphire optical fibres

Laser

Peak pulse energy (mJ)

Percentage of transmission

Attenuation (dB/m)

Germanium

Sapphire

Germanium

Sapphire

Germanium

Sapphire

Er:YSGG

 Free-running

>2,000

>1,000

76±2

88±6

0.3±0.1

0.2±0.1

 Q-switched

>25

>25

57±1

65±3

1.3±0.1

1.0±0.2

Er:YAG

 Free-running

>2,000 [28]

>1,000 [22]

68±2

77±5

0.7±0.1

0.4±0.2

 Q-switched

>15

>15

64±6

74±4

0.9±0.3

0.6±0.2

Fibre tip damage

The input tips of the fibres damaged during the high-power transmission studies of Q-switched Er:YSGG laser pulses were examined by light microscopy. The germanium oxide fibre tips showed severe damage, characterised by chipping of the fibre tip (Fig. 4a). However, damage to the sapphire fibre tip was limited, to superficial damage characterised by a melting of the fibre tip surface (Fig. 4b). Damage to the input end of sapphire fibres during high-energy, long-pulse erbium laser fibre coupling has also been reported previously and was attributed to the melting of the fibre surface at high pulse energies [22].
Fig. 4

Damage to the input fibre tip caused during high-power coupling of Er:YSGG laser pulses into the fibres. a Side view of the chipped germanium oxide fibre, b axial view of the melted sapphire optical fibre tip

Discussion

Previous studies by Papagiakoumou et al. and Serafetinides et al. (Athens, Greece) have reported the transmission of Q-switched and free-running Er:YAG laser radiation through a variety of mid-infrared delivery systems, including germanium oxide [2325], sapphire [26, 27], and fluoride glass fibres [27, 30] and hollow waveguides [30]. Laser energies approaching 10 mJ of Q-switched Er:YAG laser radiation were successfully coupled into germanium fibres, with attenuation measuring only 0.60–0.75 dB/m, resulting in approximately 8 mJ per pulse transmission through the fibres [2325].

The main purpose of this study was to build upon previous results by testing the Q-switched Er:YSGG and Er:YAG lasers with germanium and sapphire fibres at input energies significantly higher than 10 mJ, to demonstrate that sufficient Q-switched energy can be transmitted for both soft and hard tissue ablation. While less than 10 mJ is sufficient for soft tissue ablation, it might not be adequate for rapid and efficient ablation of hard tissues, as evidenced by previous reports of the ablation rates for bone and dental enamel [8, 12]. Higher incident fluence can be used on hard tissues before the onset of plasma shielding.

In this study we were able consistently to couple up to 40 mJ of Q-switched Er:YSGG laser energy into both the sapphire and germanium fibres without causing fibre damage, which resulted in fibre output energies of approximately 25 mJ per pulse. If the fibre is used in contact mode, this output energy translates into a fluence of 16–18 J/cm2. Laser energies approaching 53 mJ were coupled into some of the fibres, with peak fibre outputs reaching up to 42 mJ per pulse (26–30 J/cm2).

We were limited by the lower laser output energy of the Q-switched Er:YAG laser. However, up to 22 mJ was coupled into the fibres, with peak fibre outputs reaching up to 18 mJ per pulse and no evidence of fibre tip damage. It is likely that even higher Q-switched Er:YAG pulse energies may be transmitted through germanium and sapphire fibres. However, the Q-switched Er:YAG laser used in this study was limited to output pulse energies of 25 mJ or less. In summary, the pulse energies and fluence transmitted through the fibres tested in these studies are sufficient for both soft and hard tissue ablation by Q-switched erbium laser.

Conclusions

Q-switched Er:YSGG laser pulse energies of 25 mJ per pulse were consistently transmitted through both sapphire and germanium oxide optical fibres (n=7), with some fibres transmitting up to 42 mJ per pulse. These results represent a significant increase over previously reported results of less than 10 mJ per pulse and, thus, demonstrate that sufficient Q-switched erbium laser pulse energy can be transmitted for rapid and efficient fibre optic ablation of a variety of both soft and hard tissues.

Acknowledgments

We thank Ken Levin, Dan Tranh, and Alex Tchapyjnikov of Infrared Fiber Systems (Silver Spring, Md., USA) for providing the germanium fibres used in this study. This research was supported, in part, by an NIH phase I SBIR grant awarded to Infrared Fiber Systems: grant no. 1R43 EY13889–01, Department of Defense Prostate Cancer Research Program, grant no. DAMD17-03-0087 and NIH/NIDR grant no. 1-R01 DE14554.

Copyright information

© Springer-Verlag London Limited 2004