High power erbium-ytterbium co-doped fiber amplifier (EYDFA) has been radiated to the dose of 50krad at the dose rate of 40rad/s. Some key parameters have been measured to investigate the radiation effect on the EYDFA for space optical communication. Considering the dose of 50krad is big enough to the most of low-dose radiation environment, these experimental results will be a good reference for the low-dose inter-satellite optical communication designers.
© 2009 OSA
Fiber amplifiers are being widely deployed in terrestrial applications for long-line or highly distributed optical communications systems [1,2]. Most recently, the technique also attracted much attention from many space engineers, because of their high signal amplification performance and low weight. Especially, Yb-doped power amplifiers are envisioned to amplify the master laser of the Mars Laser Terminal (MLT) for the Mars Laser Communications Demonstration Project (MLCD), which will demonstrate one possible solution to meet NASA’s future long-haul communication needs [3,4]. However, the fiber amplifiers will face a harsh radiation environment of various types during their stay in orbit.
Considering that erbium-doped fiber amplifiers (EDFAs) are the most popular amplifiers for the ground optical communication systems, almost all the studies are focus on the radiation effect on this type of fiber amplifiers [5–8]. However, the space optical communication is different from the ground, because there are no repeaters between the optical transmitter and receiver. Therefore, high output power is the most important technical index for space optical communication. The maximum output power of commercial EDFAs is usually only about 30dBm. With the development of new space optical communication systems, EDFAs are much more difficult to satisfy the increasing optical power of new transmitters. While commercial erbium-ytterbium co-doped fiber amplifiers (EYDFAs) can provide a maximum output power of 40dBm. That will be a new way to resolve the problem of the output power for space optical communication systems.
Many research groups have note the disadvantage of EDFAs, however, little effort is spent on the study of radiation effect on EYDFAs. There are two reasons for that. Firstly, many important space optical communication programs are carried out by NASA (National Aeronautics and Space Administration) [9,10] and ESA (European Space Agency) [11–13] many years ago. And considering that the EDFA is the most mature technique of amplification in that time, many studies are focus on EDFAs for the requirements of these programs. Secondly, any type of fiber amplifiers cannot work normally again after radiation, because the radiation experiment is a failure experiment. Meanwhile, EYDFA is much more expensive than EDFA. So the cost of radiation experiment is another important problem for the study on the EYDFA. In this paper we will report the experimental results about the radiation effect on high power EYDFA in low dose environment. Some key parameters of EYDFA have been measured in the radiation experiment, such as central wavelength, half width, gain, noise figure (NF) and so on. The experimental results will be a good reference for the space communication designers.
2. Experiment set-up
The purpose of radiation experiment is to evaluate the deterioration characteristics of EYDFA for the actual space optical communication systems in low dose environment. So it is necessary to ensure the equivalence of radiation effect between the space environment and the ground. Previous studies have proved the gamma radiation, which is used for researches widely, can simulate the space radiation very well . In this viewpoint, 60Co is also used as the radiation source in our radiation experiment.
Figure 1 is the experimental apparatus for measuring the radiation effect on EYDFA. The light source is a tunable laser diode. And the peak wavelength is set to 1550.13nm by computer. The input power of signal can be controlled by the variable optical attenuator. And the deterioration characteristics of EYDFA will be measured by optical spectrum analyzer and lightwave multimeter. Considering that the measurement instruments will be also affected by gamma radiation, just EYDFA is put in the radiation chamber, while the other measurement instruments are in the testing chamber. These two parts can be connected by 15m fibers and serial port line RS232 during the testing conveniently.
The EYDFA is a benchtop C-band amplifier. And the configuration of EYDFA is shown in Fig. 2 . From the figure, it is composed of a key OEM (Original Equipment Manufacturer) amplifier and other matching equipment. There are two layers in the OEM amplifier in fact. Erbium-ytterbium co-doped fiber (EYDF) and some optical components are in the top layer. While the pump laser diode and circuit board are in the bottom layer. Considering many functions of benchtop amplifier will be not used in space optical communication, only the OEM amplifier is exposed to the radiation environment in our radiation experiment. And the other parts of benchtop amplifier will be covered by lead brick in the radiation chamber. In order to ensure the accuracy and uniformity of the radiation effect on the OEM amplifier, we also make two special bed plates to make the EYDFA stand vertically to the radiation source in the experiment. That is also shown in Fig. 2.
3. Experiment results and discussion
60Co is used as the radiation source in experiment. EYDFA is radiated to the radiation dose of 50krad at the dose rate of 40rad/s. Therefore, the EYDFA will be radiated by 60Co for 1250 seconds in the whole radiation experiment. The dose of 50krad is a typical dose level of low dose space radiation environment for several years. And the dose of 10krad, 20krad, 30krad, 40krad and 50krad are chosen as the break points in radiation experiment. The radiation will stop when each break point is arrived. Then the characteristics parameters of EYDFA will be measured in about 5 minutes.
Considering that many optical instruments are sensitive to the peak wavelength and half width of signal, these two critical parameters have been measured at each break point. Table 1 shows the radiation experimental results of peak wavelength and half width at the different dose. And the resolution of the optical spectrum analyzer is set to 0.01nm in radiation experiment. The peak wavelength and the half width of input signal show a little fluctuation characteristics because of the tunable laser diode. However, it is clear that the peak wavelength and the half width of input signal will not change after the low dose radiation. The experimental results indicate that the radiation effect on EYDFA chiefly introduces the transmission loss, but the physical mechanics of amplification is not changed by radiation. That is the same as the EDFA . Therefore, the results assure the precondition of the application for EYDFA in space optical communication systems.
The gain and noise figure (NF) are another two important characteristic parameters of EYDFA. The input power keeps 0dBm by variable optical attenuator during the measurement. And the output power of the EYDFA is 24.20dBm before radiation. Figure 3 shows the main characteristics parameters of EYDFA as a function of radiation dose from 0rad to 50krad. The gain and NF of EYDFA as a function of radiation dose are shown in Fig. 3(a) and Fig. 3(b) respectively. From the figures, the gain of EYDFA comes down 25.08dB, and the NF climbs up 3.84dB after the radiation dose of 50krad. It is obvious that the gain and NF deteriorate evidently in our radiation experiment. Especially, we also further find that the deterioration of gain represents a good linearity like the variation of the EDFA in previous researches , while the regularity of NF is not obvious. Considering the linear variation of gain, the deterioration of gain is about 0.50db/krad in our experiment. In last paragraph, we have confirmed that the radiation effect on the EYDFA dose not change the physical mechanics of amplification. It means that some color centrals, which can increase the transmission loss, has also formed in the EYDF like the EDFA during the radiation process. However, the deterioration characteristics of EYDFA are much bigger than EDFA’s under the same radiation condition. The possibility is that the extra Yb3+ will introduce more quantity of color central in the YEDF in radiation process. Unfortunately, we cannot further give the specific quantity of the color central formed in EYDFA, because we do not have the condition of electron-spin-resonance (ESR) experiment.
It is also should note that OEM amplifier is composed of the EYDF, optical isolators, the WDM optical couplers, pump laser diode and the circuit board. Therefore, the deterioration of gain and NF showed in Fig. 3(a) and Fig. 3(b) should be ascribed to all the components in fact. In this viewpoint, the radiation effect on the other components should be also investigated further here. The deterioration of the optical isolators and the WDM optical couplers can be neglected usually . However, less studies of the radiation effect on the circuit board and the pump laser diode have been reported. In order to evaluate the radiation effect on these two components, some other parameters have been further measured here. Figure 3(c) shows the monitored and measured output power as a function of radiation dose. It is clear that these two results are almost the same. It indicates that the monitoring pin of OEM amplifier can be operated normally after the radiation dose of 50krad. In another word, the monitor circuit can still provide the function of ACC (Automatic Current Control) or APC (Automatic Power Control) in this radiation environment.
In order to make sure the degree of the radiation effect on the pump laser. The variation of the current of pump laser has also been measured. From the Fig. 3(d), the current of pump laser comes down 350mA after the radiation dose of 50krad. The results confirm that the output power of the pump laser diode will deteriorate to some degree during the radiation. That is consistent with the previous researches [15,16]. However, it should also be noted that most of the EYDFAs are the high output power amplifiers in the space optical communication systems. As our experimental EYDFA, its pump current is about 5000mA when the EYDFA is in normal operation. Therefore, the drop of 350mA cannot make a big effect on the gain of EYDFA. That means there is no doubt that the radiation effect on the EYDFA can be ascribed to the factor of EYDF chiefly.
Another phenomenon should be also noted for the designers. That is an interesting phenomenon which we find in the radiation experiment accidentally. At the initial experimental program, we also intend to monitor the parameters of EYDFA by remote operator panel when the amplifier is radiated. However, once the radiation starts, the EYDFA will shut off immediately. Furthermore, the remote function of EYDFA can recover normally as usual when the radiation stops. We have tried several times in the experiment. The possible reason could be that the remote control circuit of EYDFA is sensitive to the radiation dose rate but not the total radiation dose. Therefore, although we do not make sure the relationship between the phenomenon and the dose rate, the designers should better to test this remote function at the real space radiation dose rate before the function is used.
The recovery experiment has also been carried out to measure the recovery characteristics of EYDFA with the time. Considering the dose rate is 40rad/s in radiation experiment, it is much bigger than the real space dose rate. It means we only spend very short time to achieve the same radiation dose of orbit for several years. Previous researches have shown that the fiber has an ability of anneal with time . So the experimental results are bigger than the real condition in fact. The recovery experiment is carried out in three hours later, because we have to go back to our laboratory. Figure 4(a) and Fig. 4(b) show the recovery of gain and NF in twenty-four hours after the radiation experiment. Both of them will come to steady stage gradually after several hours. It is obvious that the time cost by NF is less than the gain’s. From the Fig. 4(a) and Fig. 4(b), the gain grows up about 8dB, while the NF comes down about 1dB. Considering the correction of the results of recovery experiment, the deterioration of gain should be about 17.08db, while the deterioration of NF is about 2.84dB in the real space orbits. That will be a good correction for the designers. Figure 4(c) shows the recovery of pump current with the time. The little change indicates that the circuit of pump will not recover with the time.
Considering the requirements of actual space optical communication systems, the deterioration is a little big after the radiation dose of 50krad. Furthermore, since the dose which EYDFAs will face is less than 50krad in many low-dose orbits, the designers should be better to evaluate the deterioration of the EYDFA with our experiment results according to the total dose of actual orbit carefully. And the designers should better note the linear variation characteristics of gain in low dose radiation environment for evaluation. If the characteristics affected by the radiation cannot satisfy the requirements of the real systems, the EYDFA should be made some special radiation protection.
We have studied the radiation effects on the EYDFA by experiment for space optical communication. The EYDFA has been radiated to the total radiation dose of 50krad at the dose rate of 40rad/s in the paper. The no change of peak wavelength and half width ensures the precondition of the application for EYDFA in space communication systems. Furthermore, the experimental results show that the radiation effect on the EYDFA can be chiefly ascribed to the factor of EYDF. Considering that the deterioration is a little big, and the dose which EYDFAs will face is less than 50krad in many low-dose orbits, the designers should evaluate the deterioration of the EYDFA with our experiment results according to the total dose of actual orbit carefully. And if the characteristics of radiated EYDFA cannot satisfy the requirements of the actual systems, the EYDFA should be made some special radiation protection. Since we just have one sample of EYDFA, some comparative experiments cannot be carried out further. However, the experimental results will also be a good reference for the space optical communication designers. Considering the importance of EYDFA in space optical communication, more effort is further needed to study the radiation effect on EYDFA both in experiment and theory in future.
References and links
1. Y. Mochida, N. Yamaguchi, and G. Ishikawa, “Technology-oriented review and vision of 40-Gb/s-based optical transport networks,” J. Lightwave Technol. 20(12), 2272–2281 (2002).
2. M. Lopez-Amo, L. T. Blair, and P. Urquhart, “Wavelength-division-multiplexed distributed optical fiber amplifier bus network for data and sensors,” Opt. Lett. 18(14), 1159–1161 (1993). [PubMed]
3. D. M. Boroson, A. Biswas, and B. L. Edwards, “MLCD: Overview of NASA’s mars laser communications demonstration system,” Proc. SPIE 5338, 16–28 (2004).
4. H. Hemmati, A. Biswas, and D. M. Boroson, “30-dB data rate improvement for interplanetary laser communication,” Proc. SPIE 6877, 687707–1-687707–8 (2008).
5. G. M. Williams and E. J. Friebele, “Space radiation effects on erbium-doped fiber devices: sources, amplifiers, and passive measurements,” IEEE Trans. Nucl. Sci. 45(3), 1531–1536 (1998).
6. G. M. Williams, B. M. Wright, W. D. Mack, and E. J. Friebele, “Projecting the performance of erbium-doped fiber devices in a space radiation environment,” Proc. SPIE 3848, 271–280 (1999).
7. O. Berne, M. Caussanel, and O. Gilard, “A Model for the prediction of EDFA gain in a space radiation environment,” IEEE Photon. Technol. Lett. 16(10), 2227–2229 (2004).
8. Y. Chigusa, M. Watanabe, M. Kyoto, M. Ooe, T. Matsubara, S. Okamoto, T. Yamamoto, T. Iida, and K. Sumita, “γ-ray and neutron irradiation characteristics of pure silica core single mode fiber and its life time estimation,” IEEE Trans. Nucl. Sci. 35(1), 894–897 (1988).
9. E. Korevaar, R. J. Hofmeister, and J. Schuster et al.., “Design of satellite terminal for BMDO lasercom technology demonstration,” Proc. SPIE 2381, 60–71 (1995).
10. J. Shoemaker, P. Brooks, E. Korevaar, G. Arnold, A. Das, J. Stubstad, and R. G. Hay, “The space technology research vehicle (STRV) −2 program,” Proc. SPIE 4136, 36–47 (2000).
11. B. Laurent and O. Duchmann, “The Silex Project: The first European optical intersatellite link experiment,” Proc. SPIE 1417, 2–12 (1991).
12. R. Craig, B. Li, and B. Chan, “Laser qualification for the SILEX program,” Proc. SPIE 2123, 238–242 (1994).
13. T. Nielsenetal, “In orbit test results of the first SILEX termina,” Proc. SPIE 3615, 31–42 (1999).
14. T. S. Rose, D. Gunn, and G. C. Valley, “Gamma and proton radiation effect in erbium-doped amplifiers: Active and passive measurements,” J. Lightwave Technol. 19(12), 1918–1923 (2001).
15. H. Ohyama, E. Simoen, C. Claeys, T. Hakata, T. Kudou, M. Yoneoka, K. Kobayashi, M. Nakabayashi, Y. Takami, and H. Sunaga, “Radiation-induced lattice defects in InGaAsP laser diodes and their effects on device performance,” Physica B 273–274(1-3), 1031–1033 (1999).
16. S. L. Waterhouse, and K. K. Jobbins, “Radiation effects on laser diodes: A literary review,” Proc. SPIE 7070, 70700E–1–70700E–10 (2008).
17. W. C. Goitsos, “Radiation-induced loss studies in Er-doped fiber amplifier systems,” Proc. SPIE 2699, 304–309 (1996).