Cost-effective low timing jitter passively Q-switched diode-pumped ...

1 downloads 1 Views 84KB Size Report
A novel scheme that combines gain switching with passive Q switching of a miniature diode-pumped solid-state laser is proposed and implemented.

Cost-effective low timing jitter passively Q-switched diode-pumped solid-state laser with composite pumping pulses Jacob B. Khurgin, Feng Jin, Gregory Solyar, Chen-Chia Wang, and Sudhir Trivedi

A novel scheme that combines gain switching with passive Q switching of a miniature diode-pumped solid-state laser is proposed and implemented. A composite pumping pulse, consisting of a long, lowintensity pulse and a following short, high-intensity pulse, is used to reduce the timing jitter. A greater-than-tenfold reduction in timing jitter is demonstrated. © 2002 Optical Society of America OCIS codes: 140.3480, 140.3540, 140.3580, 140.3070, 140.5560.

1. Introduction

Compact all-solid-state lasers play a crucial role in such applications as laser radars and range finders for smart munitions, for which precision, light weight, low power consumption, and ruggedness must be achieved simultaneously. For many applications 共synchronous satellite laser1 ranging, for example兲 it is desirable to have powerful nanosecond pulses whose firing times can be controlled as well. Such pulses can be obtained by Q switching. Passive Q switching with saturable absorbers has the advantage of ruggedness, simplicity, compactness, and good control of transverse mode structure. However, the performance of a passively Q-switched laser can be substantially degraded by timing jitter. The jitter is caused by fluctuations, e.g., in temperature, pump power and wavelength, and intracavity loss, within the system. Essentially, timing jitter in a Q-switched laser is inevitable because the first photon of the oscillation mode comes from spontaneous emission of the gain medium. This initial timing jitter imposes a fundamental lower limit on the timing jitter of all Q-switched lasers. A second limitation derives from the inevitable fluctuations of the

J. N. Khurgin, F. Jin 共[email protected]兲, and G. Solyar are with the Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, Maryland 21218. C.-C. Wang and S. Trivedi are with the Brimrose Corporation of America, 7720 Belair Road, Baltimore, Maryland 21236. Received 7 May 2001; revised manuscript received 25 October 2001. 0003-6935兾02兾061095-03$15.00兾0 © 2002 Optical Society of America

pump. Temperature fluctuation, which affects pump power and wavelength and the absorption coefficient of the gain medium, usually contributes to the long-term drift of the repetition rate. An actively Q-switched solid-state laser has substantially less jitter but requires a number of additional components, which increase the size, weight, complexity, and ultimately the cost of the laser. A number of jitter-reduction schemes that involve combinations of active–passive Q switching2– 4 had been proposed, but all those schemes involved a large quantity of additional optical elements, and the reduction in jitter was not substantial. In this paper we propose and demonstrate a novel scheme to reduce timing jitter that combines gain switching with passive Q switching of a miniature diode-pumped solid-state laser by use of a composite pumping pulse that consists of a long, low-intensity pulse and a following short, high-intensity pulse. First, let us consider a simple model of a passively Q-switched laser. Let us say that the effective pumping rate of the upper laser level is peff ⫽ ␩P兾 h␯L V, where P is the electrical power input; ␩ is the pumping efficiency, which is the product of laserdiode efficiency and optical coupling efficiency; ␯L is the laser frequency; V is the laser mode volume, and h is the Planck constant. Then the equation for the population density of the upper laser level, NG, is given by NG dN G ⫺ c␴ G n p N G, ⫽ p eff ⫺ dt ␶g


where np is the density of photons in the laser mode, ␶g is the lifetime of the upper laser level, ␴G is the 20 February 2002 兾 Vol. 41, No. 6 兾 APPLIED OPTICS


emission cross section of the gain medium, and c is the speed of light. The solution in the absence of lasing 共np ⫽ 0兲 can be written as N G ⫽ p eff␶ g关1 ⫺ exp共⫺t兾␶ g兲兴.


At the same time, the equation for the photon density in the cavity is given by np dn p ⫽ c␴ G n p N G ⫺ c␴ A n p N A ⫺ ⫹ r sp, dt ␶p


where NA is the density of the saturable-absorber atoms 共in the absence of laser light it is NA0兲; ␴A is the effective absorption cross section 共it takes into account the ratio between the volumes of the active mode in the absorber and in the gain medium兲; ␶p is the photon cavity lifetime; and, finally, rsp is the rate of spontaneous emission into the laser mode. We can introduce now the threshold density of the upper 0 laser level, NG,th , with the unsaturated saturable absorber, by setting dnp兾dt ⫽ 0 and assuming that rsp ⫽ 0 共because few photons generated by spontaneous emission contribute to the laser mode that is defined by the cavity mirrors兲: 0 N G,th ⫽

1 ␴A ⫹ N A0 . c␴ G␶ p ␴G


Therefore one can obtain the time when the laser reaches its threshold by replacing NG in Eq. 共2兲 with 0 NG,th , as ␶ d ⫽ ␶ g ln共␣兾␣ ⫺ 1兲,


where ␣⫽

p eff␶ g 0 N G,th


is the ratio of pump power to the threshold pump power. Now, the larger the pump power is, the shorter the delay time. One can find the timing jitter by differentiating Eq. 共5兲: ␶ g ⌬␣ ␶ g ⌬P ⌬␶ d ⫽ ⫽ . 1⫺␣ ␣ 1⫺␣ P


Now, from Eq. 共7兲, one can see that a small change in pump power causes very large change in the delay unless ␣ is very large 关Fig. 1共a兲兴. It is, however, difficult to maintain a pump power that is so large for a prolonged period of time. We propose and demonstrate a new pumping scheme that uses composite pump pulses. Each composite pump pulse consists of two rectangular pulses: a long, low-power pulse followed by a short, high-power pulse 关Fig. 1共b兲兴. The first pump pulse brings the system close to lasing, and the second one raises the upper-level population in a much faster manner and thus triggers the Q-switched pulse. Timing jitter is in reverse proportion to the slope, dNG兾dt, within the window, i.e., peff. By incorporating two pulses into the pump pulse one can make the 1096

APPLIED OPTICS 兾 Vol. 41, No. 6 兾 20 February 2002

Fig. 1. Operating principle of jitter reduction. 共a兲 Dependence of timing jitter on pump rate. 共b兲 Proposed composite pulse pumping. 共c兲 Jitter reduction with the composite pulse.

slope much higher within the lasing window, leading to a substantial reduction in timing jitter. To obtain the proof of concept, we used a composite pulse from two separate lasers rather than one laser because the two-laser arrangement makes the electronics much simpler. As shown in Fig. 2, two laser diodes 共LDs兲 emitting at 808 nm with orthogonal po-

Fig. 2. Q-switched laser setup. I represents the intensity of the laser beam.

Fig. 3. Timing jitter 共histogram for 2 s兲 of Q-switched laser pulses when the pulse is generated 共a兲 by only the long pulse and 共b兲 by the composite pulse.

composite pulse 共the Q-switched pulse was timed to be triggered by the short pulse兲 is shown in Fig. 3. The histograms that represent timing jitter were recorded for 2 s. Comparing the widths of the histograms in Figs. 3共a兲 and 3共b兲, more than a tenfold reduction 共from 6 to 0.5 ␮s兲 in timing jitter was obtained with the introduction of a composite pumping pulse. This result is in full accordance with our theoretical predictions. In addition, the timing of the Q-switched pulse was found also to be locked to that of the short pulse for a limited range of time, which is a natural result. The effect of the power of pumping on the timing jitter of a passively Q-switched laser was also demonstrated by use of only one pumping laser diode but with constant dc injection currents of different magnitudes. The timing jitter during 2 s was measured to be 25, 10, and 4 ␮s when the dc injection current applied to that pump laser was 0.72, 0.85, and 0.96 A, respectively. The performance of the proposed composite pumping scheme can be further improved by use of only one pump laser to provide both long and short pulses and by improvement of the electronics to yield a sharper short pulse. In conclusion, we have proposed and demonstrated a simple way to reduce the timing jitter of a passively Q-switched laser with a composite pump pulse that is composed of a long, low-power pulse and a short, high-power pulse. A reduction of more than an order of magnitude in timing jitter has been obtained for a Nd:YAG兾Cr4⫹:YAG laser. References

larization were used to generate long and short pulses, respectively. The injection currents were 1 and 4 A and the pulse widths were 500 and 1 ␮s. The slope efficiencies of the two pump laser diodes were measured and found to be almost identical 共0.9 W兾A兲; thus more than 99% of the pump energy was provided by the long pulse. Two pulses were combined by a polarizing beam splitter 共PBS兲. A variable-delay generator was used to set the correct timing of the second pulse relative to the first pulse. A comparison of Q-switched pulses obtained with simple pumping 共the short pulse was timed to be delayed relative to the Q-switched pulse兲 and the

1. H. Kunimori, T. Otsubo, B. Engelkemier, T. Yoshino, and B. Greene, “Timing precision of active Q-switched mode-locked laser and fire control system for the synchronous satellite laser ranging,” IEEE Trans. Instrum. Meas. 44, 832– 835 共1995兲. 2. M. Arvidsson, B. Hansson, M. Holmgren, and C. Lindstrom, “A combined actively and passively Q-switched microchip laser,” in Solid State Lasers VII, R. Scheps, ed., Proc. SPIE 3265, 106 – 113 共1998兲. 3. S. Huang, T. Tsui, C. Wang, and F. Kao, “Timing jitter reduction of a passively Q-switched laser,” Jpn. J. Appl. Phys. 38, L239 – L241 共1999兲. 4. W. Mandeville, K. M. Dinndorf, and N. E. Champigny, “Characterization of passively Q-switched microchip lasers for laser radar,” in Laser Radar Technology and Applications, G. W. Kamerman, ed., Proc. SPIE 2748, 358 –367 共1996兲.

20 February 2002 兾 Vol. 41, No. 6 兾 APPLIED OPTICS


Suggest Documents