Many laser systems consist of an oscillator followed by one or more amplifier stages. These optical amplifiers can take several very different forms, including that of a one-pass amplifier, a multipass amplifier, or a regenerative amplifier. A number of interdependent factors determine which of these configurations is best suited for a particular laser type. These factors include the amplification required, the gain and saturation properties of the active medium, the input power from the oscillator, system cost and complexity, and the desired beam quality. This article examines how these factors influence amplifier choice and design. one-pass amplifier
Figure 1. In a one-pass amplifier, the seed pulse passes through the gain medium a single time.
A one-pass amplifier offers the greatest simplicity and lowest cost. In this type of amplifier, the output from the oscillator passes once through a gain medium that is pumped optically or electrically. At most, the optics consist of a pair of lenses or curved mirrors and do not require complex alignment. A common example is the lamp-pumped, Q-switched neodymium-doped yttrium aluminum garnet (Nd:YAG) laser used in many laboratory applications (see figure 1). These lasers often consist of an oscillator stage with one or two lamp-pumped Nd:YAG rods operating at tens of hertz. The output pulses from the oscillator pass through one or more additional lamp-pumped rods before exiting the laser head. The cross section for stimulated emission is relatively high in Nd:YAG, so the net gain from a single pass can be as high as 10. In addition, the excited state lifetime of Nd:YAG is relatively long (220 µs), so precise timing of the flashlamp pulse that pumps the amplifier with the output pulse of the laser oscillator is not critical.
Another feature of the one-pass amplifier is that the spectral, temporal, and spatial properties of the amplified pulse are largely the same as that of the input pulse. However, differences may occur. For example, when the amplifier is pumped by a laser (rather than a lamp), the spatial profile of the final pulse may more closely resemble that of the pump beam than the input beam. Another reason for differences is gain saturation. This is a concept that must be clearly understood when considering using multipass or regenerative amplifiers. gain, saturation, and cavity losses
Figure 2. In a laser amplifier, the gain coefficient begins to saturate as the input signal strength increases. This saturation is usually defined by the saturated gain coefficientthe input signal fluence at which the gain has dropped to half the small signal gain.
A number of factors determine the gain of a laser amplifier, including input signal strength. As the signal beam fluence increases, the amplifier eventually starts to saturate (see figure 2). In simple terms, the number of photons being extracted from the gain medium becomes significant in comparison to the number of excited atoms or molecules. As the fluence increases, the amplifier gain continues to fall from the so-called small-signal gain until it asymptotically reaches a limiting value of unity. Saturation is usually described numerically by the saturated gain coefficient, which is defined as the input signal fluence that yields a gain of half the small-signal gain. Since the amplifier includes some loss (reflection losses at surfaces, scatter, diffraction, etc.), the net value in a real amplifier will be less than unity when the input signal is very high.
Figure 3. As the input fluence increases, the output fluence (blue line) starts to saturate. In a multipass amplifier, the pulse energy first grows as the number of passes N (red lines) increases, then begins to fall due to cavity losses and gain decrease.
The shape of this saturation curve not only influences the choice of amplifier (one-pass versus other types) but also determines the design characteristics of that amplifier. As a laser pulse passes through the gain medium, it cannot gain infinite energy. In fact, the pulse energy will grow as a function of the dumber of passes through the gain medium (see figure 3). After peaking, the pulse energy starts to fall off from its maximum value, again because of cavity losses. The most efficient amplifier (highest overall gain) is one in which most of the available pump power is extracted. This requires an amplifier operating just below saturation. Consequently, in a one-pass amplifier, for a given input pulse energy, an optimum length exists for the gain medium (e.g., an Nd:YAG rod).
There are several potential consequences of operating a one-pass amplifier near saturation. In the spatial domain, an input beam with a Gaussian cross-section may saturate an amplifier near the beam center more than at the edges of the beam, resulting in a "top hat" profile. In the temporal domain, the leading part of the pulse will experience less saturation than the later parts of the pulse, potentially changing the temporal profile. In the spectral domain, the middle part of the spectrum may be less amplified than the wings, depending on amplifier spectral broadening characteristics. Alternatively, if the amplifier material does not have sufficient bandwidth to amplify the entire spectrum uniformly, then spectral narrowing will occur. In the case of transform-limited femtosecond pulses, this can also result in temporal pulse broadening. multipass amplification
When the output of a one-pass amplifier is too low as a result of a limited small-signal gain coefficient and/or a small input signal, the next design to consider is the multipass amplifier. Here, optics (usually mirrors) are arranged so that the input beam makes several passes through the amplifier gain medium before exiting; alternatively, electro-optic switching can be used.
Figure 4. In a multipass amplifier, the beam passes through the gain medium several times, at a slightly different angle each time.
In practice, each pass through the gain medium may travel through the same optically pumped spot in the center of the amplifier material, but with a different path (i.e., different angle; see figure 4). A folded path allows the beam to enter and exit the amplifier after a finite number of passes. The optimum number of passesand hence the number of folds in the amplifier pathdepends on the gain per pass, the amount of overall gain required, the saturated gain coefficient for the material, and the amount of optical complexity tolerable. At Spectra-Physics, we have used two-pass amplifiers in high-power, high repetition rate ultrafast titanium-doped sapphire (Ti:sapphire) laser systems to give a final stage of amplification to the high-energy output of a regenerative amplifier.
A multipass configuration is also the perfect choice for an optical parametric amplifier (OPA). In one example of such a system, the femtosecond pulse from a regenerative Ti:sapphire amplifier is split into three parts. The smallest part is tightly focused into a nonlinear crystal to produce a white-light continuum that acts as an input pulse to seed the OPA. The second part of the femtosecond pulse is used to pump the first pass through the beta barium borate (BBO) crystal that acts as the parametric gain medium, and the third and largest part is used to pump a second pass through the same crystal. Synchronized double pumping, and therefore careful timing, is very critical in this device because the optical parametric gain has no storage time, unlike conventional laser-gain media. The system is designed so that the first pass has a small beam waist in the OPA crystal to optimize gain, whereas the second pass produces a larger beam waist to avoid damage to the crystal.
The spatial and temporal properties of the OPA pulse are governed by the characteristics of the seed pulse and the pump pulses. However, the wavelength of the OPA output is controlled by the phase-matching angle of the crystal, which determines which part of the white-light continuum is amplified. Changing the phase-matching angle yields broad wavelength tunability. regenerative amplifiers
Practical issues of optical complexity limit the number of passes feasible for a multipass amplifier, so the net gain of such an amplifier cannot be increased beyond a certain level. For some applications, this gain level is not sufficient. This happens when the input signal from the laser oscillator is very weak, or when several passes are not enough to reach saturation. Both instances are often the result of a low cross-section for stimulated emission. A well-known example is the ultrafast, mode-locked Ti:sapphire laser, in which a typical commercial oscillator produces pulse energies at the nanojoule level, and the one-pass amplifier net gain is only a factor of approximately two.
Figure 5. In a regenerative amplifier, an electro-optic modulator acts as a gate to allow a single pulse to enter the amplifier cavity. After many passes, the amplified pulse is allowed to exit the cavity via an optical gate.
The solution is to use an amplifier within a cavity, such as a regenerative amplifier. In a regenerative Ti:sapphire amplifier, an electro-optic modulator is used as a gate to allow a single pulse from the oscillator to enter the amplifier cavity. This pulse then makes tens of passes through the Ti:sapphire crystal, which is pumped by a Q-switched, frequency-doubled neodymium-doped yttrium lithium fluoride (Nd:YLF) laser (see figure 5). Once most of the pump energy has been extracted from this crystal, a second (or the same) electro-optic modulator switches to deflect the pulse out of the cavity. As with other amplifier types, a regenerative amplifier is usually operated close to saturation in order to maximize efficiency and pulse-to-pulse stability. Depending on the beam diameter in the crystal and the amount of pump energy, the Ti:sapphire regenerative amplifier is capable of delivering gains as high as 106.
By the time the overall gain has passed 105, the pulse energy in a regenerative Ti:sapphire amplifier typically begins to approach the millijoule level. For femtosecond pulses, the resultant high peak power would damage the Ti:sapphire laser crystal. To avoid catastrophic damage, pulses in regenerative amplifiers are stretched in time before amplification and recompressed after. Because the gain per individual pass is low and many passes are required, it is vitally important to minimize cavity losses in order to optimize net gain per pass. This necessitates the use of low-loss optics that feature high reflectivity and very low scatter.
In terms of output characteristics, one of the major advantages of a regenerative amplifier is that the spatial profile and pointing of the output beam is defined by the cavity. With a well-designed cavity, the regenerative amplifier is capable of delivering transform-limited ultrafast pulses in a very high-quality beam.
There are a number of arrangements that can be used to amplify the output of a laser oscillator. With a high-power input beam and/or a high gain coefficient, most laser designers will opt for a single or multipass amplifier. If the gain coefficient or the input power is low, it is often better to use a regenerative amplifier. Fortunately for the laser user, most of these systems are now available in rugged, compact packages that provide access to this functionality with turnkey simplicity. oe
Arnd Krueger is marketing manager at Spectra-Physics, Mountain View, CA.