Is there a way to generate and emit specific frequencies of IR or NIR? There’s always the sledgehammer method of generating a range of frequencies and filtering for what is desired, but I would hope there’s a better way. 
Pretty vague question…
How about using a broadband diode laser and then stabilizing to a specific frequency using spectral hole burning. You probably don’t even need to use a spectral hole unless you want a very stable frequency, just use an extended laser cavity in the littrow configuration.
No, really, what frequency, what bandwidth, coherent or non coherent, what kind of power levels?
There’s always tunable dye lasers. The particular model shown here is tunable to any wavelength from shortwave UV to far IR.
10 watts would be nice.
I’d try to avoid the dye lasers if I were you. Messy buggers and can be a little temperamental if you ask me :mad:
Stick to the lovely diode lasers. 
A little toxic too!
10 Watts? That’s a lot of power! What frequency range?
You may want to go with a fiber laser. Erbium doped fiber lasers works well around 1500 nm IIRC. Ytterbium can be used around 1000 nm. I believe these can be tuned over wide range with a fiber bragg grating. This is not my field, so perhaps someone else will come along with more info.
I have never heard of a diode laser with an output (cw) that is this high in power. That does not mean they don’t exist though. I work with a high power,high brigtness, stabilized (< 1 MHz linewidth at 793 nm), diode laser with a high power tapered semiconductor amplifier that I can squeeze around 300 mW from if the phase of the moon if perfect…
If you want 10 watts, I suggest LEDs.
Look in a Newark or Digikey catalog to see if there are LEDs that have outputs at the frequencies you want, and then gang them together in parallel to achieve your output power.
WARNING - if you collimate that much power into anything resembling a tight beam, then the resulting device will NOT be eye safe, even if you can’t see the beam, so wear protection. And I don’t mean a condom. And if you point it at someone else’s eyes, you will be commiting a criminal act. So BE CAREFUL.
If you need very narrowband (e.g., less than 10 nanometer) output, then you need a laser, probably a diode laser. But you’re talking major dinero for those, especially at that much power.
And lasers are even less eye safe. The feds will probably start looking at you funny if you start inquiring about that much laser power without a darn good reason. Not to mention OSHA.
Not quite 10 watts, but there are GaAlAs diode lasers that put out up to 4 W CW at around 800 nm. Maybe too longwave for what the OP has in mind, but they’re out there. By the way, one could seriously hurt oneself with such a laser. And since you can’t see the beam or its reflections without special eyeware, it wouldn’t be advisable to play around with such a beast.
OSHA couldn’t care less what you do in your own home. They only monitor occupational safety hazards–hence the name. Oddly enough, the FDA sets the regulatory standards for lasers. Who woulda thunkit?
10 watts is pretty high. Maybe you meant 10 mW? In that case, I might suggest superluminescent diodes (SLDs). Although a lot does depend on what bandwidth you want, as SLDs generally tend to be broader. Any hints as to what this is for?
A 10W IR diode pumped laser is available, if you’ve got the cash. Cash, IIRC, meaning six-figures.
You can generate a wide range of frequencies using a 555 timer and a few resistors and capacitors, but you probably wouldn’t be able to power anything directly with it. You could easily use a power MOSFET or a darlington transistor pair to get the amount of currnet you need to light up an IR LED array.
It would help a lot if I knew what you were trying to do, and what you needed to use this for.
A friend at my last office showed me an advertisement online for a very narrow wavelength IR bulb designed to jam police IR laser speed detectors (they were dual-filament (?) headlamp bulbs, emitting broad visible and also a tight stripe of near IR spectra). Obviously, that application is illegal anywhere that the law has caught up with the technology, but if the wavelength you’re looking for happens to be close to that wavelength, these bulbs (incandescent, IIRC) might be cheaper than a laser purchase. Of course, you’d have to set up an electrical system that operated at roughly the same voltage and amperage as a car’s headlights.
Perhaps with that info, Google will be your friend.
I’m pretty sure that the OP is talking about different frequencies (thus different wavelengths) of the IR or near-IR source, hightechburrito, rather than different modulation frequencies. However, as you say, the OP needs to define his/her needs better. We’re told “10 watts”, but is that CW (continuous wave) or pulsed? Also, how tunable does it need to be?
In any case, 10W is not unusual for an infrared laser. You could be looking at the mid-five figures (i.e. around $20,000 - $70,000 US) for a new system, depending on what exactly you need.
A quick search of eBay turns up a 100W CW YAG laser (wavelength = 1064nm), apparently $50k new, starting bid $3500. Alternatively, there’s a 15W CW CO2 laser (wavelength = 10.6 microns) with a “Buy It Now” of $5995.
Neither of the above are inherently tunable, although you may be able to select specific transition lines from the ones available. You might want to look into “F-center lasers”, which are tunable, but I don’t think that you’ll get 10W CW out of one of them.
Needless to say, as has been mentioned by several posters above, 10W of infrared light can be very hazardous for the eyes.
Thanks all for your answers. Yes, I meant 10 watts, and seeing the responses, I understand that this is not a practical avenue to pursue. Unfortunately, I cannot explain my need further. Thank you all again.