The GaAs laser (904/905 nm) has been on the market for many years, first introduced by the Italian Space Company in the late 80s. GaAs technology is more complicated than the more common InGaAlP and GaAlAs lasers, so few manufacturers have been offering this wavelength.

The GaAs laser has several advantages over other wavelengths, but there are some problems which have to be addressed. But let us first have a look at the technology.

GaAs diodes are very small and if loaded with a continuous beam, they will burn. Therefore they are superpulsed. This means that the beam is turned on and off during very short periods, typically 100-200 ns (nanoseconds). Each pulse is at its peak very intense, indeed in the Watt range rather than in the milliwatt range. Typical peak powers for GaAs lasers can range from 5-100 Watts. This appears to be a very intense laser, but the ultrashort pulse should not be compared to that of a CW laser. What is of interest, however, is the average power. This means that the number of pulses is added and the average output is calculated. And since the pulses are very short and the intervals between them decide their collected energy, the actual output is in the milliwatt range. But many manufacturers prefer to indicate the peak pulse power (PPP), and 100 W seems more impressing than 10 mW! So the customer is fooled and the honest producers are cheated. For all GaAs lasers, the average power is of interest for the clinician. The other parameters are of interest for technicians. It is fair to indicate PPP, but to do it without indicating the average output is to confound the customer.

Figure showing different pulsing options, all leading to different average power.

Now, here is the other problem. The old (and many of the new) GaAs lasers have a number of settings for pulsing. This means that if the maximum average power is 10 mW, this is obtained by maximum pulsing at 10 000 Hz (pulses per second). If pulsed at 5 000 Hz, the number of pulses obtained is reduced by 50% and the average power is 5 mW. Suppose you have heard that low pulse repetition rates (frequencies) are best for pain treatment, you set the equipment at 1 000 Hz – and you have 0.5 mW! And imagine setting it at 100 Hz…

Some manufacturers have solved this problem but using “pulse train modulation”. With this technique, the average output is constant over all settings. So if you think that a particular pulse repetition rate (PRR) is the best, you can use it and still have the same power. This is illustrated by the figure below, where the pulse train modulated GaAs laser has a more or less constant output over all PRRs, whereas the traditional ones are pretty useless unless used with maximum PRR.

The knowledge about the effects of pulsing is scant. It probably matters, but we do not know how. Some studies have compared 2 or 3 settings, and one was best. But suppose the best one was 250 Hz. How about 275, 4890, 7999 Hz? But since the GaAs laser is always pulsed, we have to make a choice of PRR, and only anecdotal evidence is there. Besides, the type of pulsing here is quite different from the “chopping” of other LLLT devices. The “pulsing” in these lasers is merely shutting on and off a continuous beam.

Now, let us look at another important aspect of the GaAs laser – the penetration. This is what we have been told:

This is of course quite correct, but only if a CW laser is used. Surprisingly enough, the penetration of the superpulsed GaAs laser has not been investigated, until now. Joensen of the University of Bergen, Norway, compared the penetration of 808 nm and 904 nm through rat skin. The result was the following:

This means that the high pulse “spikes” somehow force the light deeper and deeper into the tissue, while the 808 has a constant depth of penetration, regardless of irradiation time. When it comes to treating targets deep into a body, this knowledge is important.

Here is another aspect of the GaAs laser: It requires lower energies to obtain the same effect as a CW laser. This was observed already in the late 80s by Abergel et al. So, for instance, a study comparing the effect of neural growth using different wavelength found that the GaAs had no effect, whereas others had. Well, by using the same energies for all wavelengths, the energy of the GaAs was too high and had an inhibitory effect.

Finally, let us look at the reporting of parameters in scientific studies. These are much too often incomplete, thus reducing the quality of the study. When it comes to GaAs, the average power is often missing, probably because it was not indicated in the manufacturers’ manual. Just looking at dose (J/cm2) is not acceptable! Let us look at two examples from Medline:

Adult male Wistar rats were divided randomly into three groups (n=6), namely sham (uninjured muscle), muscle injury without treatment, and muscle injury with LLLT (GaAs, 904 nm). Each treated point received 5 J/cm2 or 0.5 J of energy density (12.5 s) and 2.5 J per treatment (five regions)

Since the time and energy are reported, the ouput can be calculated to 40 mW (40 mW x 12.5 s = 500 mJ = 0.5 J). The 0.5 J is called energy density, while J is actually energy and energy density is J/cm2. In conclusion, the average power is not reported in spite of the fact that it is an important parameter and the nomenclature is incorrect. The spot size of the beam should also have been reported. Admittedly, this is difficult with an invisible beam, but necessary.

Still, things can be much worse:

“The other 25 patients were assigned into the experimental group and received conventional therapy plus low-level laser therapy (4 J/cm2) at each point over a maximum of ten painful points of shoulder region for total 5 min duration.”

Maybe the full paper has a better reporting of parameters, but after all, most persons looking for scientific evidence go to Medline only.

No names of manufacturers or researchers have been mentioned here (unless positively reported). The aim of this paper is not to point fingers at anyone, just to use examples for the sake of illustrating.



Joensen J, Ovsthus K, Reed R K, Hummelsund S, Iversen V V, Lopes-Martins R A, Bjordal J M. Skin Penetration Time-Profiles for Continuous 810 nm and Superpulsed 904 nm Lasers in a Rat Model. Photomed Laser Surg. 2012; 30 (12): 688-694.

Hashmi JT, Huang YY, Sharma SK, Kurup DB, De Taboada L, Carroll JD, Hamblin MR. Effect of pulsing in low-level light therapy. Lasers Surg Med. 2010;42(6):450-466.