|
The key objective of this article is to underscore that as engineers, we need to pay close attention in repeatedly validating
and re-validating the underlying physical processes behind a working theory that models a phenomenon we are using to
create tools and technologies. We use the test case, the prevailing mode-lock theory, to illustrate our views by identifying
existing contradictions and showing approach towards their resolution by identifying the relevant physical processes.
The current theory tells us that the Fourier summation of all the allowed cavity modes directly produces the train of
pulses. It effectively assumes that electromagnetic (EM) waves are capable of re-organizing their spatial and temporal
energy distribution to generate a train of temporal pulses while preserving the spatial mode energy distribution. The
implication is that EM waves interact with each other by themselves. Even though the theory is working, we have three
logical problems. First, in the real world, in the linear domain, waves never interact with each other. On careful analysis
of all types of interference experiments, we will recognize that only in the presence of some interacting material medium
can we observe the physical superposition EFFECT. In other words, detectors carryout the superposition effect we call
interference phenomenon, through the summation of their multiple simultaneous linear stimulations and then absorbing
energy proportional to the square modulus of the sum total stimulation. Second, a Fourier monochromatic wave, existing
in all space and time, is a non-causal hypothesis. Just because our theories are working does not mean that we have
understood the real physical interaction processes in nature. We need to build our theories based upon space and time
finite EM wave packet containing a finite amount of energy, which is a causal approach. Third, in spite of staggering
successes of Quantum Mechanics, we do not yet have a self consistent model for space and time finite model of a
photon. QM only predicts that EM energy emission (spontaneous and stimulated) takes place only in a discrete amount at
a time from atoms and molecules. It does not give us recipe about how to visualize a propagating photon as it expands
diffractively. However, Huygens-Fresnel's classical diffraction integral gives us a rigorous model, which is the
cornerstone of modeling evolution of laser cavity modes, CW or pulsed. In this paper, we highlight the contradictions
that arise out of the prevailing mode-lock theory and resolve them by using causal models, already underscored above.
For example, there are now a wide range of very successful technological applications of the frequency comb extracted
out of fs lasers. If the Fourier summation were the correct physical process, then all the cavity modes would have been
summed (converted) into a single mean frequency around the gain line center for perfectly mode-locked systems.
Further, sending such fs pulses through an optical spectrometer would have always displayed a transform limited fringe,
centering on the mean Fourier frequency, rather than generating the comb frequencies, albeit instrumentally broadened.
Output pulse train from a phase locked laser is functionally produced due to the oscillatory time-gating behavior of the
intra-cavity phase-locking devices. So, we need to pay more attention to the fast temporal behavior of the materials we
use for achieving very fast time-gating, since this material imposes phase locking on the cavity modes to enhance its own
high-contrast time-gating behavior.
|
