Mode Matching
Author: the photonics expert Dr. Rüdiger Paschotta
Definition: the precise spatial matching of the electric field distributions of laser beams and resonator modes or waveguide modes
Categories: 
DOI: 10.61835/x1d Cite the article: BibTex plain textHTML Link to this page LinkedIn
In many situations, it is necessary to spatially match a laser beam precisely to another beam or a mode in order to obtain some kind of efficient coupling. Examples are:
- A beam from a laser has to be coupled into an optical fiber or some other type of waveguide.
- A laser beam must be matched to a mode (often the fundamental mode) of a passive optical resonator which should act, e.g., as a spatial and/or spectral filter (→ mode cleaner cavities) or for resonant frequency doubling.
- For injection locking, the mode of a master laser has to be matched to that of the slave laser.
The necessary matching of modes means not only creating a good spatial overlap of the intensity profiles, but also matching the optical phase profiles. In other words, the wavefronts of the beams will then be matched. If the complex amplitude profiles of two beams (having the same optical frequency) are well matched in a certain plane, they will remain well matched during further propagation.
Mode matching can be achieved by using suitable relay optics (typically some combination of curved mirrors or lenses), provided that the beam quality of the initial beam is close to diffraction-limited.
Mathematically, the quality of mode matching can be quantified with an overlap integral. The following formula, involving the square of such an overlap integral, calculates the coupling efficiency concerning optical powers:
$$\eta = \frac{{{{\left| {\int {E_1^*{E_2}\;{\rm{d}}A} } \right|}^2}}}{{\int {{{\left| {{E_1}} \right|}^2}{\rm{d}}A\;\int {{{\left| {{E_2}} \right|}^2}{\rm{d}}A} } }}$$where <$E_1$> and <$E_2$> are the complex electric fields in a plane, referring e.g. to a laser beam and the field of a resonator or waveguide mode, and the integration spans the whole beam cross-section. That quantity is preserved during propagation in free space.
A similar overlap integral can be used for calculating complex mode amplitudes.
The equation above can used to calculate, for example, which fraction of the optical power of a Gaussian laser beam can be launched into a single-mode fiber. That can be close to 100% if the fiber's guided mode is close to Gaussian.
Mode Matching Calculations
The software RP Resonator is a particularly flexible tool for calculating all kinds of resonator mode properties. With a little script code (which we are happy to provide), you can also calculate mode overlap integrals. For example, calculate how efficiently a Gaussian input beam can be coupled to a resonator under conditions of misalignment.
If the beam from a frequency-tunable single-frequency laser hits a symmetric Fabry–Pérot interferometer and the laser frequency is tuned over the whole free spectral range of the resonator, the transmitted light can be used to analyze the degree of mode matching. For perfect matching to a cavity mode (typically the fundamental Gaussian mode), complete transmission of the resonator can be observed when the resonance condition is met, whereas other resonances (corresponding to other resonator modes) can not be excited.
When trying to launch light into a waveguide without perfect mode matching, one may get part of the light into cladding modes of the waveguide, rather than the desired guided modes.
More to Learn
Encyclopedia articles:
Suppliers
The RP Photonics Buyer's Guide contains 14 suppliers for Fabry–Perot interferometers. Among them:
Questions and Comments from Users
2021-09-17
How to couple light into a waveguide which does not have a precise mode description but only a rough mode width; how to use the formula?
The author's answer:
You can at least normally assume that the wavefronts of the modes are flat. As a first guess, you may then assume a Gaussian amplitude profile.
2021-10-28
Do the two fields E1 and E2 have to be normalized to the maximum?
The author's answer:
No – the formula contains the normalization. You can easily see, for example, that the result won't change if you double <$E_1$>.
2021-11-23
What if we do not have the phase information, but measured the intensity profiles with a camera? I guess one could work with the square root of the intensity, but that would only obtain an upper limit, wouldn't it?
The author's answer:
You need the full phase information to do any calculations on mode matching.By taking the square root of the intensity profile, you are effectively assuming a flat phase profile. That might serve as an upper limit for the possible mode matching efficiency, if the ideal output field has flat phase fronts.
2022-04-22
Is there any relation between the “overlap INTEGRAL” as described on this page and the “overlap FACTOR” (<$\eta_\textrm{m}$>) on your page concerning optical heterodyne detection?
The author's answer:
Yes, it is the same. The heterodyne signal is proportional to the amplitude of the input signal; its signal-to-noise ratio is proportional to the square of that. That's compatible with the formula for the <$\eta$> value shown in this article.
2022-09-12
Does the formula take into account the imaginary part of the electric field? Is there any problem if I only consider the real part of the electric field?
The author's answer:
The formula works with complex electric fields, as indicated. Here, you cannot simply ignore their imaginary parts. For example, a wavefront curvature as seen in the complex phase is relevant for mode matching.
2022-10-06
I am a bit confused by your answer on 2022-09-12. When we describe waves using the complex representation it is always understood that this is for algebraic convenience and only the real part has physical significance. Your answer seems to imply that there is information in the complex part, and that the complex representation is actually necessary. Is this a contradiction?
The author's answer:
An interesting and important question. The physical electric field is the real part of the product of that complex amplitude and the oscillating factor <$\exp(i \omega t)$>. Written out, I mean <$E(t) = Re(E \cdot \exp(i \omega t))$>. You see from this that the imaginary part of that complex amplitude alone does matter! You also see that from the denominator in the formula: the modulus squared appears, and that also depends on the imaginary part.
2022-10-13
Is there a good book on the topic? Especially I would be interested in mode overlap integrals for LG- and HG- higher-order modes for different waist sizes or misalignment.
The author's answer:
I would simply numerically calculate that myself without searching for literature. The equations are reasonably simple.
2023-09-21
I have a question regarding the angle under which the beams arrive. For example in a star coupler, does this angle affect the coupling? Or is the narrower field, resulting from one port being under an angle, sufficient for estimating the efficiency?
The author's answer:
No, the angle is important because any tilt is associated with a linearly varying phase change. You cannot consider the intensity profile only.
Here you can submit questions and comments. As far as they get accepted by the author, they will appear above this paragraph together with the author’s answer. The author will decide on acceptance based on certain criteria. Essentially, the issue must be of sufficiently broad interest.
Please do not enter personal data here. (See also our privacy declaration.) If you wish to receive personal feedback or consultancy from the author, please contact him, e.g. via e-mail.
By submitting the information, you give your consent to the potential publication of your inputs on our website according to our rules. (If you later retract your consent, we will delete those inputs.) As your inputs are first reviewed by the author, they may be published with some delay.
2021-02-15
How does one account for polarization in this equation?
The author's answer:
Not at all. The calculation is based on scalar fields, i.e., neglecting polarization. If you want to calculate the overlap often arbitrary polarized field with a polarized mode, you first have to decompose the field into its polarization components.