What are the best methods to measure the propagation loss of waveguides?
Cut-back method applicable for both single-mode or multimode waveguides, Interferometric method for single-mode waveguides.
Cut-back method applicable for both single-mode or multimode waveguides, Interferometric method for single-mode waveguides.
Dr. Minh Tran Anh, I am referring to you for the following article.
Title:
Measuring mode propagation losses of integrated optical waveguides: a simple method
Authors:
Yasuyuki Okamura, Shinji Yoshinaka, and Sadahiko Yamamoto »View Author Affiliations
Journal & Web location:
Applied Optics, Vol. 22, Issue 23, pp. 3892-3894 (1983)
http://dx.doi.org/10.1364/AO.22.003892
I will assume that it is a passive waveguide. One technique issue is how to cancell out the coupling efficiency which is unknown. If you change the experiment condition, such as the waveguide end surface, the gap between your objective lens and waveguide end surface etc, the coupling efficiency will be different. If you use visible light, you could monitor the scattered light intensity above the waveguide from the coupled light inside the waveguide.
Dear Minh Tran Ahn, all depends on a waveguide type. What is your waveguide?
Hi,
It is correct that all depends on a waveguid type. If you are using prism coupling method then you may refer this paper
https://www.researchgate.net/publication/243394099_Dip-coated_HNR120_photoresists_thin-film_optical_waveguides?ev=prf_pub
Article Dip-coated HNR120 photoresists thin-film optical waveguides
Hi, thank you everyone for your replies. Actually I am setting up my experiment to measure the loss of 1550 nm light propagating through my waveguides by using cut-back methods. And yah, I face the trouble of coupling efficiency variation. What is the best way to eliminate this variation?
Take readings of total loss (in dB) for some different lengths of the waveguides processed similarly, and plot a loss versus length graph. The slop of the best fit straight line of this plot will give you the waveguide propagation loss (in dB/cm), and the intercept in Y-axis will give you average i/o coupling loss (in dB). As the number of waveguides of different lengths increases the accuracy increases.
Hi,
One easy way to minimize the effect varying coupling efficiency might have on your results is to modify the cut-back method in a way that you use a set of identically processed waveguides with a different length each and grating couplers for coupling.
If you can make sure that
a) all waveguides are more or less of the same quality and
b) all grating couplers have more or less the same coupling efficiency,
which both is possible if appropriate fabrication techniques are used, that should be the most convenient and reliable way to measure optical propagation losses.
You should measure propagation losses versus waveguide length, i.e. fabricate few different length waveguides and measure the overall propagation losses. The coupling efficiency will be the same (it can change slightly depending on the quality of the waveguide faces). And, mostly that's all.
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering