The optical fiber with pure quartz core and Fluorine-doped glass cladding was made by POD (plasma outside deposition) technique in some corporations, while we used the creative technique of “overcladding F-doped tube onto quartz rod in high temperature” to make the optical fiber which has the same structure as that from POD, in order to research and compare the influence factors on the loss of the fiber, our research work includes contrast experiments on coating polymers with different refractive index and the concentricity error of the fiber core and cladding. The measurement results show us that there are great differences in the loss spectra between the different fiber samples. We made analysis of it.
The optical fiber with pure quartz core and fluorinedoped glass cladding is a new kind of large core optical fiber that appeared in recent years. Some corporations have made it by POD technique (plasma outside deposition) [
The structure of this kind of fiber is quite different from that made by conventional technique. There are two main different points: one is that the fiber core material is pure quartz, another is that the cladding thickness is greatly reduced. The influence of the first different point on the fiber loss is obvious, and what influence will come about the loss of fibre brought by another different point? We want to be quite clear on this point, make it clear, we conducted a series of researches to confirm that, which include: 1) We cover the fiber with high and low refractive index polymers separately, and compare the changes between their loss spectra; 2) The influence of the glass cladding thickness with low-refractive-index coat on fiber loss spectra; 3) The influence of the concentricity error between core and cladding on the fiber loss spectra. We found the great changes on their loss spectra in the experiments, which shows that the influence of the above factors is obvious.
Before using our technique of “rod in tube” to make the optical fiber with pure quartz core and F-doped glass cladding, we have to get firstly the F-doped tube treated by a special process, in order to avoid bubble appearing on the surface of F-doped tube in the following high temperature process. Then we use the “rod in tube” technic to make the fiber preform in high temperature on MCVD lathe, as shown in
In the experiments, we made two optical fiber preforms No. 1 and No. 2 by the technique of “rod in tube” and mechanical grinding. We used the quartz rod (type F300ES) as the core of two preforms. The F-doped quartz tubes have the same parameters. We made two optical fibers 1# and 2# by drawing No. 1 preform two times, whose inner polymer coats have separately low and high refractive index. We also drew No. 2 preform and fabricated the 3# optical fiber, and cover it with high refractive index polymer as inner coat of the fiber. The geometry parameters of 1# and 2# fibers are exactly the same. The design geometry parameters of 3# fiber is the same as them. The detailed materials and the parameters as well as the structure of the optical fibers are shown in
1# and 2# optical fibers were covered respectively with low and high refractive index polymers as the inner coat. Their structure and refractive index distribution are respectively shown in Figures 2 and 3, and their loss spectras are shown in Figures 4 and 5.
The structure and refractive index distribution of 3# optical fiber is shown in
In
In the
As to multimode optical fiber, we know the light with operation wavelength λ0 is effectively limited and transmitted in the fiber waveguide with low loss when the cladding wall thickness with low refractive index is over 3λ0. When the wavelength of the light that transfers in the optical fiber increases, according to relationship of 3λ, in order to keep the light transmitting with low loss in the fiber, it demands the thicker glass cladding wall.
Through analysis, we think the loss spectra of optical fiber 2# in
but because of the process error of mechanical grinding and fluctuation of fiber diameter, the practical result is that some cladding wall thickness of fiber 2# is less than 5µm. We can see the result from the loss spectra in
The loss spectra of 3# optical fiber in
Through analysis, we think the unusual loss spectra of the 3# fiber has something to do with the tolerance of the F-doped glass cladding wall thickness (concentricity error), which resulted from the mechanical grinding process, as is shown in
the fiber is over 3λ0 = 2.424 µm, from
As to meridian or precession light, the fiber with large concentricity error results in the result that the light transmitted in the fiber reaches to the aero nearby δ1 leaks out of the fiber and gets into the polymer coat with high refractive index, and it is absorbed and exhausted. This happens already in the range of short wavelength, so the background loss is raised up in the spectra of 3# fiber in
We know the light propagation constant , which transmits in the fiber, where n1 is refractive index of the fiber core, λ is the working wavelength, θ is the angle between vector k and the axis of the fiber, and each β is corresponding to one transmitting light module. With increase of the light wavelength λ that transmits in the fiber, β corresponding to each λ will become smaller and smaller. In terms of optics waveguide theory, when the propagation constant β of one transmitting light module reduces and exceeds one critical value β1 (i.e. optical wavelength λ increases and exceeds some value λ1), the transmitting module of light changes into leaking module, if optical wavelength further increases when the propagation constant β of one transmitting light module reduces and exceeds another critical value β2 (i.e. optical wavelength λ increases and exceeds some value λ2), the transmitting light will change its leaking module into radiation module, and lose all transmitting optical energy. This is the relationship between various transmitting optical wavelengths in the fiber and the energy carried by the corresponding module.
Through our experiment results, we can see it is still existed that relationship between the transmitting light wavelength λ and the energy carried by the corresponding module in this kind of fiber, and that because various inner polymer coat and the different glass cladding thickness and the concentricity error between the fiber core and cladding, the relationship is changed more complicated. But from another point, this maybe give us a chance, we can effectively influence the character of the fiber through adjusting some design on it. This provides us with some chances about the potential application of the special fiber.
In addition, from the experiments and discussion above, we have known that the loss of the fiber with pure quartz core and F-doped glass cladding is influenced by many factors, some of which come from the special structure of the fiber, so when someone plans to fabricate this kind of fiber by our technique of “rod in tube”, he should propose the corresponding requests on materials and fiber structure and processing geometry accuracy so that he can ensure the quality of this kind of optical fiber.