• No results found

Optical fiber coatings: high-modulus coatings for fibers with a low microbending sensitivity

N/A
N/A
Protected

Academic year: 2021

Share "Optical fiber coatings: high-modulus coatings for fibers with a low microbending sensitivity"

Copied!
6
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Optical fiber coatings: high-modulus coatings for fibers with a

low microbending sensitivity

Citation for published version (APA):

Bouten, P. C. P., Broer, D. J., Jochem, C. M. G., & Meeuwsen, T. P. M. (1989). Optical fiber coatings: high-modulus coatings for fibers with a low microbending sensitivity. Polymer Engineering and Science, 29(17), 1172-1176. https://doi.org/10.1002/pen.760291706

DOI:

10.1002/pen.760291706

Document status and date: Published: 01/01/1989 Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

Optical Fiber Coatings: High Modulus Coatings for Fibers

With a Low Microbending Sensitivity

P. C. P. BOUTEN, D.

J.

BROER, C. M.

G.

JOCHEM,and

T.

P. M. MEEUWSEN

Philips R e s e a r c h Laboratories

5600

J A

Eindhouen

The Netherlands

An important function of a n optical fiber coating is to prevent optical losses due to microbending induced by lateral forces on the fiber. To protect the fiber over a wide temperature range the modulus of the primary and the secondary coating should be low and high, respectively, and temperature independent. However, selecting the most appropriate organic coating materials introduces a

new source of optical losses. Since the linear thermal expansion coefficients of silica and the organic coatings differ by about two orders of magnitude, thermal fluctuations will cause axial stresses. Cooling may then induce bending or buck- ling of the glass fiber in the soft primary coating, resulting in increased trans- mission losses. This effect is especially pronounced when a high-modulus sec- ondary coating is selected with a glass transition temperature above 80°C. For this type of coating the difference in radial shrinkage between the buffer and the top coating during cooling from the curing temperature becomes important. The influence of primary coating thickness is discussed.

INTRODUCTION

ightwave communication based on fiber optics is

L

a n almost mature technology. Fibers can be cho- sen from a wide range of types. The most common configuration is a core of silica, doped with oxides of germanium or phosphorus to raise the refractive in- dex, a cladding of pure silica, and a n organic coating which may consist of several layers (1). The propa- gating light is confined to the core due to a difference

in refractive index. Typical diameters of the core are 50 pm for multimode fibers and 8 pm for single-mode fibers. For telecommunication purposes the diameter of the silica cladding is standardized at 125 pm. Usually, the diameter of the organic coating is 250 pm. The optical attenuation at a wavelength of 1300 nm is typically 0.55 dB/km for the multimode and 0.35 dB/km for the single-mode fibers, which permits data to be transmitted at a speed of 565 Mb/s and a bridge length of 45 km without signal amplification. On-line coating during fiber drawing is essential for protection of the fiber surface (2, 3). As the coat- ing itself must not abrade the glass surface during or after its application, thorough filtration of the liquid coating over a 1 pm filter is necessary (4, 5). To achieve fiber-drawing speeds >5 m/s the coating must be applied and solidified in a very short time, typically €0.1 s (6). The viscosity of the coating is preferably in the order of 1 P a s at the application temperature. For these reasons UV-curable acrylates are the preferred class of coating materials (7, 9).

Generally, they combine good rheological properties for filtration and high-speed extrusion and high cur- ing rates.

A very important function of the coating is to pro- tect the fiber from microbending losses (1 0). The term microbending refers to random bends with a short period (<1 mm) and small amplitude (typically a few microns). Such small period distortions may be caused by lateral forces, for instance when the fiber is wound on a drum under tension or when it is mounted in a cable. Microbending results in trans- mission losses through radiation when the optical modes leave the core. The total loss may become as

high as several dB/km (5). In order to decrease the microbending losses the fiber is surrounded by a dual coating system consisting of a soft buffer and a hard top layer (10-12). The soft coating provides the soft compliant enclosure which decouples the fiber me- chanically from its environment. The hard top coat- ing acts a s a stiffener which withstands external forces. These functions must be performed over a wide temperature range, preferably from -60 to +80"C. This imposes demands on the glass transition temperatures of the coatings, which must be outside this temperature window.

Introducing the dual coating system may increase the influence of temperature on the optical transmis- sion. Since the linear thermal expansion coefficients of silica and the organic coating usually differ by about two orders of magnitude, thermal fluctuations will cause axial stresses. A s the fiber is embedded in

(3)

P . C . P. Bouten, D.

J.

Broer, C . M. G . Jochem, and T. P. M . Meeuwsen a soft surrounding, this may lead to buckling of the

fiber at low temperatures (13-16). In this paper, we discuss how coating parameters such as modulus, glass transition temperature, and thermal expansion coefficient are related to the external force and tem- perature dependence of optical transmission proper- ties of the fiber.

special coating combination B-2/T-2 are compared with optical fibers coated with commercially avail- able coatings B-

1/T-

1. The mechanical properties of the various coatings used are summarized in Table 1 .

The microbending sensitivity of single-mode fibers was measured by winding 150 m on a drum covered with sand-paper

No. 120

and measuring the in- creased attenuation as a function of the wavelength.

EXTERNAL FORCE INDUCED OPTICAL

LOSSES

In the present state of the art, the modulus of the UV-curing coating materials used for optical fiber coating is 1 to 10 MPa for the buffer coating and 0.2 to 1 GPa for the top coating. However, as confirmed by model calculations (17), the sensitivity of the fiber

The winding force was 2 N and the diameter of-the drum 0.5 m. The thickness of the primary buffer coating was varied. The outer coating diameter was kept constant at 250 pm. Some of the test results are given in Table 2. From Table 2 it can be concluded to microbending losses under lateral pressure can be

reduced drastically by selecting a buffer coating with a n even lower modulus and a top coating with a

higher modulus. The lowest modulus of crosslinked rubbers, excluding fluids, is of the order of 1 MPa. The improvements with respect to the sensitivity to lateral forces must, therefore, be obtained by increas- ing the modulus of the top coating. The modulus of isotropic polymer glasses is normally in the range

of

2 to 3 GPa. However, these types of materials are susceptible to brittle fracture upon bending and are unsuited for optical fiber coating. As a compromise between toughness and high modulus we developed

a new UV-curable top coating with a tensile modulus of 1.6 GPa a t room temperature and a glass transition temperature, defined as the maximum in tan 6, of

100°C measured at a frequency of 1 Hz. Previously,

that the microbending sensitivity of the single-mode fibers coated with the coatings B-2/T-2 is very low when compared with the reference fiber coated with

B-l/T-l.

The influence of the buffer coating thick- ness is only minor. If no buffer-coating is applied at

all, the microbending sensitivity is comparable with the B- 1 /T- 1 combination.

In another test the optical attenuation of multi- mode fibers, which are more sensitive to microbend- ing than single-mode fibers, was measured during clamping between two parallel plates. One of the plates was provided with twelve sharp edged slots. The test length was 0.5 m and the pressure on the plates was 50 kPa. The temperature was varied be- tween -60 and +80"C. The results of this test are given in Table 3. As only relatively short fiber lengths were measured under rather severe test conditions, (6) we described the development of a low-modulus

polyetherurethane acrylate coating with a tensile modulus of about 1 MPa at room temperature. In the tests described below, optical fibers coated with this

the relation with the deformations which can occur in normal practice is less obvious than is the case with the sand-paper test. On the other hand, it is possible to perform these measurements at various

Table 1. Moduli (MPa) of the Coatings Used in This Study. The Dynamic Moduli Were Measured at a Frequency of 1 Hz. Temperature Buffer Coating Top Coaling

(C) 6-1 B-2 1-1 T-2 Tensile Modulus 20 Dyn. Modulus -60 -40 -20 0 20 40 60 80 3.0 1.3 1413 63 1 1259 89 1 85 178 22 3 6.3 1.6 1.3 1.3 1

.o

1

.o

1

.o

1

.o

345 1585 1250 930 708 398 120 23 13 1617 1995 1820 1660 1622 1585 1259 794 501

Table 2. Microbending Losses at Various Wavelengths of Single-Mode Fibers Tested With the Sand-Paper Test. Optical Attenuation

Buffer TOP Buffer Coating Increase (dB/km)

Coating Coating Thickness (pm) 1300 nm 1550 nm 1700 nm

8-1 T-1 32 1.3 2.7 5.5 8-2 T-2 33 c0.05 C0.05 0.1 8-2 T-2 27 0.1 0.1 0.6 8-2 T-2 19 0.1 0.1 0.6 8-2 T-2 12 0.1 0.1 0.6 none T-2

-

1.3 2.0 3.1

(4)

Optical Fiber Coatings temperatures. In Table 3 it can be seen that at 20°C

the B-2/T-2 coating combination is superior with respect to the resistance to the parallel plate test at

a buffer coating thickness of 35 pm. When the thick- ness of the buffer coating is 11 pm, the sensitivity to lateral pressure in this test is similar to that of the B-l/T-l coating at room temperature. However, it is considerably lower when measured both at reduced and at elevated temperatures. From Table 1 it already could be concluded that for the B-l/T-l combination the modulus of the buffer coating increases at lower temperatures, which in fact results in a pseudo-one- layer coating, whereas at high temperatures the mod- ulus of the top coating decreases to a n almost rubber- like value. This means that at -60°C as well as at

+80"C the coating conditions are far from optimum for withstanding lateral forces. Because of the ther- mally induced losses at low temperatures, which will be discussed in the next section, the optical atten- uation of the 35 pm B-2 coating was not measured as

a function of temperature.

THERMALLY INDUCED OPTICAL LOSSES A s mentioned in the introduction, the opical trans- mission of dual-coated fibers is sensitive to temper- ature fluctuations. Thermal shrinkage of the high- modulus top coating during cooling causes axial compression forces on the silica fiber which ulti- mately may lead to buckling of the fiber in its soft surrounding. I t is to be expected that this will espe- cially be the case when a top coating is selected with

a high modulus, which increases the compression forces, and when a buffer coating is selected with a

low Tg which reduces the radial counter-forces at

lower temperatures. The temperature sensitivity of

Table 3. Microbending Sensitivity (A

=

850 nm) of a Multimode F

both single-mode and multimode fibers was tested by recording the optical transmission of a loosely wound fiber on a 0.5 m drum during temperature cycling between -60 and +80"C. Table 4 summarizes the test results.

Fibers coated with the B- 1 /T- 1 combination with- stand the temperature cycling test without any ap- preciable increase of the optical attenuation. How- ever, when the B-2/T-2 combination is applied in the same coating thicknesses, the optical attenuation at low temperatures of both the single-mode and the multimode fiber increases dramatically. Only by de- creasing the thickness of the buffer coating below 20 pm are optical fibers obtained with a good tempera- ture cycling resistance.

The fact that only a small difference in thickness of the buffer coating can cause such a large differ- ence in the thermal dependence of the optical behav- ior is surprising. It becomes even more surprising when the axial compression forces are calculated and compared with the minimum force to be exceeded in order to cause thermal buckling [ 15, 16, 18). Mainly because of the incompressibility of the buffer coat- ing, i.e. the Poisson ratio u approaches 0.5, the min- imum force needed to initiate buckling is about two orders of magnitude higher than the actual axial force. Apparently, apart from the difference in ther- mal expansion coefficient in the axial direction be- tween the top coating and the silica fiber, there must be another phenomenon responsible or co-responsi- ble for the increase in optical attenuation at lower temperatures.

Microscopic observations revealed that the fibers which were sensitive to thermally induced optical attenuation also showed a tendency to delaminate at the silica-organic coating interface. The temperature

7ber at Various Temperatures as Measured by the Parallel Plates Test. Optical Attenuation

Buffer TOP Buffer Coating Increase (dB)

Coating Coating Thickness (Nm) -6OOC 20% 8OoC

B-1 T-1 33 0.69 0.21 0.37

8-2 T-2 35

-

0.05

-

8-2 T-2 11 0.38 0.1 9 0.18

Table 4. The Increase in Optical Attenuation During Temperature Cycling. Single-Mode Fibers Are Measured at 1300 nm, Multimode at 850 nm.

Optical Attenuation Buffer TOP Buffer Coating Increase (dB/km) Fiber Type Coating Coating Thickness (pm) -6OOC +8OoC

Single-Mode B- 1 T-1 32 C0.05 0.0 Single-Mode 8-2 T-2 33 >4 0.0 Single-Mode B-2 T-2 27 1.4 0.0 Single-Mode 8-2 T-2 24 0.8 0.0 Single-Mode B-2 T-2 19 0.0 0.0 Single-Mode 8-2 T-2 12 0.0 0.0 Multirnode B-1 T-1 33 0.1 C0.05 Multirnode B-2 T-2 35 >10 <0.05 Multimode 8-2 T-2 27 1.1 c0.05 Multimode 8-2 T-2 21 0.5 <0.05 Multimode B-2 T-2 13 0.1 C0.05

(5)

P . C . P . Bouten, D . J . Broer, C . M . G. Jochem, and T . P . M . Meeuwsen at which this partial delamination sets in is around

room temperature in case of the thicker buffer coat- ings (33 and 35 pm B-2) and becomes lower at de- creasing thicknesses of the buffer coating. It is ob- vious that a s soon as the fiber is no longer completely surrounded by a n incompressible rubber but partly by air, the above-mentioned calculations of the buck- ling force are no longer valid. The axial compressive force still exists, but the counter-force in radial or lateral directions, which should prevent buckling, is locally absent.

Now that the delamination phenomenon explains the existence of thermal buckling, the question arises a s to why there should be any difference between for instance a 19 pm and a 33 pm buffer coating in the case of the B-2/T-2 combination. The chemical and physical interactions in the interface are not ex- pected to be affected by the coating thickness and cannot explain why the 33 pm buffer loses its adhe- sion and the 18 pm buffer does not. The explanation can be found by considering the differences in ther- mal expansion of both coatings in the radial direc- tion. Normally, when one coating is applied, the coat- ing tends to build up a compressive force in the radial direction when cooled down from its curing temper- ature. This temperature lies between 80 and 150°C and depends on the fiber drawing rate and the length of the UV lamps for the photocrosslinking of the acrylate coatings. The latter produce not only UV light but also a considerable amount of heat. The direct cause for this build-up of compressive force is again the difference in thermal expansion coefficient of the silica and that of the organic coating. The forces become considerable when the T , of the coat- ing is passed. A s long as the fiber is concentrically coated, these forces will not affect the performance of the fiber.

However, when the coating consists of two layers with different thermal expansion coefficients, the situation becomes more complicated. When cooled down, both coatings tend to shrink faster than the silica fiber, and as the expansion coefficient of a

rubber is higher than that of a glassy polymer, the buffer coating in turn tends to shrink faster than the outer top coating. Calculations show (1 8) that there exists a discrete maximum thickness of the buffer coating up to which this coating is compressed on

the fiber. When this threshold thickness is exceeded the buffer coating will become stretched between the top coating and the silica fiber during cooling from the curing temperature to room temperature or be- low. And as the adhesion between the silica and the buffer in most cases in the weakest link, this stress will result in the observed delamination behavior. The threshold buffer coating thickness 7 can be cal-

culated from Eq 1.

and ffbuf and atop refer to the thermal expansion coef-

ficients of the buffer and top coating, respectively. The thermal expansion of the silica fiber has been ignored as it is more than two orders of magnitude smaller than that of the organic materials. The ref- erence temperature is the temperature at which the two coatings start to exhibit a different shrinkage behavior during cooling from the application temper- ature. This can either be the curing temperature itself or, the T,. of the top coating.

Fig. 1 shows the thermal expansion coefficients of the coatings studied. It can be seen that there are two distinct levels. The lower level is that of the polymer glasses and lies at 5.10-50C-', while the upper level is that of a rubbery material and lies a t about 20 to 25. 10-50C-'. It is obvious that below 90°C the coating combination B-2/T-2 with the low T , buffer coating and the high Tg top coating shows a larger mismatch in thermal expansion coefficient than combination B-1/T-1. From the above given equation it can be concluded that a larger mismatch in O( should lead to

a smaller 7 . Based on the experimental data obtained

from Fig. 1 the threshold buffer-coating thickness has been calculated for the B-2/T-2 coatings a s a function of the temperature. The result is given in F i g . 2. The reference temperature has been taken a s 90°C corresponding to the onset of the decrease in a. Fig. 2 should be read as follows. Cooling down from high temperatures, e.g. 90°C, with a certain buffer- coating thickness 7 to a compressive stress in radial

direction is developed. However, when the difference

30

n c

0

m O

20

'0

8

f- Y

10

0

-80

-40

0

-

LO

80 120

T

("C)

In this equation d, is the diameter of the silica fiber,

T the actual temperature, T , a reference temperature Fig. 1 . Thermal expansion coefficients

01 of the tested coatings uersus temperature T .

(6)

Optical Fiber Coatings

-m

h

60

I

1

$!‘

30

t

0

-30

uu15

-

40

65

T

(pml

Fig. 2. Calculated temperature T, a t which the radial compressive stress is converted to radial tensile stress as a function of the thickness of the primary bu&fer coat- ing T.

in thermal expansion coefficients of the two coatings becomes too large, the compressive stress starts de- creasing again. The curve in Fig. 2 indicates the temperature at which the compressive stress equals zero just before it becomes negative, or in other words just before it builds-up a tensile stress. In Fig. 2 it can be seen that during cooling of the fiber to -60°C. a buffer coating of less than about 20 pm always remains under a radial compressive stress. Thicker coatings are under radial tensile stress at low tem- peratures. As the stress is circumferential this does not result in strain but in delamination.

CONCLUSIONS

The moduli of the buffer and the top coating and the temperature dependence of these moduli are of vital importance for the proper functioning of a n optical fiber. The selected coating combination with a 1 MPa buffer and a 1.6 GPa top coating makes the fiber insensitive to microbending losses induced by mechanical forces even when the fiber is pressed against a rough surface like sand-paper. The Tg of the coating are chosen to be low and high for the buffer and the top coating, respectively, in order to maintain the protective properties over a broad tem- perature range, e.g. from -60°C to +80”C. As a con-

sequence of this large difference in Tg’s there exists also a large difference in thermal expansion coeffi- cient between both coatings. This difference may lead to delamination of the coating from the silica fiber at lower temperatures due to a stress build-up in the radial direction. As soon as the delamination sets in, the optical attenuation increases dramati- cally due to fiber buckling induced by the compres- sive forces in the axial direction. In this paper, it has been shown that buckling can be prevented by a n appropriate choice of the thickness of the buffer coating. Both the data and the calculations point to a maximum buffer coating thickness of 20 pm for the given coating combination, which is somewhat smaller than normally applied.

ACKNOWLEDGMENTS

The authors wish to thank G. V. A. Aben, J. W. C . van der Ligt, H. J. M. Timmermans, G. N. Mol. and H. Vonk for their technical assistance during the performance of the measurements and D. C. L. Vangheluwe for the valuable discussions at the start of the investigation.

REFERENCES

1. L. L. Blyler. Jr.. F. V. Dimarcello, A. C. Hart, and R. G. Huff, “Polymers for High Technology,” Ed. M. J. Bow- den and S . R. Turner, ACS Symp. Ser. 346, Amer. Chem. SOC., Washington, DC (1987). Ch. 34, p. 410. 2. C. R. Kurkjian, R. V. Albarino, J. T. Krause, H. N.

Vazirani, F. V. DiMarcello, S. Torza. and H. Schonhorn, Appl. Phys. Lett., 28. 588 (1976).

3. D. Kalish, P. L. Key, C. R. Kurkjian, 8. T. Tariyal, and T. T. Wang, Optical Fibre Telecommunication, Aca- demic Press (1979). Ch. 12.

4. T. J. Miller, Amer. Chern. SOC. Prep. Org. Coat. Plast. Chem. 40.217 (1979).

5. C. J. Aloiso, L. L. Blyler, Jr., and J. T. Chapin, Proc. 4th Int. Conf. Plastics in Telecommunication, The Plas- tic and Rubber Institute. London, Sept. 17-19 (1986). 6. D. J. Broer and G. N. Mol, J. Lightwave Techn., LT-

4(7), 938 (1986).

7. H. Schonhorn, C. R. Kurkjian, R. E. Jaeger, H. N. Vazirani, R. V. Albarino, and F. V. DiMarcello, Appl. Phys. Lett., 29. 712 (1976).

8. U. C. Paek and C. M. Schroeder, Appl. Optics 20, 1230 (1981).

9. R. E. Ansel and J. J. Stanton, Adv. Ceramics, Vol. 2: Physics of Fiber Optics, Ed. B. Bendow andS. S. Mitra, Am. Ceram. SOC.. Columbus (1981). p. 27.

10. D. Gloge, Bell System Techn. J., 54(2), 245 (1975).

11. D. Gloge and W. B. Gardner, Optical Fiber Telecom- munications, Ed. A. G. Chynowetz and S. E. Miller, Academic Press, New York (1979). Ch. 6.

12. M. Yoshizawa, T. Yabuta, N. Kojima, and Y. Negishi, Appl. Optics. 20. 3146 (1981).

13. Y. Katsuyama, Y. Mitsunaya, Y. Ishida, and K. Ishihara, Appl. Optics, 19, 4200 (1980).

14. T. Yabuta, N. Yoshizawa, and K. Ishihara, Appl. Op- tics, 22, 2356 (1983).

15. D. C. L. Vangheluwe, Appl. Optics, 23, 2045 (1984). 16. T. A. Lenahan, AT&T Techn. J., 64,1565 (1985). 17. D. C. L. Vangheluwe, private communication.

18. P. C. P. Bouten, D. J. Broer. C. M. G. Jochem. T. P. M. Meeuwsen, and H. J. M. Timmermans, to be published.

p. 1211.

Referenties

GERELATEERDE DOCUMENTEN

(2010) Phishing is a scam to steal valuable information by sending out fake emails, or spam, written to appear as if they have been sent by banks or other reputable organizations

(2010) Phishing is a scam to steal valuable information by sending out fake emails, or spam, written to appear as if they have been sent by banks or other reputable organizations

Only the primary source of information or study regarding the impacts was included on the score sheet Magnitude Only species for which impacts MO or higher were expected or

Het antwoord op de vraag ligt voor de hand (dat het in de ruimte de bol en in het vlak de cirkel is bedenkt de klas zonder veel aarzelen), maar toch heeft het bewijs, zcrvertel

Omdat van deze thermometers door ijking de constanten van de tem~ • peratuur-weerstandrelatie bekend zijn, kan nu door de weerstand te meten de temperatuur van

IBP, inflammatory back pain; NSAIDs, Non-Steroidal Anti-Inflammatory Drugs; IBD, inflammatory bowel disease; HLA-B27, Human Leucocyte Antigen B27; ESR, erythrocyte sedimentation

Error t-Statistic Prob.. Error t-Statistic

initial projected savings of approximately R20 million (±$3.3 million) per year (Mckenzie and Wegelin, 2005) were in fact exceeded and after the first full year of