2526 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 31, NO.

15, AUGUST 1, 2013

Sensitivity Characteristics of Fabry-Perot

Pressure Sensors Based on Hollow-Core
Microstructured Fibers
Long Jin, Bai-Ou Guan, and Huifeng Wei

Abstract—In this paper, the sensitivity characteristics of the transverse geometry is greatly simplified [3], [4]. Its guid-
Fabry-Perot (F-P) hydrostatic pressure sensor based on two ance relies on the antiresonance between the core mode and the
different hollow-core (HC) microstructured optical fibers are mode of the silica wall around the core.
experimentally and theoretically investigated. The sensors are
fabricated by simply splicing a length of HC fiber to single- HC-PBGFs have been exploited as photonic sensors in the
mode fibers. Hydrostatic pressure is measured by monitoring recent years, taking advantage of their unique guidance mecha-
the wavelength shifts of the interferometric fringes as a result nism and microstructured geometry [5]. For example, an inter-
of the two reflection beams at the splicing points. The measured ferometric fiber optic gyro has been implemented based on this
pressure sensitivities of F-P sensors fabricated on the simplified fiber, which has reduced Kerr effect, temperature sensitivity and
HC microstructured fiber and hollow-core photonic bandgap
fiber (HC-PBGF) are 17.3 and 23.4 pm/MPa, respectively. Faraday effect [6]. Fiber grating sensors can be fabricated on the
Theoretical investigation is then carried out based on the analysis bandgap fibers through heat treatment. The long period grating
of elastic properties for the individual fibers. The calculated result sensors, which contain periodic -laser-induced deforma-
suggests that the pressure sensitivities are dominantly determined tions, are sensitive to axial strain and insensitive to environ-
by the induced changes in cavity length. In comparison, the con- mental temperature [7], [8]. Hydrostatic pressure measurement
tribution of mode-index change is slight. The mechanisms behind
the mode-index changes for the two fibers are clarified by ana- has been realized with a HC-PBGF by monitoring the spectral
lyzing the deformation of the fiber structure under pressure. The shift of its transmission band or the change in phase difference
holey microstructure is highly deformable compared to the solid between orthogonal polarizations [9]–[11].
fiber, which provide another dimension for the implementation of In-line Fabry-Perot sensors have also been implemented on
tunable photonic devices.
HC-PBGFs for the measurement of temperature, axial stain vi-
Index Terms—Fabry–Perot sensors, fiber optic sensors, photonic bration, ultrasound signal, and even magnetic field [12]–[16].
bandgap fibers, pressure sensors. The sensors present advantages including low cost, simple fab-
rication process and the ability to work under high temperature.
I. INTRODUCTION The low-loss transmission in the PBGF is beneficial for its mul-
tiplexing capability. The air guiding also allow the F-P etalon to
have a length of mm or cm order to obtain dense interferometric

H OLLOW-CORE photonic bandgap fiber (HC-PBGF) is

a kind of optical fiber which guides light in an air core
relying on photonic bandgap effect, rather than total internal re-
fringes. In contrast, the F-P cavities based on silica capillaries
are typically tens of microns in length [17], [18]. F-P sensors
based on simplified HC microstructured fibers have also been
flection [1]. The establishment of the photonic bandgap guiding
demonstrated for the measurement of refractive index [19], [20].
depends on its periodic cladding which consists of a hexagonal
Note that the existence of the microstructure greatly changes
array of air holes. Low-loss transmission can be realized within
the elasticity of a silica fiber and the effect of the microstruc-
one or more spectral windows. The cladding can be also engi-
ture to the sensitivity is not totally clear so far. In this paper,
neered to have a “Kagome” profile to obtain much wider trans-
the pressure sensitivities of F-P sensors based on the simplified
mission range [2]. Recently, an alternative air-core fiber has
bandgap fiber and HC-PBGF are experimentally and theoret-
been proposed, which has only one ring of air-hole cladding and
ically investigated, respectively. Theoretical investigations are
carried out for the individual fibers based on the analysis of their
Manuscript received January 22, 2013; revised April 16, 2013 and May
27, 2013; accepted June 11, 2013. Date of publication June 17, 2013; date of
elastic properties. The calculated result suggests that the pres-
current version June 28, 2013. This work was supported by National Natural sure sensitivities for the F-P sensors are mainly resulted from the
Science Foundation of China (Grant No. 11104117), in part by the National changes in cavity length. In contrast, the effective-index change
High Technology Research and Development Program of China under grant
2010CB735904, and in part by the National High Technology Research and induced by the fiber deformation is extremely small. The dif-
Development Program of China under Grant 2012AA041203. ference in mode-index changes between the two fibers are an-
L. Jin and B. O. Guan are with Institute of Photonics Technology, Jinan Uni- alyzed based on the description of the deformations of the mi-
versity, Guangzhou 510632, China.
H. Wei is with the State Key Laboratory of Optical Fiber and Cable Manufac- crostructures under pressure. Theoretical analysis suggests that
ture Technology, Yangtze Optical Fiber and Cable Company Ltd. R&D Center, the holey microstructure can be highly deformable, especially
Wuhan 430073, China. for directional stresses, as a result of the large air-filling ratios,
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. which offers the feasibility of further implementation of tunable
Digital Object Identifier 10.1109/JLT.2013.2269136 photonic devices based on the air-core fibers.

0733-8724/$31.00 © 2013 IEEE

JIN et al.: SENSITIVITY CHARACTERISTICS OF FABRY-PEROT PRESSURE SENSORS BASED ON HOLLOW-CORE MICROSTRUCTURED FIBERS 2527

Fig. 1. Microscopic images of the transverse geometry of (a) simplified mi-

crostructure fiber and (b) HC-PBGF.

II. EXPERIMENT

Fig. 1 shows the microscopic cross sections of the two HC

fibers. Fiber I is a simplified bandgap fiber, which is fabricated
by Yangtze Optical Fiber and Cable Company Ltd. The fiber
has an air core surrounded by a ring of thin silica wall. The air
core and the outer silica cladding are connected with six bridges.
Light can be guided in the air core due to the antiresonance be-
tween the core mode and the modes of the inner silica wall. The
outer and inner radii of the outer silica cladding are
and , respectively. The distance from the fiber
center axis to the wall of the air core is about 16 . The sim-
plified PBGF presents a loss peak at around 770 nm, as a re-
sult of the resonance of the thin wall. The loss wavelength is
determined by the thickness of the wall between adjacent air
holes via the the relation [4]. Substi-
tuting and , we can obtain the thickness of
the wall . Fiber II is the HC-PBGF (NKT Photonics,
HC-1550-2). It has an air core surrounded by a holey cladding
with periodic air holes. The spacing or pitch between the adja-
cent air holes is . The air-filling ratio, i.e., the ratio
between the cladding-hole diameter and the pitch is .
The outer radius and the holey-cladding radius are
and , respectively. The radius of the air core
is about . The periodic structure establishes a
photonic bandgap and therefore the fiber presents a transmission
window ranges from 1400 to above 1700 nm. The transmission
wavelength range is determined by the geometrical parameters
of the air-silica cladding, including the pitch and the air-filling
ratio.
F-P cavities are fabricated by splicing a section of HC fiber
to singlemode fibers at both ends. The arc fusion parameters in-
cluding arc duration and strength have been optimized to obtain
higher extinction ratio. The polymer jackets have been removed
before the splicing. Three F-P sensors with cavity lengths of
12 and 20 mm are fabricated with fibers I and II, respectively.
Fig. 2(a) shows the setup to measure the pressure responses of
the individual F-P sensors. Hydrostatic pressure is subjected to Fig. 2. (a) Experimental setup for the measurement of the pressure response.
the sensor by use of a pressure chamber with a maximum pres- (b) Measured reflection spectra for F-P cavities based on the two hollow core
fibers. (c) Measured wavelength shifts as a function of applied pressure for the
sure of 10 MPa. A broadband light source and an optical spec- F-P sensors.
trum analyzer are used to record the fringe shift. The light is
launched onto the sensor and the reflection signal is retrieved
via an optical circulator. Fig. 2(b) shows the measured spectra of Fig. 2(c) shows the measured wavelength shifts of the transmis-
the F-P cavities based on hollow-core fibers at around 1550 nm. sion dip closest to 1550 nm of the three sensors. The fringes
2528 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 31, NO. 15, AUGUST 1, 2013

blue shift with applied pressure for all the three sensors. The
measured sensitivities for fibers I and II are 17.3 and 23.4
pm/MPa, respectively.

III. THEORETICAL ANALYSIS

The fringes in the reflection spectrum are a result of the inter-
ference between the two reflection beams at the two interfaces
between the SMF and the HC fibers. The spectrum is determined
by the phase difference between the two beams

(1) Fig. 3. Ressembled fiber transverse geometries for the calculation of the stress
and strain distributions. (a) Simplified HC bandgap fiber. (b) HC-PBGF. The
where is the cavity length, is wavelength and is the ef- holey cladding is considered as a uniform material in elasticity.
fective index of the fundamental mode of the HC fiber. The in-
terference signal reaches its maximum intensity when the phase
difference satisfies , where is integral, and the and . The parameter represents
peak wavelengths are determined by the ratio between the area defined by the outer radius and the
actual cross sectional area of the silica ring. With the conditions
(2)
(6a)
When the cavity is perturbed, the phase difference changes as a
(6b)
result of the changes in cavity length and effective index. The
(6c)
pressure sensitivity for the th peak can be expressed by
the constants A, C and D can be found and the stress and strain
(3) field can be expressed as follows

The former term in the bracket depicts the change in cavity

length under pressure and the latter one represents the index (7)
change. In the following text, these two effects are discussed,
respectively.
(8)
A. The Effect of Cavity-Length Change
1) FiberI: Fig. 3(a) shows the constructed transverse geom-
Based on (3) and (8), the contribution of the cavity-length
etry for the simplified HC fiber to calculate the response of the
change to the pressure sensitivity is
F-P pressure sensor. Considering the extremely thin walls be-
. Compared with the phase-change expression for a solid
tween the air holes, the deformation of the fiber is dominantly
fiber, the sensitivity is enhanced by times. Substituting
determined by the outer silica cladding. The elasticity of fiber
, , and ,
I, i. e., the simplified bandgap fiber can be analyzed with a
the cavity-length change contribute 17.7 pm/MPa to the
single-layer model with outer and inner radii and . The
pressure sensitivity.
stress and strain over silica can be expressed by [21]
2) FiberII: For fiber II, i. e., the photonic bandgap fiber, a
two-layer model described in [22]–[24] can be applied to calcu-
(4) late the stress and strain fields, as shown in Fig. 3(b). The inner
layer is the honeycomb cladding, with its inner and outer radii of
and . The outer layer is silica glass with inner and outer
and (See equation at bottom of page) where and are the radii of and . The holey cladding can be considered as
Young’s Modulus and Poisson’s ratio of silica glass. We define a uniform material since it contains an air hole structure with

(5)
JIN et al.: SENSITIVITY CHARACTERISTICS OF FABRY-PEROT PRESSURE SENSORS BASED ON HOLLOW-CORE MICROSTRUCTURED FIBERS 2529

tens of layers. The material is inhomogeneous and its Young’s contributes 21.2 pm/MPa out of the pressure sensitivity. We
modulus and Poisson’s ratio can be expressed by [22] also found that if the holey cladding is removed, the sensitivity
only changes by 0.5%, through further calculation. This is be-
cause the holey cladding has a large air-filling ratio and small
(9) effective Young’s modulus, and the longitudinal strain is almost
determined by the outer silica cladding.
and
B. The Effect of Effective-Index Change
(10) The applied hydrostatic pressure subjected to the HC fiber
can also induce a mode-index change, which can possibly con-
tributes to the round-trip phase change and the pressure sensi-
The small amplitudes of and suggest the fact that the tivity, according to (3). The mode-index change comes from the
honeycomb structure is highly compressible for a directional changes in fiber transverse geometry and the refractive-index
in-plane stress. is proportional to the actual cross sectional change of the silica walls. In this subsection, this effect is dis-
area of silica glass. and cussed for both fibers.
indicates that the applied directional in-plane strain can induce 1) FiberI: For fiber I, considering the cell walls are too thin,
identical strain along the orthogonal in-plane direction but can the deformation of the microstructure and the refractive-index
hardly induce a change in longitudinal strain. The stress over change of the silica walls are determined by the deformation of
the individual regions can also be expressed by the outer silica ring. The calculation of pressure-induced mode-
index change is performed with the following steps: First, the
deformation of the silica ring are calculated with the single-layer
(11) structure. Second, the geometrical parameters and the mate-
rial-index of the microstructure are determined by the deformed
silica ring. Third, mode solving is carried out for the deformed
Substituting their respective Young’s Moduli and Poisson’s ra- microstructure.
tios, the strain field over the silica outer cladding can be ex- The radial strain at the inner boundary of the silica ring is cal-
pressed by culated as 2.95 for an applied pressure of 1 MPa, based on
(8). That means the diameter of the silica ring becomes slightly
larger when the cavity is under pressure, because the longi-
tudinal component of the applied pressure tends to enlarge it,
(12-a) due to the positive Poisson’s ratio. The silica ring stretches the
and over the holey cladding bridges between the core and the ring, which causes a defor-
mation of the microstructure and corresponding material-index
change of silica walls. Fig. 4 shows a one-sixth model for fiber
(12-b) I. The applied displacement at point C is determined by
. Since the angles between AD, BD and CD are all
120 , the silica bridges and the core walls experience normal
The amplitudes of , , and can be found with the fol- stresses with identical amplitude . Assume the radial dis-
lowing conditions, including the boundary conditions at the placement at point D is . For section CD, we have
inner and outer surface

(13-a) (14)

(13-b)
For sections AD and BD,
the requirement of continuity
(15)
(13-c)
(13-d) With (14) and (15), the amplitudes of and can be obtained.
Note that the thickness of the walls are also changed considering
and the requirement at the end face the Poisson’s ratio. The mode index of the deformed microstruc-
ture is calculated and its contribution to the pressure sensitivity
(13-e) is estimated as . We found that the con-
tribution of the refractive-index change to mode-index change
With the plane strain approximation,
is about two orders lower than the effect of the microstructure
(13-f) deformation and can be ignored.
2) FiberII: For fiber II, the bandgap guidance in the air core
With the found constants , and , the longitudinal strain relies on the cladding structure and the mode property is largely
can be determined. We found that the change of cavity length determined by the geometry of the holey cladding. In addition,
2530 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 31, NO. 15, AUGUST 1, 2013

Fig. 4. Calculated deformation of the silica walls and first order principle stress
over the microstructure for fiber III when the radial displacement of the inner
boundary of the silica ring is 1 .
Fig. 5. Left: The ressembled fiber transverse geometry for the calculation
of mode indexes for fiber II. Right: Schematic deformations of two selected
cladding cells. The dashed curves define the orignal cells and the solid curves
the mode index is also affected by the size and shape of the represents the deformed ones.
air core. Generally speaking, larger core results in higher mode
index (closer to 1). The microstructure change under pressure is
quite complicated for the two-layer structure. The strain distri- of 6.25 . The effective index of the fundamental mode is
bution can be characterized by (11) and (12). With the calculated first calculated by use of finite element method. Then the mode
, and , we found that the amplitude of the latter terms solving process is performed for a fiber structure with a slightly
are much smaller than the former ones and the in-plane strain reduced core, but the cladding structure remains unchanged, in
distribution can be approximately expressed by order to estimate the index change under pressure. The pres-
sure sensitivity contributed from the index change is estimated
(16) as only .

IV. DISCUSSION
where is calculated based on (11)–(13).
Eq. (16) suggests that the in-plane strain is nonuniform over We found that the cavity-length change plays a dominate
the holey cladding. The region near the core experiences much role for both the sensors by comparing the calculated and mea-
higher strain than at near the silica outer cladding, since the sured result in Fig. 2(c). The pressure-induced cavity length
strain is proportional to . For example, both the radial and is largely determined by the cross sectional area of the outer
azimuth strain at (at the inner boundary of the silica cladding. The enhancement can be measured with the
holey cladding) is about 30 times higher than at (at amplification factor . The definition of indicates that the
the outer boundary). pressure sensitivity can be effectively enhanced by reducing
Since has a negative value, the cladding cells are com- the thickness of the silica outer cladding.
pressed along the azimuth direction and stretched along the ra- In contrast, contributions of mode-index change to the sensi-
dial direction. Fig. 5 demonstrates the schematic deformation of tivities are much smaller. Note that the simplified bandgap fiber
two selected cells when the cavity is under hydrostatic pressure. presents a much higher and positive index change rate than the
However, since the in-plane strain is of microstrain order with HC-PBGF and we intend to discuss the different mechanisms
an applied pressure of 1 MPa, the deformation can hardly affect of mode-index changes between the two fibers. The elementary
the bandgap. The cell walls are bent when the holey cladding is unit of the microsturcture is the hexagonal air-hole cell for the
deformed and the refractive index of the silica walls may change both fibers. This difference is mainly due to the existence of
as a result of the elasto-optic effect. However, this change is the defect core for fiber II, which is created by omitting seven
relatively small, compared with the silica-air index difference air holes from the holey structure. If the air core is excluded in
which is a critical parameter in the determination of the profile the microstructure, i. e., the fiber contains an ideally uniform
of the bandagp. This analysis indicates that the induced defor- hexagon array, all the cell walls experience normal stresses and
mation of the holey cladding is too small to induce the bandgap strains when the fiber is under pressure, like the case for fiber
shift and mode-index change. I. The air-hole cells can keep their hexagonal profile rather than
On the other hand, the air core also deforms when the cavity deforms like those in Fig. 5. The introduction of the defect core
is under pressure. The calculated radial strain at the inner wall of significantly changes the elasticity of the fiber. The longitudinal
the air core is based on (16), which suggests that component of the applied pressure tends to induce expansion of
the core shrinks due to the deformation of the holey cladding. the holey cladding. The defect core provides a freedom for the
With the composed transverse fiber geometry in Fig. 5, we cal- expansion. As a result, the air holes surrounding the core can
culated the mode index for the HC-PBGF. The core has a radius not hold ideal hexagonal shape.
JIN et al.: SENSITIVITY CHARACTERISTICS OF FABRY-PEROT PRESSURE SENSORS BASED ON HOLLOW-CORE MICROSTRUCTURED FIBERS 2531

In comparison, the F-P pressure sensors fabricated in solid- REFERENCES

core microstructured optical fibers present much lower pres-
sure sensitivities than the HC ones [25]–[27]. This can be at- [1] C. M. Smith, N. Venkataraman, M. T. Gallagher, D. Müller, J. A. West,
N. F. Borrelli, D. C. Allan, and K. W. Koch, “Low-loss hollow-core
tributed to two reasons. First, the solid-core fibers have lower silica/air photonic bandgap fibre,” Nature, vol. 424, no. 6949, pp.
air-filling ratios. Second, the mode-index change induced by 657–659, Aug. 2003.
applied pressure significantly compensates the effect of cavity [2] F. Couny, F. Benabid, and P. S. Light, “Large-pitch kagome-struc-
tured hollow-core photonic crystal fiber,” Opt. Lett., vol. 31, no. 24,
length change. pp. 3574–3576, Dec. 2006.
It is now interesting to discuss how to enhance the microstruc- [3] S. Février, B. Beaudou, and P. Viale, “Understanding origin of loss in
ture deformation and induce higher mode-index rates for the two large pitch hollow-core photonic crystal fibers and their design simpli-
fication,” Opt. Exp., vol. 18, no. 5, pp. 5142–5150, Mar. 2010.
fibers. Note that the honeycomb structure is highly deformable [4] F. Gérôme, R. Jamier, J. Auguste, G. Humbert, and J. Blondy, “Simpli-
by in-plane stress due to the thin cell walls and the consequent fied hollow-core photonic crystal fiber,” Opt. Lett., vol. 35, no. 8, pp.
low effective Young’s modulus. That means one can possibly 1157–1159, Apr. 2010.
[5] W. Jin, H. F. Xuan, and H. L. Ho, “Sensing with hollow-core photonic
induce significant deformation of the microstructure with much
bandgap fibers,” Meas. Sci. Technol., vol. 21, no. 9, Sep. 2010, Article
lower pressure in principle. The main limitation factor is the Number: 094014.
silica outer cladding. Due to its high stiffness, the silica re- [6] M. Digonnet, S. Blin, H. K. Kim, V. Dangui, and G. Kino, “Sensitivity
gion acts like a shield and greatly reduces the effect of the ap- and stability of an air-core fibre-optic gyroscope,” Meas. Sci. Technol.,
vol. 18, no. 10, pp. 3089–3097, Oct. 2007.
plied pressure. For the microstructure of fiber I, the method is [7] Y. P. Wang, W. Jin, J. Ju, H. F. Xuan, H. L. Ho, L. M. Xiao, and D.
to increase the radial displacement at point C, i. e., the inner N. Wang, “Long period gratings in air-core photonic bandgap fibers,”
boundary of the silica ring. Based on (8), the radial strain at Opt. Exp., vol. 16, no. 4, pp. 2784–2790, Feb. 2008.
[8] Z. Wu, Z. Wang, Y. Liu, T. Han, S. Li, and H. Wei, “Mechanism and
point C can be expressed by . This indi- characteristics of long period fiber gratings in simplified hollow-core
cates that the mode-index change rate can be enhanced by in- photonic crystal fibers,” Opt. Exp., vol. 19, no. 18, pp. 17344–17349,
creasing the amplitude of , by reducing the thickness of the Aug. 2011.
[9] R. E. P. de Oliveira, C. J. S. de Matos, J. G. Hayashi, and C. M. B.
silica ring while keeping the inner radius constant. In con- Cordeiro, “Pressure sensing based on nonconventional air-guiding
trast, the change rate of the core radius of fiber II is determined transmission windows in hollow-core photonic crystal fibers,” J.
by multiple parameters, based on the two-layer model. The re- Lightw. Technol., vol. 27, no. 11, pp. 1605–1609, Jun. 2009.
[10] G. Statkiewicz, T. Martynkien, and W. Urbanczyk, “Measurements of
duction in thickness of the outer silica cladding does not nec- birefringence and its sensitivity to hydrostatic pressure and elonga-
essarily induce higher index change. An alternative method is tion in photonic hollow core fiber with residual core ellipticity,” Opt.
to increase the air-filling ratio and therefore a lower effective Commun., vol. 255, no. 4–6, pp. 175–183, Nov. 2005.
[11] R. E. P. de Oliveira, C. J. S. de Matos, G. E. Nunes, and I. H. Bech-
Young’s modulus, based on (14). told, “Visible transmission windows in infrared hollow-core photonic
Note that the calculated result for HC bandgap fiber is dif- bandgap fiber: Characterization and response to pressure,” J. Opt. Soc.
ferent with the polymer-jacketed ones described in [23], due to Amer. B, vol. 29, no. 5, pp. 977–983, May 2012.
[12] Y. J. Rao, M. Deng, T. Zhu, and H. Li, “In-line Fabry-Perot etalons
the significant difference in elasticity between silica glass and based on hollow-core photonic bandgap fibers for high-temperature ap-
the polymer. The jacketed fibers elongate with applied pressure, plications,” J. Lightw. Technol., vol. 27, no. 19, pp. 4360–4365, Oct.
and the bare ones are shortened in contrast. On the other hand, 2009.
[13] Q. Shi, F. Y. Lv, Z. Wang, L. Jin, J. J. Hu, Z. Y. Liu, G. Y. Kai, and
the holey cladding of the jacketed fiber experiences positive ra-
X. Y. Dong, “Environmentally stable Fabry-Perot-type strain sensor
dial strain and the air core is enlarged. based on hollow-core photonic bandgap fiber,” IEEE Photon. Technol.
Lett., vol. 20, no. 4, pp. 237–239, Feb. 2008.
[14] T. Ke, T. Zhu, Y. J. Rao, and M. Deng, “Accelerometer based on
all-fiber Fabry-Perot interferometer formed by hollow-core pho-
V. CONCLUSION tonic crystal fiber,” Microw. Opt. Technol. Lett., vol. 52, no. 11, pp.
2531–2535, Nov. 2010.
[15] Y. J. Rao, W. Wang, T. Zhu, and D. Duan, “In-line fiber-optic Fabry-
In conclusion, we have investigated the sensitivity charac- Perot ultrasound sensor formed by hollow-core photonic-crystal fiber,”
teristics of F-P pressure sensors based on two different hollow Proc. IEEE Sensors, pp. 858–860, 2009, Article Number: 5398249.
core microstructured optical fibers. Despite their respective mi- [16] Y. Zhao, R. Lv, Y. Ying, and Q. Wang, “Hollow-core photonic crystal
crostructure profile, the pressure sensitivities are dominantly de- fiber Fabry-Perot sensor for magnetic field measurement based on mag-
netic fluid,” Opt. Laser Technol., vol. 44, no. 4, pp. 899–902, Jun. 2012.
termined by the cavity-length change, which is relevant with an [17] J. S. Sirkis, D. D. Brennan, M. A. Putman, T. A. Berkoff, A. D. Kersey,
amplification factor . In contrast, the mode-index change in- and E. J. Friebele, “In-line fiber etalon for strain measurement,” Opt.
duced by the applied pressure is extremely small. However, we Lett., vol. 18, no. 22, pp. 1973–1975, Nov. 1993.
[18] D. H. Wang, S. J. Wang, and P. G. Jia, “In-line silica capillary tube
intended to give a detailed investigation on the physical mech- all-silica fiber-optic Fabry-Perot interferometric sensor for detecting
anisms behind the mode-index changes for the HC bandgap high intensity focused ultrasound signals,” Opt. Lett., vol. 37, no. 11,
fibers. Due to the highly deformable holey structure, the air-core pp. 2046–2048, Jun. 2012.
[19] Y. Wang, D. N. Wang, C. R. Liao, T. Hu, J. Guo, and H. Wei, “Temper-
bandgap fibers may offer a new dimension for the implementa- ature-insensitive refractive index sensing by use of micro Fabry-Pérot
tion of tunable photonic devices. cavity based on simplified hollow-core photonic crystal fiber,” Opt.
Lett., vol. 38, no. 3, pp. 269–271, Feb. 2013.
[20] D. J. J. Hu, Y. Wang, J. L. Lim, T. Zhang, K. B. Milenko, Z. Chen, M.
Jiang, G. Wang, F. Luan, P. P. Shum, Q. Sun, H. Wei, W. Tong, and
ACKNOWLEDGMENT T. R. Wolinski, “Novel miniaturized Fabry-Perot refractometer based
on a simplified hollow-core fiber with a hollow silica sphere tip,” IEEE
Sensors J., vol. 12, no. 5, pp. 1239–1245, May 2012.
The authors would like to thank J. Ma, F. Yang, Z. Quan and [21] W. D. Pilkey, Peterson’s Stress Concentraton Factors, 2nd ed. New
M. Lin for their kind help and fruitful discussion. York, NY, USA: Wiley, 1999.
2532 JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL. 31, NO. 15, AUGUST 1, 2013

[22] V. Dangui, H. Kim, M. Digonnet, and G. Kino, “Phase sensitivity to Bai-Ou Guan received the B.Sc. degree in applied physics from Sichuan Uni-
temperature of the fundamental mode in air-guiding photonic-bandgap versity, Chengdu, China, in 1994, and the M.Sc. and Ph.D. degrees in optics
fibers,” Opt. Exp., vol. 13, no. 18, pp. 6669–6684, Sep. 2005. from Nankai University, Tianjin, China, in 1997 and 2000, respectively.
[23] M. Pang and W. Jin, “Detection of acoustic pressure with hollow-core From 2000 to 2005, he was with the Department of Electrical Engineering, the
photonic bandgap fiber,” Opt. Exp., vol. 17, no. 13, pp. 11088–11097, Hong Kong Polytechnic University, Hong Kong, first as a Research Associate,
Jul. 2009. then as a Postdoctoral Research Fellow. From 2005 to 2009, he was with School
[24] M. Pang, H. Xuan, J. Ju, and W. Jin, “Influence of strain and pressure to of Physics and Optoelectronic Engineering, Dalian University of Technology,
the effective refractive index of the fundamental mode of hollow-core Dalian, China, as a full professor, where he established the PolyU-DUT Joint
photonic bandgap fibers,” Opt. Exp., vol. 18, no. 13, pp. 14041–14055, Research Center for Photonics. In 2009 he joined Jinan University, Guangzhou,
Jul. 2010. China, where he founded the Institute of Photonics Technology. His current
[25] C. Wu, B. O. Guan, Z. Wang, and X. Feng, “Characterization of pres- research interests include fiber-optic biochemical sensors, micro/nanofiber op-
sure response of Bragg gratings in grapefruit microstructured fibers,” tical sensors, polarimetric fiber grating laser sensors, microstructured optical
J. Lightw. Technol., vol. 28, no. 9, pp. 1392–1397, May 2010. fiber sensors, photonic microwave generators, and photonic components for
[26] C. Wu, H. Y. Fu, K. K. Qureshi, B. O. Guan, and H. Y. Tam, “High- sensing and telecommunication. He has authored and coauthored more than
pressure and high-temperature characteristics of a Fabry-Perot interfer- 160 technical papers and presented more than 10 invited talks at international
ometer based on photonic crystal fiber,” Opt. Lett., vol. 36, no. 3, pp. conferences.
412–414, Feb. 2011. He is a member of IEEE and OSA, served as the General Co-Chair of the
[27] S. H. Aref, M. I. Zibaii, M. Kheiri, H. Porbeyram, H. Latifi, F. M. 10th International Conference on Optical Communications and Networks
Araújo, L. A. Ferreira, J. L. Santos, J. Kobelke, K. Schuster, and O. (ICOCN2011), the General Co-Chair of the 2nd Asia-Pacific Optical Sensors
Frazão, “Pressure and temperature characterization of two interfero- Conference (APOS2010), and the Technical Program Committee Co-Chair of
metric configurations based on suspended-core fibers,” Opt. Commun., the 5th Asia-Pacific Microwave Photonics Conference 2010 (APMP2010).
vol. 285, no. 3, pp. 269–273, Feb. 2012.

Huifeng Wei received the M.S. degree from the College of Physics Science,
Nankai University, Tianjin, China, in 2006. Currently, he is with the State Key
Long Jin received the B.S. degree in applied physics and the Ph.D. degree in Laboratory of Optical Fiber and Cable Manufacture Technology, Yangtze Op-
fiber optics from Nankai University, China, in 2003 and 2008, respectively. He tical Fiber and Cable Company Ltd. (YOFC), Wuhan, China, and he is the
joined the Department of Electrical Engineering, Hong Kong Polytechnic Uni- leader of photonic crystal fiber project with YOFC, and the executive leader
versity in 2008, as a research assistant and then a Postdoctoral Research Fellow. of the project of the national Key Basic Research and Development Program
Since 2010, he has been with Institute of Photonics Technology, Guangzhou, of China under Grant 2010CB735904. His research interests include fabrica-
China as an associate professor. He has published more than 50 journal and tion and characterization of photonic crystal fiber, photonic crystal fiber-based
conference papers. His research interests include fiber grating devices, photonic super continuum source, and research and development of Rare Earth-doped
crystal fibers and optical fiber sensors. fiber and fiber laser.