sensors
Article
Refractive Index Fiber Laser Sensor by Using a Fiber Ball Lens
Interferometer with Adjustable Free Spectral Range
Ricardo Iván Álvarez-Tamayo 1, *
Citation: Álvarez-Tamayo, R.I.;
Prieto-Cortés, P. Refractive Index
Fiber Laser Sensor by Using a Fiber
Ball Lens Interferometer with
Adjustable Free Spectral Range.
Sensors 2023, 23, 3045. https://
doi.org/10.3390/s23063045
Academic Editors: Luís C. Coelho
and José Manuel Almeida
Received: 18 February 2023
Revised: 1 March 2023
Accepted: 6 March 2023
Published: 11 March 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Sensors 2023, 23, 3045. https://doi.org/10.3390/s23063045
and Patricia Prieto-Cortés 2
1 Faculty of Mechatronics, Bionics and Aerospace, Universidad Popular Autónoma del Estado de Puebla,
Puebla 72410, Mexico
2 Mechatronics Division, Universidad Tecnológica de Puebla, Puebla 72300, Mexico
* Correspondence: ricardoivan.alvarez01@upaep.mx
Abstract: In this work, a fiber laser refractometer based on a fiber ball lens (FBL) interferometer is
proposed. The linear cavity erbium-doped fiber laser uses an FBL structure acting as a spectral filter
and sensing element for determining the RI of a liquid medium surrounding the fiber. The optical
interrogation of the sensor is the wavelength displacement of the generated laser line as a function of
the RI variations. For the proposed FBL interferometric filter, the free spectral range of its wavelength-
modulated reflection spectrum is adjusted to maximum in order to obtain RI measurements in a range
of 1.3939 to 1.4237 RIU, from laser wavelength displacements in a range from 1532.72 to 1565.76 nm.
The obtained results show that the wavelength of the generated laser line is a linear function of the RI
variations on the medium surrounding the FBL with a sensitivity of 1130.28 nm/RIU. The reliability
of the proposed fiber laser RI sensor is analytically and experimentally investigated.
Keywords: optical fiber; laser sensors; fiber ball lens; interferometric spectral filters; refractometers;
EDF fiber lasers
1. Introduction
Fiber refractometers have been of persistent interest for sensing applications in chemi-
cal, biological, medical instrumentation systems, and marine environment monitoring [1–5]
because of their intrinsic advantages such as compactness, high sensitivity, and low-cost
implementation. Among the various fiber optic refractometers, those based on the use
of interferometric filters have proven their reliability due to their ease of implementation,
variety of interference methods, and flexibility in the setting of operating parameters [6–8].
For this purpose, the use of micro-machined special fibers has allowed the implementation
of robust, compact, stable, and flexible interferometers suitable to be integrated to more
complex optical systems. Moreover, the use of fiber ball lenses (FBL) has been proven
for the development of functional optical resonators satisfying the principle of minimum
energy [9], and as high efficiency light couplers between fibers [10,11]. Recently, Jasim
et al. [12] proposed the implementation of an interferometer based on the use of an FBL
and a reflecting mirror. The optical setup acts as a two-path multi-beam interferometer
due to a couple of reflective surfaces generating a wavelength-modulated reflection output
spectrum. Based on the FBL interferometer (FBLI), broadband fiber sensors have been
reported for the measurement of different variables such as displacement [12], glucose in
ionized water [13], humidity [14,15], and formaldehyde [16,17], with improved and repeat-
able results. For these reported approaches, the interrogation method is the determination
of the free spectral range (FSR) variation as a function of the measured variable. Then, the
adjustable FSR can be an attractive characteristic of the FBLI for its use as a spectral filter
in different applications including tunable and multi-wavelength fiber lasers [18,19] and,
potentially, for optical sensors with different interrogation methods.
Moreover, laser sensors have proven the improvement of wavelength interrogation
optical sensors compared with broadband-spectrum sensors due to their well-defined
https://www.mdpi.com/journal/sensors
Sensors 2023, 23, 3045 2 of 12

narrow band spectrum which enables the high precision determination of the generated
laser line [20–22]. Here, the straightforward determination of the wavelength displacement
of a narrow-width laser spectrum is precisely determined, unlike traditional broadband
spectrum sensors where the wavelength at which the maximum intensity is obtained is not
easy, leading to measurement errors. In addition, the quality of the sensor is significantly
improved as it can be estimated by the Q value [22]. In this regard, the use of interferometers
and other wavelength tunable optical filters along with a fiber laser cavity represent a
suitable configuration to develop high precision fiber sensors [23–26]. However, the use
of the FBLI has been unexplored for the development of fiber laser refractometers. In this
case, the use of the FBLI structure represents a feasible alternative due to its advantages of
ease of construction and flexibility in adjusting the operation parameters [12,13].
In this paper, for the first time, to the best of our knowledge, we propose a fiber
laser refractometer by using an FBLI interferometer as a spectral filter. By the wavelength
shift of the laser line, the linear cavity erbium-doped fiber (EDF) laser uses the FBL as a
sensing element for the determination of the RI of a liquid medium surrounding the fiber
structure. For the used FBLI, the FSR of its reflection spectrum is adjusted to obtain the RI
estimation in a broad laser emission wavelength range. The proposed sensor using an FBL
interferometer within a laser cavity significantly improves the refractive index detection
compared with the reported sensors using the FBL interferometer for detection directly
from the interference spectrum. The determination of the wavelength displacement as the
optical interrogation is more precise and straightforward to determine from a laser line
than from a broadband spectrum.

2. Fiber Ball Lens Interferometer

The FBL consist of a quasi-spherical-shaped tip attached to the fiber body. Typically,
the fiber ball is fabricated by a controlled electric arc discharge on the fiber tip, from the
manual operation of a fusion splicer. The discharge parameters such as time and power are
optimized to control the ball diameter and shape, from both single-mode and multimode
fibers [27–30]. Nowadays, fusion splicers with features for special fibers fabrication, such
as fiber ball lenses and tapered fibers, are currently available. In our case, instead of
manual fabrication, the FBL used in our approach was fabricated from a multimode fiber
with a core diameter of 50 µm (Thorlabs FG050LGA, Newton, NJ, USA), by using the
dedicated ball lens arc fusion program (AFL Fiber Processing Software FPS, ver. 1.2b,
Tokyo, Japan) of a fusion splicer with special fiber processing features (Fujikura ArcMaster
FSM-100M, Tokyo, Japan), where the fabrication process is controlled and monitored with
high precision ensuring the consistency and repeatability of the FBL fabrication. Based
on the performance results obtained by Li et al. [27] for the fabrication of MMF-based
FBL structures, the ball diameter was adjusted to 300 µm with minimal ellipticity. The
results obtained in ref. [27] shows that large-diameter FBLs are recommended as follow-up
Sensors 2023, 23, x FOR PEER REVIEW 3 of 13
experimental optical devices. The microscope image of the fabricated FBL is shown in
Figure 1.

Figure1.1.Microscope
Figure Microscopeimage
imageof
ofthe
thefabricated
fabricatedmultimode
multimodeFBL
FBLwith
withaaball
balldiameter
diameterof
of300
300µm.
µm.

The FBL interferometer (FBLI) is based on the multiple-beam interference

configuration proposed by A. A. Jasim et al. [12], shown in Figure 2. A reflecting mirror is
placed in front of the FBL cavity at an adjustable distance 𝑑. The input light 𝐼 from the
fiber core is propagated in multiple modes within the FBL cavity. Then, two different
Sensors 2023, 23, 3045 Figure 1. Microscope image of the fabricated multimode FBL with a ball diameter of 300 µm. 3 of 12

The FBL interferometer (FBLI) is based on the multiple-beam interference

configuration
The FBLproposed by A. A.
interferometer Jasim
(FBLI) is et al. [12],
based shown
on the in Figure 2. interference
multiple-beam A reflecting mirror is
configura-
placed in front of
tion proposed bythe FBL
A.A. cavity
Jasim at an
et al. adjustable
[12], Figure 2.𝑑.AThe
shown indistance input light
reflecting 𝐼 isfrom
mirror placedthein
fiber
frontcore is propagated
of the FBL cavity atinanmultiple
adjustable modes within
distance theinput
d. The FBL cavity.
light IinThen, twofiber
from the different
core is
interfaces
propagated act in
as multiple
reflectingmodes
surfaces: the the
within interface formed
FBL cavity. by the
Then, two FBL surfaceinterfaces
different with the act
freeas
reflecting
space, and surfaces: the interface
the one formed between formed by the
the free FBLand
space surface with theAfree
the mirror. space,
portion ofand
the the one
input
light with intensity 𝐼 meet the conditions to be reflected by the FBL inner surface back to
formed between the free space and the mirror. A portion of the input light with intensity
I1 fiber.
the meet Onthe the
conditions to beanother
other hand, reflected partbyofthe
theFBL
lightinner
withsurface
intensity 𝐼 to
back the fiber.
surpasses theOnFBLthe
other hand,
surface, anotherthrough
propagates part of the
thelight
spacewithwith lengthI2𝑑,
intensity surpasses the FBL
is reflected surface,
by the propagates
mirror, and is
throughback
coupled the space
to the with
fiber length d, isFBL
using the reflected by the
as a light mirror, and is coupled back to the fiber
coupler.
using the FBL as a light coupler.

Figure 2. Schematic of the multiple-beam FBLI.

Figure 2. Schematic of the multiple-beam FBLI.
The reflected beams from I1 and I2 interfere back to the fiber with a phase difference
dueThe reflected
to the beams
optical path from 𝐼 between
difference and 𝐼 two interfere
beams,backgivento bythe
[12]fiber with a phase
difference due to the optical path difference between two beams, given by [12]
4π
∆φ = 4𝜋 n g d + φ0 , (1)
∆𝜙 = λ 𝑛 𝑑 𝜙 , (1)
𝜆
where λ is the operating wavelength and n g is the refractive index (RI) of the medium
where 𝜆 is thethe
surrounding operating wavelength
gap between the FBLandand𝑛theismirror.
the refractive index (RI) of the medium
surrounding the gap between the FBL and the mirror.
The light propagation through the FBLI was simulated by using the Optiwave Op-
The light
tiFDTD propagation
software. throughofthe
The parameters theFBLI was simulated
fabricated FBL were by using
used. the Optiwave
A fiber ball with a
OptiFDTD software. The parameters of the fabricated FBL were
diameter of 300 µm and an MMF with a core diameter of 50 µm and cladding diameterused. A fiber ball with a
diameter
of 125 µm were simulated. The input source was normalized with maximal intensityofat
of 300 µm and an MMF with a core diameter of 50 µm and cladding diameter
125 µmnm.
1550 wereThesimulated. The input
main objective source
of the beamwas normalized
trace simulation with
is tomaximal
confirmintensity at 1550
the experimental
nm. The main objective of the beam trace simulation is to confirm the experimental
results obtained by Jasim et al. [11] for the microscope image describing the FBL-reflected results
obtained
light frombyits
Jasim
inneretsurface.
al. [11] for thethe
Then, microscope
objective image describingthe
is to investigate the FBL-reflected
second reflectedlight
inten-
from
sity its
frominner
the surface.
reflector Then,
mirrorthe objective
in order is to investigate
to obtain the reflectiontheinterference
second reflected
back to intensity
the fiber.
from the3areflector
Figure shows the mirror in order
xz plane to obtain
distribution ofthe
the reflection interference
light propagation backthe
through tooptical
the fiber.
axis.
Figure 3a shows the xz plane distribution of the light propagation through
The input light propagates as different modes through the MMF core; then, when the fiber the optical axis.
ball is reached, it acts as coupling element. As a result, most of the light intensity from
the multiple modes remain their propagation at the central area of the XY plane, without
significant dispersion within the ball. Part of the light is reflected at the inner surface of the
FBL whereas a portion of light passes the FBL’s surface being propagated through the free
space, reaching the mirror where it is reflected and coupled back to the FBL. As a result, the
interference between the two parts of light reflected with different optical paths is observed.
An amplitude maximum is observed near to the central point of the fiber ball according
to the FBL cavity length, also shown in the 3D image of the light propagation through the
FBLI in Figure 3b. The multiple beam interference is observed all over the FBL surface
at the XZ plane, with a high amplitude at the x axis center region. As it can be observed,
for the FBL dimensions and the reflector mirror distance, the focusing point where the
maximal intensity of the interfered light is near to the center of the FBL is at ~400 nm of the
light propagation through the optical axis. The quality of the interfered beam inside the
FBL is investigated in Figure 3c where the two-dimensional light intensity distribution (XY
 through the FBLI in Figure 3b. The multiple beam interference is observed all over the
FBL surface at the XZ plane, with a high amplitude at the x axis center region. As it can be
observed, for the FBL dimensions and the reflector mirror distance, the focusing point
where the maximal intensity of the interfered light is near to the center of the FBL is at
~400 nm of the light propagation through the optical axis. The quality of the interfered
Sensors 2023, 23, 3045 beam inside the FBL is investigated in Figure 3c where the two-dimensional light intensity 4 of 12
distribution (XY plane) cut at 390 µm is shown. As it can be observed, the light intensity
distribution exhibits wavelength modulation due to the interference between the two
reflecting surfaces
plane) cut atwith
390 an
µmHWHM
is shown.of As
132.8 µm.beThe
it can results the
observed, confirm that the FBL
light intensity is actingexhibits
distribution
as a coupling element as light is propagated back and forth, where reflected
wavelength modulation due to the interference between the two reflecting surfaces beams with an
interfereHWHM
back to of
the132.8
fiber.µm. The results confirm that the FBL is acting as a coupling element as
light is propagated back and forth, where reflected beams interfere back to the fiber.

Figure 3.Figure Beamsimulation

Beam3.trace trace simulation of intensity
of intensity distribution
distribution for for
thethe FBLI:
FBLI: (a)(a)XZ
XZplane
plane intensity
intensity distribu-
tion; (b) 3D intensity distribution; (c) XY plane cut at 390 µm of z length.
distribution; (b) 3D intensity distribution; (c) XY plane cut at 390 µm of z length.

The FSR of the interference periodical wavelength modulation can be expressed as [18]:
λ2
∆λ = (2)
2d
The numerical simulation for a two-beam optical interference model, with the phase
difference described in Equation (1), is shown in Figure 4. As it can be observed from
Figure 4a and described in Equation (2), the FSR of the FBLI periodical-reflected output
spectrum can be varied by the fine adjustment of the distance d. The FSR of the interference
spectrum decreases with the increase of the gap. For the simulated results, a high FSR
variation of 31.6 nm for d changes in a range from 20 to 80 µm is noticed. Moreover,
Figure 4b shows the behavior of the FBLI output spectrum for variations on the RI of the
gap between the FBL and the reflector mirror. As a result, the wavelength displacement of
the interference spectrum towards longer wavelengths as a linear function of the RI of the
gap increase is observed. The wavelength of the interference spectrum shifts from ~22 nm
for RI variations in a range from 1.4 to 1.42 RIU.
 high FSR variation of 31.6 nm for 𝑑 changes in a range from 20 to 80 µm is noticed.
Moreover, Figure 4b shows the behavior of the FBLI output spectrum for variations on
the RI of the gap between the FBL and the reflector mirror. As a result, the wavelength
displacement of the interference spectrum towards longer wavelengths as a linear
function of the RI of the gap increase is observed. The wavelength of the interference
Sensors 2023, 23, 3045 5 of 12
spectrum shifts from ~22 nm for RI variations in a range from 1.4 to 1.42 RIU.

Figure 4. Numerical
Figure 4. Numerical simulation simulation
of the FBL of the FBL interference
interference reflected spectrum (a,b).
reflected spectrum.
3. Experimental Setup
3. Experimental Setup
The experimental setup of the proposed fiber laser refractometer based on the use of a
The experimental
fiber ball lens setup
is shownof the proposed
in Figure fiber
5a. As thelaser
gainrefractometer basedcavity
medium, the linear on thelaser
use of
uses a 2.2 m
a fiber balllong
lenssingle-mode
is shown inerbium-doped
Figure 5a. As fiber
the gain
(EDF) (CorActive, EDF-L 1500) with a core aabsorption
medium, the linear cavity laser uses
2.2 m longofsingle-mode
21 dB/m at 1530 erbium-doped
nm, cladding fiber (EDF) (CorActive,
absorption of 12 dB/mEDF-L 1500)
at 980 nm, with
and a core aperture
numerical
absorptionof of0.25. The EDF is pumped through a 980/1550 nm wavelength divisionand
21 dB/m at 1530 nm, cladding absorption of 12 dB/m at 980 nm, multiplexer
numerical(WMD)aperture byofa 980
0.25.nmThelaser
EDFdiode
is pumped
(LD) with through
powera of980/1550
200 mW. nmAtwavelength
one end, the cavity
division multiplexer
is limited by (WMD)
a 90/10by aoptical
980 nmcoupler
laser diode
with(LD) with poweroutput
interconnected of 200 mW.
portsAt one
forming a 64%
end, the cavity is limited by a 90/10 optical coupler with interconnected output
transmission fiber loop mirror (FLM). At the other end, the cavity is limited by an FBLI, ports
forming aformed
64% transmission
by the FBLfiber
alongloop
withmirror
a flat (FLM). At the other
NIR broadband end, the
dielectric cavity
mirror is limitedBB03-E04).
(Thorlabs,
by an FBLI, formed by the FBL along with a flat NIR broadband dielectric
The FBLI acts as a spectral filter for the laser generation and as a sensing element mirror in contact
(Thorlabs,withBB03-E04).
the liquidThe FBLIthe
filling acts as aAsspectral
gap. it can befilter for thethe
expected, laser generation and as a spectrum
wavelength-modulated
sensing element
of the FBLI in contact
selects with the emission
the laser liquid filling
at thethe gap. As itfor
wavelength can
thebemaximum
expected,amplitude.
the In
addition, the generated laser line is wavelength tuned as a function of the refractive index
variations on the medium surrounding the FBLI fulfilling the gap. Figure 5b shows a
schematic detail of the FBLI. By using a V-Groove fiber holder, the FBL was attached and
fixed to an XYZ micrometric translation stage, perpendicularly aligned in front of the
reflector mirror surface. The mirror was placed on the bottom face of a container into which
the sensed liquid is poured. Then, by micrometrical vertical displacement on the FBL,
the distance between the FBL and the reflector mirror is varied to adjust the free spectral
range (FSR) of the interference spectrum. Once the distance is adjusted, the container is
fulfilled with the liquid, covering the FBL, to perform the refractive index measurement of
the FBL-surrounding medium. The unconnected port of the 90/10 optical coupler is the
laser output, where an optical spectrum analyzer (OSA) with a maximum resolution of
0.03 nm is used to measure the laser wavelength displacement as the optical interrogation
for the liquid refractive index variations.
 to adjust the free spectral range (FSR) of the interference spectrum. Once the distance is
adjusted, the container is fulfilled with the liquid, covering the FBL, to perform the
refractive index measurement of the FBL-surrounding medium. The unconnected port of
the 90/10 optical coupler is the laser output, where an optical spectrum analyzer (OSA)
Sensors 2023, 23, 3045 with a maximum resolution of 0.03 nm is used to measure the laser wavelength 6 of 12
displacement as the optical interrogation for the liquid refractive index variations.

Figure 5.
Figure 5. Experimental
Experimental setup
setup of
of the
the FBL
FBL interference
interference fiber
fiber laser
laser refractometer:
refractometer: (a)
(a) Configuration
Configuration of
of
the linear cavity laser configuration; and (b) detailed schematic of the FBLI implementation.
the linear cavity laser configuration; and (b) detailed schematic of the FBLI implementation.

4. Results and Discussion

With the
With thepurpose
purpose of of optimizing
optimizing the the
FBLIFBLI
for itsfor
useitsasuse as a spectral
a spectral filter for filter for laser
laser emission,
emission,
the theoutput
reflected reflected output
signal signal
of the of the modulated
modulated spectrum spectrum was characterized.
was characterized. Thus, theThus,
FLM
the FLM
was was disconnected
disconnected from the experimental
from the experimental setup (between setup the(between the and
coupler port coupler port and
the WDM) to
the WDM)
open to open
the cavity. At thethe cavity. At port
unconnected the of
unconnected
the WDM, the portreflected
of the spectrum
WDM, the of reflected
the FBLI
due to the EDF
spectrum amplified
of the FBLI due spontaneous
to the EDF emission (ASE)spontaneous
amplified was measured by using(ASE)
emission an OSA. was
For the
measured bydesign
using anof OSA.
laser sensors based on the use of interferometric filters for inter-
rogation
For by
thedisplacement
design of laserof the laser wavelength,
sensors based on the it isuse
suitable to set a long FSR
of interferometric of the
filters for
modulated
interrogation spectrum that allows
by displacement for laser
of the obtaining a greater
wavelength, it operating
is suitable range
to set aoflong
the sensor,
FSR of
limited by the dual
the modulated laser emission.
spectrum that allowsInfor
this regard, by
obtaining meansoperating
a greater of micrometric
range displacement
of the sensor,
in the translation stage, the distance d was reduced to a minimum in order to obtain the
maximum FSR of the FBL filter. Figure 6 shows the experimental results of the FBLI re-
flected coefficient for liquid refractive index variations by using the EDF ASE as the input
signal. The FSR of the wavelength-modulated spectrum is ~32 nm. The laser emission is
expected to be generated at the reflection minimum within the ASE wavelength range. The
reflection signal was normalized as the reflection coefficient from the measured EDF ASE
spectrum shown in the inset of Figure 6. The insertion losses of the interferometric filter are
of ~55%. A set of characterized glycerol solutions was used to vary the RI of the medium
surrounding the FBL. For the spectrum characterization, the RI was varied from 1.3939 to
1.4095 RIU. As it can be observed, the increase on the RI of the gap wavelength displaces
the reflected periodical spectrum toward longer wavelengths.
 emission is expected to be generated at the reflection minimum within the ASE
wavelength range. The reflection signal was normalized as the reflection coefficient from
the measured EDF ASE spectrum shown in the inset of Figure 6. The insertion losses of
the interferometric filter are of ~55%. A set of characterized glycerol solutions was used to
vary the RI of the medium surrounding the FBL. For the spectrum characterization, the RI
Sensors 2023, 23, 3045 7 of 12
was varied from 1.3939 to 1.4095 RIU. As it can be observed, the increase on the RI of the
gap wavelength displaces the reflected periodical spectrum toward longer wavelengths.

Figure 6. Reflection
Reflection coefficient of the FBLI wavelength-modulated spectra
spectra for variations
variations of RI on the
reflector mirror
gap between the FBL and the reflector mirror with
with EDF
EDF ASE
ASE as
as the
the input
input signal.
signal. Inset:
Inset: Measured
spectrum of
spectrum of the
the EDF
EDF ASE.
ASE.

The operation
operation of of the
the fiber
fiberlaser
laserrefractometer
refractometerisisshownshownininFigure
Figure7. 7.
AA setset
of eight RI
of eight
variations
RI variationswere carried
were out out
carried in ainrange fromfrom
a range 1.3939 to 1.4237
1.3939 RIU.RIU.
to 1.4237 Figure 7a shows
Figure the laser
7a shows the
spectra
laser measured
spectra at theatlaser
measured output
the laser port with
output port anwithOSA. As it As
an OSA. canitbe observed,
can the laser
be observed, the
laser line is wavelength displaced towards longer wavelengths
line is wavelength displaced towards longer wavelengths as the refractive index of theas the refractive index of
the FBL-surrounding
FBL-surrounding mediummedium is increased.
is increased. For theFor theRI
lower lower
valueRIofvalue
1.3939,ofthe
1.3939, the laser
laser emission
emission
exhibits dualexhibits dual wavelength
wavelength laser generation
laser generation with a higher with a higher
power at thepower
shorterat wavelength
the shorter
wavelength
laser line atlaser line nm.
~1530.3 at ~1530.3 nm. Conversely,
Conversely, for the higher for the
RI higher
value RI of value
1.4237,of the
1.4237, the
shorter
shorter
wavelength wavelength laser
laser line of line of thewavelength
the dual dual wavelength laser generation
laser generation startsstarts to appear.
to appear. Then, Then,
the
the
lowest and highest RI values, for which the dual emission with equal powers would be
lowest and highest RI values, for which the dual emission with equal powers would
obtained,
obtained, set setthe
theoperating
operating limits of theofrefractometer.
limits This operating
the refractometer. range is inrange
This operating agreementis in
with
agreement with the FSR of the FBLI. Figure 7b depicts the central wavelength oflaser
the FSR of the FBLI. Figure 7b depicts the central wavelength of the generated the
line as a function
generated of the
laser line as arefractive
function of index variations.index
the refractive As it variations.
can be observed, thebe
As it can wavelength
observed,
at
thewhich the laseratline
wavelength is obtained
which the laser actsline
as the
is high precision
obtained acts optical
as the interrogation
high precision foroptical
the RI
variations. The wavelength displacement of the laser line as a function
interrogation for the RI variations. The wavelength displacement of the laser line as a of the RI changes
on the medium
function of the RIsurrounding
changes onthe theFBL can besurrounding
medium linearly fittedthewith
FBLacanslope
be of 1130.28
linearly nm/RIU.
fitted with
The residual wavelength error is also shown for each RI variation
a slope of 1130.28 nm/RIU. The residual wavelength error is also shown for each compared with the linear
RI
fit value. The
variation worst-case
compared witherror observed
the linear over the
fit value. TheRI testing range
worst-case errorisobserved
of 0.93 nm,overobtained
the RI
for the RI
testing valueisofof1.3976.
range 0.93 nm,Theobtained
calculated formean
the absolute
RI value error is 0.292
of 1.3976. Thenm.calculated mean
In order
absolute errortois characterize
0.292 nm. the repeatability of the measured laser central wavelength
as a function of the RI variations, a set of 15 sequential measurements from the lower to
the higher RI were carried out. Then, the calculated mean absolute error (MAE) from the
average wavelength for each RI measurement was obtained. From the obtained results,
the repeatability of the laser line central wavelength for the RI variations is shown in
Figure 8. The MAE values are shown at the corresponding error bar for each tested RI.
As it can be observed, the worst-case wavelength error value observed was ±0.0371 nm
corresponding to a liquid RI of 1.3976 nm. The calculated MAEs for the 15 RI index scans
are around the OSA resolution; as a result, it can be inferred that, due to the wavelength
interrogation method, the resolution of the proposed fiber laser sensor is limited to the OSA
wavelength resolution.
Sensors 2023, 23, x FOR PEER REVIEW 8 of 13
Sensors 2023, 23, 3045 8 of 12

Figure 7. Operation of the FBL-based laser refractometer: (a) the output signal of the laser sensor for
RI variations on the medium surrounding the FBL; and (b) laser central wavelength displacement
as a function of the RI changes.

In order to characterize the repeatability of the measured laser central wavelength as

a function of the RI variations, a set of 15 sequential measurements from the lower to the
higher RI were carried out. Then, the calculated mean absolute error (MAE) from the
average wavelength for each RI measurement was obtained. From the obtained results,
the repeatability of the laser line central wavelength for the RI variations is shown in
Figure 8. The MAE values are shown at the corresponding error bar for each tested RI. As
it can be observed, the worst-case wavelength error value observed was ±0.0371 nm
corresponding to a liquid RI of 1.3976 nm. The calculated MAEs for the 15 RI index scans
Figure 7.are around
Figure
Operation ofthe
the OSA
7. Operation resolution;
of the
FBL-based FBL-based aslaser
a result, it can
refractometer:
laser refractometer: (a) be(a)inferred
the output thesignalthat,
output thedue
of signal oftothe
laser the wavelength
laser
sensor forsensor for
interrogation
RI variations
RI variations on the mediummethod,
on the medium the resolution
surrounding
surrounding of
the FBL;the the
FBL;
and proposed
(b)and fiber
(b)central
laser laser
laser central sensor
wavelength
wavelength is limited to the
displacement
displacement as
aOSA
as a function ofwavelength
function
the of
RIthe resolution.
RI changes.
changes.

In order to characterize the repeatability of the measured laser central wavelength as

a function of the RI variations, a set of 15 sequential measurements from the lower to the
higher RI were carried out. Then, the calculated mean absolute error (MAE) from the
average wavelength for each RI measurement was obtained. From the obtained results,
the repeatability of the laser line central wavelength for the RI variations is shown in
Figure 8. The MAE values are shown at the corresponding error bar for each tested RI. As
it can be observed, the worst-case wavelength error value observed was ±0.0371 nm
corresponding to a liquid RI of 1.3976 nm. The calculated MAEs for the 15 RI index scans
are around the OSA resolution; as a result, it can be inferred that, due to the wavelength
interrogation method, the resolution of the proposed fiber laser sensor is limited to the
OSA wavelength resolution.

Figure 8.
Figure 8. Repeatability
Repeatability of
of wavelength
wavelength displacement
displacement for
for the
the RI
RI fiber
fiber laser
laser sensor.
sensor.

Figure
Figure 99 shows
shows the
the experimental
experimental results on thethe laser
laser line
line stability
stability of
of the
the sensor.
sensor. For
For
two different RI levels, the measurements
two different RI levels, the measurements grouped in a
a set of 15 laser spectra captured
set of 15 laser spectra captured
every
every 55 min
min were
were obtained
obtained to
to discuss
discuss the
the long-term
long-term stability
stability ofof the
the sensor.
sensor. Figure
Figure 9a
9a shows
shows
the
the obtained
obtained captures
captures of
of the
the generated
generated laser
laser line
line spectra
spectra for
for aa liquid
liquid medium
medium surrounding
surrounding
the fiber structure with an RI of 1.4009, whereas Figure 9b shows the spectra for laser
lines generated with an RI of 1.4095. The obtained results are shown in linear scale. Good
stability of the output power and the wavelength for each set of measurements is observed.
An in-depth analysis for the output power and wavelength stability of the generated laser
line for each RI level of 1.4009 and 1.4095 is shown in Figures 9c and 9d, respectively.
The output peak power and the central wavelength of the captured laser line spectra are
depicted for each RI level. For the RI of 1.4009, the average power of 118.1 mW exhibits
Figure 8. Repeatability of wavelength displacement for the RI fiber laser sensor.
variations within the 4.7%, whereas for the RI of 1.4095, the peak power variations are
within 6.7% from an average power of 120.8 mW. However, due to the interrogation method
Figure 9 shows the experimental results on the laser line stability of the sensor. For
of the sensor, the most important indicator lies in the stability of the wavelength of the
two different RI levels, the measurements grouped in a set of 15 laser spectra captured
generated laser line. Thus, for the RI of 1.4009, the wavelength variation is in a range of
every 5 min were obtained to discuss the long-term stability of the sensor. Figure 9a shows
0.46 nm with an average wavelength of 1540.63 nm, which is 0.024 nm away from the
the obtained captures of the generated laser line spectra for a liquid medium surrounding
theoretical value obtained by the linear fit. For the RI of 1.4095, the wavelength variations
 mW exhibits variations within the 4.7%, whereas for the RI of 1.4095, the peak power
variations are within 6.7% from an average power of 120.8 mW. However, due to the
interrogation method of the sensor, the most important indicator lies in the stability of the
wavelength of the generated laser line. Thus, for the RI of 1.4009, the wavelength variation
Sensors 2023, 23, 3045 is in a range of 0.46 nm with an average wavelength of 1540.63 nm, which is 0.024 nm 9 of 12
away from the theoretical value obtained by the linear fit. For the RI of 1.4095, the
wavelength variations are in a range of 0.41 nm in an average central wavelength of the
laser line are
of 1550.32 nm,ofwhich
in a range is 0.021
0.41 nm in annm away central
average from the theoreticalof
wavelength value.
the laser line of 1550.32 nm,
which is 0.021 nm away from the theoretical value.

Figure 9. Stability characterization of the RI fiber laser sensor: (a) set of output spectra for RI of
Figure 9. Stability characterization of the RI fiber laser sensor: (a) set of output spectra for RI of
1.4009; (b) set spectra
1.4009; (b) set of output of output
forspectra for RI (c)
RI of 1.4095; of 1.4095; (c) plot
plot of the peakofpower
the peak powerand
stability; stability;
(d) plotand
of (d) plot of
the stability for the wavelength of the
the stability for the wavelength of the laser line. laser line.

It is worth noting that the sensitivity of the FBL to temperature variations was recently
studied by Zhang et al. in ref. [21]. For the reported fiber sensor, the crosstalk temperature
of around 40 pm/◦ C compromised the liquid level measurement. In our case, we assume
that the low sensitivity to temperature of the FBL structure can be neglected since a high
sensitivity to the refractive index of 1130.28 nm/RIU was obtained for the proposed fiber
laser refractometer.
In the case of optical sensors by interrogation by wavelength displacement, the main
advantage of laser sensors over broad-spectrum ones lies fundamentally in the ability to
determine with greater precision the wavelength to which the spectrum is shifted, due
to the narrow bandwidth of the laser emission spectrum. This can be quantified through
the measurement of the Q value which describes the quality of the sensor in terms of the
sensing parameters and the spectral bandwidth of the optical spectrum, as follows [22]:
KVS2
Q= (3)
FWHM
Sensors 2023, 23, 3045 10 of 12

Q is a dimensionless factor where S is the sensitivity of the sensor and V is the spectrum
visibility. K is a normalizing unit coefficient. As it can be noticed, a laser sensor significantly
improves the Q value since the FWHM and the visibility are highly decreased and increased,
respectively, compared with broadband spectrum optical sensors. For the proposed fiber
laser refractometer, the generated laser line exhibits a visibility of ~45.2 dB and a FWHM of
~0.018 nm. Then, the estimated quality factor Q is ~3.2 × 109 . Then, in order to compare
the reliability of the proposed fiber sensor, the obtained results are compared with reported
fiber-based refractometers operating in the 1.55 µm, in terms of the quality factor Q, as
it is shown in Table 1. As it can be observed, the Q value obtained from the fiber laser
refractometers is highly increased compared to the broadband spectrum sensors due to the
narrow FWHM and the high visibility of the laser line. Furthermore, the fiber structure
used determines the sensitivity of the sensor. In this regard, plasmonic sensors based on
material coated no-core fibers (NCF) exhibit high sensitivity but at the cost of a broadened
optical spectra compromising the wavelength shift measurement. In the case of ref. [23],
a high sensitivity of a fiber sensor is reported; however, the fiber structure is not easy to
construct, compromising the reproducibility of the sensor. Additionally, the narrow FSR
limits the measurement to a short refractive index range.

Table 1. Comparison in terms of Q-factor of some reported fiber-based refractometers and the
proposed approach.

Sensitivity FWHM Visibility

Ref. Q Value RI Range Structure Notes
(nm/RIU) (nm) (dB)
[31] 259.85 ~3.7 35 6.39 × 105 1.333–1.381 SMS
[32] 286.2 ~1.1 16 1.19 × 106 1.33–1.45 SMF-Etched SMS
[33] 261.9 ~0.3 18 4.12 × 106 1.3333–1.3737 Multi-tapered SMS
Mach-Zehnder SMSMS
[34] 188 ~2.8 27 3.41 × 105 1.33–1.40
interferometer
Cascaded NCF with long
[35] 580.269 ~0.9 17 6.63 × 106 1.4–1.45
period fiber grating
SMF Fabry-Perot
[36] 1130 ~3 15.5 6.60 × 106 1.3333–1.3434
interferometer tip
Lossy mode resonance
[37] 1214.7 ~12 17 2.09 × 106 1.3678–1.4009
AZO-coated NCF
Surface plasmon Visible
3.2 (48%
[38] 2848 ~70 3.71 × 105 1.3328–1.3853 resonance silver-coated wavelength
transmission)
NCF range
Core-offset
1.33302–
[23] −2953.44 0.0164 42 2.23 × 1010 Mach-Zehnder Laser sensor
1.33402
interferometer
Fiber ball lens
This work 1130.28 0.018 45.2 3.21 × 109 1.3939–1.4237 Laser sensor
interferometer

5. Conclusions
In this paper, a fiber laser refractive index sensor based on the use of an FBLI was
demonstrated. The fiber refractometer estimates the RI of the liquid medium surrounding
the FBL structure by wavelength shift interrogation of the generated laser line. With a
wavelength-modulated of the FBLI with an FSR of ~32 nm, estimations of the RI in a range
from 1.3939 to 1.4237 RIU were obtained by laser central wavelength shift in a range from
1532.72 to 1565.76 nm. The laser wavelength exhibits a linear dependence to RI variations
with a sensitivity of 1130.28 nm/RIU and a quality factor Q of ~3.21 × 109 . The reliability
of the proposed fiber laser refractometer with advantages of ease of implementation and
interrogation, flexibility, robustness, and accuracy were experimentally demonstrated.

Author Contributions: R.I.Á.-T. and P.P.-C. conceived and designed the experiments and performed
the experiments; R.I.Á.-T. analyzed the data; R.I.Á.-T. and P.P.-C. wrote the paper; R.I.Á.-T. contributed
with materials/analysis tools. All authors have read and agreed to the published version of the
manuscript.
Funding: This research was funded by Fondo de Investigación UPAEP 2022.
Sensors 2023, 23, 3045 11 of 12

Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data are unavailable due to privacy or ethical restrictions.
Acknowledgments: R.I. Álvarez-Tamayo wants to thank Fondo de publicación UPAEP for support-
ing the publication of the article. P. Prieto-Cortés was supported in part by CONACyT Estancias
Posdoctorales por México 2022–2023.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Min, R.; Liu, Z.; Pereira, L.; Yang, C.; Sui, Q.; Marques, C. Optical fiber sensing for marine environment and marine structural
health monitoring: A review. Opt. Laser Technol. 2021, 140, 107082. [CrossRef]
2. Guo, T.; Liu, F.; Guan, B.-O.; Albert, J. Tilted fiber grating mechanical and biochemical sensors. Opt. Laser Technol. 2016, 78, 19–33.
[CrossRef]
3. Tearney, G.J.; Brezinski, M.E.; Southern, J.F.; Bouma, B.E.; Hee, M.R.; Fujimoto, J.G. Determination of the refractive index of highly
scattering human tissue by optical coherence tomography. Opt. Lett. 1995, 20, 2258–2260. [CrossRef]
4. Urrutia, A.; del Villar, I.; Zubiate, P.; Zamarreño, C.R. A Comprehensive Review of Optical Fiber Refractometers: Toward a
Standard Comparative Criterion. Laser Photonics Rev. 2019, 13, 1900094. [CrossRef]
5. Liang, H.; Peric, B.; Hughes, M.; Podoleanu, A.G.; Spring, M.; Roehrs, S. Optical Coherence Tomography in archaeological and
conservation science—A new emerging field. In Proceedings of the 1st Canterbury Workshop on Optical Coherence Tomography
and Adaptive Optics, Canterbury, UK, 8–10 September 2008; p. 7139.
6. Yang, M.; Dai, J. Fiber optic hydrogen sensors: A review. Photonic Sens. 2014, 4, 300–324. [CrossRef]
7. Shaimerdenova, M.; Ayupova, T.; Sypabekova, M.; Tosi, D. Fiber optic refractive index sensors based on a ball resonator and
optical backscatter interrogation. Sensors 2020, 20, 6199. [CrossRef]
8. Álvarez-Tamayo, R.I.; Prieto-Cortés, P.; Durán-Sánchez, M.; Ibarra-Escamilla, B.; Barcelata-Pinzón, A.; Kuzin, E.A. Laser
Wavelength Estimation Method Based on a High-Birefringence Fiber Loop Mirror. Photonic Sens. 2018, 9, 89–96. [CrossRef]
9. Lee, C.-C.; Hsiao, S.-Y.; Fang, W. Implementation of a micro ball lens on a silicon optical bench using insoluble two-phase liquid
immersion technology. J. Micromechanics Microengineering 2010, 20, 085015. [CrossRef]
10. Dong, Y.; Wang, K.; Jin, X. Package of a dual-tapered-fiber coupled microsphere resonator with high Q factor. Opt. Commun. 2015,
350, 230–234. [CrossRef]
11. Cai, M.; Painter, O.; Vahala, K.J.; Sercel, P.C. Fiber-coupled microsphere laser. Opt. Lett. 2000, 25, 1430–1432. [CrossRef]
12. Jasim, A.A.; Zulkifli, A.Z.; Muhammad, M.Z.; Harun, S.W.; Ahmad, H. A new compact micro-ball lens structure at the cleaved tip
of microfiber coupler for displacement sensing. Sens. Actuators A Phys. 2013, 189, 177–181. [CrossRef]
13. Harun, S.W.; Jasim, A.A.; Rahman, H.A.; Muhammad, M.Z.; Ahmad, H. Micro-ball lensed fiber-based glucose sensor. IEEE Sens.
J. 2012, 13, 348–350. [CrossRef]
14. Ali, U.U.M.; Johari, M.A.M.; Jusoh, Z.; Rahman, H.A.; Harun, S.W. A study of relative humidity sensor on micro-ball resonator. J.
Phys. Conf. Ser. 2019, 1371, 012009. [CrossRef]
15. Jali, M.H.; Rahim, H.R.A.; Johari, A.M.; Ahmad, A.; Yusof, H.H.M.; Johari, S.H.; Baharom, M.F.; Thokchom, S.; Dimyati, K.; Harun,
S.W. Integrating microsphere resonator and ZnO nanorods coated glass for humidity sensing application. Opt. Laser Technol. 2021,
143, 107356. [CrossRef]
16. Sharif, M.; Johari, M.; Al Noman, A.; Khudus, M.A.; Harun, S. PMMA microfiber and Microball Resonator for fomaldehyde
liquid sensing. Sens. Actuators A Phys. 2020, 304, 111828. [CrossRef]
17. Jali, M.H.; Rahim, H.R.A.; Johari, A.M.; Ali, U.; Johari, S.H.; Mohamed, H.; Harun, S.W.; Yasin, M. Formaldehyde sensor with
enhanced performance using microsphere resonator-coupled ZnO nanorods coated glass. Opt. Laser Technol. 2021, 139, 106853.
[CrossRef]
18. Duran-Sanchez, M.; Alvarez-Tamayo, R.I.; Pottiez, O.; Kuzin, E.A.; Ibarra-Escamilla, B.; Pinzón, A.B. Fiber laser strain sensor
based in the measurement of a Sagnac interferometer optical power spectrum. In Fiber Optic Sensors and Applications XI.; SPIE
Press: Bellingham, WA, USA, 2014. [CrossRef]
19. Álvarez-Tamayo, R.I.; Durán-Sánchez, M.; Prieto-Cortés, P.; Salceda-Delgado, G.; Castillo-Guzmán, A.A.; Selvas-Aguilar, R.;
Ibarra-Escamilla, B.; Kuzin, E.A. All-fiber laser curvature sensor using an in-fiber modal interferometer based on a double clad
fiber and a multimode fiber structure. Sensors 2017, 17, 2744. [CrossRef] [PubMed]
20. Bai, X.; Fan, D.; Wang, S.; Pu, S.; Zeng, X. Strain sensor based on fiber ring cavity laser with photonic crystal fiber in-line
Mach-Zehnder interferometer. IEEE Photon. J. 2014, 6, 1–8.
21. Zhang, J.; Yang, J.; Hou, L.; Ran, L. Temperature compensated fiber-optic liquid level sensor by micro-sphere based in-line
Michelson interferometer. Opt. Fiber Technol. 2022, 68, 102817. [CrossRef]
22. Bender-Pérez, C.E.; Castillo-Guzmán, A.A.; Alvarez-Tamayo, R.I. Tunable erbium doped fiber laser based on a spherical micro-ball
lens as a Fabry–Perot interference filter. Laser Phys. 2021, 31, 033001. [CrossRef]
Sensors 2023, 23, 3045 12 of 12

23. Niu, P.; Jiang, J.; Wang, S.; Liu, K.; Ma, Z.; Zhang, Y.; Chen, W.; Liu, T. Optical fiber laser refractometer based on an open
microcavity Mach-Zehnder interferometer with an ultra-low detection limit. Opt. Express 2020, 28, 30570–30585. [CrossRef]
24. Zhao, J.; Niu, P.; Zhang, C.; Sun, X.; Bai, H.; Miao, C. Temperature-compensated fiber laser refractometer based on compact
hybrid S-taper/FBG structure. Opt. Fiber Technol. 2018, 45, 300–305. [CrossRef]
25. Zhao, J.; Wang, J.; Zhang, C.; Xu, W.; Sun, X.; Bai, H.; Chen, L. Fiber laser refractometer based on tunable bandpass filter tailored
FBG reflection. Opt. Laser Technol. 2018, 99, 101–106. [CrossRef]
26. Gaspar-Ramirez, O.; Álvarez-Tamayo, R.I.; Prieto-Cortés, P.; Durán-Sánchez, M. Fiber laser refractometer operating in the 2 µm
spectral region based on the MMI effect. Supl. Revista Mex. Física 2021, 2, 66–71. [CrossRef]
27. Li, M.; Wang, Y.; Zhu, G.; Wang, Z.; Singh, R.; Zhang, B.; Kumar, S. Optimization, fabrication, and performance analysis of
SMF-/MMF-based microfiber ball structure. Appl. Opt. 2021, 60, 1196–1200. [CrossRef]
28. Yu, H.; Huang, Q.; Zhao, J. Fabrication of an optical fiber micro-sphere with a diameter of several tens of micrometers. Materials
2014, 7, 4878–4895. [CrossRef] [PubMed]
29. Guzowski, B.; Lisik, Z.; Tosik, G. Realization of optical fibers terminated with ball lenses. Bull. Pol. Acad. Sci. Tech. Sci. 2016, 64,
279–282. [CrossRef]
30. Álvarez-Tamayo, R.I.; Gaspar-Ramírez, O.; Prieto-Cortés, P.; Garcia-Mendez, M.; Barcelata-Pinzón, A. TiOxNy thin film sputtered
on a fiber ball lens as saturable absorber for passive q-switched generation of a sin-gle-tunable/dual-wavelength er-yb double
clad fiber laser. Nanomaterials 2020, 10, 923. [CrossRef]
31. Chen, Y.; Han, Q.; Liu, T.; Xiao, H. Wavelength Dependence of the Sensitivity of All-Fiber Refractometers Based on the
Singlemode–Multimode–Singlemode Structure. IEEE Photon. J. 2014, 6, 1–7.
32. Zhao, Y.; Cai, L.; Li, X.-G.; Meng, F.-C.; Zhao, Z. Investigation of the high sensitivity RI sensor based on SMS fiber structure.
Sensors Actuators A Phys. 2014, 205, 186–190. [CrossRef]
33. Zhao, Y.; Cai, L.; Hu, H.-F. Fiber-Optic Refractive Index Sensor Based on Multi-Tapered SMS Fiber Structure. IEEE Sens. J. 2015,
15, 6348–6353. [CrossRef]
34. Ma, Y.; Qiao, X.; Guo, T.; Wang, R.; Zhang, J.; Weng, Y.; Rong, Q.; Hu, M.; Feng, Z. Mach-Zehnder interferometer based on a
sandwich fiber structure for refractive index measurement. IEEE Sensors J. 2012, 12, 2081–2085. [CrossRef]
35. Zhang, J.; Pu, S.; Rao, J.; Yao, T. Refractive index and temperature sensors based on no-core fiber cascaded with long period fiber
grating. J. Mod. Opt. 2018, 65, 1098–1103. [CrossRef]
36. Ran, Z.; Rao, Y.; Zhang, J.; Liu, Z.; Xu, B. A miniature fiber-optic refractive-index sensor based on laser-machined Fabry-Perot
interferometer tip. J. Light. Technol. 2009, 27, 5426–5429.
37. Prieto-Cortés, P.; Álvarez-Tamayo, R.I.; García-Méndez, M.; Durán-Sánchez, M. Lossy mode resonance generation on sputtered
aluminum-doped zinc oxide thin films deposited on multimode optical fiber structures for sensing applications in the 1.55 µm
wavelength range. Sensors 2019, 19, 4189. [CrossRef] [PubMed]
38. Feng, Y.; Li, H.; Li, S.; Liu, Y.; Meng, X. A high-sensitivity SPR refractive index sensor based on no-core fiber with ag-cu composite
films. Sensors 2021, 21, 7000. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.