Article

Numerical Study of a Dual-Mode Optical Sensor for Temperature

and Refractive Index Sensing with Enhanced Temperature Range
Muhammad Favad 1 , Muhammad Zakwan 2 , Saleem Shahid 3, * , Ahsan Sarwar Rana 1 ,
Muhammad Mahmood 4 and Wolfgang Bösch 3

1 Department of Electrical and Computer Engineering, Air University, Islamabad 44230, Pakistan;
favad.qadir@au.edu.pk (M.F.); ahsan.sarwar@au.edu.pk (A.S.R.)
2 Institute of Avionics and Aeronautics, Air University, Islamabad 44230, Pakistan; mzakwan@au.edu.pk
3 Institute of Microwave and Photonic Engineering, Graz University of Technology, 8010 Graz, Austria;
wbosch@tugraz.at
4 Department of Mechatronic Engineering, Atlantic Technological University, F91 YW50 Sligo, Ireland;
muhammad.ali@atu.ie
* Correspondence: saleem.shahid@tugraz.at

Abstract
This study presents a photonic integrated optical sensor based on a dual-polarization
microring resonator with angular gratings on a silicon-on-insulator (SOI) waveguide,
enabling simultaneous and precise refractive index (RI) and temperature measurements.
Due to the distinct energy distributions for transverse electric (TE) and transverse magnetic
(TM) modes in SOI waveguides, these modes show distinct sensitivity responses to the
variation in ambient RI and temperature. Simultaneous measurements of both temperature
and RI are enabled by exciting both these transverse modes in the microring resonator
structure. Furthermore, incorporating angular gratings into the microring resonator’s
inner sidewall extends the temperature measurement range by mitigating free spectral
range limitations. This work presents a novel approach to dual-polarization microring
resonators with angular gratings, offering an enhanced temperature measurement range
and detection limit in optical sensing applications requiring an extended temperature
range. The proposed structure is able to yield a simulated temperature measurement
range of approximately 35 nm with a detection limit as low as 2.99 × 10−5 . The achieved
Academic Editor: Mario Iodice
temperature sensitivity is 334 pm/°C and RI sensitivity is 13.33 nm/RIU for the TE0 mode,
Received: 4 May 2025
while the TM0 mode exhibits a temperature sensitivity of 260 pm/°C and an RI sensitivity
Revised: 12 June 2025
of 76.66 nm/RIU.
Accepted: 25 June 2025
Published: 26 June 2025
Keywords: SOI waveguide; integrated optical sensors; dual polarization; simultaneous
Citation: Favad, M.; Zakwan, M.;
measurement
Shahid, S.; Rana, A.S.; Mahmood, M.;
Bösch, W. Numerical Study of a
Dual-Mode Optical Sensor for
Temperature and Refractive Index
Sensing with Enhanced Temperature 1. Introduction
Range. Sensors 2025, 25, 3999. https:// In optical sensing, the simultaneous measurement of multiple parameters is a sig-
doi.org/10.3390/s25133999
nificant area of ongoing research, with numerous structures based on fiber optics and
Copyright: © 2025 by the authors. integrated optics employed to achieve the accurate detection of these parameters. Tem-
Licensee MDPI, Basel, Switzerland. perature and RI are among the most critical parameters due to their interdependence,
This article is an open access article
necessitating simultaneous and precise measurements. Various structures, including op-
distributed under the terms and
tical fibers [1–5] and planar waveguides [6], have been utilized for this purpose. Recent
conditions of the Creative Commons
Attribution (CC BY) license
advancements have introduced complex configurations, such as those involving artificial
(https://creativecommons.org/ neural networks [7,8], plasmon resonance [9], interferometers [10,11], the Fabry–Perot cav-
licenses/by/4.0/). ity, [12–14], fiber SPR [15], and photonic crystal-based nanobeam cavities [16,17]. However,

Sensors 2025, 25, 3999 https://doi.org/10.3390/s25133999

Sensors 2025, 25, 3999 2 of 18

optical microcavities like microring resonators (MRRs) stand out due to their simplicity
and robustness. In MRRs, the resonating wavelength shifts in response to the variations in
ambient temperature and RI; however, this shift is constrained by the free spectral range
(FSR) [18], limiting the simultaneous sensing range for RI and temperature.
Expanding the sensing range is crucial for comprehensive environmental monitoring
and industrial applications, where the accurate detection of diverse and dynamic changes in
RI and temperature is essential. Given that these parameters are often interdependent, their
variations can significantly impact system performance and safety. A broader sensing range
allows integrated optical sensors to provide more detailed and reliable data, enhancing
decision-making processes and operational efficiency. Although numerous studies have
explored methods to extend the measurement range of MRR-based sensors by incorporating
angular gratings [19], cascaded ring resonators [20], and variations in the effective group
index [21], there has been little focus on improving the accuracy of these sensors under
varying temperature conditions [1]. However, by simultaneously measuring the RI and
temperature in an MRR sensor, the accuracy of these parameters can be improved.
This study presents a dual-polarization angular-grating microring resonator (DP AG-
MRR) designed for the precise detection of both RI and temperature over an extended
measurement range. Unlike the dual-resonance approach utilized in metal-clad ridge
waveguide (MCRW) structures, which is primarily focused on RI sensing [22], the pro-
posed DP AG-MRR employs a simplified silicon-on-insulator (SOI) platform capable of
simultaneously measuring both RI and temperature. The dual-mode excitation of TE0 and
TM0 is made possible through a directional coupler-like bend access waveguide, whereas
the extended temperature measurement capability is enabled by embedding angular grat-
ings in the sidewalls of the MRR, which predominantly induce the Bragg effect for the
TE0 mode. For simplicity of design in the coupling section, a directional coupler-like bend
waveguide is employed instead of more complex asymmetric coupling designs. The effec-
tiveness of the designed DP AG-MRR sensor is evaluated using the 2.5D VarFDTD method.
The sensor’s response to temperature and RI changes is characterized by the differential en-
ergy distributions of the fundamental TE0 and TM0 modes in the SOI waveguide, resulting
in distinct spectral peaks for each polarization.

2. Materials and Methods

2.1. Structure Design and Working Principle
To construct the optical sensor, an SOI-based photonic platform was utilized, featuring
a silicon device layer with a thickness of 220 nm and an insulator layer of 2 µm, topped
with air cladding. The refractive indices of the silicon device, insulator layer, and air
cladding were 3.48, 1.44, and 1.0, respectively. Figure 1a illustrates a schematic of the
proposed structure
In the design of the optical sensor, the input bus waveguide width was set at 500 nm to
facilitate single-mode TE0 propagation. Near the coupling section, a narrower bus waveg-
uide width WB of approximately 295 nm was employed to enhance coupling efficiency
with the ring waveguide. An adiabatic taper section bridges the bus waveguides at the
input and the coupling sections. The ring waveguide itself was approximately 550 nm wide,
supporting efficient propagation of both TE0 and TM0 modes. It features a bend radius of
10 µm, with reference to the circle’s center of the ring waveguide. At the coupling section,
the gap between the ring and bus waveguides gc was approximately 0.16 µm. The angular
grating dimensions with length LG , height Hc , and grating gap Wg were 0.568 µm, 0.045 µm,
and 0.042 µm, respectively, with a grating period ( LΛ ) of about 0.61 µm, as depicted in
Figure 1b. The waveguide sections with widths WR + Hc and WR were 595 nm and 550 nm,
Sensors 2025, 25, 3999 3 of 18

respectively. The duty cycle D of the grating was selected at approximately 93%. The overall
device dimensions were less than 42 µm × 25 µm.

Figure 1. (a) Structural layout of the proposed DP AG-MRR. (b) Upper view taken at the center of
the coupling section.

The operating principle of the microring resonator is predicated on the simultaneous

excitation of the TE0 and TM0 modes within the microring cavity (R1,1 ). This can be
achieved by exciting a continuous-wave broadband light source. The input waveguide,
with a thickness of 500 nm, supports only the TE0 mode, after which the light enters the
taper section and couples with the ring waveguide. The width of the ring waveguide was
set to 550 nm, enabling the simultaneous support of both TE0 and TM0 modes, as shown
in Figure 2. Furthermore, it was constructed to accommodate fabrication imperfections.
Even with usual deviations during fabrication [23], the mode coupling efficiency was
maintained, and both TE0 and TM0 modes could be excited and propagated without
significant degradation. To mitigate alignment sensitivities, a taper was employed to
ensure a gradual transition for stable coupling into the microring waveguide. The taper’s
position relative to the ring waveguide was also optimized to facilitate the efficient coupling
of both TE0 and TM0 modes.
To simplify the phase-matching conditions in the coupling section, a directional
coupler-like structure was employed, featuring a zero coupling length and a coupling
gap of approximately 0.16 µm. For the optical microring resonator, the resonance condition
is described as follows:
mλres = neff · L (1)

where m denotes the azimuthal mode number associated with the longitudinal resonances
(m = 1, 2, 3,...), λres represents the resonating wavelength, and neff · L signifies the effective
optical length of the microring resonator. To achieve this dual mode propagation, the width
of the ring waveguide was set at 550 nm when the cladding was air, as shown in Figure 2.
This shows that TE1 modes also propagate in ring waveguides, but they exhibit poor mode
confinement within the ring waveguide and tend to diminish over time. Consequently,
only the TE0 and TM0 modes persisted, as illustrated in Figure 1a.
The coupling strength between the fundamental TE0 mode of the bus waveguide
and that of the ring waveguide was notably higher compared to that between the TE0
mode of the bus waveguide and TM0 mode of the ring waveguide. This discrepancy arises
because coupling between identical polarization modes is inherently more efficient than
between different polarization modes. Figure 3 displays the power coupling ratio of the
TE0 mode in the bus waveguide to both the TE0 and TM0 modes supported by the ring
waveguide and highlights the distinct extinction ratios attributable to their differing power
coupling efficiencies.
Sensors 2025, 25, 3999 4 of 18

Figure 2. Dependence of the effective refractive index (neff ) on waveguide width for air cladding.

Figure 3. Power coupling ratio from bus waveguide mode to the ring mode.

For sensing applications, where shifts in the resonance wavelength are critical, the sens-
ing range is governed by the free spectral range (FSR), which corresponds to the spectral
spacing between successive longitudinal modes of the ring resonator. However, this range
is subject to certain limitations, as shown below:

δλres < FSR (2)

where δλres indicates the variation in the resonance wavelength. To enhance the mea-
surement range, this shift must be greater than the FSR between the longitudinal modes.
To achieve this shift, the incorporation of angular gratings on the inner sidewall of the
microring resonator has been proposed. This modification enables selective wavelength
operation independent of the FSR and hence facilitates the measurement of target parame-
ters over an expanded range. Our target is to achieve the dominant mode that is able to
measure a larger temperature range while measuring RI simultaneously. So, we designed
the angular gratings on the inner wall of the microring resonator in such a way that they
will facilitate Bragg reflection predominantly for the TE0 mode and not for the TM0 mode.
Consequently, the device’s response selectively filters out side modes surrounding the
dominant mode for TE0 polarization, leading to a more pronounced resonance for the TE0
Sensors 2025, 25, 3999 5 of 18

mode relative to the TM0 mode, as shown in Figure 4. This figure illustrates the simulated
normalized power transmission of the dual-polarization angular-grating microring res-
onator (DP AG-MRR) when the cladding is air. The transmission spectrum is quantified as
the ratio of the output-to-input power integrals. The observed variations in the extinction
ratios for the TE0 and TM0 modes are attributed to their distinct power coupling ratios,
as discussed previously. The FSR for the TE0 mode is extended to approximately 35 nm
due to the suppression of side modes. The FSR for the TM0 mode is approximately 7.6 nm,
displaying no dominant mode.

Figure 4. Simulated normalized transmission spectrum of DP AG-MRR. R2,4 .

Figure 5 shows dual resonances visible in the output transmission spectrum corre-
sponding to the TE0 and TM0 modes. Figure 6a illustrates the normalized electric field
profile of the TE0 mode in the input waveguide, positioned near the coupling region. Addi-
tionally, the mode profiles for the TE0 and TM0 polarizations, corresponding to the relative
widths of the waveguides, are depicted in Figure 6b–e.
The proposed dual-polarization microring resonator with angular gratings (DP AG-
MRR) could be fabricated on a 220 nm thick silicon-on-insulator wafer by following a
standard fabrication process for silicon photonic devices. The process would begin with
spin-coating a 250–300 nm thick electron beam photoresist layer and then defining the
microring resonator and waveguide structures using electron beam lithography. This
patterned layout is subsequently transferred onto the silicon device layer using inductively
coupled plasma reactive ion etching (ICP-RIE), using an C4F8 and SF6 gas mixture [24,25].
Given that manufacturing tolerances can impact the performance of conventional microring
resonators [23], careful process calibration would be essential. Lastly, grating couplers
could be incorporated at the input/output waveguide facets using an overlay exposure
and a shallow etch to facilitate efficient fiber-to-chip coupling [26]. A potential approach
for experimentally evaluating the proposed device is to couple light from a tunable laser
or a continuous-wave (CW) broadband source, typically operating near 1550 nm, into the
chip via polarization-sensitive grating couplers. The chip’s output transmission spectrum
can then be recorded using an optical spectrum analyzer [27].
Sensors 2025, 25, 3999 6 of 18

Figure 5. (a) TM0 . (b) TE0 .

Figure 6. (a) Normalized electric field distribution at input bus waveguide near coupling section.
(b) Quasi TE01 fundamental mode profile at wavelength 1550 nm with waveguide of 550 nm and
(c) 595 nm. (d) Quasi TM01 mode profile with waveguide width of 550 nm. (e) TM01 mode profile
with waveguide width of 595 nm.

2.2. Parametric Analysis

The device’s performance was optimized through the application of the 2.5-dimensional
variational finite-difference time-domain (var-FDTD) approach. For broadband simulations,
the variational effective index method was employed (R1,8 ). Mesh accuracy was maintained
at a value of 4, with a minimum mesh step size of 0.00025 µm selected to effectively balance
precision, memory requirements, and simulation duration. The mesh resolution is defined
by a minimum step size of 250 pm. In the x, y, and z dimensions, perfectly matched
layer (PML) boundaries were employed to absorb incident light with minimal reflection.
To further enhance simulation robustness, a stabilized PML profile along the x-axis was
implemented, supplemented with additional layers. This configuration utilizes a stretched
coordinate PML, improving absorption efficiency and reducing reflection at the boundaries.
The dual-polarization angular-grating microring resonator (DP AG-MRR) was evalu-
ated by injecting light from a fundamental TE-mode source into the input port of the bus
waveguide, as described in [28]. Structural parameters, including gc = 160 nm, Hc = 40 nm,
and D = 93%, were meticulously optimized. The device demonstrated a pronounced
resonant peak for the TE0 mode at the operational wavelength of 1550 nm, within a spectral
range of 1500–1600 nm. The conventional ring resonator featured a free spectral range
(FSR) of approximately 9.5 nm, whereas the FSR for the dominant and secondary peaks of
the DP AG-MRR was about 35 nm, nearly threefold that of the conventional design for the
TE0 mode.
Sensors 2025, 25, 3999 7 of 18

2.3. Quality Factor, Extinction Ratio, and Side-Mode Suppression

The extinction ratio (ER) and quality factor are critical metrics for assessing the detec-
tion capabilities of optical sensors, particularly in terms of sensitivity and detection limits.
To attain enhanced sensitivity and minimal detection limits, it is imperative to rigorously
evaluate the ER and QF of the proposed sensor. The quality factor is defined as

λ
Q= (3)
∆λ

where λ denotes the resonant wavelength, and ∆λ represents the full width at half max-
imum (FWHM) of the resonance. To achieve a superior quality factor ( QF ), it is crucial
to minimize losses, which in the DP AG-MRR primarily arise from propagation and cou-
pling. The coupling length is effectively set to zero by opting for a ring-shaped rather than
racetrack-shaped resonator, as demonstrated in [29], thereby reducing losses to a minimal
extent (R1,9 ). The scattering losses due to angular gratings are below 13% [30]. The excess
bending losses for a 550 nm wide ring waveguide are less than 0.009 dB/90° [31] for a
bend radius of 10 µm, which accounts for mode mismatch losses. A bent waveguide with a
width of 295 nm was utilized to optimize coupling, with a coupling gap gc of approximately
160 nm. Other parameters such as the D and Hc were also carefully selected at 93% and
40 nm, respectively.
The dependence of the extinction ratio on the coupling gap gc is depicted in Figure 7.
Within this figure, it is evident that the ER for the TE0 mode decreases as gc increases.
Conversely, the extinction ratio for the TM0 mode exhibits a more gradual variation. A gc
of 160 nm was strategically selected to facilitate clear differentiation between the extinction
ratios of the TE0 and TM0 modes.

Figure 7. Extinction ratio and coupling gap (gc ).

The relationship between quality factor QF and coupling gap gc is illustrated in

Figure 8. The QF for the TM0 mode exhibits minimal variation and remains nearly constant,
while QF for the TE0 mode shows a rapid increase at lower gc values before stabilizing at
higher values. A gc of approximately 160 nm was selected to optimize the QF, achieving an
elevated level of approximately 1.55 × 105 , which is particularly advantageous for targeted
TE0 mode temperature sensing.
Sensors 2025, 25, 3999 8 of 18

Figure 8. Quality factor and coupling gap (gc ).

R1,15 The duty cycle is a key design parameter in the development of the dual-
polarization angular-grating microring resonator (DP AG-MRR). The higher duty cycle
generally increases the grating strength, which in turn enhances coupling efficiency and
side-mode suppression. However, excessively high duty cycles may result in increased cou-
pling between TE and TM modes, consequently reducing side-mode suppression. Figure 9
illustrates the relationship between the QF and Wg for the transverse electric and transverse
magnetic modes. It is observed that an increase in the duty cycle corresponds to an increase
in QF, albeit with a concomitant decrease in the grating gap Wg . Given the complexity of
fabricating smaller grating heights, a trade-off between QF and the grating gap is necessary.
Consequently, a duty cycle of 93% with a grating height Hc of about 45 nm was selected.
The duty cycle for the TM0 mode displays a non-linear relationship and is not the focus of
this study. However, it is noteworthy that the maximum QF for the TM0 mode occurs at a
93% duty cycle.

Figure 9. Quality factor as a function of the grating gap for the TE0 and TM0 modes.
Sensors 2025, 25, 3999 9 of 18

The influence of grating height on the QF and side-mode suppression ratio (SMSR)
for the TE0 mode was evaluated by varying the grating height from 30 nm to 60 nm.
As illustrated in Figure 10, the optimal SMSR is achieved at a grating height of 45 nm, which
is critical for distinguishing between the TE0 and TM0 modes. Notably, the QF remains
relatively constant for the TM0 mode throughout this range. gc and D are maintained at
160 nm and 93%, respectively.
R1,10 The effect of fabrication tolerance [23] on the the features of traditional DP AG-
MRR is presented in Figures 11 and 12. For the proposed device, the effects associated
with the angular grating parameters such as Wg , Hc , and coupling gap gc are discussed.
The optimal device parameters, corresponding to a D of 93%, include a Wg of 0.042 µm, Hc of
0.045 µm, and gc of 0.16 µm. With these parameters, the QF is 1.55 × 105 , and the ER is 3 dB
for the target TE0 mode. Figures 11 and 12 illustrate the effects of changes in Wg , gc , and Hc
on the deviations QF and ER from their optimal values. From Figure 11, it is observed that
the variation in QF and ER is less than 20% for the change in Wg and gc . The deviation in
QF and ER is consistent for these variations. However, as the tolerance of ∆Hc increases,
the deviation in ∆QF also increases, exhibiting opposite behavior for lower tolerance values.
The variation in ER from lower to higher tolerance in ∆Hc is inconsistent, which can be
attributed to the dual-mode overlapping behavior of TE0 and TM0 modes. To achieve
better results and minimize fabrication tolerance, advanced fabrication technologies with
precision on the scale of a few nanometers are required.

Figure 10. QF and SMSR as a function of grating height for TE0 mode.
Sensors 2025, 25, 3999 10 of 18

Figure 11. Deviation in QF from the optimal value.

Figure 12. Impact of dimension variations in DP AG-MRR for extinction ratio.

3. Results and Discussion

3.1. Resonant Wavelength Shift
Figure 13 presents the simulated transmission spectra for environments of air (n = 1)
and chemical gases (n = 1 to 1.03). The analysis indicates a resonance wavelength shift
of 2.4 nm for the TM0 mode, which is limited by the FSR and is not the primary focus of
this study. In contrast, a smaller shift of 0.36 nm is observed for the TE0 mode. Notably,
the TM0 mode exhibits greater light confinement within the cladding region compared to
the waveguide region. As depicted in Figure 6, the sensitivities for different polarization
types are quantified as SRI, TM0 = 76.66 nm/RIU and SRI, TE0 = 13.33 nm/RIU.
Sensors 2025, 25, 3999 11 of 18

Figure 13. Transmission intensity of the proposed structure for range of air (1) to chemical gases
(n = 1 to 1.03).

Figure 14 presents the variation in the resonance wavelength of the TE0 mode with
respect to temperature, observed with air cladding. The spectral shift of the resonance
wavelength corresponding to the TE0 is recorded at 1.67 nm, which is greater than the
1.30 nm shift observed for the TM0 mode. This variation is due to the significantly
higher thermo-optic coefficient of silicon 1.8 × 10−4 × K −1 [17], in contrast to that of air
cladding 9.8 × 10−7 × K −1 [32]. Consequently, the temperature sensitivity of the TE0 mode,
quantified at ST, TE0 = 334 pm/°C exceeds that of the TM0 mode, which is measured at
SRI, TM0
ST, TM0 = 260 pm/°C. Furthermore, the sensitivity ratios for the RI, SRI, TE0 = 5.75 and
ST, TM0
temperature = 0.78 differ significantly, facilitating the simultaneous detection of both
ST, TE0
parameters. Notably, the RI sensitivity is relatively low, indicating that the dual-resonance
approach is well-suited for applications where temperature measurement is of primary
importance. To enhance RI sensitivity while maintaining temperature sensing capabilities,
a viable approach is to incorporate a dual-polarization, subwavelength-based structure.
This method will be explored and presented in our future work.

Figure 14. Transmission response of the proposed structure under varying ambient temperatures.
Sensors 2025, 25, 3999 12 of 18

3.2. Wavelength Selection

Wavelength selectivity for the fundamental TE0 mode was investigated using eigen-
solver analysis [33]. This analysis was applied to waveguide sections with widths WR + Hc
and WR , with widths of 595 nm and 550 nm, respectively, as depicted in Figure 1b.
To demonstrate wavelength selectivity, Figure 15 features a solid red line representing
the equivalent wavelength of the dual-polarization angular-grating microring resonator
(DP AG-MRR), calculated using Rytov’s formula [34], as below:

n2avg = D · n2eff1 + (1 − D ) · n2eff2 (4)

Figure 15. Dominant TE0 mode selection of DP AG-MRR.

Here, neff1 and neff2 represent the effective refractive indices of the waveguide seg-
ments with widths WR + Hc and WR , respectively. The solid blue line in the diagram
represents the Bragg wavelength, intersecting the red line at 1550 nm. The dashed black,
blue, and green lines correspond to mode numbers 103, 102, and 101, respectively. The wave-
lengths for modes 101 and 103 are significantly divergent from the Bragg wavelengths;
therefore, only the mode proximate to the Bragg wavelength, specifically mode number 102
(dashed blue line), will be selected as the dominant mode. The wavelength selectivity of
the TM0 mode will be explored in our forthcoming research, which will include investiga-
tions into the dual-polarization angular-grating microring resonator (DP-AG MRR) (R1,2 ).
Figure 16 demonstrates that angular gratings have no impact on the TM0 mode, resulting
in no wavelength selection from the TM0 mode. The dashed red line represents the average
effective index of the waveguide, which corresponds to the same wavelength range as the
TE0 mode, as both modes propagate within the same ring waveguide. The dashed blue
line in Figure 16 indicates the Bragg wavelength within the 1.03 to 1.12 µm range, which is
significantly different from the 1.5 to 1.6 µm wavelength range. Although mode 102 is close
to the Bragg wavelength for the TM0 mode, it is not selected because it does not intersect
with the dashed red line representing the average effective index of the waveguide. This
behavior is evident in the transmission spectrum shown in Figure 4. Notably, this study
marks the first successful demonstration of a dominant TE0 mode using angular gratings
within a dual-polarization microring resonator. Table 1 provides a comparative analysis of
the current study on the simultaneous detection of RI and temperature through MRR-based
optical sensing.
Sensors 2025, 25, 3999 13 of 18

Figure 16. Non-selection of dominant TM0 mode in DP AG-MRR (R1,2 ).

Table 1. Comparative analysis with other MRR structures.

RI T
Structure Range RI; T LOD (/RIU);
References Parameters Sensitivities Sensitivities
Type (nm) (/°C)
(nm/RIU) (pm/°C)
[19] AG-MRR Single 95.27 pm/% NA 1 50.2 (RI) 0.0032
Limited by
[35] DP-MRR Dual 104; 319 78.7; 34.1 3.8 × 10−4 ; 0.5
FSR
Limited by
[36] SWGMRR Single 366 NA1 NM 2
FSR
AG-
[37] Single 672.8 NA 1 NM 2 6.69 × 10−5
SWGMRR
7.6; 35 (FSR 1.14 × 10−2 ;
This Work DP-AGMRR Dual 76.66; 13.33 260; 334
Free for T) 2.99 × 10−5
1 Not applicable. 2 Not mentioned.

3.3. Sensing Analysis

The sensing characteristics of DP AG-MRR are quantified using the sensitivity matrix
SMT,n defined below: " #
ST,TE0 SRI,TE0
SMT,RI = (5)
ST,TM0 SRI,TM0

where ST,TM0 and SRI,TM0 denote the sensitivities to temperature and RI, respectively,
under TM0 polarization, while ST,TE0 and SRI,TE0 represent the corresponding sensitivities
under TE0 polarization. The variations in RI and temperature can be determined using the
following equations: " # " #
∆T −1 ∆λTE0
= SMT,RI · (6)
∆n ∆λTM0

SRI,TM0 · ∆λTE0 − SRI,TE0 · ∆λTM0

∆T = (7)
det(SMT,RI )
ST,TE0 · ∆λTM0 − ST,TM0 · ∆λTE0
∆n = (8)
det(SMT,RI )
Sensors 2025, 25, 3999 14 of 18

From Equations (7) and (8), it is clear that changes in temperature and RI are calculated
using the combined effect of wavelength deviation represented as ∆λTE0 and ∆λTM0 due to
TE0 and TM0 polarizations, respectively, which ensures accurate measurement.
The sensitivity of SOI-based MRR is attributed to the corresponding deviation in the
effective refractive index, which is induced by variations in the surrounding temperature
and RI. This relationship was previously established in [38].
  !
δλ p δneff,p δλ p
Sensitivityvar,p = = × (9)
δvar δvar δneff,p

where δvar represents the change in the variable, either RI or temperature; p denotes the
polarization, either TM0 or TE0 ; λ p denotes the resonance wavelength corresponding to
polarization state p; and δneff,p indicates the change in the effective refractive index for
polarization p.
The ambient variations change the neff , and this can be described by

δneff,p δn δnc δn
(λ) = Aclad,p (λ) · clad (λ) + Ac,p (λ) · (λ) + Abuff,p (λ) · buff (λ) (10)
δvar δvar δvar δvar

Here, Aclad,p , Ac,p , and Abuff,p represent the confinement factors in the cladding, core,
and buffer regions, respectively, for polarization p, as illustrated in Figure 12. If the ambient
RI changes, no change is induced in (neff ) for the core and buffer layers, as δn c
δn ( λ ) = 0 and
δnbuff
δn ( λ ) = 0, respectively. Regarding the effect of temperature changes on the cladding,
silicon, and silica layers, it is essential to consider the thermo-optic coefficients of these
materials. Given that the thermo-optic coefficients [39,40] for liquid solvents and silicon
layers are significantly higher than those for silica, the impact of the silica layer on tem-
perature variation can be disregarded. Consequently, using Equation (6), we derive the
following equations:

δnclad δλ p
Sensitivityn,p = Aclad,p (λ) · (λ) · (11)
δn δneff,p

δnclad δnc δλ p
SensitivityT,p = Aclad,p (λ) · (λ) + Ac,p (λ) · (λ) (12)
δT δT δneff,p
For the equations presented, Sensitivityn,TM0 will be greater than Sensitivityn,TE0 . This
discrepancy arises because Sensitivityn,p is strongly influenced by the confinement factor
within the cladding region, and the TM0 polarization exhibits greater field intensity within
this region. Considering the substantially lower thermo-optic coefficient of air compared
to that of silicon [32,41], SensitivityT,TE0 is correspondingly lower in air than in the silicon
layer. Additionally, SensitivityT,TM0 remains low in silicon due to the predominance of TE0
fields within the silicon core, as opposed to TM0 fields, which are more concentrated in the
cladding region.
Consequently, the sensitivities to both RI and temperature vary between TE0 and
TM0 polarizations. The incorporation of angular gratings facilitates the simultaneous
measurement of the RI and temperature, enabling a broader range of temperature assess-
ment. Utilizing Equations (5) and (6) and incorporating sensitivity values corresponding to
specific polarizations, we formulate the following equation:
" # " # −1 " #
∆T 13.33 nm/RIU 334 pm/◦ C ∆λTE0
= (13)
∆n 76.66 nm/RIU 260 pm/◦ C ∆λTM0
Sensors 2025, 25, 3999 15 of 18

One key parameter for assessing sensor performance is the limit of detection (LOD).
In this work, we primarily focus on extending the FSR for temperature measurement while
achieving an improved LOD, simultaneously monitoring both the RI and temperature.
Using the expression presented in [42], our structure achieves an LOD of approximately
2.99 × 10−5 for the TE0 mode, which is employed for temperature sensing, while the LOD
for the TM0 mode is 1.14 × 10−2 :
λres
LOD = (14)
QS
As shown in Table 1, our structure exhibits a superior LOD and sensitivity for tempera-
ture variations, making it particularly suitable for applications where ambient temperature
changes are of prime importance.
R1,14 To assess the viability of simultaneous measurements with our device, we first
set the temperature Tset to 27 °C and the refractive index RIset ) to 1 (the refractive index
of air) in the simulation setup. The corresponding resonant wavelengths, λT,set and λRI,set ,
were measured to be 1550.07 nm and 1549.81 nm, respectively. Using Equation (13), we
then varied the values of temperature and refractive index (RI) and calculated ∆λTE0 and
∆λTM0 for each variation group with respect to the set conditions. Table 2 presents the
measured values of ∆λTE0 and ∆λTM0 for six distinct groups, enabling the determination of
changes in temperature (∆T) and RI (∆n).

Table 2. Measurement of temperature and RI simultaneously.

∆Tbase (◦ C) ∆nbase ∆λ T E0 (nm) ∆λ T M0 (nm) ∆ncal ∆Tcal (◦ C)

0 0.03 0.4 2.3 0.030 0.001
3 0.01 1.51 3.24 0.010 3.33
5 0 1.4 1.6 0.003 5.55
8 0.02 3.5 7.0 0.028 8.05
12 0.015 4.6 6.9 0.025 11.98
13 0.025 5.33 9.65 0.044 12.87

4. Conclusions
In this study, we introduced a dual-polarization angular-grating microring resonator
(DP AG-MRR)-based structure incorporating angular gratings and a directional coupler-like
coupling section, designed to extend the measurement range for temperature variations.
For the first time, we explored the capability of the angular-grating MRR (AG-MRR) to
simultaneously detect changes in temperature and RI across dual polarization modes
within its transmission spectra, thereby achieving a comprehensive measurement range
for ambient temperature variations. Through parameter optimization, we established a set
of resonance pairs with distinct sensing ratios for temperature and RI. The simulated RI
sensitivities of 13.33 nm/RIU and 76.66 nm/RIU, along with temperature sensitivities of
334 pm/°C and 260 pm/°C, were obtained by analyzing the resonance shift in the modes
in response to RI variations at different temperatures. The simulations yielded an FSR of
about 35 nm with a detection limit as low as 2.99 × 10−5 , significantly enhancing the mea-
surement range for temperature while concurrently sensing the ambient RI. Furthermore,
we assessed the impact of dimensional deviations in the angular gratings and coupling
region parameters, such as the coupling gap gc , grating gap Wg , and duty cycle (D), on the
performance of the DP AG-MRR. Given its straightforward configuration and compatibility
with various silicon-on-insulator (SOI) devices, this design proves advantageous for ap-
plications requiring simultaneous detection with a single measurement including a larger
temperature range.
Sensors 2025, 25, 3999 16 of 18

Author Contributions: Conceptualization, M.F.; methodology, M.F.; software, M.Z.; validation,

S.S.; formal analysis, S.S.; investigation, M.F. and A.S.R.; resources, M.M.; data curation, M.F.;
writing—original draft preparation, M.F. and A.S.R.; writing—review and editing, M.Z., W.B., and
S.S.; visualization, S.S. and M.M.; supervision, M.Z. and S.S.; project administration, M.F.; funding
acquisition, S.S. and W.B. All authors have read and agreed to the published version of the manuscript.

Funding: Open Access Funding by the Graz University of Technology, Austria.

Data Availability Statement: All original findings presented in this study are contained within
the article.

Conflicts of Interest: The authors declare no conflicts of interest.

Abbreviations
The following abbreviations are used in this manuscript:
SOI Silicon-On-Insulator;
TE Transverse Electric;
TM Transverse Magnetic;
RI Refractive Index;
SPR Surface Plasmon Resonance;
MRR Microring Resonator;
FSR Free Spectral Range;
MCRW Metal-Clad Ridge Waveguide;
DP AG-MRR Dual-Polarization Angular-Grating microring resonator;
ICP-RIE Inductively Coupled Plasma Reactive Ion Etching;
CW Continuous Wave;
PML Perfectly Matched Layer;
ER Extinction Ratio;
QF Quality Factor;
LOD Limit of Detection.

References
1. Alonso-Murias, M.d.C.; Velázquez-González, J.S.; Monzón-Hernández, D. SPR Fiber Tip Sensor for the Simultaneous Measure-
ment of Refractive Index, Temperature, and Level of a Liquid. J. Light. Technol. 2019, 37, 4808–4814. [CrossRef]
2. Liu, M.; Leng, X.; Ni, W.; Shum, P. Simultaneous and Ultrasensitive Measurement of Refractive Index and Temperature Based on
SPR of Hollow-Core Anti-resonant Fibers. Plasmonics 2024, 19, 3091–3100. [CrossRef]
3. Zhang, Y.; Liu, M.; Zhang, Y.; Liu, Z.; Yang, X.; Zhang, J.; Yang, J.; Yuan, L. Simultaneous measurement of temperature and
refractive index based on a hybrid surface plasmon resonance multimode interference fiber sensor. Appl. Opt. 2020, 59, 1225–1229.
[CrossRef]
4. Yu, F.; Xue, P.; Zhao, X.; Zheng, J. Simultaneous Measurement of Refractive Index and Temperature Based on a Peanut-Shape
Structure In-Line Fiber Mach–Zehnder Interferometer. IEEE Sens. J. 2019, 19, 950–955. [CrossRef]
5. Freitas, A.I.; Pereira, D.; Bierlich, J.; Marques, J.C.; Ferreira, M.S. Simultaneous measurement of refractive index and temperature
using a balloon-type optical fiber sensor. Optik 2024, 309, 0030–4026. [CrossRef]
6. Butt, M.A. Plasmonic Sensor System Embedded with Orthogonal Mode Couplers for Simultaneous Monitoring of Temperature
and Refractive Index. Plasmonics 2025, 20, 459–469. [CrossRef]
7. Geetha, T.S.; Chellaswamy, C.; Kaliraja, T. Optimized convolutional neural network-based temperature and refractive index fiber
optic sensor. J. Opt. 2024. [CrossRef]
8. Gou, Z.; Wang, C.; Han, Z.; Nie, T.; Tian, H. Artificial neural networks assisting the design of a dual-mode photonic crystal
nanobeam cavity for simultaneous sensing of the refractive index and temperature. Appl. Opt. 2022, 61, 4802–4808. [CrossRef]
9. Dhara, P.; Singh, V.K.; Kumar, A.; Olivero, M.; Perrone, G . Reflection based silicon incorporated silver coated fiber optic SPR
sensor for refractive index and temperature measurement. Microsyst. Technol. 2024, 30, 913–922. [CrossRef]
10. Duan, C.; Li, J.; Zhang, K.; Tian, M. Optical fiber dual-parameter sensors based on different kinds of interferometers for measuring
refractive index and temperature: A review. Opt. Quant. Electron. 2024, 56, 262. [CrossRef]
11. Yu, X.; Zuo, S.; Zhang, Y.; Ma, Y.; Wang, R.; Yang, W.; Tian, K.; Geng, T.; Wang, P. Mach–Zehnder interferometer based on peanut
structure for temperature and refractive index measurement. Opt. Quant. Electron. 2024, 56, 492–497. [CrossRef]
Sensors 2025, 25, 3999 17 of 18

12. Liu, Q.; Wang, D.; Li, X.; Gao, H.; Yu, D. Simultaneous measurement of temperature and liquid refractive index based on fiber
open Fabry-Pérot cavity and Bragg grating. Optoelectron. Lett. 2024, 20, 477–482. [CrossRef]
13. André, R.M.; Warren-Smith, S.C.; Becker, M.; Dellith, J.; Rothhardt, M.; Zibaii, M.I.; Latifi, H.; Marques, M.B.; Bartelt, H.; Fraz, O.
Simultaneous measurement of temperature and refractive index using focused ion beam milled Fabry-Perot cavities in optical
fiber micro-tips. Opt. Express 2016, 24, 14053–14065. [CrossRef]
14. Zhu, Y.; Zhang, Y.; Tang, Q.; Li, S.; Zheng, H.; Liang, D.; Xiao, H.; Du, C.; Yu, Y.; Ruan, S. Double-Cavity Fabry–Perot Interferometer
Sensor Based on Polymer-Filled Hollow Core Fiber for Simultaneous Measurement of Temperature and Gas Pressure. Sensors
2025, 25, 2396. [CrossRef] [PubMed]
15. Lu, X.; Yu, X.; Zhou, J.; Chang, M.; Lu, D. An Ultra-Wide Range D-Shaped Fiber SPR Sensor with a Nanostructure of Gold–MoS2
and Sodium for the Simultaneous Measurement of Refractive Index and Temperature. Sensors 2025, 25, 377. [CrossRef] [PubMed]
16. Yang, S.; Wu, Y.; Yang, Y.; Wang, C.; Tian, H. High sensitivity and anti-external interference dual-parameter sensor based on a
multimode slotted photonic crystal nanobeam cavity. J. Mod. Opt. 2021, 68, 357–364. [CrossRef]
17. Liu, P.; Shi, Y. Simultaneous measurement of refractive index and temperature using cascaded side-coupled photonic crystal
nanobeam cavities. Opt. Express 2017, 25, 28398–28406. [CrossRef]
18. Claes, T.; Molera, J.G.; Vos, K.D.; Schacht, E.; Baets, R.; Bienstman, P. Label-Free Biosensing With a Slot-Waveguide-Based Ring
Resonator in Silicon on Insulator. IEEE Photonics J. 2009, 1, 197–204. [CrossRef]
19. Ma, T.; Sun, L.; Yuan, J.; Sang, X.; Yan, B.; Wang, K.; Yu, C. Integrated label-free optical biochemical sensor with a large
measurement range based on an angular grating-microring resonator. Appl. Opt. 2016, 55, 4784–4790. [CrossRef]
20. Kim, H.-T.; Yu, M. Cascaded ring resonator-based temperature sensor with simultaneously enhanced sensitivity and range. Opt.
Express 2016, 24, 9501–9510. [CrossRef]
21. Addya, S.; Dey, S.; Mandal, S. Optical Ring Resonator Based Temperature Sensor. Sens. Imaging 2017, 18, 33. [CrossRef]
22. Saha, N.; Kumar, A. A novel dual resonance long period waveguide grating based highly sensitive refractive index sensor with
reduced temperature sensitivity. Opt. Commun. 2020, 474, 126092. [CrossRef]
23. Prinzen, A.; Waldow, M.; Kurz, H. Fabrication tolerances of SOI based directional couplers and ring resonators. Opt. Express 2013,
21, 17212–17220. [CrossRef] [PubMed]
24. Selvaraja, S.K.; Sleeckx, E.; Schaekers, M.; Bogaerts, W.; Thourhout, D.V.; Dumon, P.; Baets, R. Low-loss amorphous silicon-on-
insulator technology for photonic integrated circuitry. Opt. Commun. 2009, 282, 1767–1770. [CrossRef]
25. Watts, M.R.; Trotter, D.C.; Young, R.W.; Lentine, A.L. Ultralow power silicon microdisk modulators and switches. In Proceedings
of the 2008 5th IEEE International Conference on Group IV Photonics, Sorrento, Italy, 17–19 September 2008. [CrossRef]
26. Roelkens, G.; Thourhout, D.V.; Baets, R. High efficiency grating coupler between silicon-on-insulator waveguides and perfectly
vertical optical fibers. Opt. Lett. 2007, 32, 1495–1497. [CrossRef]
27. Vos, K.D.; Bartolozzi, I.; Schacht, E.; Bienstman, P.; Baets, R. Silicon-on-Insulator microring resonator for sensitive and label-free
biosensing. Opt. Express 2007, 15, 7610–7615. [CrossRef]
28. Marcatili, E.A.J. Dielectric rectangular waveguide and directional coupler for integrated optics. Bell Syst. Tech. J. 1969,
48, 2071–2102. [CrossRef]
29. Tseng, C.-W.; Tsai, C.-W.; Lin, K.-C.; Lee, M.-C.; Chen, Y.-J. Study of coupling loss on strongly-coupled, ultra compact microring
resonators. Opt. Express 2013, 21, 7250–7257. [CrossRef]
30. Cai, X.; Wang, J.; Strain, M.J.; Johnson-Morris, B.; Zhu, J.; Sorel, M.; O’Brien, J.L.; Thompson, M.G.; Yu, S. Integrated compact
optical vortex beam emitters. Science 2012, 338, 363–366. [CrossRef]
31. Selvaraja, S.K.; Jaenen, P.; Bogaerts, W.; Thourhout, D.V.; Dumon, P.; Baets, R. Fabrication of Photonic Wire and Crystal Circuits
in Silicon-on-Insulator Using 193-nm Optical Lithography. J. Light. Technol. 2009, 27, 4076–4083. [CrossRef]
32. Bauld, R.; Choi, D.-Y.W.; Bazylewski, P.; Divigalpitiya, R.; Fanchini, G. Thermo-optical characterization and thermal properties of
graphene–polymer composites: A review. J. Mater. Chem. C 2018, 6, 2901–2914. [CrossRef]
33. Shi, W.; Wang, X.; Zhang, W.; Yun, H.; Lin, C.; Chrostowski, L.; Jaeger, N.A.F. Grating-coupled silicon microring resonators. Appl.
Phys. Lett. 2012, 100, 121118. [CrossRef]
34. Liu, W.; Lai, Z.; Guo, H.; Liu, Y. Guided-mode resonance filters with shallow grating. Opt. Lett. 2010, 35, 865–867. [CrossRef]
35. Liu, P.; Shi, Y. Simultaneous measurement of refractive index and temperature using a dual polarization ring. Appl. Opt. 2016,
55, 3537–3541. [CrossRef]
36. Tu, Z.; Gao, D.; Zhang, M.; Zhang, D. High-sensitivity complex refractive index sensing based on Fano resonance in the
subwavelength grating waveguide micro-ring resonator. Opt. Express 2017, 25, 20911–20922. [CrossRef]
37. Wu, N.; Xia, L. Side-mode suppressed filter based on an angular grating-subwavelength grating microring resonator with high
flexibility in wavelength design. Appl. Opt. 2019, 58, 7174–7180. [CrossRef] [PubMed]
38. Fu, H.-w.; Zhao, H.; Qiao, X.-g.; Li, Y.; Zhao, D.-z.; Yong, Z. Study on a novel photonic crystal temperature sensor. Optoelectron.
Lett. 2011, 7, 419–422. [CrossRef]
Sensors 2025, 25, 3999 18 of 18

39. Kamikawachi, R.C.; Abe, I.; Paterno, A.S.; Kalinowski, H.J.; Muller, M.; Pinto, J.L.; Fabris, J.L. Determination of thermo-optic
coefficient in liquids with fiber Bragg grating refractometer. Opt. Commun. 2008, 281, 621–625. [CrossRef]
40. Lee, C.-L.; Ho, H.-Y.; Gu, J.-H.; Yeh, T.-Y.; Tseng, C.-H. Dual hollow core fiber-based Fabry Perot interferometer for measuring the
thermo-optic coefficients of liquids. Opt. Lett. 2015, 40, 459–462. [CrossRef]
41. Frey, B.J.; Leviton, D.B.; Madison, T.J. Temperature-dependent refractive index of silicon and germanium. Proc. Optomech. Technol.
Astron. 2006, 6273, 62732J. [CrossRef]
42. Chrostowski, L.; Grist, S.; Flueckiger, J.; Shi, W.; Wang, X.; Ouellet, E.; Yun, H.; Webb, M.; Nie, B.; Liang, Z.; et al. Silicon photonic
resonator sensors and devices. In Proceedings of the SPIE 8236, Laser Resonators, Microresonators, and Beam Control XIV,
San Francisco, CA, USA, 6 February 2012. [CrossRef]

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.