Optical sensors based on lossy-mode resonances

TLDR: Lossy-mode resonance (LMR)-based optical sensing technology has emerged in the last two decades as a nanotechnological platform with very interesting and promising properties as mentioned in this paper, which enables one to tune the position of the resonance in the optical spectrum, to excite the resonance with both transverse electric (TE) and transverse magnetic (TM) polarized light.

Abstract: Lossy-mode resonance (LMR)–based optical sensing technology has emerged in the last two decades as a nanotechnological platform with very interesting and promising properties. LMR complements the metallic materials typically used in surface plasmon resonance (SPR)–based sensors, with metallic oxides and polymers. In addition, it enables one to tune the position of the resonance in the optical spectrum, to excite the resonance with both transverse electric (TE) and transverse magnetic (TM) polarized light, and to generate multiple resonances. The domains of application are numerous: as sensors for detection of refractive indices voltage, pH, humidity, chemical species, and antigens, as well as biosensors. This review will discuss the bases of this relatively new technology and will show the main contributions that have permitted the optimization of its performance to the point that the question arises as to whether LMR–based optical sensors could become the sensing platform of the near future.

Figures

Fig. 2. (a) SPRs and LMRs obtained with the same Kretschmann configuration at different incidence angles (63.7º and 86.5º) and polarizations (TE and TM). (b) Lossy-mode resonances obtained for different values in the imaginary part of the refractive index and different coating lengths. Figure reproduced with permission from: a, ref. [15].

Fig. 3. Spectral sweep as a function of coating thickness, resonance wavelengths (numerical data are shown by a solid line and experimental data by squares) as a function of coating thickness, and two spectra obtained for two coating thickness values. Materials analyzed: (a) TiO2/PSS (refractive index: 1.92-2.11) and (b) PAH/PAA (refractive index: 1.47-1.5). Figure reproduced with permission from ref. [42], © 2012 OSA. In video S1 the experimental and theoretical evolution of LMRs in the PAH/PAA-coated cladding removed multimode fiber (CRMF) as a function of coating thickness is shown.

Fig. 5. (a) Coated CRMF-based sensor with a 110 nm SnO2 coating. The LMR shifts to the red as a function of the external refractive index. A photograph of the output light with different external refractive indices is shown. (b) CRMF-based sensor, with a 1200 nm TiO2 coating. The thicker coating permits one to obtain several LMRs with different sensitivities. Figure reproduced with permission from: b, ref. [19], © 2010 OSA.

Figure 1. (a) Generation of LMR and SPR with a nanocoated optical prism (Kretschmann configuration) and a nanocoated D-shaped optical fiber. (b) Conditions for LMR and SPR generation in both configurations.

Fig. 6. (a) Detection of organic vapors. (b) Humidity sensor. (c) Voltage detection. (d) PH sensor. (e) H2S sensor. (f) Immunosensor. Figure reproduced with permission from: a, ref. [72], © 2013 Elsevier; b, ref. [22], © 2010 Elsevier; c, ref. [75], © 2013 OSA; d, ref. [47], © 2011 Elsevier; e, ref. [34], © 2011 Elsevier; f, ref. [52], © 2014 Elsevier.

Fig. 4. (a) Comparison of LMRs obtained with a CRMF structure and a tapered single-mode fiber. (b) Comparison of LMRs obtained with a CRMF structure and a D-shaped fiber. (c) ITO-coated LMR in a CRMF structure (TE and TM LMRs overlap and only one attenuation band is obtained). (d) Indium oxide–coated LMR in a CRMF structure (TE and TM LMRs do not overlap and two attenuation bands are obtained). Figure reproduced with permission from: a,b ref. [60], © 2015 IEEE; c, d ref. [18], © 2010 IOP.

Full text

1
Optical sensors based on lossy-mode resonances
Ignacio Del Villar,
1
* Francisco J. Arregui,
1,2
Carlos R. Zamarreño,
1
Jesus M. Corres,
1
Candido Bariain,
2
Javier Goicoechea,
2
Cesar Elosua,
2
Miguel Hernaez,
2
Pedro J. Rivero,
2
Abian B. Socorro,
1
Aitor Urrutia,
2
Pedro Sanchez,
2
Pablo Zubiate,
2
Diego Lopez,
2
Nerea De
Acha,
2
Joaquin Ascorbe,
2
Ignacio R. Matias
1
1
Institute of Smart Cities (ISC), Public University of Navarra, 31006 Pamplona, Spain
2
Electrical and Electronic Engineering Department, Public University of Navarra, 31006
Pamplona, Spain.
* Corresponding author: ignacio.delvillar@unavarra.es
Abstract: Lossy-mode resonance (LMR)–based optical sensing technology has emerged in
the last two decades as a nanotechnological platform with very interesting and promising
properties. LMR complements the metallic materials typically used in surface plasmon
resonance (SPR)–based sensors, with metallic oxides and polymers. In addition, it enables
one to tune the position of the resonance in the optical spectrum, to excite the resonance
with both transverse electric (TE) and transverse magnetic (TM) polarized light, and to
generate multiple resonances. The domains of application are numerous: as sensors for
detection of refractive indices voltage, pH, humidity, chemical species, and antigens, as
well as biosensors. This review will discuss the bases of this relatively new technology and
will show the main contributions that have permitted the optimization of its performance to
the point that the question arises as to whether LMR–based optical sensors could become
the sensing platform of the near future.
Keywords: optical sensor, resonance, thin-film, waveguide, hydrogel, biosensor
1. INTRODUCTION
In recent years, the deposition of thin films has permitted the development of numerous
applications in important domains such as optical communications, optical microscopy, and
photovoltaics [1,2]. In the field of sensor research, the development of the first surface plasmon
polariton resonance-based sensor in 1982 was a scientific breakthrough [3]. It used the

2
Kretschmann-Raether configuration [4] (see Fig. 1(a)), which basically consists of an optical
prism on which a 56-nm-thick silver coating had been deposited. This setup permits surface
plasmon polaritons to be generated at the metal-dielectric interfaces and to couple light at
specific wavelength ranges. For this reason, this phenomenon is called surface plasmon polariton
resonance or, for the sake of simplicity, surface plasmon resonance (SPR). The position of the
resonance in the spectrum is very sensitive to the thin-film thickness and to the surrounding
medium. If a layer sensitive to biological or chemical species is set on top of the metallic layer,
then a biosensor or a chemical sensor is obtained [5]. In view of these interesting properties, the
number of publications has increased exponentially, especially for detecting chemical and
biological species [6,7], and there are some companies exploiting these devices especially for
biosensing (Biacore http://www.biacore.com, Bionavis http://www.bionavis.com, and Xantec
Bioanalytics http://www.xantec.com/).
Unfortunately, the sensitivity limit for SPR sensors seems to have been attained [8]. However,
there is still another phenomenon that can be obtained with the Kretschmann configuration. In
the same year as the first SPR sensor was developed, it was proved using a dielectric waveguide
with a semiconductor waveguide clad that attenuation maxima in the transmission spectrum
could be obtained for specific thickness values of cladding [9].

3
Figure 1. (a) Generation of LMR and SPR with a nanocoated optical prism (Kretschmann configuration) and a nanocoated D-shaped
optical fiber. (b) Conditions for LMR and SPR generation in both configurations.

4
On this basis it was proved later that it is possible to obtain a sensor for humidity, water, and
alcohol vapor, or even n-heptane and iso-octane vapors, by using an anisotropic polymer
deposited on a waveguide [10–12]. The basic principle of the measurements was to record the
phase shift between the two index-matched modes after propagating through the waveguide.
However, the previous technique does not use a wavelength-based detection system, an idea
explored in 1993 by Marciniack [13]. In that publication the lossy-mode resonances (LMRs)
were observed with a wavelength sweep and explained as a coupling between dielectric
waveguide modes and a lossy mode (a guided mode with a complex effective index) of the
semiconductor-clad waveguide. Surprisingly, it was necessary to wait until 2005 to find a
practical sensor configuration where LMRs are tracked as a function of wavelength [14]. Since
that moment, a great number of works have been published, including the Kretschmann
configuration setup [15,16] and a fiber-optic–based one [17–28] (see Fig. 1(a)), which has given
rise to the necessity of presenting this topic to a broad scientific audience.
Waveguide modes can be classified as guided, leaky, and radiated [29]. Though some authors
use the term guided-mode resonance for the phenomenon explained in this review [30], in view
of the fact that a lossy mode is a specific type of guided mode, the term lossy-mode resonance
has become popular in recent years [17,31–34]. That is why this term will be used henceforth.
In this review, we will discuss the conditions for generation and optimization of the LMR
sensors, and we will show the main applications developed during recent years.
2. BASIC CONCEPTS IN LOSSY-MODE RESONANCES
SPRs and LMRs can be observed in two similar structures: a thin-film-coated optical prism or a
thin-film-coated optical fiber (see Fig. 1a). The optical spectrum position of the SPRs and LMRs
depends on the external refractive index, and both phenomena can be used for the fabrication of
highly sensitive refractometric sensors. In fact, the similarity in shape of LMRs and SPRs in the
optical spectrum has led in some cases to false recognition of an SPR [35–37].
The first step in order to obtain an LMR is to choose an adequate material for the thin film. The
conditions are very different from SPR generation. SPRs are obtained when the real part of the
thin-film permittivity is negative and higher in magnitude than both its own imaginary part and
the permittivity of the material surrounding the thin film, whereas LMRs occur when the real

5
part of the thin-film permittivity is positive and higher in magnitude than both its own imaginary
part and the material surrounding the thin film [30].
According to this, the expressions for SPR and LMR generation as function of permittivity
( ) and refractive index (n is the real part and k the imaginary part) are presented in
Fig. 1(b) under the assumption of a substrate refractive index of 1.45 (with small variations, this
is the value of silica in a broad range of the optical spectrum [38]) with a surrounding medium
refractive index of 1 (air) [39].
Also in Fig. 1(b) a map is plotted containing the regions where LMRs and SPRs can be obtained.
As long as the real part of the thin-film permittivity is positive and the absorption coefficient (k)
is low, many materials except pure metals (typical of SPRs) can induce LMRs. For instance,
several contributions using metal oxides [17–23,40–46], polymer coatings [42,47–50], two-
layer-coated structures combining both materials [22], or even immunosensors consisting of
multilayers of polymers and antibodies [51,52] have successfully produced LMRs so far.
It is important to consider here that there are hybrid materials (e.g., indium tin oxide, ITO),
which satisfy the conditions for LMR generation at certain wavelengths and the conditions for
SPR generation at others [15,17,18]. Precisely with ITO, a material that permits the generation of
both LMRs and SPRs, it has been observed that the optimum angle for exciting LMRs and SPRs
differs (see Fig. 2(a)) [15]. This optimum angle is based on the influence of two parameters:
refractive index and thickness. However, in the setup [15] the same thickness has been analyzed
both for LMR generation at shorter wavelengths and for SPR generation at longer wavelengths.
The result has been that the LMR is excited only at angles approaching 90º, whereas the SPR can
be obtained at angles approaching 90º, but in a not optimal way. At 60º a much better resonance
is attained. In fact this last question agrees with other works where the best angle range for SPR
excitation is 40–75º [53,54], whereas for LMRs the optimum angles approach 90º [55,56]. Even
though it is possible to move away from 90º by modifying the refractive index (increasing the
imaginary part) or changing the thickness, it is easier to obtain the LMR for angles approaching
90º [55]. That is why the simplest and most effective structure for LMR generation consists of
the optical fiber configuration, where the incident angles approach 90º (see Fig. 1(a)).
As an example of optical fiber configuration, several transmission spectra from a cladding-
removed multimode fiber coated with different materials are presented in Fig. 2(b). It is

Frequently asked questions

Q1. What have the authors contributed in "Optical sensors based on lossy-mode resonances" ?

This review will discuss the bases of this relatively new technology and will show the main contributions that have permitted the optimization of its performance to the point that the question arises as to whether LMR–based optical sensors could become the sensing platform of the near future. 

Q2. What future works have the authors mentioned in the paper "Optical sensors based on lossy-mode resonances" ?

In view of the success obtained by LMR-based devices in such a short period and in important domains, such as the biosensors market, the LMR platform, with its continuous sensitivity improvement during recent years, is an ideal candidate platform for the development of the sensors of the future. 

Q3. Why is there a moment when the LMR can no longer be monitored?

Since the LMR peak is progressively redshifted as a function of the refractive index and the thickness of the thin film, there is a moment when the LMR can no longer be monitored due to the limited spectral range of the spectrum analyzer. 

Q4. How many studies have been published in the last five years?

Lossy-mode resonance–based sensors is still a young research field, but it must be underlined that during the last five years more than 50 studies have been published in different journals. 

Q5. What is the sensitivity of a refractometer with a silicon coating?

With a silicon coating, a sensitivity of 5700 nm/RIU in the range of 1.33–1.38 was obtained [14], whereas with a ZnO coating, a sensitivity of 3000–6700 nm/RIU in the range of 1.33-1.4 was attained [27]. 

Q6. What are the design rules for optimized devices?

The design rules for optimized devices were established for devices coated with a single thin film: increasing the refractive index and thickness of the thin film and the surrounding medium refractive index (SRMI), provided the first LMR is tracked. 

Q7. What is the recent development of thin films?

In recent years, the deposition of thin films has permitted the development of numerous applications in important domains such as optical communications, optical microscopy, and photovoltaics [1,2]. 

Q8. Why is the general rule of increasing the thickness for a higher sensitivity not applicable?

Due to ITO’s special characteristics, the general rule of increasing the thickness for a higher sensitivity is not applicable when the LMR band approaches the SPR region. 

Q9. Why are refractometers typically tested with oil or glycerol solutions?

The reason is that refractometers are typically tested with oil or glycerol solutions, which adhere to the rough surface of the polymer, which prevents an adequate characterization. 

Q10. What is the first possibility of the combination of ITO and TiO2?

A first possibility is the combination of ITO and TiO2 [32], which permits a twofold increase in sensitivity compared to the results obtained with a single ITO layer. 

Q11. What is the general rule for obtaining a higher sensitivity?

As a general rule, the thin-film refractive index and thickness, and SRMI must be increased in order to obtain a higher sensitivity [42]. 

Q12. What is the thickness of the coating in Fig 2(b)?

The coating thickness in all the spectra of Fig 2(b) is 200 nm and the real part of the refractive index is 1.55, a typical value for polymeric coatings. 

Q13. What is the optical spectrum of SPRs and LMRs?

The optical spectrum position of the SPRs and LMRs depends on the external refractive index, and both phenomena can be used for the fabrication of highly sensitive refractometric sensors. 

Q14. How many nm/ppm can be detected with a thin film?

In addition to this, in 2012 it was possible to detect, with a thin film including an organic-metallic compound with chemical structure ,vapors from ethanol, methanol, and isopropanol, with sensitivities of 0.067, 0.131, and 0.074 nm/ppm, respectively [71]. 

Q15. What type of coating was used to study sensitivity to hydrogen gas?

In another study, sensitivity to hydrogen gas (H2) was studied with three different types of coatings: a single ITO thin film, a single ITO nanoparticle-based layer, and an ITO thin film plus an ITO nanoparticle layer. 

Q16. Why is it difficult to overcome electronic temperature sensors?

Temperature has only been studied theoretically with LMR-based sensors [70], because it is very difficult to overcome electronic temperature sensors, a technology that, without the need of a thin film, is simpler and effective.