- Email: [email protected]
Sensing RF and microwave energy with fiber Bragg grating heating via soft ferromagnetic glass-coated microwires
Sensors and Actuators A 210 (2014) 25–31
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
Sensing RF and microwave energy with fiber Bragg grating heating via soft ferromagnetic glass-coated microwires P. Colosimo a,∗ , A. Chen a,b , J. Devkota b , H. Srikanth b , M.H. Phan b,∗∗ a b
Applied Physics Laboratory, University of Washington, Seattle, WA 98105, USA Department of Physics, University of South Florida, Tampa, FL 33620, USA
a r t i c l e
i n f o
Article history: Received 30 July 2013 Received in revised form 31 December 2013 Accepted 25 January 2014 Available online 1 February 2014 Keywords: Fiber optic Sensor Magnetic Microwire
a b s t r a c t We present results from a fiber Bragg grating-based microwave energy sensor. The sensor relies on a soft ferromagnetic glass-coated microwire that is bonded to the cladding of the grating. The microwire absorbs microwave energy and heats up thus raising the temperature of the fiber Bragg grating. Compared to a similar sensor that uses gold to absorb electromagnetic radiation, the microwire yields a sensor with greater sensitivity (∼10 times at f = 3.25 GHz) relative to the perturbation of the microwave field. With the sensor reported here, the best sensitivity to electromagnetic radiation corresponds to AC electric fields that have an average electric energy density of approximately 1.3 J/m3 . © 2014 Elsevier B.V. All rights reserved.
1. Introduction The fiber Bragg grating (FBG) is the basis of numerous sensors [1–3]. For the most part, strain and temperature are the primary environmental parameters that can be detected with FBGs. Other variables can be measured by using a probe design that converts the desired variable to a strain or temperature change. For example, an FBG bonded to the wall of a vacuum chamber might be used to measure pressure if the wall strain vs. pressure calibration were known. We present results from a new type of microwave energy sensor that relies on Joule heating of a soft ferromagnetic glasscoated microwire to change the temperature of an FBG. Using optical fibers to sense electromagnetic (EM) fields is a promising approach and has been the subject of extensive research. A small sampling of the many published designs rely on surrounding the FBG core with polymer-dispersed liquid crystals [4], fluorescence from a phosphor that has a temperature-dependent lifetime [5], and resonant coupling of light between the fiber’s core and a nonlinear electro-optical slab waveguide [6,7]. Many more complex EM field sensors are reviewed in [8].
∗ Corresponding author at: Applied Physics Laboratory, University of Washington, 1013 NE 40th Street, Box 355640, Seattle, WA 98105-6698, USA. Tel.: +1 206 543 3148; fax: +1 206 543 6785. ∗∗ Corresponding author. E-mail addresses: [email protected] (P. Colosimo), [email protected] (M.H. Phan). 0924-4247/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sna.2014.01.038
Previous EM field sensors based on fiber optic technology are rather large (cm dimensions [9,10]) and/or point sensors [5,11]. To the best of our knowledge, previous reports of fiber optic schemes to detect EM fields have not reported the degree to which those probes perturbed the field being measured. The sensor presented here has a greater sensitivity compared to one that uses gold to absorb electromagnetic radiation. It can be deployed as a distributed sensor, and often only minimally perturbs the field being measured (i.e. produced a small increase of S11 in the test waveguide). Given that FBGs of very short length (0.35 mm) can be made [12], there is potential to construct very small probes based on this design.
2. Device theory and fabrication An FBG is a periodic variation of the index of refraction of an optical fiber. Wavelengths of light at and near the Bragg condition will be reflected while other wavelengths will pass through the grating with their amplitudes virtually unaffected. For a fiber grating designed to reflect a single wavelength, the Bragg condition can be written FBG = 2n where FBG is the Bragg wavelength, n is the effective index of refraction of the core, and is the grating pitch. As the temperature of an FBG changes, the center wavelength of the grating will shift according to the formula [1]:
FBG = 2n ˛ +
(dn/dT ) T n
26
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
where ˛ is the coefficient of thermal expansion, n is the effective index of refraction, and T is the temperature change. This wavelength shift is exploited to use an FBG as a temperature sensor. For wavelengths around 1550 nm and optical fiber (Corning SMF-28) used here, the conversion from FBG to temperature is well-known and about 10 pm/◦ C [1]. By monitoring the transmitted or reflected optical spectrum, we can determine the average temperature of the FBG. To create the probes, the microwave-absorbing soft ferromagnetic glass-coated microwires were glued to the cladding of a commercial FBG. For the probes that use gold to absorb microwave energy, approximately 120 nm of the metal was sputtered onto the FBG’s cladding. Any temperature increase of the microwire or gold due to Joule heating was transmitted to the FBG and appeared as a shift of the notch in the FBG’s transmission spectrum. Soft ferromagnetic amorphous glass-coated microwires of Fe4.97 Co64.63 B16 Si11 Cr3.4 Ni0.02 were prepared by a modified TaylorUlitovsky process [13]. The metallic diameter of the wire was about 14 m surrounded by about 3 m thick glass layer. The wires used were approximately 3 cm long. This length was chosen to make each microwire slightly longer than the length of the FBG. This alloy composition has a vanishing magnetostriction constant ( ∼ 0) that
determines its soft ferromagnetic characteristic. These microwires show a giant magneto-impedance effect making them attractive for use in magnetic field sensors [14–16]. Since the microwires also show excellent microwave absorption properties [17], they have recently been exploited for making novel microwire-based composites for structural health monitoring and self-sensing applications [18,19]. In this study, we demonstrate that using these microwires as a microwave absorber it is possible to fabricate a fiber Bragg grating-based microwave energy sensor with improved sensitivity and less perturbation of the microwave field. 3. Experiments and results Light from a broadband amplified spontaneous emission source (JDSU M/N BBS1560+1FP) was launched into the optical fiber that contained the FBG. Light transmitted through the optical fiber was monitored with an optical spectrum analyzer (HP M/N 70951B). The sensor was placed into a homemade 50 microstrip transmission line. A microstrip transmission line produces a quasi-Transverse Electromagnetic (TEM) mode that we treat as a pure TEM mode here. The microstrip design was chosen to facilitate insertion and removal of the probe. We will refer to the microstrip transmission
Fig. 1. Diagram of experimental setup. (a) Setup used for sensor response data. Optical (microwave) components and paths are shown in gray (black). Abbreviations are as follows: ASE – amplified spontaneous emission; TEM cell – 50 microstrip transmission line; OSA – optical spectrum analyzer; D.C. – directional coupler. (b) Setup used for S11 measurements. The difference between S11 with only ordinary fiber in the TEM cell and S11 with the sensor in the TEM cell was used to quantify the S11 increase due to the sensor. (c) Cross-section of gold-based probe (not to scale). (d) Cross-section of microwire-based probe (not to scale).
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
line as “TEM cell” here. The TEM cell comprises two parallel copper plates with the geometry needed to produce 50 impedance. Electromagnetic energy propagates down the length of the TEM cell, the electric field is mostly confined to the volume between the metal plates, and the magnetic field is perpendicular to the length of the TEM cell, which is strongest between the two plates. Our configuration results in a magnetic field that is predominantly parallel to the axis of the optical fiber probe. For more details on the microstrip transmission line, see [20–22]. As microwave energy of different frequencies and powers was sent through the transmission line, the optical spectrum transmitted by the FBG was recorded. After a step change of the microwave frequency or power, the OSA signal took up to 1–2 s to stabilize. The microwaves were generated by an HP M/N 8703A, amplified by either a Mini-Circuits M/N ZHL-42W or Avantek M/N APT-10555, and monitored with an HP M/N E4419B. A schematic diagram of the experimental set-up for response measurements is given in Fig. 1a. To record S11 due to each different probe, the setup shown in Fig. 1b was used. The HP M/N 8703A was connected to the TEM cell via a coaxial cable which was terminated with a 50 load. S11 due to a section or ordinary optical fiber was recorded as a reference scan. This reference scan was subtracted from an S11 scan taken with the sensor in the TEM cell. For the S11 scans, care was taken to not disturb the setup while the optical fiber was being moved. The remainder of Fig. 1 shows how the various microwave absorbers were bonded to the FBG. Fig. 2 illustrates the geometry of the sensor within the TEM cell as well as the electric field distribution of the TEM cell (at one instant in time). The shift of FBG as a function of average electric energy density at one microwave frequency is shown in Fig. 3 (open triangles). The data clearly indicate that FBG increases as the amount of microwave energy in the transmission line increases. Our most sensitive results (able to detect average electric energy density of 1.3 J/m3 RMS, at 10.25 GHz) correspond to measuring a temperature change of approximately 0.22 ◦ C. Data was taken with a bare FBG in the transmission line to determine whether or not FBG was influenced by factors other than the microwire (filled diamonds in Fig. 3). We found a small response to microwave energy with the bare FBG. This response of the bare FBG was most likely due to heating of the transmission line being coupled to the FBG. During our early studies, the response of the microwire probe
a
27
Top conductor
V Bottom conductor Sensor
b
Fig. 2. Details of TEM cell. (a) The sensor was perpendicular to the length of the top and bottom conductors of the TEM cell. The 3 cm long microwire was centered on the 2.5 cm long FBG. The sensor was positioned such that approximately equal lengths of the sensor protruded beyond the 3 mm long top conductor. (b) Electric field distribution around a cross-section of the microstrip. The electric field (black lines) between the TEM cell conductors (gray rectangles) is most concentrated between the two conductors. The density of black lines is intended to qualitatively indicate field strength. The alternating voltage is applied to the top conductor and electrical current flows out of the page. The bottom conductor is grounded. The magnetic field is in the plane of the page.
was found to be dependent on repositioning the sensing region in the transmission line. However, very careful removal and installation of the probe resulted in reproducible microwave energy sensing. We attribute the initial behavior to twist of the sensing region since various electromagnetic properties of the magnetic microwire material are known to be sensitive to twist [23]. Additionally, we recorded the S-parameter S11 to quantify the impedance change of the transmission line due to the presence of each microwave absorber. This data allowed us to compute a
Fig. 3. Comparison of sensor performance with and without magnetic microwire bonded to FBG. Also shown are linear fits to the response data. In general, there was a linear relationship between the microwave power delivered to the transmission line and FBG at all microwave frequencies studied. The response of the bare FBG is most likely due to unintentional heating of the FBG that was close to the TEM cell’s top conductor (see Fig. 2a). The microwave frequency was 7.75 GHz.
28
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Fig. 4. S11 measurements made with the magnetic microwire probe fiber. Shown are the measurements with the microwave transmission line empty, occupied with ordinary SMF28e fiber that still has its factory coating, and occupied with the probe.
Fig. 5. S11 increase in the transmission line due to presence of microwave absorber. The increase is measured by subtracting S11 due to the presence of ordinary (coated) SMF28e fiber by S11 due to the microwave absorber/FBG combination. A larger S11 indicates a greater perturbation of the EM field in the TEM cell.
figure-of-merit (FOM) that characterizes the magnitude of FBG relative to the perturbation of the EM field. The S11 measurements were made with the HP 8703A and the increase of S11 relative to that due to the presence of ordinary SMF28e is used. After calibrating with the transmission cell empty, S11 values were better than −30 dB for the frequency range 130 MHz–12 GHz. Repeatability of the measurements was confirmed by taking repeated scans as the relevant fiber section was inserted and removed from the transmission line. Fig. 4 shows one set of S11 measurements of the magnetic microwire-based probe. Results with the gold probe were similar. Fig. 5 shows an S11 comparison for the probe based on the microwire to the probe on based on gold. From the figure we see that for microwave frequencies less than 11 GHz, the gold
probe usually disturbs the EM field more than the microwire since increase of S11 for gold probe is usually greater than the increase of S11 for the microwire probe. 4. Discussion As expected for a particular microwave frequency, we found that FBG shifted to longer wavelengths when increasing amounts of microwave energy were delivered to the microwave cell. To determine the location of FBG , a Gaussian was fit to the background-subtracted OSA scan. For a fixed microwave frequency, typically there was a linear relationship between FBG and the amount of microwave power delivered to the transmission line.
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
29
Fig. 6. FBG vs. microwave energy density at several microwave frequencies for ferromagnetic microwire probe. In general, there was a linear relationship between FBG and the microwave power delivered to the transmission line. Linear fits to each dataset are shown.
Fig. 7. FOM of the microwire-based probe and the gold-based probe. The best performance of the microwire-based probe relative to the gold-based probe is ∼10 at f = 3.25 GHz. Note that the microwire had poor thermal coupling to the FBG (see Fig. 1) therefore we expect that the ferromagnetic wire compares more favorably than this graph suggests.
Fig. 6 shows the performance of the microwire probe at several microwave frequencies. The figure demonstrates that the slope of FBG vs. microwave energy density is a function of the microwave frequency. In Table 1, we show the response and electrical S11 measurements at several representative microwave frequencies. We find that if we consider only the response of the different sensing materials there is no clear choice of one material over the other. Though the table lists only microwave frequencies where the microwire outperforms gold, we remind the reader of Fig. 5 that indicates
some frequencies where the gold probe perturbs the EM field less than the microwire. To better compare the performance of the probes, we compute a figure-of-merit (FOM) at a particular microwave frequency for each probe in the following way: FOM =
linear slope of FBG vs. microwave power 10(S11/10)
where S11 is the increase of S11 of the TEM cell with the probe relative to S11 of the TEM cell with ordinary optical fiber (see Fig. 5).
30
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Table 1 Performance of gold and ferromagnetic microwire probes at several representative EM frequencies. The Response columns represent the slope of the plot of FBG vs. microwave energy density. The S11 increase columns show the increase of that electrical scattering parameter as the optical fiber is translated such that the sensing portion of the fiber moves into the TEM cell (i.e. the difference between S11 with the microwave absorber and S11 with only the optical fiber in the TEM cell). Microwave freq. (GHz)
Response gold (pm/(mJ/m3 ))
S11 increase gold (dB)
Response microwire (pm/(mJ/m3 ))
S11 increase microwire (dB)
0.5 1 2 3 5 6 7 9.5
31.91 86.15 166.88 322.82 1574.42 1162.39 4500.43 3977.52
7.54 20.80 17.17 16.78 25.95 23.21 24.45 18.92
31.82 136.32 235.12 330.83 1447.19 1326.86 1556.41 2697.43
5.86 18.00 12.90 8.53 21.83 20.15 17.22 15.05
Fig. 8. OSA scans demonstrating chirp induced in the FBG with one of the microwires bonded to it as the microwave power is increased. The traces have been offset vertically for clarity. The wavelength span of the top trace suggests that the temperature varied by about 250 ◦ C across the ∼2.5 cm long FBG.
The results of the FOM characterization of the microwire probe relative to the gold probe are shown in Fig. 7. From the figure we find that at many microwave frequencies the microwires have a larger FOM than the gold-coated FBG. Note that there is a significant difference between the gold-based and microwire-based probe configurations. Recall from Fig. 1 that the gold has much better thermal contact with the FBG’s cladding. Given this difference, we expect the microwires to compare even more favorably to gold if the probes are similarly constructed. With the equipment used for the work presented here, the best sensitivity to EM radiation corresponds to AC electric fields that have average energy densities of approximately 1.3 J/m3 . This value is extracted from the greatest FBG vs. microwave energy density slope and the standard deviation () of nominally identical OSA scans. Specifically, for more than 150 “identical” OSA scans the uncertainty in determining the location of FBG was FBG . We assume that a shift of 6 FBG is needed to reliably determine a shift of the FBG wavelength due to the external EM field. Most data obtained suggest that the FBG temperature was increasing by a maximum of approximately 50 ◦ C (i.e. FBG ∼ = 0.5 nm). One interesting effect that we witnessed under certain conditions was that some of the OSA scans revealed that a strong spatial chirp was being induced in the FBGs. One set of OSA scans demonstrating a chirped FBG is shown in Fig. 8. Given that the FBG spectrum of the top trace of spans approximately 2.5 nm, we conclude that the temperature varied by about 250 ◦ C along the FBG for that scan. Note that the plots in Fig. 8 have been offset
vertically for clarity. Without these offsets, the baselines of the three traces would overlap. 5. Conclusion We have presented an optical probe based on soft ferromagnetic glass-coated microwires that can be used to measure RF and microwave energy. Applications that require a strong response and minimal perturbation of the EM-field would benefit from such a probe. Additionally, many sensing regions could be multiplexed along a single optical fiber using established distributed sensing techniques such as Optical Frequency Domain Reflectometry [24]. Additionally, the microwave resonances of the microwires are tunable and so one can imagine sensors that are even more appropriate at particular microwave frequencies. Finally, recall from Fig. 1d that the microwires are in relatively poor thermal contact with the FBG. We expect the probe to perform even better with improved thermal coupling to the cladding of the FBG. Role of the funding sources Funding for this work was provided by USAMRMC (grant W81XWH-07-1-0708) and the University of South Florida. No funding source was involved in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Acknowledgements The work at USF was supported by the Florida Cluster for Advanced Smart Sensor Technologies (FCASST) and USAMRMC through grant number W81XWH-07-1-0708. Microfir Tehnologii Industriale (www.microwires.com) is acknowledged for providing the soft ferromagnetic glass-coated microwires.
References [1] A.D. Kersey, M.A. Davis, H.J. Patrick, M. LeBlanc, K.P. Koo, C.G. Askins, M.A. Putnam, E.J. Friebele, Fiber grating sensors, J. Lightwave Technol. 15 (8) (1997) 1442–1463. [2] K.T.V. Grattan, T. Sun, Fiber optic sensor technology: an overview, Sens. Actuators A: Phys. 82 (2000) 40–61. [3] A. Othonos, Fiber Bragg gratings, Rev. Sci. Instrum. 68 (12) (1997) 4309–4341. [4] V.M.N. Passaro, F. Dell’Olio, F. De Leonardis, Electromagnetic field photonic sensors, Prog. Quantum Electron. 3 (2006) 45–73. [5] J. Randa, M. Kanda, R.D. Orr, Thermo-optic designs for electromagnetic-field probes for microwaves and millimeter waves, IEEE Trans. Electromagn. Compat. 33 (3) (1991) 205–214. [6] D.G.R. Perry, B. Schreeve, S. Schultz, D. Selfridge, Multi-axial fiber-optic electric field sensor, in: Proc. SPIE 7648, Smart Sensor Phenomena, Technology, Networks, and Systems 2010, San Diego, 2010. [7] S. Chadderdon, R. Gibson, R.H. Selfridge, S.M. Schultz, W.C. Wang, R. Forber, J. Luo, A.K.-Y. Jen, Electric-field sensors utilizing coupling between a D-fiber and an electro-optic polymer slab, Appl. Opt. 50 (20) (2011) 3505–3512. [8] V.M.N. Passaro, Broadband electric field sensors, in: A. Chen, E. Murphy (Eds.), Broadband Optical Modulators: Science, Technology, and Applications, CRC Press, Boca Raton, FL, 2011, pp. 383–398. [9] K.P. Koo, J.G.H. Sigel, An electric field sensor utilizing a piezoelectric polyvinylidene fluoride (PVF2) film in a single-mode fiber interferometer, IEEE J. Quantum Electron. 18 (4) (1982) 670–675.
31
[10] H. Liu, S.W. Or, H.Y. Tam, Magnetostrictive composite-fiber Bragg grating (MCFBG) magnetic field sensor, Sens. Actuators A Phys. 173 (2012) 122–126. [11] K. Yang, L.P.B. Katehi, J.F. Whitaker, Electric field mapping system using an optical-fiber-based electrooptic probe, IEEE Microw. Wireless Compon. Lett. 11 (4) (2001) 164–166. [12] M. Ando, M. Yamauchi, K. Nakayama, K. Moriyama, K. Fujita, Y. Masuda, M. Kimura, Y. Mizutani, S. Kimura, T. Yokouchi, Y. Suzaki, K. Nakagawa, S. Ejima, Dependence of fiber Bragg grating characteristics on its length, Jpn. J. Appl. Phys. 43 (7A) (2004) 4234–4235. [13] V.S. Larin, A.V. Torcunov, A. Zhukov, J. González, M. Vazquez, L. Panina, Preparation and properties of glass-coated microwires, J. Magn. Magn. Mater. 2 (2002) 39–45. [14] M.-H. Phan, H.-X. Peng, Giant magnetoimpedance materials: fundamentals and applications, Prog. Mater. Sci. 53 (2) (2008) 323–420. [15] M. Vázquez, Advanced magnetic microwires, in: Handbook of Magnetism and Advanced Magnetic Materials, John Wiley & Sons, Ltd., Hoboken, NJ, USA, 2007. [16] V. Zhukova, M. Ipatov, A. Zhukov, Thin magnetically soft wires for magnetic microsensors, Sensors 9 (11) (2009) 9216–9240. [17] H. García-Miquel, M.J. Esbrí, J.M. Andrés, J.M. García, J.M. García-Beneytez, M. Vázquez, Power absorption and ferromagnetic resonance in Co-rich metallic glasses, IEEE Trans. Magn. 37 (1) (2001) 561–564. [18] H.X. Peng, F.X. Qin, M.H. Phan, J. Tang, L.V. Panina, M. Ipatov, V. Zhukova, A. Zhukov, J. Gonzalez, Co-based magnetic microwire and field-tunable multifunctional macro-composites, J. Non-Cryst. Solids 355 (24–27) (2009) 1380–1386. [19] F.X. Qin, N. Pankratov, H.X. Peng, M.H. Phan, L.V. Panina, M. Ipatov, V. Zhukova, A. Zhukov, J. Gonzalez, Novel magnetic microwires-embedded composites for structural health monitoring applications, J. Appl. Phys. 107 (9) (2010) 09A314. [20] D.D. Grieg, H.F. Engelmann, Microstrip – a new transmission technique for the Klilomegacycle range, Proc. IRE 40 (12) (1952) 1644–1650. [21] F. Assadourian, E. Rimai, Simplified theory of microstrip transmission systems, Proc. IRE 40 (12) (1952) 1651–1657. [22] J. Kostriza, Microstrip components, Proc. IRE 40 (12) (1952) 1658–1663. [23] F. Qin, H.-X. Peng, Ferromagnetic microwires enabled multifunctional composite materials, Prog. Mater. Sci. 58 (2013) 183–259. [24] B.J. Soller, D.K. Gifford, M.S. Wolfe, M.E. Froggatt, High resolution optical frequency domain reflectometry for characterization of components and assemblies, Opt. Express 13 (2) (2005) 666–674.
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
Sensing RF and microwave energy with fiber Bragg grating heating via soft ferromagnetic glass-coated microwires P. Colosimo a,∗ , A. Chen a,b , J. Devkota b , H. Srikanth b , M.H. Phan b,∗∗ a b
Applied Physics Laboratory, University of Washington, Seattle, WA 98105, USA Department of Physics, University of South Florida, Tampa, FL 33620, USA
a r t i c l e
i n f o
Article history: Received 30 July 2013 Received in revised form 31 December 2013 Accepted 25 January 2014 Available online 1 February 2014 Keywords: Fiber optic Sensor Magnetic Microwire
a b s t r a c t We present results from a fiber Bragg grating-based microwave energy sensor. The sensor relies on a soft ferromagnetic glass-coated microwire that is bonded to the cladding of the grating. The microwire absorbs microwave energy and heats up thus raising the temperature of the fiber Bragg grating. Compared to a similar sensor that uses gold to absorb electromagnetic radiation, the microwire yields a sensor with greater sensitivity (∼10 times at f = 3.25 GHz) relative to the perturbation of the microwave field. With the sensor reported here, the best sensitivity to electromagnetic radiation corresponds to AC electric fields that have an average electric energy density of approximately 1.3 J/m3 . © 2014 Elsevier B.V. All rights reserved.
1. Introduction The fiber Bragg grating (FBG) is the basis of numerous sensors [1–3]. For the most part, strain and temperature are the primary environmental parameters that can be detected with FBGs. Other variables can be measured by using a probe design that converts the desired variable to a strain or temperature change. For example, an FBG bonded to the wall of a vacuum chamber might be used to measure pressure if the wall strain vs. pressure calibration were known. We present results from a new type of microwave energy sensor that relies on Joule heating of a soft ferromagnetic glasscoated microwire to change the temperature of an FBG. Using optical fibers to sense electromagnetic (EM) fields is a promising approach and has been the subject of extensive research. A small sampling of the many published designs rely on surrounding the FBG core with polymer-dispersed liquid crystals [4], fluorescence from a phosphor that has a temperature-dependent lifetime [5], and resonant coupling of light between the fiber’s core and a nonlinear electro-optical slab waveguide [6,7]. Many more complex EM field sensors are reviewed in [8].
∗ Corresponding author at: Applied Physics Laboratory, University of Washington, 1013 NE 40th Street, Box 355640, Seattle, WA 98105-6698, USA. Tel.: +1 206 543 3148; fax: +1 206 543 6785. ∗∗ Corresponding author. E-mail addresses: [email protected] (P. Colosimo), [email protected] (M.H. Phan). 0924-4247/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.sna.2014.01.038
Previous EM field sensors based on fiber optic technology are rather large (cm dimensions [9,10]) and/or point sensors [5,11]. To the best of our knowledge, previous reports of fiber optic schemes to detect EM fields have not reported the degree to which those probes perturbed the field being measured. The sensor presented here has a greater sensitivity compared to one that uses gold to absorb electromagnetic radiation. It can be deployed as a distributed sensor, and often only minimally perturbs the field being measured (i.e. produced a small increase of S11 in the test waveguide). Given that FBGs of very short length (0.35 mm) can be made [12], there is potential to construct very small probes based on this design.
2. Device theory and fabrication An FBG is a periodic variation of the index of refraction of an optical fiber. Wavelengths of light at and near the Bragg condition will be reflected while other wavelengths will pass through the grating with their amplitudes virtually unaffected. For a fiber grating designed to reflect a single wavelength, the Bragg condition can be written FBG = 2n where FBG is the Bragg wavelength, n is the effective index of refraction of the core, and is the grating pitch. As the temperature of an FBG changes, the center wavelength of the grating will shift according to the formula [1]:
FBG = 2n ˛ +
(dn/dT ) T n
26
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
where ˛ is the coefficient of thermal expansion, n is the effective index of refraction, and T is the temperature change. This wavelength shift is exploited to use an FBG as a temperature sensor. For wavelengths around 1550 nm and optical fiber (Corning SMF-28) used here, the conversion from FBG to temperature is well-known and about 10 pm/◦ C [1]. By monitoring the transmitted or reflected optical spectrum, we can determine the average temperature of the FBG. To create the probes, the microwave-absorbing soft ferromagnetic glass-coated microwires were glued to the cladding of a commercial FBG. For the probes that use gold to absorb microwave energy, approximately 120 nm of the metal was sputtered onto the FBG’s cladding. Any temperature increase of the microwire or gold due to Joule heating was transmitted to the FBG and appeared as a shift of the notch in the FBG’s transmission spectrum. Soft ferromagnetic amorphous glass-coated microwires of Fe4.97 Co64.63 B16 Si11 Cr3.4 Ni0.02 were prepared by a modified TaylorUlitovsky process [13]. The metallic diameter of the wire was about 14 m surrounded by about 3 m thick glass layer. The wires used were approximately 3 cm long. This length was chosen to make each microwire slightly longer than the length of the FBG. This alloy composition has a vanishing magnetostriction constant ( ∼ 0) that
determines its soft ferromagnetic characteristic. These microwires show a giant magneto-impedance effect making them attractive for use in magnetic field sensors [14–16]. Since the microwires also show excellent microwave absorption properties [17], they have recently been exploited for making novel microwire-based composites for structural health monitoring and self-sensing applications [18,19]. In this study, we demonstrate that using these microwires as a microwave absorber it is possible to fabricate a fiber Bragg grating-based microwave energy sensor with improved sensitivity and less perturbation of the microwave field. 3. Experiments and results Light from a broadband amplified spontaneous emission source (JDSU M/N BBS1560+1FP) was launched into the optical fiber that contained the FBG. Light transmitted through the optical fiber was monitored with an optical spectrum analyzer (HP M/N 70951B). The sensor was placed into a homemade 50 microstrip transmission line. A microstrip transmission line produces a quasi-Transverse Electromagnetic (TEM) mode that we treat as a pure TEM mode here. The microstrip design was chosen to facilitate insertion and removal of the probe. We will refer to the microstrip transmission
Fig. 1. Diagram of experimental setup. (a) Setup used for sensor response data. Optical (microwave) components and paths are shown in gray (black). Abbreviations are as follows: ASE – amplified spontaneous emission; TEM cell – 50 microstrip transmission line; OSA – optical spectrum analyzer; D.C. – directional coupler. (b) Setup used for S11 measurements. The difference between S11 with only ordinary fiber in the TEM cell and S11 with the sensor in the TEM cell was used to quantify the S11 increase due to the sensor. (c) Cross-section of gold-based probe (not to scale). (d) Cross-section of microwire-based probe (not to scale).
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
line as “TEM cell” here. The TEM cell comprises two parallel copper plates with the geometry needed to produce 50 impedance. Electromagnetic energy propagates down the length of the TEM cell, the electric field is mostly confined to the volume between the metal plates, and the magnetic field is perpendicular to the length of the TEM cell, which is strongest between the two plates. Our configuration results in a magnetic field that is predominantly parallel to the axis of the optical fiber probe. For more details on the microstrip transmission line, see [20–22]. As microwave energy of different frequencies and powers was sent through the transmission line, the optical spectrum transmitted by the FBG was recorded. After a step change of the microwave frequency or power, the OSA signal took up to 1–2 s to stabilize. The microwaves were generated by an HP M/N 8703A, amplified by either a Mini-Circuits M/N ZHL-42W or Avantek M/N APT-10555, and monitored with an HP M/N E4419B. A schematic diagram of the experimental set-up for response measurements is given in Fig. 1a. To record S11 due to each different probe, the setup shown in Fig. 1b was used. The HP M/N 8703A was connected to the TEM cell via a coaxial cable which was terminated with a 50 load. S11 due to a section or ordinary optical fiber was recorded as a reference scan. This reference scan was subtracted from an S11 scan taken with the sensor in the TEM cell. For the S11 scans, care was taken to not disturb the setup while the optical fiber was being moved. The remainder of Fig. 1 shows how the various microwave absorbers were bonded to the FBG. Fig. 2 illustrates the geometry of the sensor within the TEM cell as well as the electric field distribution of the TEM cell (at one instant in time). The shift of FBG as a function of average electric energy density at one microwave frequency is shown in Fig. 3 (open triangles). The data clearly indicate that FBG increases as the amount of microwave energy in the transmission line increases. Our most sensitive results (able to detect average electric energy density of 1.3 J/m3 RMS, at 10.25 GHz) correspond to measuring a temperature change of approximately 0.22 ◦ C. Data was taken with a bare FBG in the transmission line to determine whether or not FBG was influenced by factors other than the microwire (filled diamonds in Fig. 3). We found a small response to microwave energy with the bare FBG. This response of the bare FBG was most likely due to heating of the transmission line being coupled to the FBG. During our early studies, the response of the microwire probe
a
27
Top conductor
V Bottom conductor Sensor
b
Fig. 2. Details of TEM cell. (a) The sensor was perpendicular to the length of the top and bottom conductors of the TEM cell. The 3 cm long microwire was centered on the 2.5 cm long FBG. The sensor was positioned such that approximately equal lengths of the sensor protruded beyond the 3 mm long top conductor. (b) Electric field distribution around a cross-section of the microstrip. The electric field (black lines) between the TEM cell conductors (gray rectangles) is most concentrated between the two conductors. The density of black lines is intended to qualitatively indicate field strength. The alternating voltage is applied to the top conductor and electrical current flows out of the page. The bottom conductor is grounded. The magnetic field is in the plane of the page.
was found to be dependent on repositioning the sensing region in the transmission line. However, very careful removal and installation of the probe resulted in reproducible microwave energy sensing. We attribute the initial behavior to twist of the sensing region since various electromagnetic properties of the magnetic microwire material are known to be sensitive to twist [23]. Additionally, we recorded the S-parameter S11 to quantify the impedance change of the transmission line due to the presence of each microwave absorber. This data allowed us to compute a
Fig. 3. Comparison of sensor performance with and without magnetic microwire bonded to FBG. Also shown are linear fits to the response data. In general, there was a linear relationship between the microwave power delivered to the transmission line and FBG at all microwave frequencies studied. The response of the bare FBG is most likely due to unintentional heating of the FBG that was close to the TEM cell’s top conductor (see Fig. 2a). The microwave frequency was 7.75 GHz.
28
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Fig. 4. S11 measurements made with the magnetic microwire probe fiber. Shown are the measurements with the microwave transmission line empty, occupied with ordinary SMF28e fiber that still has its factory coating, and occupied with the probe.
Fig. 5. S11 increase in the transmission line due to presence of microwave absorber. The increase is measured by subtracting S11 due to the presence of ordinary (coated) SMF28e fiber by S11 due to the microwave absorber/FBG combination. A larger S11 indicates a greater perturbation of the EM field in the TEM cell.
figure-of-merit (FOM) that characterizes the magnitude of FBG relative to the perturbation of the EM field. The S11 measurements were made with the HP 8703A and the increase of S11 relative to that due to the presence of ordinary SMF28e is used. After calibrating with the transmission cell empty, S11 values were better than −30 dB for the frequency range 130 MHz–12 GHz. Repeatability of the measurements was confirmed by taking repeated scans as the relevant fiber section was inserted and removed from the transmission line. Fig. 4 shows one set of S11 measurements of the magnetic microwire-based probe. Results with the gold probe were similar. Fig. 5 shows an S11 comparison for the probe based on the microwire to the probe on based on gold. From the figure we see that for microwave frequencies less than 11 GHz, the gold
probe usually disturbs the EM field more than the microwire since increase of S11 for gold probe is usually greater than the increase of S11 for the microwire probe. 4. Discussion As expected for a particular microwave frequency, we found that FBG shifted to longer wavelengths when increasing amounts of microwave energy were delivered to the microwave cell. To determine the location of FBG , a Gaussian was fit to the background-subtracted OSA scan. For a fixed microwave frequency, typically there was a linear relationship between FBG and the amount of microwave power delivered to the transmission line.
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
29
Fig. 6. FBG vs. microwave energy density at several microwave frequencies for ferromagnetic microwire probe. In general, there was a linear relationship between FBG and the microwave power delivered to the transmission line. Linear fits to each dataset are shown.
Fig. 7. FOM of the microwire-based probe and the gold-based probe. The best performance of the microwire-based probe relative to the gold-based probe is ∼10 at f = 3.25 GHz. Note that the microwire had poor thermal coupling to the FBG (see Fig. 1) therefore we expect that the ferromagnetic wire compares more favorably than this graph suggests.
Fig. 6 shows the performance of the microwire probe at several microwave frequencies. The figure demonstrates that the slope of FBG vs. microwave energy density is a function of the microwave frequency. In Table 1, we show the response and electrical S11 measurements at several representative microwave frequencies. We find that if we consider only the response of the different sensing materials there is no clear choice of one material over the other. Though the table lists only microwave frequencies where the microwire outperforms gold, we remind the reader of Fig. 5 that indicates
some frequencies where the gold probe perturbs the EM field less than the microwire. To better compare the performance of the probes, we compute a figure-of-merit (FOM) at a particular microwave frequency for each probe in the following way: FOM =
linear slope of FBG vs. microwave power 10(S11/10)
where S11 is the increase of S11 of the TEM cell with the probe relative to S11 of the TEM cell with ordinary optical fiber (see Fig. 5).
30
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Table 1 Performance of gold and ferromagnetic microwire probes at several representative EM frequencies. The Response columns represent the slope of the plot of FBG vs. microwave energy density. The S11 increase columns show the increase of that electrical scattering parameter as the optical fiber is translated such that the sensing portion of the fiber moves into the TEM cell (i.e. the difference between S11 with the microwave absorber and S11 with only the optical fiber in the TEM cell). Microwave freq. (GHz)
Response gold (pm/(mJ/m3 ))
S11 increase gold (dB)
Response microwire (pm/(mJ/m3 ))
S11 increase microwire (dB)
0.5 1 2 3 5 6 7 9.5
31.91 86.15 166.88 322.82 1574.42 1162.39 4500.43 3977.52
7.54 20.80 17.17 16.78 25.95 23.21 24.45 18.92
31.82 136.32 235.12 330.83 1447.19 1326.86 1556.41 2697.43
5.86 18.00 12.90 8.53 21.83 20.15 17.22 15.05
Fig. 8. OSA scans demonstrating chirp induced in the FBG with one of the microwires bonded to it as the microwave power is increased. The traces have been offset vertically for clarity. The wavelength span of the top trace suggests that the temperature varied by about 250 ◦ C across the ∼2.5 cm long FBG.
The results of the FOM characterization of the microwire probe relative to the gold probe are shown in Fig. 7. From the figure we find that at many microwave frequencies the microwires have a larger FOM than the gold-coated FBG. Note that there is a significant difference between the gold-based and microwire-based probe configurations. Recall from Fig. 1 that the gold has much better thermal contact with the FBG’s cladding. Given this difference, we expect the microwires to compare even more favorably to gold if the probes are similarly constructed. With the equipment used for the work presented here, the best sensitivity to EM radiation corresponds to AC electric fields that have average energy densities of approximately 1.3 J/m3 . This value is extracted from the greatest FBG vs. microwave energy density slope and the standard deviation () of nominally identical OSA scans. Specifically, for more than 150 “identical” OSA scans the uncertainty in determining the location of FBG was FBG . We assume that a shift of 6 FBG is needed to reliably determine a shift of the FBG wavelength due to the external EM field. Most data obtained suggest that the FBG temperature was increasing by a maximum of approximately 50 ◦ C (i.e. FBG ∼ = 0.5 nm). One interesting effect that we witnessed under certain conditions was that some of the OSA scans revealed that a strong spatial chirp was being induced in the FBGs. One set of OSA scans demonstrating a chirped FBG is shown in Fig. 8. Given that the FBG spectrum of the top trace of spans approximately 2.5 nm, we conclude that the temperature varied by about 250 ◦ C along the FBG for that scan. Note that the plots in Fig. 8 have been offset
vertically for clarity. Without these offsets, the baselines of the three traces would overlap. 5. Conclusion We have presented an optical probe based on soft ferromagnetic glass-coated microwires that can be used to measure RF and microwave energy. Applications that require a strong response and minimal perturbation of the EM-field would benefit from such a probe. Additionally, many sensing regions could be multiplexed along a single optical fiber using established distributed sensing techniques such as Optical Frequency Domain Reflectometry [24]. Additionally, the microwave resonances of the microwires are tunable and so one can imagine sensors that are even more appropriate at particular microwave frequencies. Finally, recall from Fig. 1d that the microwires are in relatively poor thermal contact with the FBG. We expect the probe to perform even better with improved thermal coupling to the cladding of the FBG. Role of the funding sources Funding for this work was provided by USAMRMC (grant W81XWH-07-1-0708) and the University of South Florida. No funding source was involved in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
P. Colosimo et al. / Sensors and Actuators A 210 (2014) 25–31
Acknowledgements The work at USF was supported by the Florida Cluster for Advanced Smart Sensor Technologies (FCASST) and USAMRMC through grant number W81XWH-07-1-0708. Microfir Tehnologii Industriale (www.microwires.com) is acknowledged for providing the soft ferromagnetic glass-coated microwires.
References [1] A.D. Kersey, M.A. Davis, H.J. Patrick, M. LeBlanc, K.P. Koo, C.G. Askins, M.A. Putnam, E.J. Friebele, Fiber grating sensors, J. Lightwave Technol. 15 (8) (1997) 1442–1463. [2] K.T.V. Grattan, T. Sun, Fiber optic sensor technology: an overview, Sens. Actuators A: Phys. 82 (2000) 40–61. [3] A. Othonos, Fiber Bragg gratings, Rev. Sci. Instrum. 68 (12) (1997) 4309–4341. [4] V.M.N. Passaro, F. Dell’Olio, F. De Leonardis, Electromagnetic field photonic sensors, Prog. Quantum Electron. 3 (2006) 45–73. [5] J. Randa, M. Kanda, R.D. Orr, Thermo-optic designs for electromagnetic-field probes for microwaves and millimeter waves, IEEE Trans. Electromagn. Compat. 33 (3) (1991) 205–214. [6] D.G.R. Perry, B. Schreeve, S. Schultz, D. Selfridge, Multi-axial fiber-optic electric field sensor, in: Proc. SPIE 7648, Smart Sensor Phenomena, Technology, Networks, and Systems 2010, San Diego, 2010. [7] S. Chadderdon, R. Gibson, R.H. Selfridge, S.M. Schultz, W.C. Wang, R. Forber, J. Luo, A.K.-Y. Jen, Electric-field sensors utilizing coupling between a D-fiber and an electro-optic polymer slab, Appl. Opt. 50 (20) (2011) 3505–3512. [8] V.M.N. Passaro, Broadband electric field sensors, in: A. Chen, E. Murphy (Eds.), Broadband Optical Modulators: Science, Technology, and Applications, CRC Press, Boca Raton, FL, 2011, pp. 383–398. [9] K.P. Koo, J.G.H. Sigel, An electric field sensor utilizing a piezoelectric polyvinylidene fluoride (PVF2) film in a single-mode fiber interferometer, IEEE J. Quantum Electron. 18 (4) (1982) 670–675.
31
[10] H. Liu, S.W. Or, H.Y. Tam, Magnetostrictive composite-fiber Bragg grating (MCFBG) magnetic field sensor, Sens. Actuators A Phys. 173 (2012) 122–126. [11] K. Yang, L.P.B. Katehi, J.F. Whitaker, Electric field mapping system using an optical-fiber-based electrooptic probe, IEEE Microw. Wireless Compon. Lett. 11 (4) (2001) 164–166. [12] M. Ando, M. Yamauchi, K. Nakayama, K. Moriyama, K. Fujita, Y. Masuda, M. Kimura, Y. Mizutani, S. Kimura, T. Yokouchi, Y. Suzaki, K. Nakagawa, S. Ejima, Dependence of fiber Bragg grating characteristics on its length, Jpn. J. Appl. Phys. 43 (7A) (2004) 4234–4235. [13] V.S. Larin, A.V. Torcunov, A. Zhukov, J. González, M. Vazquez, L. Panina, Preparation and properties of glass-coated microwires, J. Magn. Magn. Mater. 2 (2002) 39–45. [14] M.-H. Phan, H.-X. Peng, Giant magnetoimpedance materials: fundamentals and applications, Prog. Mater. Sci. 53 (2) (2008) 323–420. [15] M. Vázquez, Advanced magnetic microwires, in: Handbook of Magnetism and Advanced Magnetic Materials, John Wiley & Sons, Ltd., Hoboken, NJ, USA, 2007. [16] V. Zhukova, M. Ipatov, A. Zhukov, Thin magnetically soft wires for magnetic microsensors, Sensors 9 (11) (2009) 9216–9240. [17] H. García-Miquel, M.J. Esbrí, J.M. Andrés, J.M. García, J.M. García-Beneytez, M. Vázquez, Power absorption and ferromagnetic resonance in Co-rich metallic glasses, IEEE Trans. Magn. 37 (1) (2001) 561–564. [18] H.X. Peng, F.X. Qin, M.H. Phan, J. Tang, L.V. Panina, M. Ipatov, V. Zhukova, A. Zhukov, J. Gonzalez, Co-based magnetic microwire and field-tunable multifunctional macro-composites, J. Non-Cryst. Solids 355 (24–27) (2009) 1380–1386. [19] F.X. Qin, N. Pankratov, H.X. Peng, M.H. Phan, L.V. Panina, M. Ipatov, V. Zhukova, A. Zhukov, J. Gonzalez, Novel magnetic microwires-embedded composites for structural health monitoring applications, J. Appl. Phys. 107 (9) (2010) 09A314. [20] D.D. Grieg, H.F. Engelmann, Microstrip – a new transmission technique for the Klilomegacycle range, Proc. IRE 40 (12) (1952) 1644–1650. [21] F. Assadourian, E. Rimai, Simplified theory of microstrip transmission systems, Proc. IRE 40 (12) (1952) 1651–1657. [22] J. Kostriza, Microstrip components, Proc. IRE 40 (12) (1952) 1658–1663. [23] F. Qin, H.-X. Peng, Ferromagnetic microwires enabled multifunctional composite materials, Prog. Mater. Sci. 58 (2013) 183–259. [24] B.J. Soller, D.K. Gifford, M.S. Wolfe, M.E. Froggatt, High resolution optical frequency domain reflectometry for characterization of components and assemblies, Opt. Express 13 (2) (2005) 666–674.