Invited Paper

Pressure measurement with fiber-optic sensors:

Commercial technologies and applications
ric Pinet
FISO Technologies Inc., 500-195, Ave. St-Jean-Baptiste, Qubec (Qc) Canada G2E 5R9 Tel: (+1) 418-688-8065; Fax: (+1) 418-688-8067; www.fiso.com; eric.pinet@fiso.com

Abstract: Mainly three technologies are presently commercially available for pressure measurement with
fiber-optic sensors: intensity-based, fiber Bragg gratings and Fabry-Prot. The first one is probably the simplest and the cheapest but it is limited to applications where having 2 fixed or up to 4 flexible fibers is not an issue, whereas the two other technologies require only one fiber. With generally low sensitivity to pressure and prohibitive cost for non multiplexed measurements, fiber Bragg grating pressure sensors are still limited to marginal applications. Fabry-Prot technology is the best compromise offering at affordable price a great flexibility in terms of pressure ranges, high sensitivity and miniature size suitable for most applications including disposable medical devices.
Keywords: Fiber optics sensors; Sensors; Pressure; Intensity-based; Fiber Bragg grating; Fabry-Prot

1. Introduction
Pressure can be measured over a very broad range from 10-12 Pa (in extreme vacuum) up to 10+12 Pa (in explosions) and a very large number of electrical sensor types have been commercialized so far. To convert pressure induced mechanical deformation into an electrical digital signal, such sensor usually rely on different physical principles such as piezoresistive or piezoelectrical principles, inductive or capacitive principles, or even thermo-electrical or acoustic principles. All of these principles have a large number of technological limitations which restrict their full utilization for sensing pressure in environments with important electromagnetic interferences (EMI), elevated temperatures, harsh chemicals or explosion sensitivity. Most of those constraints are actually circumvolved by optical fiber sensors (OFS) which often offer also better long-term reliability and sometimes with the benefit of a smaller size. Being however slightly more expensive than conventional technologies based on electrical and mechanical principles, OFS are usually reserved to niche critical applications, where they remain the best choice. As a consequence, only the leaders of the OFS companies are presently successful to penetrate this exclusive market Despite challenges associated with the reality of those markets, an increasing number of OFS starts to become commercial success stories. Although temperature OFS are certainly the most widely spread in the industry with well-known applications ranging from industrial process control, energy, civil engineering to medical [1, 2], pressure measurement is probably the second most important physical parameter that is successfully addressed with OFS. Actually pressure OFS have gained new market shares only in the last past five years as new applications involving mainly medical disposable sensors are now commercialized. This paper will present the three main pressure OFS technologies that are commercially available as well as some associated applications.

2. Intensity-based pressure sensors

Intensity-based pressure sensors were probably one of the first pressure OFS to be commercialized. The principle behind such sensors is quite simple: the light emitted by a multimode optical fiber tip is collected by another after reflecting on a diaphragm deflecting with pressure. The intensity of the collected light is dependent on the core diameter and position of the optical fibers as well as on their numerical aperture, but it is also directly related to the distance separating the fiber ends and the reflecting diaphragm, allowing direct pressure measurement. The light intensity increases rapidly from zero (when the diaphragm is in contact with the fiber-ends plane) to reach a maximum when the diaphragm plane matches the position where the solid angles of numerical aperture of both fibers overlap the most. As the diaphragm plane moves further away, less light could be collected and the light intensity decreases slowly back to zero. By properly designing the sensor for a selected pressure range providing that only a monotone region of the light intensity curve is used, a unique pressure could be derived from light intensity once the sensor has been calibrated. The main advantage of this technology resides in the fact that it does not require expensive optical devices to operate. Simple light source such as cheap light emitting diode (LED) and light detector such as photodiode are sufficient to inject and collect light into pig-tailed large core multimode fibers. Even the optical connector usually used in most OFS technologies could be avoided since all the inexpensive optical components could be integrated into the sensor element, allowing a more convenient electrical interface with the data processing unit.

21st International Conference on Optical Fiber Sensors, edited by Wojtek J. Bock, Jacques Albert, Xiaoyi Bao, Proc. of SPIE Vol. 7753, 775304 2011 SPIE CCC code: 0277-786X/11/$18 doi: 10.1117/12.895536

Proc. of SPIE Vol. 7753 775304-1

Although it seems that such technology could have the best of both worlds (advantages of optical sensing combined with ease of electrical connection), it is not without a high price to pay due to the intrinsic nature of the measurement itself: light intensity is always dependent on a lot of factors which are often hard to control. For instance, the measured intensity is directly linked to the light source which could fluctuate due to aging or temperature changes and the same apply also for the light detector. Changes in the reflectivity of the diaphragm mirror due to thermal effects or oxidation affects also the sensor long term reliability, especially if the sensor is subjected to high temperatures (>150C). The thermal expansion of the diaphragm, which could be made of metal or polymers depending on the application, could also induce unexpected apparent pressure shift, furthermore that positioning and orientation of the two fibers relative to the diaphragm could also be affected by temperature changes. The design and the encapsulation of the flexible membrane is therefore often critical and probably the biggest challenges to produce reliable sensors using this technology. This is particularly true if the sensor is used in a harsh environment such as for instance the one encountered in combustion engine chambers, where direct pressure measurement could help to improve the energetic yield by better sparkplug ignition timing capacity. In such harsh chemical environment, the sensor may be subjected to pressures up to 30 MPa with instantaneous gas temperature of about 1 500C and continuous temperature up to 300C. Besides long term (>10 years) reliability requirements with hundreds of millions cycles, the highest pressure on such application is probably the price: each cent decrease is significant for the automotive industry. This hurdle truly complicates the market penetration of this OFS technology for which a target price should be bellow 10$... Another source of intensity unexpected changes is directly linked to fiber losses due for instance to fiber bending, especially when large core fibers are used as often the case in order to maximize light intensity output. Such major drawback could however be avoided by fixing the optical fibers pathway (which is possible for instance in the above cited application) but such option is not always acceptable. For instance in medical instrumented catheters, the fibers have to be flexible and prediction of the intensity losses due to tortuous path is not possible. To solve this problem, one or two additional non-sensing fibers have to be added next to the other fibers to estimate the bending losses [3, 4]. Although theoretically not perfect due to local micro-bending issues, such approach is practically acceptable for adequate compensation, such the case for intensity-based sensors used for intra-cranial pressure (ICP) monitoring. This important application probably still represents the highest market for pressure OFS in terms of number of units sold per year. Since the brain is contained in the skull, a rigid container, any liquid accumulation such as blood or cephalospinal fluid (CSF) or mass lesions such as tumors, pus or hematoma may increase ICP. High ICP is a common cause of death in neurological patients and sustained high ICP suggests poor prognosis [5]. Forty percent of patients admitted unconscious have high ICP. In this group, high ICP will be the leading cause of death in half of cases [6] and effective treatment of high ICP was proven to reduce mortality [7]. As ICP varies continuously and especially at high ICP, a single measurement may be misleading and therefore, a continuous record of the ICP wave is necessary to avoid missing a sudden rise in ICP [5].

3. Fiber Bragg grating pressure sensors

A lot of research is done on fiber Bragg grating (FBG) sensors, especially concerning pressure measurement, but since FBG intrinsic pressure sensitivity in not very high, those sensors are always designed to amplify the pressure measurement indirectly by sensing the strain instead. Two approaches are commonly used: one consists of attaching the FBG fiber to a flexible diaphragm either orthogonally or in the diaphragm plane in areas where the strain is maximal. In both cases, such designs always imply bulky sensors, often limited to high pressure ranges which are however acceptable for applications in civil engineering or in the oil and gas industry [8] where sensor size is not a real issue. Another interesting approach consists of mounting the FBG sensor in cylindrical assemblies so that increased pressure sensitivity is achieved though mechanical amplification schemes. Many designs are proposed in the literature with variations in coatings and assembly. They are always compromising size and sensitivity to achieve sensor outer diameters (typically in the range of 1 mm or less) usually much smaller than the first approach. Since the length of the FBG itself is generally in the 5-10 mm range, encapsulated FBG pressure sensors are not really suitable for true point-sensing pressure in very small regions. Also the lack of very high sensitivity to hydrostatic pressure of such sensors limits their use in most applications requiring better performances. However some interesting ones involving FBG sensors could be found in the biomedical field, mostly related to evaluation of pressure in rigid structures involving bones [9, 10] or dental implants [11]. For instance, inter-vertebral disc pressure could be measured with such technology [12, 13]. One important difficulty related to the FBG sensing technology relies in the fact that the sensing grating is about equally sensitive to strain (measuring indirectly the pressure) than to temperature. In applications where temperature is not constant and could vary a lot, this represents a practical problem that has to be addressed either by complex athermal encapsulation or by complicated temperature compensation strategies. Although performances of FBG pressure sensors could be acceptable for many applications, the fact that most of the time only one pressure measurement is done on single fiber actually limits the potential of such technology. Compared to other pressure OFS technologies, the cost of the FBG interrogator is still too high, even though it reduced significantly over the past decade. Such technology performs well, on a cost-per-measurement basis, only when several measurements are multiplexed on the same fiber. Although some spatially distributed pressure sensing using FBG have been already proposed [14], no high-volume potential application has reached commercial maturity so far.

Proc. of SPIE Vol. 7753 775304-2

4. Fabry y-Prot press sure sensors

From the c commercial po of view, on of the best O technologies for pressure measurement is definitely the one involving oint ne OFS g Fabry-Pro (F-P) point s ot sensors. Beside offering also other multiple parameters me es easurements [15 (such as tem 5] mperature, strain n, displacement, refractive i index) it allo the highest design flexibility as far as the sizes and the p ows e pressure ranges are concerned s d. es F-P nsors, a reflectiv membrane is assembled above a vacuume cavity with a semi-reflective ve s ed In all case of extrinsic F pressure sen layer at its bottom formin a F-P cavity whose length is changing w pressure fle s ng y with exing the memb brane. The interference pattern n created by the F-P cavity could be used to measure pr y y d recisely the dia aphragm deflect tion and thus th pressure cha he anges. The ligh ht used for F-P cavity inter F rrogation is car rried on by an optical fiber (e either single or multimode) linking the inter r rrogator and the pressure se ensor. Differ F-P interro rent ogation techniqu are commer ues rcially available and differ ma ainly on the ligh source: single wavelength o ht or narrow-ba light source will give a periodic patter shifting wit pressure. Alt and es rn th though applica able only for re elative pressure measurem ments (since the sensor has to be referenced each time it is optically c e o t connected to th interrogator this patented he r), d interrogati technique [16] commercia ion alized by FISO Technologies, is best fitted fo monitoring v or very fast occurr ring events such h as explosio or blasts [1 18]. Thanks to the use of e ons 17, s efficient light so ource and detec ctor, the acquis sition rate is typ pically 0.2 MHz z but it could be extended t 2 MHz if dat post-processing is possible. to ta broadband ligh source, such as white-ligh is used, the periodic patte changes as the interferen at differen ht h ht ern s nce nt If a b wavelengt become rapi ths idly destructive except for the zero order wh e, here all wavele engths are in ph hase. By using a Fizeau wedge that create a linear varia es ation of thickne esses, a cross-correlated interf ference pattern could be gener rated in the F-P interrogator to P o simplify d processing since the maxim data mum intensity peak position t then correspond to the exact p ds position where the optical path h difference equals the one created at the F-P cavity. Thi patented field e is d-proven techn nology commerc cialized by FIS Technologie SO es 4 ntage of providi an absolute measurement o the F-P cavity length with s ing of sub-nanometer range precision n since 1994 has the advan over sever decades of m ral micrometer spa thus giving a very interestin dynamic ran With interr an, ng nge. rogators that co ould be as smal ll as the size of a <1 inch th credit card with OEM com e hick d mpetitive price, have an acquis , sition rate up to 15 kHz or hav the capability o ve y to offer m multiple channe such techn els, nology addresse a broad var es riety of applica ations and be m more easily an transparently nd y integrated into more com mplex systems re equiring OFS pr ressure measurement. design of the ex xtrinsic F-P sen nsor strongly in nfluences its ap pplicability in ta argeted applica ations. For civil engineering o l or The d other indu ustrial applicatio ons, the memb brane could be machined from metal to crea bulky and r m ate rough sensors u usually for high h pressure ra anges. Other m materials for mem mbrane design such as sapphir are also inter re resting to impro the reliabili of the senso ove ity or to very high temperature such as the ones in many o and gas we applications. Although the size of the pre es oil ell . essure sensor is ot m ds asily accommod 1-2 inches outer diameter sensor, smalle date r er usually no a real problem for those kind of applications that could ea pressure sensor with equ uivalent or bette performances opens new op er s pportunities. Fo instance the new miniature F-P piezomete or er ed mented geo-texti or inserted without problem ile w m which has been recently proposed [19] could be more easily integrate into instrum small diameter t tube temporary wells used trad y ditionally for w water level asses ssment during soil compaction monitoring. n into very s

550 m m 260 m 125 m m

Fi igure 1: Example of F-P miniatur pressure sensor commercialize by FISO Techn es re rs ed nologies. Left: FOP-MIV ( 550 m); Cente FOP-M260 ( 260 m); Right FOP-F125 ( 125 m) L er: t:

The tr rend toward fur rther reduction of sensor size is particularly t true for medica applications. The added valu of integrating al ue g miniature pressure OFS into instrument catheters is now clearly a ted s accepted in field like cardiolo where the sensor provides ds ogy ssure measurem ments. Such app plications involving disposabl medical devi le ices, like for in nstance in intra aprecise in situ blood pres loon pumping t therapy [20] (I IABP) or in fra actional flow re eserve (FFR) e evaluation, requ low-cost still high-quality uire y aortic ball sensors m manufacturing ca apabilities. Suc challenges h ch have been solve by FISO Technologies whi now comm ed ich mercializes high hvolume of F-P pressure sensors for sev f veral OEM customers. Those sensitive senso (with flexib diaphragm made from thin ors ble n silicon) ar mass-produce on wafers u re ed using well-contr rolled photolith hographic techn nologies and ind dividually assem mbled at the tip p

Proc. of SPIE Vol. 7753 775304-3

of multimode optical fibers using fully automated assembly lines. As presented in Figure 1, several models with decreasing diameters from 550 m (FOP-MIV) down to 125 m (FOP-F125) are commercially available. For this FOP-F125 world smallest pressure sensor pushing further the size limits, the membrane is made from glass, which is more flexible than silicon, and assembled over a cavity carved at the tip of the optical fiber using patented technologies [21-24].

5. Conclusion
Several technologies are available commercially for pressure OFS. Each one has its own advantages and drawbacks as presented in this paper, but each one has found also some niche applications where it seems to be the best solution. However real commercial success for such sensor will come from high-volume repeated business and so far only medical applications involving disposable medical devices seems to offer such opportunity. For this specific market, F-P seems to be best positioned since it offers, at a price compatible with the market, great sensitivity sensors with miniaturization appropriate for simplified integration.

6. References
[1] [2] [3] [4] [5] [6] [7] [8] [9] Pinet ., Hamel C., Glii B., Inaudi D. & Miron N., Health monitoring with optical fiber sensors: from human body to civil structures in Health Monitoring of Structural and Biological Systems, vol. 6532 of Proc. SPIE, pp. 653219-1 to 653219-12 (2007) Pinet ., Ellyson S. & Borne F. Temperature fiber-optic point sensors: commercial technologies and industrial applications Proc. 46th Int. Conf. Microelectron. Devices Mat. (MIDEM 2010), pp. 31-43, ISBN 978-961-9233-0-0 (2010) Knute W.L. Fiber optic measurement system having a reference conductor for controlling the energy level of the light source, US patent #5,065,010 (1991) Knute W.L. & Bailey W.H. Fiber-optic transducer apparatus, US patent #5,107,847 (1992) Reilly P. & Bullock R. in Head Injury, Pathophysiology and Management (Reilly P. & Bullock R. Eds.) ISBN10 0-340-80724-5 (2005) Miller J. D., Becker D.P., Ward J.D., Sullivan H.G., Adams W.E. & Rosner M.J. Significance of intracranial hypertension in severe head injury, J. Neurosurg., Vol. 47 (4), pp 503516 (1977) Marshall L. F., Smith R.W. & Shapiro H.M. The outcome with aggressive treatment in severe head injuries. Part II: Acute and chronic barbiturate administration in the management of head injury J. Neurosurg., Vol. 50 (1), pp 2630 (1979) Song Dongcao Optical fiber pressure sensor based on fiber Bragg grating, Ph.D. thesis, Stevens Institute of Technology, 102 pages; AAT 3428858, http://proquest.umi.com/pqdlink?did=2191533691&Fmt=7&clientI d=79356&RQT=309&VName=PQD (2009) Mohanty L., Tjin S.C., Lie D.T.T., Panganiban S.E.C. & Chow P.K.H. Fiber grating sensor for pressure mapping during total knee arthroplasty, Sensors Act. A, Vol. 135, pp. 323-328 (2007)

[10] Talia P.M., Ramos A., Abe I., Schiller M.W., Lopes P., Nogueira R.N., Pinto J.L., Claramunt R. & Simoes J.A. Plated and intact femur strains in fracture fixation using fibre Bragg gratings and strain gauges, Exp. Mech., Vol. 47, pp. 355-363 (2007) [11] Carvalho L., Silva J.C.C., Nogueira R.N., Pinto J.L. Kalinowski H.L. & Simoes J.A. Application of Bragg grating sensors in dental biomechanics, J. Strain Anal., Vol. 41, pp. 411-416 (2006) [12] Dennison C.R., Wild P.M., Wilson D.R., Cripton P.A. & Dvorak M. Pressure sensor for biological fluids and use thereof US Patent application # 2009/0247899 A1 (2009) [13] Dennison C.R., Wild P.M., Wilson D.R. & Cripton P.A. A minimally invasive in-fiber Bragg grating sensor for intervertebral disc pressure measurements, Meas. Sci. Technol., Vol. 19, 085201 (12pp) DOI:10.1088/0957-0233/19/085201 (2008) [14] Kanellos G.T., Papaioannou G., Tsiokos D., Mitrogiannis C., Nianos G. & Pleros N. Two dimensional polymer-embedded quasidistributed FBG pressure sensor for biomedical applications, Optics Express, Vol. 18(1), pp. 179-186 (2010) [15] Pinet . Fabry-Prot fiber-optic sensors for physical parameters measurement in challenging conditions, Journal of sensors, Vol. 2009, Article ID 720980, 9 p., http://www.hindawi.com/journals/js/2009/720980.html (2009) [16] Van Neste R., Belleville C., Pronovost D. & Proulx A. System and method for measuring an optical path difference in a sensing interferometer US patent #6,842,254 B2 (2005) [17] Chavko M., Koller W.A., Prusaczyk W.K & McCarron R.M. Measurement of blast wave by miniature fiber optic pressure transducer in the rat brain, J. Neurosci. Meth., Vol. 159 (2), pp. 277-281 (2007) [18] Bauman R.A., Ling G., Tong L., Januskiewicz A., Agoston D., Delanerolle, N., Kim Y., Ritzel, D., Bell R., Ecklund J., Armonda R., Bandak F. & Parks S. An introductory characterization of a combat-casualty-care relevant swine model of closed head injury resulting from exposure to explosive blast, J. Neurotrauma, Vol. 26, pp. 841-860, DOI: 10.1089/neu.2008.0898 (2009) [19] Rodrigues C., Inaudi D., Juneau F. & Pinet . Miniature Fiber-Optic MOMS piezometer Geotech. Instr. News, pp. 10-13 (Sept. 2010) [20] Pinet ., Pham A. & Rioux S., Miniature fiber optic pressure sensor for medical applications: an opportunity for intra-aortic balloon pumping (IABP) therapy, Proc. SPIE, Vol. 5855, pp. 234-237 (August 2005) [21] onlagi D., Cibula E. & Pinet . Fabry-Prot optical sensor and method of manufacturing the same US patent application #2008/0159687 A1 (2008) [22] onlagi D., Cibula E. & Pinet . Single piece Fabry-Prot optical sensor and method of manufacturing the same US patent #7,684,657 B2 (2010) [23] Pinet ., Cibula E. & onlagi D. Ultra-miniature all-glass Fabry-Prot pressure sensor manufactured at the tip of a multimode optical fiber, Proc. SPIE, Vol. 6770, pp. 6770DU-1 to -8, DOI# 10.1117/12.734888 (2007) [24] Cibula E., Pevec S., Lenardili B., Pinet . & onlagi D. Miniature all-glass robust pressure sensor Optics Express, Vol. 17 (7), pp. 5098-5106 (2009)

Proc. of SPIE Vol. 7753 775304-4