1

The Use of Smart Materials Technologies in Radiation Environment and Nuclear Industry

Victor Giurgiutiu*, Andrei Zagrai

University of South Carolina, Mechanical Engineering Department, SC 29208

ABSTRACT Application of smart materials technology in nuclear industry offer new opportunities in this industrial sector for safety enhancement, reduced personal exposure, life-cycle cost reduction, and performance improvement. However, the radiation environments associated with nuclear operations represent a unique challenge to the testing, qualification and use of smart materials. The present study assesses the physical limitations of smart materials applications to nuclear industry and identifies areas of where such technologies could be reliably used. In our investigation, we considered a spectrum of smart materials technologies and a variety of irradiation environments. A literature survey to identify prior experimental data related to irradiation effects on smart-materials showed significant irradiation induced alterations in performance. However, the available database was limited and additional tests will be required before smart materials technologies can be introduced in most radiation environments. This paper summarizes a test program that was developed for the Accelerator Production of Tritium project. This program was designed to identify specific irradiation areas where the application of smart materials technologies is practical.

Keywords: piezoelectric sensors, TRIP steel, shape memory alloys, irradiation, test procedure, health monitoring, damage detection, failure prevention, life-cycle cost reduction.

1. INTRODUCTION The operational safety of nuclear facilities could benefit from the use of smart materials technologies in both: the monitoring and inspection processes. The use of smart materials technologies in such facilities could result in safety enhancement, life-cycle costs reduction, and performance improvement while decreasing personal exposure. However, the use of such smart materials in nuclear facilities requires knowledge about the materials respond to irradiation, and how this response is influenced by the radiation dose. Considerable work has been done on material respond to irradiation, but little is directly applicable to smart materials. The work summarized in this paper was done by the University of South Carolina (USC) under United States Department of Energy sponsorship in connection with the Accelerator Production of Tritium (APT) project. The general scope of the work was to determine possible applications of smart materials technology for monitoring and inspection of nuclear facilities, with special focus on the APT project. A literature search was performed to identify previous work and to provide a technical basis for an experimental research and testing program. The experimental program was to include introducing smart materials specimen packages in the Los Alamos Neutron Science Center (LANSC). The packages were to contain specimens of active materials whose properties were to be investigated before and after irradiation. Thus, two phases of the project were planned: (1) measurements before irradiation, and (2) measurements after irradiation. The APT project has been canceled, thus, only Phase 1 of the projected has been completed. This paper describes the work performed in Phase 1 and discusses possibilities for continuing the Phase 2 (irradiation of samples and post irradiation measurements) in cooperation with the Savannah River Site (SRS) and/or other DoE facilities.

The project started with the selection of types of smart materials and technologies utilized by other industries for structural health monitoring and accident prevention applications were prime candidates. Eventually, four types of smart materials were selected: (a) piezoelectric ceramics (PZT); (b) shape memory alloys (SMA); (c) peak strain sensors (TRIP technology); and (d) fiber optics. An extensive literature search was performed to identify previous work done on the behavior of these materials under irradiation. Various smart materials samples were purchased from commercial vendors and processed at USC into specimens. Tests plans for the smart materials specimens were developed in cooperation with SRS and LANL. The specimens were to be divided into 2 identical sets, “exposure” and “control”. The exposure set would be introduced in *Correspondence: Email: victorg@sc.edu; WWW: http://www.me.sc.edu/faculty/giurgiut.htm; Tel.: 803-777-8018; Fax: 803-777-0106

SPIE's 7th International Symposium on Smart Structures and Materials and 5th International Symposium on Nondestructive Evaluation and Health Monitoring of Aging Infrastructure, 5-9 March 2000, Newport Beach, CA. paper # 3985-103

2

prototypical components (to be produced in collaboration with SRS and LANL) and subjected to irradiation in LANSC. Clearly, these materials would not be exposed to irradiation conditions that would produce significant displacement damage but would only see how gamma irradiation affects them. The control set was to be retained at USC. Both sets, irradiated and control, were to be subjected to a battery of tests to determine the effect of irradiation on performance.

The specimens fabrication and testing were performed at USC in the Laboratory for Adaptive Materials and Smart Structures (LAMSS). Testing included: (a) physical properties; (b) mechanical properties; (c) smart-materials properties (sensing/actuating); (d) frequency response and resonance frequency (where applicable). The properties of each specimen were entered in properties bulletin and in a database. Characterization of specimens after irradiation was planned to be performed at DOE facilities (SRS/LANL).

2. STATE OF KNOWLEDGE ON IRRADIATION EFFECTS ON SMART MATERIALS The literature survey to identify the effect of irradiation on smart materials found results for piezoelectric materials, shape memory alloys (SMA), and fiber optics materials. No data was found on the effect of irradiation on TRIP steel peak-strain sensors. These observations are summarized below.

2.1 Piezoelectric materials. Glower and Schlosser, (1969) reported research on detectors made from ferroelectric material use for monitoring of nuclear facilities. Augusto de Carvalho and Mascarenhas (1992) measured the energy flux of a single pulse of X-radiation using photoacoustic, piezoelectric, and pyroelectric radiation dosimeters. The effect of irradiation on piezoelectric ceramic and quartz resonator response was also investigated. The durability of piezoceramic in near reactor environment was discussed by Baranov et al (1982). Broomfield (1984) reported radiation damage in Strontium Niobate, damage recovery commenced when the material was annealed at about 4000C. Meleshko et al (1984) studied the electrophysical parameters of various types of piezoceramics under gamma irradiation and determined the relationship between the resistance, capacitance, and tan delta of the piezoceramics employed in piezoelectric transducers, together with the relationships between the signals generated by utilized transducers and the power of the gamma radiation and neutron fluxes to which they are exposed. These authors also investigated the effect of thickness on the radiation resistance of the piezoelectric ceramics. It was noticed that with increasing flux, the resistance of all piezoceramic transducers monotonically decreased and absolute values of resistance were independent of the size or type of the piezoceramic (Meleshko et al, 1986). In the same way, an increase in the reactor power was accompanied by a reduction in the capacitance of all specimens. However, radiation had less effect on the capacitance of relatively thick specimens and depended on type of piezoceramic used. The effect of irradiation on the sensitivity of transducer was also studied by feeding standard pressure pulses to the exposed transducer and measuring the signal amplitude over time. It was observed that amplitude of the signal decrease monotonically while reactor was in operation. Additionally, it was observed that radiation exposure reduced the modulus of elasticity and Poisson’s ratio in piezoceramic material (Pozdeeva and Ul’yanov, 1989).

2.2 Shape Memory Alloys Shape memory alloys (SMA) have been used as active elements in mechanical devices for monitoring nuclear facilities. Bychkov et al (1989) investigated the effect of gamma dose on the structure of shape memory effect in Titanium Nickelide. The parameters of shape memory effect and intensity of X- ray diffraction varied nonmonotonically with increasing of radiation dose. The possible application of SMA in fission and fusion reactor devices was considered by Hoshiya et al. (1990). Three kinds of Ti-Ni shape memory alloy were neutron irradiated with different doses and their electrical resistance measurements were carried out after isothermal and isochronal annealing. The temperature dependence of the electrical resistance for the irradiated alloys was found to be similar to that for unirradiated alloys after annealing. Hoshiya et al. (1991) found that martensitic transformation temperature of Ti-Ni alloy seemed to decrease with the irradiation and unusual elastic behaviour with a recoverable strain in tensile testing was found. Ibrahim (1991) conducted experiments where Nitinol with a memorized shape retained that shape when exposed to fast neutron flux. Hoshiya et al. (1992) discussed transformation properties and deformation behavior of Ti-Ni type shape memory alloys, which were neutron-irradiated and subsequently post-annealed. The properties of alloys were measured by electrical resistance measurements and tensile tests. Abrupt changes in martensitic transformation temperatures and stress-strain curves of specimens were observed. In addition, it was shown that the irradiated state of this alloy can be represented by two conflicting processes; ordering and disordering, which depends on temperature, displacement and displacement rate. The key temperature of that state was also found and defined as a temperature at which the ordering becomes predominant over disordering and brings about the occurrence of a restoration

3

phenomena. The proposed fields of application for SMA in the nuclear power engineering were discussed in the paper presented by Ionajtis, Kotov and Shchukin (1995). Authors mentioned the passive safety systems, elements for tightening and hermetic connection of tubes and electric conductors, thermo-mechanical drives and engines, dampers, flow rate controllers, thermodetectors, direct-acting emergency devices, elements of electric lines as possible directions of the SMA application. Thus, specific features of SMA using in the nuclear power engineering were considered. It was also concluded that the SMA unique properties open wide possibilities for application in nuclear power engineering. Al-Aql and Dughaishi (1996) reported the results of electrical resistivity measurements of Nitinol shape memory alloy under the effect of various proton-beam energy levels. It was observed that increasing the proton beam energy enhances the creation of defects and, thus, increases the electrical resistivity at the martensitic transformation temperature. The resistance increase in the martensitic transformation in Nitinol alloy was reported to be sharper and higher when the proton beam energy had certain value. Dughaishi (1997) suggested that these results provide a new method of producing SMA materials of low martensitic transformation temperature and seems to have an important application in the robots industry, controls, sensors etc. In addition, he proposed that the results of his work will extend the applications of the shape memory alloys of low martensitic transition temperatures. Ionajtis (1997) described shape memory devices which have the most widespread application in nuclear industry. It emphasizes applications of shape memory alloys that would increase the reliability and safety of nuclear facilities and would reduce the risk of personnel irradiation. Microstructure, mechanical properties and transformation temperature of neutron irradiated NiTi shape memory alloy were investigated by Matsukawa, Suda, Ohnuki and Namba (1999). The results indicate that amorphous phase dominates the suppression of the martensitic transformation, and causes the change in the mechanical properties.

2.3 Optical fibers Greenwell and Lyons (1990) describer the evolution of component test procedures required to document potential radiation effects on some fiber optic systems. Bekjaev et al (1991) studied the stability of optical fiber by reflectometry method. It was shown, however, that the method cannot be used to study pulse and quick processes. The pulse irradiation was taken into account by Henschel, Koehn, and Schmidt (1992) in their experiments on measuring radiation sensitivity of different integrated optic devices under standardized test conditions. Henschel, Koehn, and Schmidt (1993) presented results of measurement of radiation-induced losses of several types of optical fibers as well as of different kinds of connectors, couplers, and multiplexers The properties and the application possibilities of fiber-optic dosimeters based on radiation-induced losses are discussed by Bueker and Haesing (1994). Of particular interest were the requirements that have to be met by fiber-optic dosimeters for application in radiation therapy. In 1994, the American Society for Testing and Materials (ASTM) published standards to qualify fiber optic spectroscopic systems for use in adverse nuclear environments with an emphasis on nuclear waste storage facilities and nuclear generating stations. A fiber optic sensor system assessment in gamma radiation environments was also presented by Spencer et al. (1994). They discussed the basic design, trade-offs, and performance characteristics of general purpose analog fiber optic link. Very interesting results on the use of remote fiber optic spectroscopic instrumentation systems in monitoring and characterizing nuclear waste were given by Greenwell et al (1995). Authors discussed test procedures and broadband test results to qualify fiber optic components for use in adverse nuclear waste environments and present some design options for spectroscopic instrumentation systems configurations. Van Uffelen, Jucker, and Fenaux (1997) discussed concepts for optical fiber's behavioral models under radiation exposure together with results of irradiation tests. Henschel et al (1997) investigated the influx of gamma radiation on sensitivity of fibre amplifiers in order to find fibers with extremely high loss increase for applications in dosimetry of low radiation levels. Measurements confirmed the applicability of a simple dose rate transformation method also to rare-earths doped fibers. Deparis et al (1998) critically discussed the implementation of optical fiber technology in nuclear industry was presented by. Berghmans et al (1998) reported irradiation experiments on different types of commercially available fiber-optic sensors. For temperature sensors, the authors reported results which show that gamma radiation does not interfere with the basic sensing mechanism, but the most critical component turns out to be the optical fiber itself. Semiconductor absorption temperature sensors showed no degradation up to total certain doses, whereas the characteristics of Fabry-Perot type sensors and fluorescence temperature sensors were already dramatically influenced at the low doses. The Bragg-grating resonance wavelength was found to shift with irradiation, but the temperature sensitivity of the sensor remained unaltered. Authors conclude that the use of Bragg grating fiber-optic sensors as well as of Fabry-Perot type fiber-optic strain sensors may become degraded by neutron radiation. Gao et al (1999) reported work done on the design of double-option-fiber sensor for measuring the radioactive character and harm of the nuclear material in storing state.

4

3. PREDICTED RADIATION EFFECTS ON ACTIVE MATERIALS

3.1 Predicted Irradiation Effects on Piezoelectric Material Piezoelectric materials are expected to undergo modification of their mechanical parameters such as modulus of elasticity and Poisson’s ration when irradiated (Pozdeeva and Ul’yanov, 1989). Since, in many applications, the piezoelectric sensor is used as a resonator, its basic properties such as resonance frequency, shape and amplitude of a resonance peak, and the coupling coefficient will be subject to variations during and after irradiation. In summary, three main characteristics of an piezoelectric resonator should be modified under irradiation conditions: (i) maximum resistance of PZT at resonance and as a function of damping; (ii) the shape of the peak and of the curve on either side of resonance; (iii) frequency at which resonance occurs. Furthermore, we expected to see variations in the value of electro-mechanical coupling coefficient of PZT resonator. In order to achieve this, we took advantage of two factors. The first is that the behavior of a material when it is being driven at or near resonance is a sensitive measure of the materials properties. The second is that, at resonance, certain mathematical approximations, which are only locally valid, can be used to simplify the calculations. Another important factor is that we are working within an engineering environment where research must be directed toward well-defined objectives, in this case structural health monitoring. Since the greatest level of transduction takes place within the PZT at resonance, it follows that sensor systems utilizing PZT designed to operate near or at resonance will maximize this effect. The problem that arises with utilizing PZT resonators testing is to account the effect of other environmental factors such as temperature and handling damage. The effect of these factors has to be effectively removed, and the changes due to them need to made separate from the changes due to irradiation effects. In general, the fundamental predicted change is a reduction of energy transferred into mechanical from electrical at resonance. This reduction will manifest itself as a shifting of the resonance peak to a slightly lower frequency and a rounding of its profile when plotted in the impedance vs. frequency domain.

3.2 Predicted TRIP Steel Behavior under Irradiation Exposure Although no information could be found in the literature regarding TRIP steel sensors behavior under irradiation, we used the physical-metallurgical principles of their construction to propose investigative methods of identifying such changes in our experiments. We proposed to look at how this material handles magnetic field before and under irradiation. If one changes the magnetic field magnitude and direction constantly the material will act as a reservoir for the changing fields energy. The capacity of such reservoir and the maximum energy flow rate to and from the reservoir are the variables of interest. (Grayson, 1999). Therefore, the change in magnetic susceptibility change with strain could be predicted to change with irradiation.

3.3 Predicted Shape Memory Alloy Behavior under Irradiation Exposure Prediction of Shape Memory Alloy (SMA) behavior under irradiation were done base on the literature review results. Changes of the strain recovery as a function of temperature during the phase transformation, with accompanying changes in the mechanical properties are expected. Of major interest are (a) the decrease in martensitic transformation temperature reported by Hoshiya et al. (1991) and (b) the increase in electrical resistively at the martensitic transformation temperature reported by Dughaishi (1997). Thus, we concluded that the study of martensitic transformation temperature under radiation exposure should be performed. If the differences in temperature history can be corrected for, then it would be possible to isolate the irradiation effects causing differences in the ratio of austenite to martensite phase in SMA material at various temperatures, as reported by Hodgson et al. (1998).

3.4 Predicted Optical Fiber Behavior under Irradiation Exposure Optical fiber can be affected in different ways depending on the kind of radiation and its dose rate as well as the material of which the fiber is made. Optical fiber is known turn black under severe radiation. At lower irradiation, the fiber attenuation will increases with the level and duration of irradiation. After low levels of irradiation, a tendency to recover over time exists once the fiber has been removed from the irradiation source. In this case, the attenuation may reduce back to the pre-irradiation value, or close to it. The recovery period varies however for different fibers, and in function of the irradiation conditions to which the specimen has been subjected. The material type is a significant factor because each will absorb radiation in differing amounts. For example, doping elements such a Boron exhibit a large cross sectional area under neutron bombardment, and this results in a larger production of defect centers that permanently diminish light transmission and thus increase the measured attenuation. Electromagnetic radiation also produces transient effects; their recovery rate depends on the optical fiber materials, that can include photon production within the fiber due to the photoelectric effect or increased attenuation due to "induced loses". Overall gamma ray irradiation does not permanently degrade optical fibers unless exotic photosensitive materials are used in doping the fiber. The latter may be the case with the Bragg grating optical fiber sensors.

5

The permanent increase in attenuation is predominantly a function of the neutron flux rate and total dose received by the material. The number of variables and factors to take into account when conducting radiation tests on optical fiber are large. Based on the literature survey, we predict the following irradiation effects: (a) an increase in the amount of light attenuation within the fiber, possibly to the point of opaqueness (Greenwell, 1994) and (b) a resonance wavelength shift of the Bragg-gratings used in the fiber optic strain sensors (Berghmans et al, 1998).

(a)

0.5mm

8.0mm

109.2mmVolume A

Volume C0.5mm

7.0mm

109.2mmVolume B

8.2mm

2.0mm

53.3mm

(b)

Figure 1 (a) schematic of the sample volumes available in the experimental APT facility; (b) example of simulated volume cavity constructed at the University of South Carolina to test sample packing hypotheses.

4. PREPARATION OF EXPERIMENTS

4.1 Design and constructions of specimens When performing the design and construction of our specimens we had to confirm to stringent space constraints. Figure 1a shows the target-blanked area of the experimental APT facility at the Los Alamos National Laboratory, while Figure 1b indicates the initially planned insertion of the smart materials specimens in the target-blanket region. The insertion of the smart material specimens was planned together with the insertion of other material samples. The space volumes made available for this insertion were very tight. Specification changes were encountered regarding these experimental volumes. Final volume types are shown in Figure 2. Note that the available thickness varies from 0.5 mm for volumes type A and B, to 2 mm for volume type C. Creative and innovative specimen design and construction was required to meet these tight constraints. The constraints imposed by the available experimental volumes had to be precisely met. To this purpose, sample containers, containing dummy experimental volumes, were fabricated (Figure 2b). These sample containers were used to verify that the smart materials specimens developed during the project would fit well in the actual APT experimental facility. Specimens were acquired/constructed for the four active materials selected in the project. The specimens size was restricted by the need to fit in the experimental volumes planned at Los Alamos National Laboratory APT facility (Figure 2). To achieve this objective, ingenious technologies had to be elaborated. The main design drivers included: (a) safe and reliable positioning of the specimens inside the volume; (b) good contact with the outside walls to ensure appropriate cooling; and (c) easy handling even with the remote manipulators specific to nuclear “hot cells”.

Piezoelectric specimens. The piezoelectric specimens were constructed in the form of small PZT wafers cut from 73 mm (2.85-in) square, 0.19 mm (0.0075-in) thick sheets that also had electrode layer of Nickel (Ni) vapor-deposited on both sides. To fit into the Type C experimental volume, the PZT specimens size was selected to be 15 mm square. To facilitate establishing electrical contact while using manipulators in the hot cell, special connectors were fabricated from thin copper foil and attached with thin wires to the opposite faces of the PZT wafer. The fabrication technology involves several stages (Figure 4): cutting of piezoceramic and cooper foil, assembling of a specimen in a special jig, and soldering process. Capacitance tests were conducted immediately after fabrication. These tests allowed to weed out rejects and sort out specimens with good and close capacitance values. (Capacitance of the specimens was found to vary in the range 5.5nF -- 10.5nF.) A history of specimen production and quality control was kept for each specimen and entered in the specimen own bulletin. The final configuration of a piezoelectric sensor is shown on Figure 3.

Figure 2 Piezoelectric specimen

6

(a)

Perspexsheets

PZT wafer

Diamond tip scriber

Small Pressureapplied

(b)

1/38"gauge wire

Aluminumplate (c)

15mm

8mm

SolderingPaste

(d)

Special jig

Figure 3 The fabrication stages for a PZT specimen: (a) cutting size out of PZT sheet , (b) eliminating wire isolation, (c) cutting cooper foil and producing electrodes for the sensors, (d) assembling of the sensor on special jig.

TRIP steel specimens. TRIP steel sensors (Figure 5) were obtained from manufacturer (Grayson, 1999) and utilized as TRIP steel specimens for the irradiation experiment. Each individual TRIP specimen would fit in a 109.2mm × 8.0mm × 0.5mm volume. No fabrication operations, like bending or trimming, were required. It was projected that up five sensors could be fitted in a Type B experimental volume.

(a)

Length = 85 mm

Active region =13 mm

Thickness = 0.152mm

Width = 1.3 mmActive region width = 0.6 mm

(b) (c)

Figure 4 TRIP steel specimens: (a) schematic drawing of TRIP steel sensor. (b) actual TRIP steel sensor; (c) measurement of the TRIP steel specimen in a calibration coil.

Shape memory alloys (SMA) specimens. The shape memory alloys (SMA) specimens were designed and constructed for an initially specified experimental volume of circular form with 12.5 mm diameter and 2 mm thickness. (This volume was subsequently discontinued by APT.) The SMA specimen consisted of a 200-mm SMA ribbon that was coiled into a flat spiral with 12.5mm maximum diameter (Figure 11). The coiling up of the spiral was performed at a temperature far below the austenite start transformation temperature, As. A syringe type tool with a winding and holding mechanism was specially fabricated for this operation. Once the length of ribbon is coiled completely inside the tool and the center core is removed, the syringe nature of the tool is utilized to insert the coil in the simulated experimental volume, which was a 12.5 mm diameter circular depression in the surface of the sample container.

Fiber optics specimens. The fiber optics specimen consisted of an optical fiber that would have to be coiled up and inserted in one of the experimental volumes. The transmission and attenuation properties of the fiber would be measured before and after irradiation exposure using Optical Time-Domain Reflectometer (OTDR) equipment. Preliminary discussion with the optical fiber manufacturers revealed that at least 200 meters of fiber, including two 1-meter leads, are needed to perform relatively accurate measurements. Tight curvatures should be also avoided in the specimen construction. For these reasons, it was finally decided that the tight experimental volumes made available at the APT facility would not be able to accommodate a credible fiber optics specimen. However, a 200-meter reel of optical fiber was prepared for irradiation at a separate facility.

7

4.2 Experimental Plans and Procedures The objectives of experiments were to obtain characterization of the smart materials specimens before and after irradiation. The results of these tests were to be recorded and entered into a database for further processing. Two identical groups of specimens, one “control” and the other “exposure” were created. Characterization experiments were performed on both groups in the USC LAMSS facility. Details follow.

Piezoelectric Specimens: In our experiments on piezoelectric specimens, we studied their behavior electromechanical resonators. Basic resonator properties (resonance frequency, shape and amplitude of a resonance peak, and electromechanical coupling coefficient) were expected to vary during and after irradiation. The study was focused on identifying these changes. However, these resonator parameters are directly related to the fundamental properties of the piezoelectric material; hence, more in depth investigation on the irradiation effects on the fundamental properties of piezoelectric materials may be warranted for future studies. The step by step testing and data acquisition process is described in Figure 6a. The major equipment used in these experiments was the HP 4194A Hewlett Packard Gain-Phase Impedance Analyzer (Figure 6b). An A. C. voltage of 1-Volt nominal value was applied across the thickness of the PZT resonator through its face electrodes, and the corresponding current was measured by the analyzer. The resulting impedance was calculated inside the analyzer, displayed on its screen, and transferred to a PC through a GPIB interface. Frequency sweeps were performed in the 100 -- 1000 kHz range.

(a)

Connect wires to impedance phasegain analyzer.

Initialize program to enable data tobe logged into PC

Test PZT wafer between 200kHzand 380khz.

Move PZT Wafer to next orientation.

Extract foil electrodes from clampsthen stow away sample.

End

No

Start

Place foil electrodes into insulatedclamps.

Ensure each white wire has goodcontact with the foil

Test PZT wafer capacitance acrosswires and directly across electrodes.

Are reading similarto with≈ 0.2nF?

Yes

Experimental Method Flow Chart

Test PZT wafer between 200kHzand 380khz.

(b)

Figure 5 (a) Experimental method flowchart for piezoelectric sensors (b) HP 4294A Impedance Analyzer utilized in the PZT and TRIP steel specimen characterization experiments.

TRIP Steel Specimens: The TRIP steel specimens were tested under conditions of varying magnetic field using an experimental coil connected to the HP 4194A Hewlett Packard Gain-Phase Impedance Analyzer. The coil was provided by the TRIP steel sensor manufacturer (Grayson, 1999). The coil inductance depends, among other factors, on magnetic permeability of the core material. In our case, the magnetic core is the TRIP steel specimen. By measuring the impedance of the coil with the TRIP steel specimen inside it, we actually measured its inductance and, indirectly, the magnetic permeability of the TRIP steel specimen. As for the PZT samples, the data resulting from these measurements is in the form of impedance vs. frequency.

SMA Specimens: For the shape memory alloy (SMA) specimens, we measured the strain recovery as a function of temperature during the phase transformation. The experimental data was gathered by capturing images of the SMA spiral uncoiling in a heated bath. The temperature of the bath during experiments was slowly increased to simulate a sequence of steady state conditions. The coils was constructed of one-way shape memory alloy.

Fiber Optics Specimens: The 500 meter long reel of optical fiber was taken to Pirelli North America facility in Lexington, SC and subjected to an OTDR calibration test to determine its baseline properties. The experiment would be continued with two identical fiber specimens enclosed in two separate yet identical volumes, one to be kept as control in normal laboratory conditions, the other to be subjected to irradiation, In parallel, a independent experiment was initiated at USC LAMSS to determine the operational behavior of fiber optics strain sensors using Bragg grating technology. A Bragg gratings fiber optics device was applied onto a 500 mm long, 20 mm wide, 3 mm thick metal beam. The fiber optics device was 650 mm

8

long and consisted of 5 Bragg strain sensors mounted on a single fiber at 100 mm intervals. Collocated with the optical sensors were conventional electrical resistance strain gauges in half-bridge per measurement point wiring (Figure 7a). Thus, the strain applied to the beam through loading was simultaneously monitored with the two methods. The fiber optics strain gauges were monitored with an AFSS-PC interrogation system (Figure 7b), which delivers the data to a computer by means of a National Instruments data acquisition card. The readings obtained with such a system are to be compared with the readings obtained with the conventional strain gauges taken with conventional instrumentation.

(a) (b)

Figure 6 (a) a section of a beam with strain gauge and fiber optic sensor installed in order to obtain data for comparison; (b) Interrogation system for fiber optic strain.

5. EXPERIMENTAL RESULTS To date, only the results from the first-phase experiments have been collected. The first-phase experiments were conducted in USC LAMSS under laboratory conditions. The two sets of active material specimens, “control” and “exposure”, have been tested. Results of these tests are presented next. The measured parameters and characteristics of these specimens were stored in the project database. For each specimen tested, a data bulletin was also created. The data collected and processed during this phase provides complete information about the behavior of the piezoelectric sample. These data would later be compared with results taken after irradiation exposure.

a)

Wafer A1 Re(z) vs. Frequency

0

500

1000

1500

2000

2500

3000

200 250 300 350 400Frequency (kHz)

Re(

Z) (O

hms)

LooseTight

b)

Impedance Domain Plot of Wafer A1

-2000

-1500

-1000

-500

0

500

1000

1500

0 1000 2000 3000

Re(Z) (Ohms)

Im(Z

) (O

hms)

LooseTight

c)

Wafer A7 Re(z) vs. Frequency

0

200

400

600

800

1000

1200

1400

200 250 300 350 400Frequency (kHz)

Re(

Z) (O

hms)

LooseTight

Figure 7 Electro-mechanical impedance characterization curves of the PZT active sensor specimen: (a) Re(Z) curve; (b) Impedance plane curve; (c) Re (Z) curve of PZT sensor with poor impedance characteristics.

5.1 Experimental Results for Piezoelectric Specimens The electromechanical impedance of the piezoelectric specimens was measured over the 100—1000 kHz range. Complex (real + imaginary) impedance values vs. frequency were recorded and displayed over the 200--380kHz range (Figure 8). This range is chosen as it contains the first resonance peak of the PZT specimen. The data was processed into plots frequency of real impedance (Figure 8a), imaginary impedance, impedance magnitude, and impedance phase. Plots of real vs. imaginary impedance were also produced (Figure 8b). The piezoelectric resonator parameters (resonant frequency, antiresonant frequency, resonant bandwidth, value of absolute magnitude, etc.) were determined from these plots and the resulting values were entered into the specimen bulletin (Figure 9). Besides the impedance measurement results, the specimen bulletin also contains important descriptive data (precise dimensional measurements, amount of soldering on electrode surfaces, etc.) that may be used in explaining why the experimental data deviates from the ideal behavior.

9

5.1.1 Resonator Modeling of the Piezoelectric Specimen

As a resonator, the rectangular piezoelectric specimen has three fundamental 1-DOF modes: lengthwise, widthwise, and thickness wise. The electromechanical admittance for the resonator operating in the lengthwise mode can be derived as (Ikeda, 1996):

( )2 231 311 tan /

2 2pzt E EL LY i C k k

c cω ωω � � �= ⋅ − + ⋅� � �

� � (1)

Where ( )( ) 12 231 31 33 11 1T Ek d s iε η δ ηδ

−�= − + −� is a coupling factor for

piezoelectric sensor, Ec is the complex sound velocity in the material,

pztC is the complex capacitance of a sensor, and L is length. Similar formulae can be derived for widthwise and thickness wise operation. In our experiments, the length and width wise modes were relatively close, while the thickness mode was widely separated from the other two. If the length and width dimensions were considerably different, the length and width modes would also decouple. To take into account all the three degrees of freedom, a simple superposition can be applied, in the first approximation.

Figure 10 shows the results of our modeling superposed on experimental data. The modeling was performed using a average dimensions obtained from measuring a 12-specimen statistical sample. The experimental and theoretical curves have the same shape. The experimental curve display the expected features associated with a resonance. This indicates that the model described by Equation (1) is adequate for describing the electromechanical impedance. It can serve as a basis for predicting the impedance changes that would result from exposing the piezoelectric specimens to irradiation.

The peak amplitude at resonance and the sharpness of the resonance curve are associated to the amount of damping. This coefficient of damping above 0.1% clearly influences to the shape of resonance curve. By varying the complex coefficient of internal looses one can fit the theoretical curves onto the experimental values. The exact value of resonance frequency, however, strictly depends on the properties of the piezoelectric material, the sound velocity in the piezoceramic, and the geometric dimensions. The shape of the resonance response curves also depends on the PZT specimen shape and on minute manufacturing details. Specimens which accidentally do not have parallel sides (with even minute deviations) were found to display poorer resonance characteristics than those with precise rectangular shapes (Figure 8c). Besides the dominant resonance, these specimens display additional modes. In our investigation, the best results were obtained with perfect rectangular PZT plate.

(a)0 100 200 300 400 500 600 700 800 900 10000

1

2

3

4

Frequency K Hz.

Impe

danc

e K

Ohm

.

b)0 100 200 300 400 500 600 700 800 900 10004

2

0

2

Frequency K Hz.

Impe

danc

e K

Ohm

.

Figure 9 Real (a) and imaginary (b) parts of impedance for theoretical modeled (blue) and a real piezoceramic plate (red)

Regarding the correlation between theory and experiment, it should be noted that, due to the laboratory conditions in which our specimens were fabricated, their actual dimensions showed considerable spread. As shown in Figure 10, the difference between theoretical and experimental results is ∆f = 62 kHz. This value is an indication of the uncertainties associated with

Samples I.D. number A 1Manufacturingdate

12/2/98

Size mm. Length 7.743 mm W idth 6.907 mm Thickness 0.1905

mmSurface 1 Surface 2Wires 30 mm 30 mm

Soldering l = 0.91 mm, w = 1.17 mm, t =0.27 l = 1.35 mm, w = 1.23 mm, t = 0.36

Cooper electrodes l = 15.00 mm, w = 7.64 mm, t = 0.01mm l = 15.06 mm, w = 7. 56 mm, t = 0.01

Capacitance tight Capacitance loose Month/date/yearCapacitance9.47 nF 9.26 nF 12/4/98

Resonance frequency IZI Hz floose = 266150Parameters ofAmplitude- frequencycharacteristics Resonance frequency IZI Hz ftight = 266150

Max. Amplitude IZI

Am = 2.60417;

2.01734

Data Info. Files A1 Dec21.xls ;

Comments

Wafer A1 Re(z) vs. Frequency

0

500

1000

1500

2000

2500

3000

200 250 300 350 400Frequency (kHz)

Re(

Z) (

Ohm

s)

LooseTight

Figure 8 A page of specification Bulletin for

tested piezoelectric specimens.

10

our specimen fabrication methodology. The influence of manufacturing factor is high. Theoretical predictions base on Equation (1) indicate that dimensional deviation of 0.5-1mm in the specimen length can produce errors in the resonance frequency as high as 100kHz.

These results show that distribution of resonances has strong dependence on geometrical dimensions of sensor as well as on properties of piezoelectric material. The exact sensor dimensions directly influence its resonance characteristics and give it basic behavior. The bigger the resonant dimension, the lower the value of resonant frequency. The closer the dimensions of a plate to each other, the closer the resonant peaks in the resonance region. When these dimensions almost match, the two curves coalesce into a twin-peak curve. This curve shape is unstable, and further changes in resonance due to the physical factors could strongly modify it. Hence, such specimens were weeded out from our batch and carried over into the Phase 2 of the program. Only specimens that displayed the most stable characteristics were selected for the Phase 2 experiments that would be conducted under irradiation conditions. The criteria of selection were the value of impedance magnitude, smoothness of the characteristic curves, and freedom from spurious (noise) peaks in frequency response curves. Only specimens with clearly defined dominant peak were selected for the Phase 2 experiments in DOE facilities. Attenuation for higher modes is relatively fast due to redistribution of energy among harmonics. Additional small peaks at the resonance curve exist due to asymmetric in geometry and rough edges of PZT specimen, as well as due to excessive soldering in the manufacturing process, etc.

5.1.1.1 Experimental Results for TRIP Steel Specimens. For the testing of sensors made from Transformation Induced Plasticity Steel (TRIP steel) we use the same electric impedance measurement technique and the HP 4194A Hewlett Packard Gain-Phase Impedance Analyzer (Figure 5c). The data acquisition process is the same as for piezoelectric specimens and data was similarly recorded and stored. The behavior of the TRIP steel sensor in the off-resonant region presented regular behavior typical of a metal resonator.

5.1.1.2 Experimental Results for Shape Memory Alloys Specimen The strain recovery as a function of temperature during the phase change transformation was investigated using the spiral-shaped SMA specimen. The As and Af temperatures for this specimen were in the 30-400C range. Experiments were conducted in water on a hot plate, with electric thermometer probe inserted next to the specimen. A rectangular background grid was also inserted. The whole setup was photographed at 0.1-0.5 °C increments using a CDD-stills camera. As the SMA ribbon undergoes phase transformation, it unwound (Figure 11). The raw image data, in the form of individual CCD camera frames, was extracted and processed with appropriate software. The resulting pictures were stitched together, in order of increasing temperature and the stack of pictures was optimized into an animated movie. Through this process, we also removed all redundant, i.e. unchanging, pixels from one picture to the next, sets the color depth, and defines the frame rate per second. From processing of these images, data about the strain and strain rate per degree C could be extracted.

5.1.1.3 Experimental Results for the Fiber Optics Specimen An attenuation test using Optical Time-Domain Reflectometer (OTDR) was conducted at the Pirelli North America facility with a 500 meters optical fiber specimen (Sach, 1999). The testing was conducted at 1310 nanometers wavelength. The total increase in attenuation was of about –1.72 dB. The average loss per kilometer was about -0.791 dB. Because attenuation is also increased by the length of the specimen, it was noticed that the slope of the graph gradually drops as it approaches the total length of the fiber. A major signal loss was noticed at the connection of the specimen to the OTDR. Another important feature is the asymptotical increase of attenuation as the end of the fiber is approached. This was due to reflective distortion.

6. CONCLUSION The research under the APT/SRS project has been conducted during summer 1998 – spring 1999. On the first stage we determined the ability of smart material to reflect changing in fundamental properties due to the radiation exposure and their possible application to the monitoring of DOE facilities. In order to obtain recent up-to date information on this subject the literature survey was done. The results of literature survey summarized in the form of prediction of behavior of active

Figure 10 Uncoiling SMA ribbon.

11

materials such as piezoelectrics, TRIP steel, shape memory alloys, and optical fiber under radiation exposure. For these materials the special containers that simulate space volume available in DOE facility were designed and manufactured. At the same time active material specimens were constructed under requirements of these volumes. The material specimens have different design, shape and other characteristics. However, specimens inside one group of materials have stable properties and small variation of characteristics due to unique of each specimen itself. To investigate these characteristics the procedure of experimental set up, data reception, data procession and data analysis was elaborated. Then, the characteristic parameters of each group of sensors were obtained through experiments utilized at LAMSS laboratory, USC. The database for specimens of each group was created in different forms and updated.

ACKNOWLEDGMENTS

The financial support of Department of Energy through the APT Southeast University Consortium, contract DE-FG04-97AL77994 is thankfully acknowledged. The authors wish to express their gratitude to the collegial support and encouragement received from the Savannah River Site Technology Center and Los Alamos National Laboratory.

REFERENCE

AL-AQL, A. A.; DUGHAISH, Z.H. (1996) “Effect of energy variation of proton beam on sharpening nitinol electrical resistivity increase at the martensitic transformation” Physica B: Condensed Matter, v 229 n 1, p 91-95, Dec 1 1996

AUGUSTO DE CARVALHO, Aparecido; MASCARENHAS, Sergio (1992) “Piezo And Pyroelectric Radiation Dosimetry”, IEEE Transactions on Electrical Insulation v 27, n 4, p 835-842, Aug 1992

BARANOV, V. M.; MARTYNENKO, S. P.; SHARAPA, A. I. (1982) “Durability Of Ztl Piezoceramic Under The Action Of Reactor Radiation” Soviet Atomic Energy (English translation of Atomnaya Energiya) v 53, n 5, p 803-804, Nov 1982.

BEKJAEV, R.A.; BONDAREV, A. I.; DOLGOV, I. I.; IGNATENKO, M. V.; TATSENKO, V. G. (1991) “Application of optic reflectometry method to investigation of radiation stability of optical fibers” Radiotekhnika, n 8, p 87-90, Aug 1991

BERGHMANS, Francis; FERNANDEZ, Alberto Fernandez; BRICHARD, Benoit; VOS, Frans; DECRETON, Marc; GUSAROV, Andrei; DEPARIS, Olivier; MEGRET, Patrice; BLONDEL, Michel; CARON, Serge; MORIN, Andre (1998) “Radiation hardness of fiber-optic sensors for monitoring and remote handling applications in nuclear environments” Proceedings of SPIE 1998 Process Monitoring with Optical Fibers and Harsh Environment Sensors Symposium, Nov 3-Nov 4 1998, v 3538, p 28-39, 1999 Boston, MA, USA.

BROOMFIELD, G. H. (1984) “High-Temperature And Post-Irradiation Piezoelectric Properties Of Strontium Niobate” Metallurgist and Materials Technologist, v 16, n 10, p 515-520, Oct 1984.

BUEKER, Harald; HAESING, Friedrich W. (1994) “Fiber optic radiation sensors” Proceedings of SPIE - The International Society for Optical Engineering Optical Fibre Sensing and Systems in Nuclear Environments Sep 17 1994, v 2425, 1994

BUSEL, J.; LINDSAY, Karen (1997) “Emerging Market Report – The Issue of Durability," Composites Design and Applications, May/June 1997.

BYCHKOV, V. A.; KOPYLOV, R. V.; MILOSERDIN, V. YU.; PERLOVICH, YU. A.; FESENKO, V. A.; KHROMOVA, L. P.; SHIROKINSKIY, S. YU. (1989) “Influence Of Gamma-Quantum Irradiation On The Structural State And Characteristics Of The Shape-Memory Effect Of Titanium Nickelide” Physics of Metals and Metallography (English translation of Fizika Metallov i Metallovedenie), v 68, n 3, p 145-149, 1989

DEPARIS, O.; MEGRET, P.; BLONDEL, M.; BERGHMANS, F.; DECRETON, M (1998) “Fibre-optics in nuclear industry: issues, problems and potential solutions” Revue HF Tijdschrift, n 2, p 65-72, 1998.

DUGHAISH, Z.H. (1997) “Effect of variation of proton beam energy on the martensitic transformation temperature of shape memory nitinol alloy” Materials Letters v 32 n 1, p 29-32, Jul 1997

GAO, Xinbao; YU, Xin; GAO, Tingru; ZHU, Yunpu (1999) “Design and application of an optical fiber sensor for measuring radiation” Proceedings of the International Symposium on Test and Measurement Proceedings of the 1999 3rd International Symposium on Test and Measurement, ISTM-99 Jun 2-Jun 4 1999, p 947-950, 1999, Xi'an, China.

GLOWER, D. D.; MILLER, D. W.; SCHLOSSER, P. A. (1969) “Use Of Ferroelectrics In The Detection Of Nuclear Radiation”, Int J Nondestruct Test, v 1 n 3, p 191-212, Oct. 1969.

GRAYSON, Paul (1999) Personal communication, Strain Monitoring Systems, Inc. 7001 Peachtree Industrial Blvd. NE, Suite 209, Norcross, GA 30092, http://www.strainmonitor.com/index2.html

GREENWELL, Roger A.; LYONS, Peter B. (1990) “Development of radiation test procedures for fiber optic systems” Proceedings of SPIE - The International Society for Optical Engineering Fibre Optics '90 Apr 24-26 1990 v 1314 1990 London.

GREENWELL, Roger A.; SAGGESE, Steven J. (1994) ”Development of American Soc. for Testing and Materials (ASTM) test procedures to measure induced ionizing radiation effects in broadband and remote Raman fiber optic spectroscopic systems” Proceedings of SPIE - The International Society for Optical Engineering Optical Fibre Sensing and Systems in Nuclear Environments Sep 17 1994, v 2425, 1994.

GREENWELL, Roger A.; SAGGESE, Steven; HATCH, Joel (1995) “Use of remote fiber optic spectroscopic instrumentation to characterize nuclear waste” Proceedings of the International Instrumentation Symposium Proceedings of the 41st International Instrumentation Symposium May 7-11 1995, Aurora, CO, , USA, 1995.

12

HENSCHEL, H.; KOEHN, O.; SCHMIDT, H. U.; KIRCHHOF, J.; UNGER, S. (1997) “Radiation-induced loss of rare earth doped silica fibres” IEEE Transactions on Nuclear Science Proceedings of the 1997 4th European Conference on Radiation and its Effects on Components and Systems, RADECS'97 Sep 15-19 1997, v 45, n 3, pt 3, p 1552-1557, Jun 1998 Cannes, Fr.

HENSCHEL, Henning; KOEHN, Otmar; SCHMIDT, Hans U. (1992) “Radiation-induced transmission loss of integrated optic waveguide devices” Proceedings of SPIE - The International Society for Optical Engineering Integrated Optical Circuits II Sep 10-11 1992, v 1794, 1993 Boston, MA, USA,

HENSCHEL, Henning; KOEHN, Otmar; SCHMIDT, Hans U. (1993) “Radiation sensitivity of passive fiber optic components” Proceedings of SPIE - The International Society for Optical Engineering Passive Fibre Optic Components and Their Reliability Apr 6-8 1993, v 1973, 1993 Berlin.

HODGSON, Darel E.; WU, Ming H.; BIERMANN, Robert J. (1998), Shape Memory Applications, Inc., Memory Technologies, and Harrison Alloys, Inc, http://www.sma-inc.com/SMAPaper.html

HOSHIVA, Taiji; DEN, Shoji; ITO, Haruhiko; TAKAMURA, Saburo; ICHIHASHI, Yoshinori (1991) “Effect Of Neutron Irradiation On The Transformation And Mechanical Properties In Ti-Ni Shape Memory Alloys” Journal of the Japan Institute of Metals v 55 n 10 Oct, p 1054-1062, 1991

HOSHIY A., TAKADA, F.; SHIMAKAWA, S.; ICHIHASHI, Y.; NISHIKAWA, M. (1992) “Restoration phenomena of Ti-Ni shape memory alloys in a neutron irradiation environment” Journal of Nuclear Materials v 191-94 pt B, p 1070-1074, Sep 1992 .

HOSHIY A.; TAKADA, F.; ICHIHASHI, Y.; PAK, H.-R. (1990) “Restoration phenomena of neutron-irradiated Ti-Ni shape memory alloys” Materials Science & Engineering A: Structural Materials: Properties, Microstructure and Processing v A130 n 2, p 185-191, Dec 1990.

IBRAHIM, E. F. (1991) “Neutron Irradiation Of Nitinol Memory Alloy” Journal of Nuclear Materials v 185 n 3 Dec 1, p 302-303, 1991. IKEDA, Takuro (1996) Fundamental of Piezoelectricity, Oxford University Press, 1996. IONAJTIS, R. R. “Possibilities and peculiarities of shape memory alloy application in nuclear engineering” Atomnaya Energiya v 82 n 5, p

392-396, May 1997 IONAJTIS, R. R.; KOTOV, V.V.; SHCHUKIN, I. M. (1995) “Application of shape memory alloys in the nuclear power engineering”

Atomnaya Energiya v 79 n 4, p 304-306, Oct 1995 MATSUKAWA, Y.; SUDA, T.; OHNUKI, S.; NAMBA, C. (1999) “Microstructure and mechanical properties of neutron irradiated TiNi

shape memory alloy” Journal of Nuclear Materials v 271-272, p 106-110, May 1999 MELESHKO, YU. P.; BABAEV, S. V.; KARPECHKO, S. G.; NALIVAEV, V. I.; SAFIN, YU. A.; SMIRNOV, V. M. (1984) “Studying

The Electrophysical Parameters Of Piezoceramics Of Various Types In An Ivv-2m Reactor” Soviet Atomic Energy (English translation of Atomnaya Energiya), v 57, n 2, p 544-548 , Aug 1984

MELESHKO, YU. P.; KARPECHKO, S. G.; LEONT’EV, G. K.; NALIVAEV, V. I.; NIKIFOROV, A. D.; SMIRNOV, V. M. (1986) “Radiation Resistance Of The Piezoelectric Ceramics Tsts-21 And Tnv-1” Soviet Atomic Energy (English translation of Atomnaya Energiya), v 61, n 1, p 561-564, Jul 1986

POZDEEVA, E.V.; UL'YANOV, V. L. (1989) “Elastic Properties Of Neutron-Irradiated Steatite Ceramics” Glass and Ceramics (English translation of Steklo i Keramika), v 45, n 9-10, p 395-396,May 1989

SACH, John R. (1999) Personal communication, Pirelli Cables and Systems North America, 710 Industrial Drive, Lexington, SC 29072. SPENCER, J. W.; SMITH, Richard V.; SMITH, S. R.; Humphries, J.; JONES, Gordon R.; FITZGERALD, A.; SHARP, R. E.; PATER, L.

(1994) “Fibre optic sensor system assessment in radiation environments” Proceedings of SPIE - The International Society for Optical Engineering Optical Fibre Sensing and Systems in Nuclear Environments Sep 17 1994, v 2425, 1994

VAN UFFELEN, M.; JUCKER, Ph.; FENAUX, Ph (1997) “Radiation resistance of fiberoptic components and predictive models for optical fiber systems in nuclear environments” IEEE Transactions on Nuclear Science Proceedings of the 1997 4th European Conference on Radiation and its Effects on Components and Systems, RADECS'97 Sep 15-19 1997, v 45, n 3, pt 3, p 1558-1565, Jun 1998, Cannes, FR.