Appendix C

Complementary Laboratory Experiments

A system of group projects was developed during the evolution of the subject matter of this book when used for teaching purposes. One format involved the use of weekly problem sets for the fundamental part of the material (Chapters 2 through 10), similar in type and level to the questions found at the end of these chapters. During the second part of the course, two alternative schemes were used. One involved the assignment of term papers on a special topic, examples of which are given at the end of this section. The other, and more elaborate approach, consisted of experimental projects. These projects were open-ended as opposed to set-piece laboratory experiments. What was actually done depended on the students backgrounds, availability of equipment, and qualied instructors. Hence it is stressed that the notes given below should be seen as guidelines or suggestions as to how a suitable laboratory component could be set up and not as formal, readyto-use laboratory methodology descriptions. For this second part of the course, students were divided into teams of two or three. A term project was carried out by each team, enabling the students to go more in depth in a given area than they could have done otherwise. Students were asked to divide up tasks in theory/computer calculation on the one hand and experimental testing on the other. Typical subject areas are given below. The approach was very exible, a particular aspect being worked out in consultation with the teacher, and the actual work carried out under the guidance of a graduate student. The projects were for approximately 1 month, after which the group compiled a single report synthesising the work of all of the participants. The work was then presented in a series of short oral presentations; instruction was given to assist in preparing the report and making the presentation, which was of a length and style similar to that of conference presentations. The advantage of this approach was that students were generally very motivated to learn the theoretical part and to carry out a successful project. Learning to work in a team and acquiring communication skills were other advantages of this approach. The required material was largely accessible from research laboratories. Computing requirements were modest and in all cases could be met with

2002 by CRC Press LLC

438

Fundamentals and Applications of Ultrasonic Waves

the departmental PCs. The laboratory equipment available included: 1. HP Model 4195A Network/Spectrum Analyzer 2. One of the following: a. MATEC RF tone burst ultrasonic generator and receiver (10 to 90 MHz) b. RITEC RAM 10000 tone burst ultrasonic generator and receiver (1 to 100 MHz) c. UTEX UT 320/340 Pulser/receiver or equivalent, such as those produced by Panametrics or Metrotek (tone burst systems are ideal for this type of experiment as they allow easy control and variation of the frequency and quantitative verication of frequency-dependent effects) 3. Standard RF attenuators, cables, etc. 4. Laboratory oscilloscope, ideally digital scope with FFT capability, such as the 300 MHz LeCroy digital oscilloscope A list of typical projects is given below, with notes on particular aspects that can be easily investigated and compared with theory. This list is by no means exhaustive, and it is easy to extend it by the procurement of modest additional resources, such as focusing transducers, additional buffer rods, means of temperature variation and control, magnetic eld etc. 1. Transducer characterization It is useful to obtain a collection of piezoelectric transducers from various sources. Commercially packaged resonators can easily be obtained in the range 1 to 20 MHz, as can unmounted transducers, longitudinal or transverse, with either fundamental or overtone polish from suppliers such as Valpey Fisher Inc. In the latter case, LiNbO3 transducers with a fundamental in the range of 5 to 15 MHz and with overtone polish are the most convenient choice, typically 5 or 6 mm in diameter. Transducer characterization is best made with respect to a welldened equivalent circuit. This could be a series resonant circuit in parallel with the static capacitance (ButterworthVan Dyke equivalent circuit for resonators) or the full Mason Model for a loaded transducer. Suggested experiments include: a. Characterization of the resonance of an unloaded transducer (resonator) using the network analyzer; determination of transducer parameters by measurement of amplitude and phase response, as well as series and parallel resonant frequencies; identication of harmonic frequencies; effects of liquid loading on the resonance for both longitudinal and transverse polarization.

2002 by CRC Press LLC

Appendix C

439

b. Frequency response of a transducer glued to a buffer rod, with air loading on the opposite face. Points to verify include: (i) Frequency response of the odd harmonics. (ii) Use of inductors/RF transformers to increase the transducer response. (iii) Observation of echoes in the buffer rod. (iv) Comparison of shape of the rst echo with that of the exciting RF pulse; effect of bond quality on the echo shape. 2. Bulk acoustic wave (BAW) propagation Experiments in this section are based around the use of a transducer mounted on the end of a buffer rod. Ideally, buffer rods made of materials such as fused quartz, sapphire, etc. can be obtained with end faces optically polished and parallel from suppliers such as ValpeyFisher. Otherwise, for studies in the low MHz range, it is possible to machine and polish the end faces of materials such as perspex, duraluminium, brass, stainless steel, etc., using standard workshop practices to obtain usable echo trains. Duraluminium is particularly useful due to its low attenuation and its machinability. The buffer rod should have dimensions of the order of 1 cm in length and 1 cm in diameter; these dimensions are not critical and should be chosen so that the rod diameter is signicantly greater than that of the transducer, with the buffer long enough so that clearly separated, nonoverlapping echoes are observed on the oscilloscope. Longitudinal transducers with overtone polish and a fundamental frequency of 5 or 10 MHz are recommended for the experiments of this section. Such experiments include: a. Mount the transducer on the end of the buffer rod with a suitable ultrasonic couplant; vacuum grease or silicon oil are convenient, as they give a good bond at room temperature which is stable for a few hours and is easily changed. The transducer bond can be improved by wringing it onto the buffer surface using a soft rubber eraser, for example. b. Tuning the generator to the transducer fundamental frequency; observing echoes. Existence or not of an exponential decay of the echo amplitudes should be registered. Transducer bond can be optimized to give maximum echo amplitude. c. Estimation of VL and comparison with the handbook value; estimation of absolute and relative error. d. Using the same transducer bond as above, steps (b) and (c) should be repeated at odd harmonic frequencies up to the maximum attainable values with the ultrasonic generator used. Variation of the overall modulation of the echo train and the number of

2002 by CRC Press LLC

440

Fundamentals and Applications of Ultrasonic Waves echoes is particularly signicant. How can these be explained for the particular buffer rod used? e. For a machined buffer rod, remachine one end face so that there are now nonparallel end faces to within a degree or so. Repeat step (d) and explain any observed variation in the modulation of the echo train. 3. BAW reection and transmission These experiments are most conveniently carried out with a buffer rod with the end opposite the transducer partially immersed in a liquid. In this conguration it is possible to measure reection at normal incidence and transmission and reection from a plate immersed in the liquid. The appropriate theoretical values can be calculated using the theory of Chapter 7. Recommended experiments are: a. Use a 5- or 10-MHz longitudinal wave transducer bonded to one end of the buffer rod as in experiment #2; prepare buffer rods of plexiglass, duraluminum, and stainless steel, which form a convenient trio of buffer rods that have low, medium, and high acoustic mismatch to liquids such as water; design and construct sample holders to enable the far end to be immersed in a uid bath. b. Pulse echo experiments at low frequency in bare buffer rod; adjustment for obtaining maximum number of echoes. c. Exposure of the end of the buffer rod to the uid in question; recording of the echo pattern and comparison with that for the unexposed rod; calculation of the reection coefcient for each echo; draw conclusions on the accuracy of the method vs. echo number. d. Systematic study of the three buffer rods against three different liquids with signicantly different acoustic impedances; compare with theory. e. For a given liquid-solid combination at a given frequency, calculate the material and thickness of the layer needed to minimize the reected signal; attempt to verify this result experimentally. f. Repeat (c) for the case where there is a reecting plate immersed in the liquid; trace possible ray paths for various returning echoes in the buffer; compare with experiment to identify all observed echoes; estimate the reection coefcient at the uidplate interface. 4. SAW device fabrication, measurement, and sensor applications IDTs operating at about 50 MHz can be made very easily in a standard darkroom using photolithography techniques using the following materials; Y-Z LiNbO3 SAW plates, about 15 mm long, 10 mm wide, and 0.5 mm thick; mask for standard transmitterreceiver transducer

2002 by CRC Press LLC

Appendix C

441

design, required to have an impedance of 50 when used with the chosen substrate; 10 nger pairs for two transducers about 10 mm apart, aperture approximately 5 mm for Y-Z lithium niobate. The steps for transducer fabrication are as follows: a. Clean the substrate with acetone and soak in methanol. b. Deposit approximately 200 nm lm of aluminum by ash evaporation. c. Deposit a photo-resist lm by pipette on the substrate in yellow light conditions. Incline the substrate to drain off excess photoresist. d. Bake the photo-resist lm at 120C for at least 15 min to harden the lm. e. Clamp the mask on top of the photo-resist lm and expose to ultraviolet light for the recommended time. f. Remove the mask in darkness and dip the substrate for a few moments in NaOH to remove the exposed portions of the photoresist. The remaining photo-resist protects the aluminum during etching. g. Etch the plate in a solution of HNO3 , HCl, and H2O, removing it rapidly at the required moment to avoid overetching. h. Thoroughly rinse the plate and then remove excess photo-resist with a small amount of NaOH. If sufcient time and facilities are not available for in-house fabrication, then nished SAW plates with IDTs can be bought from the manufacturer. A number of instructive experiments can be carried out using the SAW device. These include: a. Testing the frequency response with the network analyzer: a power splitter can be used to provide a reference signal, enabling tracing of the insertion loss as a function of frequency. The result should be compared with the expected theoretical response. b. Transducer matching: if the impedance is 50 , then it remains to tune out the static capacitance, here about 0.3 pF. This is most conveniently done with a variable inductance in series with the transducer. c. Timing ight measurement: the transmitting transducer is excited by a low-amplitude tone burst. To prevent burnout of the IDTs it is advisable to use a xed attenuator (PAD) of 10 or 20 dB in series with the input if high power sources such as the Matec are used. The source and receiver are tuned to the IDT central frequency. Absolute and relative Rayleigh wave velocity of the substrate can be measured in this way. Compare the measured value with that given in the tables.
2002 by CRC Press LLC

442

Fundamentals and Applications of Ultrasonic Waves d. Liquid loading by leaky waves can be demonstrated very effectively by putting a drop of water on the substrate between the electrodes; the propagated acoustic signal immediately disappears. It is instructive to repeat the experiment with liquids of lower acoustic impedance and increased volatility, such as acetone. e. Transforming the SAW device into an oscillator is easily accomplished by placing an RF amplier into a feedback loop connected between the two IDTs, in series with an RF attenuator. The attenuator setting must be low enough so that the loop gain exceeds the losses. Interesting conclusions can be drawn from the behavior of the signal across the device observed on an oscilloscope at high and low values of attenuation. The oscillation frequency should be measured with a frequency counter. f. Using the SAW device as a temperature sensor is possible due to the temperature dependence of the sound velocity in LiNbO3, which gives rise to a predicted temperature variation of the propagation time as 94 ppm/C. In light of the discussion in Chapter 13, this can easily be measured as the frequency shift of the oscillator in the preceding section, which is directly proportional to the delay time, hence the velocity variation. The SAW substrate can be placed on a cold plate and then a hot plate to cover a temperature range of about 100C, around room temperature. A calibrated thermometer should be attached to the SAW substrate, which should then be cycled slowly in temperature. Readings of the frequency shift at various xed temperatures should be made; the frequency shift vs. temperature should give a linear variation of a value close to that predicted. 5. Advanced experiments There are a number of more advanced experiments of potential interest, but they rely on the availability of specialized equipment. These possibilities will be mentioned only briey here; they have been found to be relatively easy to set up and to be instructive, even if carried out at an elementary level. a. Acoustic radiation measurement by hydrophone and water tank If an ultrasonic immersion test bath with x-y-z micropositioners is available, then this provides a suitable means for measuring the acoustic radiation patterns of immersion transducers. Immersion transducers can be purchased from vendors such as Panametrics. Detection is carried out by a needle hydrophone which contains a small pointlike piezoelectric detector such that it does not perturb the acoustic eld. Measurement of the radiation pattern of a transducer and comparison with theory for both near eld and far eld is feasible.

2002 by CRC Press LLC

Appendix C

443

b. Acoustic microscopy: if a low-frequency acoustic microscope is available, there are a number of simple experiments that can be performed with few complications. The most direct of these is experimental verication of the resolution of an acoustic lens. The lens is focused on the edge of a plate and scanned in a direction perpendicular to the plate edge at constant height. It is important that the lens axis be vertical and the plate accurately adjusted to be horizontal. Over the plate the reected amplitude is constant, and it then decreases continuously to zero as the focal point is scanned away from the plate edge into the bulk liquid. The width of the resulting curve gives the resolution. This can then be compared with theory for the lens opening and frequency used. A second instructive experiment, done in the same conguration as above, is the measurement of a V(z) curve. The lens axis is centered on the middle of the plate, roughly in the focal position. In this case the x,y coordinates of the lens are held xed, and the plate is scanned along the z axis toward the plate. A series of maxima and minima are observed as described in Chapter 14. The result can be used to deduce the Rayleigh wave velocity in the plate, which can then be compared to the tabulated value. c. Schlieren imaging: if a Schlieren imaging system is available, then it is the tool of choice to image the propagation paths of ultrasonic waves. Typical operation is at 10 MHz in a water bath. Phenomena such as direct reection and Schoch displacement are easily observable, as is the imaging of a focused acoustic beam. 6. Topics for term papers If suitable ultrasonic equipment is not available for experimental projects, then term papers involving literature searches and summaries on specic topics are useful. Possible topics include: Ultrasonic tomography Fresnel acoustic lens SAW biosensors SAW gas sensors SAW temperature senors Acoustic spectrum analyser Laser generation of ultrasound Equivalent circuit model of IDTs Acoustoelectric effect

2002 by CRC Press LLC