Introduction

In the determination of durability parameters (e.g. frost resistance or alkali- aggregate reaction) non-destructive acoustic test procedures such as ultrasonic scanning,resonance frequency and natural-period-of-vibration techniques show different pictures of the changes occurring in the microstructure of building parts or laboratory specimens.This is due to the following aspects:

• The depths of damage vary depending on the environmental conditions and/or on the loading regime in the laboratory (e.g. changes in temperature and moisture), the material composition and the density of the concrete surface.
• Depending on the testing procedure chosen the propagation of mechanical vibrations is measured in different parts of the specimen.
• The geometry (ratio of longitudinal to lateral dimensions) of the laboratory specimen must be able to show depths of damage that are relevant in practice;changes in the dynamic transverse dilatation may influence the calculation of the dynamic modulus of elasticity in different ways.

In the ultrasonic scanning procedure (arrangement of vibrator for direct irradiation) the propagation velocity of ultrasonic impulses between transmitter and receiver along a defined irradiation axis is determined. The application of the resonance procedure, on the other hand, permits the assessment of the vibration behaviour of the entire specimen. Considering the material changes, the differences between the testing procedures depend to a large degree on how fast the damage proceeds in the specimen (progress of damage). After the damage has reached the chosen irradiation axis, the test results resemble each other. This fact can clearly be observed in the determination of durability parameters of concrete (e.g. frost resistance) (fig. 1 and 2). When specimens are stored under changing climatic conditions to study the delayed formation of ettringite in concrete, the progress of damage depends on the penetration of water which is required for the reaction causing the damage to the microstructure. In this case the progress of damage depends on the intensity of the drying and rewetting. Data on the achieved depth of damage have not yet been available. For a practice-relevant assessment of the damage caused to the microstructure by storage under changing climatic conditions the recording of the depth of damage in the first approximation as a 2- dimensional representation on laboratory specimens would be very important. That means that principally new requirements have to be made on a non-destructive testing procedure with regard to the distance between measuring points (spatial resolution), the measuring speed and the coupling of the sensors so that the distribution of the elastic parameters decribing the state of the microstructure may be recorded and graphically represented. In current research projects it is studied how the durability parameters are effected by a concrete microstructure predamaged by cracks. This requires knowledge about the distribution of the elastic parameters with high spatial resolution depending on the load, e.g. compression, tensile bending or tensile splitting stresses [1, 2].

Fig 1: Gradient of damage of a concreteprism 8 x 8 x 25 cm3 after frost attack (measurement of impulse transit time)

Fig 2: Change of natural frequency of the bending wave during the test on frost resistance, concrete prism 8 x 8 x 25 cm3

Requirements on the ultrasonic scanning procedure

The requirements on the measurement technique for recording:

• the distribution of elastic parameters (sound velocity, dynamic modulus of elasticity) for the assessment of the homogeneity of building parts / laboratory specimens,
• gradients of damage and/or the depth of damage in laboratory specimens as a result of durability tests,
• the distribution of elastic parameters (sound velocity, attenuation, distribution of frequencies) in laboratory specimens depending on a compressive, tensile bending or tensile splitting load

may be met by the systematic further development of the ultrasonic transmission procedure to a non-contacting measuring procedure. The application of laser techniques for the excitation and reception of sound waves will gradually solve the problems mentioned above. The requirements on an optical ultrasonic measuring system may be defined as follows:

• high speed of measurements by eliminating the mechanical coupling,
• short distances between measuring points (high optical resolution),
• controllable signal attenuation,
• excitation of different types of waves (P- and S- waves ) in one measuring process

The development of an ultrasonic laser technique suitable for the building site would offer the advantage that the time-consuming marking of congruent measuring grids might be abolished and that measurements in building construction and civil engineering could be carried out without expensive scaffolding. The possible distances between excitation laser and reception laser are at the moment under investigation. The stepwise development of an optical ultrasonic procedure is carried out in a Collaborative Research Center 524 at the Bauhaus University in Weimar in the project part (TP) B1, promoted by Deutsche Forschungsgemeinschaft (DFG).

Laser-induced excitation of sound waves

The laser-induced excitation of sound waves is described for metallic materials by KRAUTKRÄMER & KRAUTKRÄMER [3], SCRUBY [4], DEUTSCH et al. [5] and MEYER et al. [6]. The mechanism of action is based on the partial absorption of the electromagnetic energy in an extremely thin surface layer of the specimen which is transformed into heat (interior photo effect). This transformation process proceeds in a period of time that is very short compared to the laser pulse duration and the time required for thermal expansion. In the area of the laser spot thermomechanical stresses are produced that excite vibrations in the adjacent volume elements. Density of energy and pulse duration have a decisive influence on the developing type of wave and its frequency content. If the excitation of vibrations is accompanied by the formation of plasma the longitudinal direction of the vibrations predominates. Acc. to [6] the range of transition for aluminium lies at > 10 MW/cm², the optimum range of excitation intensities is between 200 and 400 MW/cm². HABERER et al. [7] produced ultrasonic impulses in aluminium alloys by a Nd:YAG laser with pulse power densities between 400 and 600 MW/cm² in the high-temperature range up to a maximum of 1400° C.Publications on the laser-induced excitation of sound waves in concrete are limited to single cases. POPOVICE/ ROSE [8] refer only to energy densities with resulting formation of plasma (ablative process) which excite sound waves in the concrete. The required pulse power density given is 2000 MW/cm².

The first experiments in the framework of SFB/TP B1 were carried through in cooperation with the Fraunhofer Institute for Material and Beam Technology (IWS) in Dresden. The principal proof of the excitability of sound waves in building materials was to be furnished by commercially available laser equipments, using the experience gathered in the laser cleaning of natural stone surfaces and in the safety measures taken for the operation with laser beams on building sites. In the first basic tests the laserinduced excitation of vibrations was demonstrated in various materials (concrete, aerated concrete, brick, marble). The vibrations were excited with a Neodym-YAG laser with a wave length of the laser beam of 1064 nm and laser energies of 200 to 400 mJ. In this test stage the signals were received by a broad-band vibration receiver of the type UPE of the enterprise GEOTRON by mechanical coupling. Fig. 3 shows as an example the ultrasonic signal produced and the amplitude spectrum of the tested concrete. The height of the first amplitude and the distribution of the amplitudes in the frequency range of up to 130 kHz is typical of concrete and corresponds to the sound excitation of conventional piezo crystals. The sound velocity is v_L = 4856 m/s, the sound way s = 250 mm. Another advantage of the laser excitation is the technical possibility of exciting P- and S-waves in one measuring process. The evaluation of the S-wave velocity with a laser vibrometer is at the moment under investigation.

Fig 3: Laser-induced sound waves in concrete: a) graph of vibration of ultrasonic signal, b) FFT.

In further tests with the same measuring set-up the influence of the beam parameters on the pulse power density was investigated:

• wave length [l]
• laser power [P]
• irradiation area [A] (parallel beam, focussing, incidence angle)
• pulse duration [t], pulse repetition frequency [f]

The test results were published in [2] and [9]. The most important result of the investigations is that the range of pulse power density required for the excitation of sound waves in concrete can now be limited to 100 to 300 MW/cm². Repeated pulsing on the same point of excitation leads after 128 pulses to a change in the first amplitude of ± 2 mV (4%). The advantages of the laser excitation as to the frequency content (fig.4) of sound waves become, however, only evident in the implementation of the noncontacting signal reception (optical measuring system).

Laser reception of sound waves

The investigations into the laser reception of sound waves were made with a laser scanning vibrometer PSV 300 F. This type of single-point interferometer is a modified form of the Mach-Zehnder interferometer. The high-power scanning head is mounted on a heavy tripod with motor-operated swivel-tilting head. The maximum deflection of the scanner mirror is horizontally and vertically ± 20°. So far the distances between scanner head and measured object for recording the frequency and time signals have been between 1 and 4 m. By means of the measuring mode "time" the impulse transit time may be recorded at any measuring point in the measuring grid (fig. 4,6).

Fig 4: PSV 300 F.

Fig 5: Computer-controlled test board with concrete specimen.

Since in the ultrasound signals the beginning of the vibration of the P wave has to be automatically recognised and evaluated during the scanning process, the signal-to-noiseratio in the vibrometer signal is of great importance. Three main aspects are under investigation to permit the evaluation of the signals:

• Application of suitable coatings (flat coating) to provide for the retroreflecting effect. For this purpose micro-glass-balls are bonded in a thin layer of reflecting paint. Up to now repeated measurements after different kinds of storage (heat at T = 60° C, water, cloud chamber) have not shown any remarkable changes in the reflected signal intensity. On a test area of 7 x 7 cm² with a measuring grid of 8 x 8 measuring points on the surface of a concrete prism of 8 x 25 cm² a mean voltage signal of 4.26 ± 0.05 V was measured (corresponds to the intensity of retro sheets).The resulting quality of the ultrasonic signal can be seen in fig. 6. The present state of the tests seems to be promising for studies of microstructures in a limited field (e.g. in loading tests). So far commercial self-adhesive retro sheets meet the requirements best, but they can only be employed to a limited extent. The reasons are the higher signal attenuation (layer of adhesive), high purchase costs, and the peeling off of the sheet when stored in changing climates and/or under mechanical loading,
• Application of the software module "averaging of time signals" so that by the function "signal enhancement" dropouts in the time data acquisition may be avoided,
• Extension of the PSV software by adaptation of the speckle tracking (deflection of the laser beam around its central position) under consideration of the topology of the concrete surface. At present the deflection of the laser beam is 50 mm related to its central position. For the exact assessment of the influence of the topology of the concrete surface, concrete specimens lying on a computer-controlled test board (fig.5) were driven along a laser beam at a constant rate of 1mm/s. As an example, the intensity distribution of a laser beam reflected from a shuttered concrete surface is shown as a section (2 mm of 50 mm) in fig. 7. It was found out that the mean spacing of dropouts was about 100 to 200 mm. In cooperation with Dr. Berendt from the enterprise POLYTEC the tests were carried out on shuttered, sawn and finished concrete surfaces as well as on brick and marble specimens [10].

Fig 6: Ultrasonic signal of a concrete surface covered with micro-glass-balls without low-pass filter.

For making the time signals evaluable for the determination of the impulse transit time in the investigation of structures, priority should be given to the time data averaging, and the reflexion properties of the test area should be improved by the application of developer sprays.The remainders may easily be removed by blowing them off with air.

Fig 7: Intensity of the laser beam reflected from a shuttered concrete surface. Movement of concrete specimen 1 mm/s.

Non-contacting ultrasonic measuring system

Laser techniques were step by step included in the recording of impulse transit times:

• Stationary mechanical coupling of a sound transmitter by coupling device or adhesive for the excitation of sound. Scanning of an opposite test area by the PSV.For the assessment of the distribution of the impulse transit times the time measurement and the laser position were triggered.
• Implementation of the laser ultrasonic procedure by the excitation of sound with a Nd:YAG laser and reception with the PSV 300F (fig. 8). The laser-induced sound excitation in the tested specimens led to broad-band ultrasonic signals with amplitudes in the frequency spectrum up to 300 kHz (fig. 9). The range of velocity of ± 0.006 m/s of the time signal (fig. 9a) corresponds to a voltage range of ± 240 mV.

Fig 8: Laboratory tests with the laser-induced ultrasonic procedure on concrete cubes of 10 x 10 x 10 cm3.

Fig 9: Vibrometer signal of a laser-induced sound wave after passing a concrete cube of 10 x 10 x 10 cm3 a) time signal, b) FFT.

The laser-induced ultrasonic procedure is controlled by the data management system (fig. 10). The laser beam of the excitation laser is coupled either directly or by special transmission systems in a scanner with position feedback. Correspondingly to the scanning head of the laser vibrometer this scanner is mounted on a tripod with motoroperated swivel-tilting head and is centrally controlled. Both scanning systems provide the coordinates of the excitation and the reception points in a data file. Together with the automatically recorded impulse transit time they are stored in an appropriate table of measured values. Depending on the scanning mode used (one to n scanning points of the vibrometer may be allocated to an excitation point) this table of measured values may be used for the graphic representation of a 2 or 3-dimensional distribution of elastic parameters in a separate software and for their evaluation in different sectional areas (different depths in the specimen). This procedure will first be tested in the laboratory within the framework of the already discussed investigations on the durability of concrete. The long-term objective is the application of the ultrasonic laser technique on the building site.

Fig 10: Data stream pattern of the laser-induced ultrasonic procedure.

Conclusions

The conventional ultrasound transmission procedure for the determination of the distribution of elastic parameters (dynamic modulus of elasticity) in laboratory specimens has considerable disadvantages due to the mechanical coupling of ultrasonic transducers: it is highly time consuming to achieve the required density of data and measuring points and the varying attenuation of the signal cannot be mastered. As a rule, the investigation of building parts (e.g. concrete columns) requires additional efforts as the test areas have to be scaffolded and the measuring points have to be marked to achieve congruent measuring grids.

When building parts in use are repeatedly exposed to drying and rewetting a gradient of damage starting from the concrete surface may occur depending on the penetration depth of moisture. The assessment of the microstructure of this concrete layer plays an important role in investigations into the durability of concrete (e.g. frost resistance,alkali - aggregate reaction).

In order to eliminate the influence of the mechanical coupling and to obtain high speeds of measurements and short distances between measuring points, laser techniques are used which allow the development of ultrasonic scanning procedures into noncontacting measuring procedures. An additional advantage is that building parts can be examined from a greater distance without scaffolding and marking of measuring points.In the Collaborative Research Center 524 at the Bauhaus University in Weimar a laserinduced ultrasonic measuring procedure is being developed. Various methods of exciting and receiving sound waves have been tested. While maintaining the mechanical coupling a measuring process was conceived by which the test areas are scanned by a laser vibrometer. To obtain the complete series of measurements a Nd:YAG laser produced sound waves in concrete specimens at a distance of 2 to 3 meters which were received without contact by a laser scanning vibrometer at the same distance. The mean pulse power density for the vibrational excitation was 100 to 300 MW/cm². The aim in the future is to develop a compact measuring system suited for the building site.

References

1. AUBERG, R.: Zuverlässige Prüfung des Frost- und Frost-Tausalz-Widerstandes von Beton mit dem CDF- und CIF-Test. In: Fachbereich Bauwesen der Universität GH Essen: Dissertation 1998
2. ERFURT, W.: Erfassung von Gefügeveränderungen in Beton durch Anwendung zerstörungsfreier Prüfverfahren zur Einschätzung der Dauerhaftigkeit. In: F. A. Finger-Institut für Baustoffkunde, Bauhaus-Universität Weimar: Dissertation 2002
3. KRAUTKRÄMER, J.; KRAUTKRÄMER, H.: Werkstoffprüfung mit Ultraschall. 5. Auflage Berlin, Heidelberg, New York, London, Paris, Tokyo: Springer-Verlag, 1986
4. SCRUBY, C. B.: Some Applications of Laser Ultrasound. In: Ultrasonics 27 (1989) Nr. 7, S. 195-209
5. DEUTSCH, V.; PLATTE, M.; VOGT, M.: Ultraschallprüfung, Grundlagen und industrielle Anwendungen. Heidelberg, Berlin, New York: Springer-Verlag, 1997
6. MEYER, E. H.; RADTKE, U.; CROSTACK, H.-A.: Zerstörungsfreie Werkstoff- und Bauteilprüfung mittels laserinduziertem Ultraschall und holographischer Schallfeldabbildung. In: Univ. Dortmund, FB Maschinenbau, Lehrstuhl für Qualitätswesen: Forschungsverbundprojekt 1997
7. HABERER, B,; PAUL, M.; WILLEMS, H.; ARNOLD, W.: Measurement of internal friction in polycrystalline materials using laser-generated ultrasound. In: Journal of Alloys and Compounds 211 (1994) H. 636, S. 636-639
8. POPOVICS, J. S.; ROSE, J. L.: A Survey of Developments in Ultrasonic NDE of Concrete. In: IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 41 (1994) Nr. 1, S. 140-143
9. ERFURT, W.; KÖHLER, W.; STARK, J.: Berührungslos arbeitende Ultraschallmeßverfahren für die Bauwerksdiagnose. In: Thesis, Wiss. Z. d. Bauhaus-Universität Weimar 47 (2001) Nr. 5/6, S. 196-203 Erfurt, Köhler, Stark S. 10
10. BEHRENDT, R.: Laserversuche Polytec Berlin, 2002