Direct and Indirect Ultrasonic Measurements on Plastered and Unplastered Concrete Columns: A Case Study of an Inhabited Building in Italy

NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 1 Direct and Indirect Ultrasonic Measurements on Plastered and Unplastered Concrete Columns: A Case Study of an Inhabited Building in Italy Vincenza A.M. LUPRANO1, Antonio PALERMO2, Valerio PFISTER1, Tommaso MARCIANÒ1, Saverio MAZZARELLI1, Angelo TATÌ3, Ivan ROSELLI3, Fernando SAITTA3, Alessandro MARZANI2 1 ENEA, Brindisi Research Centre, Cittadella della Ricerca S.S.7 km 706, Brindisi, Italy; [email protected], [email protected], [email protected], [email protected] 2 Department of Civil, Chemical, Environmental and Materials Engineering - DICAM, University of Bologna, viale del Risorgimento 2, Bologna, Italy; [email protected], [email protected] 3 ENEA, Casaccia Research Centre, via Anguillarese 301, S. Maria di Galeria – Rome, Italy; [email protected], [email protected], [email protected] Abstract Assessing the condition of concrete in place is an important part of building safety verification programmes. For the study of existing concrete structures, it is important to know the strength of different elements. The assessment of the compressive strength of structures can be carried out by destructive coring tests in varying quantities and calibrated indirect methods, combining destructive coring with nonor semi-destructive techniques. When planning ultrasonic indirect measurements, it must be taken into account that in inhabited houses the columns are covered with plaster that often cannot be easily removed and restored (even more so if, in addition to being a protective covering, it also has a decorative function): indirect measurements can be performed, but it is necessary to evaluate the impact of the presence of plaster. An analytical model has been developed to predict the surface wave propagation velocities in concrete columns, both with and without a plaster covering. The aim is to support the interpretation of time-of-flight measurements in indirect ultrasonic testing. Keywords: concrete assessment; ultrasonic measurements; building safety; analytical models; ultrasonic dispersive curve. 1. Introduction Assessing the condition of concrete in place is an important part of building safety verification programmes. For the study of existing concrete structures, it is important to know the strength of different elements. The assessment of the compressive strength of structures can be carried out by destructive coring tests in varying quantities and calibrated indirect methods, combining destructive coring with nonor semi-destructive techniques. The use of core drilling is timeconsuming, labour-intensive and may weaken the existing concrete structure. Alternatively, several nonand semi-destructive techniques are available for the in-situ determination of compressive strength that must be appropriately correlated with destructive testing. Within the group ‘TC249-ISC on non-destructive in situ compressive strength assessment of concrete’ of RILEM (The International Union of Laboratories and Experts in Construction Materials, Systems and Structures), the topic of non-destructive in-situ compressive strength assessment of concrete was addressed internationally, in particular for the estimation of the mean and standard deviation compressive strength. A new method for selecting the location of test points from which to extract cores was validated to minimise the risk of error in strength assessment [1]. In the European REHOUSE (“Renovation packagEs for HOlistic improvement of EU’s bUildingS Efficiency, maximizing RES generation and cost-effectiveness”) project it has developed an integrated methodology to define the real state of the Italian case study sites useful More info about this article: https://www.ndt.net/?id=31676 e-Journal of Nondestructive Testing - ISSN 1435-4934 - www.ndt.net © 2025 The Authors. Published by NDT.net under License CC-BY-4.0 https://creativecommons.org/licenses/by/4.0/ https://doi.org/10.58286/31676 NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 2 both for carrying out the redevelopment project and for constructing the digital twin of the building useful for the maintenance phase. The case study is a social building located in Margherita di Savoia (Southern Italy). The importance of developing an effective methodology, as non-destructive as possible, for assessing the strength of concrete also has the aim of minimising inconvenience to tenants, being absolutely non-invasive in measurements and still being accurate [2]. An extensive and tailored campaign of non-destructive measurements can reduce the invasiveness (i.e. a high number of core drillings on the structural elements, columns and/or beams, and destructive tests on the extracted samples) that the Standards [3] would require with the evaluation of the compressive strength. In order to be able to apply this methodology on an inhabited building, where it is necessary not to create damage such as halos on the plaster (due to ultrasound investigations) or small bumps (due to sclerometric investigations), where it is not possible to remove the plaster in order to be able to take measurements directly on the concrete as the standard requires, where the columns are often not accessible from both sides in order to be able to take direct ultrasound measurements, it was used indirect ultrasonic technique. When performing ultrasonic testing from a single accessible side of a concrete wall, with the aim of estimating wave velocity and, subsequently, the concrete’s elastic modulus, it is important to consider that both surface and bulk waves may propagate. To fully exploit the potential of ultrasonic testing in such conditions, it is essential to understand and characterize the dispersive and multimodal wave propagation within the plaster–concrete wall waveguide. This paper reports the results of indirect ultrasound measurements carried out on concrete columns of a scale model of a building, in the presence of infill walls, before and after the application of a plaster coating. To verify the results obtained, numerical and analytical simulations were performed to understand the potential and effectiveness of the indirect ultrasonic measurement technique even in the presence of plaster. 2. Materials and Methods Ultrasonic velocity is a parameter that can help in the evaluation of the compressive strength of concrete, for existing structures. The standard [4] provides for the possibility of carrying out indirect measurements for the characterisation of a concrete structural element by repeating several indirect ultrasonic measurements at different distances between the probes. The possibility of carrying out indirect measurements is an important aspect when conducting assessments on an existing structure, even more so when it is inhabited: often external wall stratigraphy, or internal infill, does not allow access to two opposite faces of a column and, consequently, to finalize direct ultrasonic measurements [2] unless significant interventions are carried out on the wall structure. When planning ultrasonic indirect measurements, it must be taken into account that in inhabited houses the columns are covered with plaster that often cannot be easily removed and restored (even more so if, in addition to being a protective covering, it also has a decorative function): indirect measurements can be performed using as coupling medium a rubber material, but it is necessary to evaluate the impact of the presence of plaster. The support of numerical analyses allows for a better interpretation of the measurements and the definition of the aspects to pay attention to when it is necessary to proceed with in-field measurements on a structure where it is not possible to remove and restore the plastering of the structural elements. NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 3 In this work, ultrasonic indirect measurements were performed on columns of a scale model of a building (Section 2.3), in the presence of infill walls, before and after the application of a plaster coating. 2.1 Indirect Ultrasonic Tests Ultrasonic tests were performed using a low-frequency instrument IMG5200CSD (IMG Ultrasuoni srl, Mandello Lario, Lecco, Italy) equipped with 50 kHz probes. A pacometer investigation was conducted on each column to obtain information on the position of the reinforcing bars before starting the ultrasonic velocity measurements. Data processing consists of measuring flight times from which the ultrasonic velocity of the investigated element was calculated. For each measurement configuration, at least 3 distinct measurements are carried out and their signals recorded: the velocity at the column is the average value of the measurement values. Indirect measurements were performed along the column axis: one of the probes was kept fixed and measurements were carried out by modifying the distance of the second probe each time. When performing ultrasonic testing from a single accessible side of a concrete wall (i.e., using indirect transmission), with the aim of estimating wave velocity and, subsequently, the concrete’s elastic modulus, it is important to consider that both surface and bulk waves may propagate from the transmitter (TX) to the receiver (RX). The recorded signal at the RX may therefore represent a superposition of multiple wave modes, with their relative contributions depending on the TX–RX spacing along the wall surface. This scenario becomes even more complex in the presence of surface plaster layers, which introduce additional guided wave phenomena characterized by dispersion and multimodal propagation. To fully exploit the potential of ultrasonic testing in such conditions, it is essential to understand and characterize the dispersive and multimodal wave propagation within the plaster–concrete wall waveguide. To this end, the following subsection presents a closed-form dispersion relation developed for this specific layered configuration. 2.2 Analytical and Numerical Methods An analytical model is developed to predict the surface wave propagation velocities in concrete columns, both with and without a plaster covering. The aim is to support the interpretation of time-of-flight measurements in indirect ultrasonic testing. The model is based on a twodimensional, plane-strain, isotropic elastic formulation, in which a thin plaster layer overlays a concrete substrate modeled as a semi-infinite half-space. This assumption is reasonable considering that the column is not free. but has the infill wall as highlighted in Section 2.3. This configuration is chosen to isolate and characterize the surface-guided wave mode supported by the layered system, while neglecting bulk wave propagation (shear and longitudinal modes), which typically dominates in direct-path ultrasonic measurements. The propagation of vertically polarized (Rayleigh-like) surface waves is analyzed in a semiinfinite domain occupying the x–z plane for z > 0. The system consists of a surface layer of plaster (Layer 1), with finite thickness H, overlying a semi-infinite half-space of concrete (Layer 2), as illustrated in Figure 1. Both materials are modeled as homogeneous, isotropic, and linear elastic solids. The plaster layer is characterized by a density 𝜌𝜌1, longitudinal wave speed NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 4 𝑐𝑐𝐿𝐿,1 , and transverse wave speed 𝑐𝑐𝑇𝑇,1 . The underlying concrete half-space is characterized by density 𝜌𝜌2, and bulk velocities 𝑐𝑐𝐿𝐿,2 and 𝑐𝑐𝑇𝑇,2. The formulation of the dispersion law follows a standard derivation for surface waves in layered media. To analyze surface wave propagation, a time-harmonic wave is assumed with angular frequency ω and wavenumber k, propagating along the x-axis. The displacement field in each layer is polarized in the x–z plane and is given by: 𝑢𝑢𝑗𝑗=�𝑢𝑢𝑗𝑗, 0, 𝑤𝑤𝑗𝑗�, 𝑗𝑗= 1,2 (1) The displacement fields are constructed from scalar �Φ𝑗𝑗� and vector �Ψ𝑗𝑗� potentials: 𝑢𝑢𝑗𝑗=𝜕𝜕Φ𝑗𝑗 𝜕𝜕𝜕𝜕 −𝜕𝜕Ψ𝑗𝑗 𝜕𝜕𝜕𝜕 ,𝑤𝑤𝑗𝑗=𝜕𝜕Φ𝑗𝑗 𝜕𝜕𝜕𝜕 +𝜕𝜕Ψ𝑗𝑗 𝜕𝜕𝜕𝜕 (2) These potentials satisfy the Helmholtz-type wave equations. Assuming harmonic wave motion of the form 𝑒𝑒𝑖𝑖(𝜔𝜔𝜔𝜔−𝑘𝑘𝜕𝜕), the potentials take the following forms for the plaster layer (Layer 1): Φ1(𝑥𝑥,𝑧𝑧,𝑡𝑡)=�𝑎𝑎1 𝑑𝑑𝑒𝑒𝑖𝑖𝑘𝑘𝑟𝑟1𝜕𝜕+𝑎𝑎1 𝑢𝑢𝑒𝑒−𝑖𝑖𝑘𝑘𝑟𝑟1𝜕𝜕�𝑒𝑒𝑖𝑖(𝜔𝜔𝜔𝜔−𝑘𝑘𝜕𝜕) , Ψ1(𝑥𝑥,𝑧𝑧,𝑡𝑡)=�𝑏𝑏1 𝑑𝑑𝑒𝑒𝑖𝑖𝑘𝑘𝑠𝑠1𝜕𝜕+𝑏𝑏1 𝑢𝑢𝑒𝑒−𝑖𝑖𝑘𝑘𝑠𝑠1𝜕𝜕�𝑒𝑒𝑖𝑖(𝜔𝜔𝜔𝜔−𝑘𝑘𝜕𝜕) (3) and for the concrete half-space (Layer 2): Φ2(𝑥𝑥,𝑧𝑧,𝑡𝑡)=𝑎𝑎2 𝑑𝑑𝑒𝑒−𝑘𝑘𝑟𝑟2 ∗(𝜕𝜕−𝐻𝐻)𝑒𝑒𝑖𝑖(𝜔𝜔𝜔𝜔−𝑘𝑘𝜕𝜕) , Ψ2(𝑥𝑥,𝑧𝑧,𝑡𝑡)=𝑏𝑏2 𝑑𝑑𝑒𝑒−𝑘𝑘𝑠𝑠2 ∗(𝜕𝜕−𝐻𝐻)𝑒𝑒𝑖𝑖(𝜔𝜔𝜔𝜔−𝑘𝑘𝜕𝜕) (4) with: 𝑟𝑟𝑗𝑗=�� 𝜔𝜔 𝑘𝑘𝑐𝑐𝐿𝐿,𝑗𝑗�2−1�1 2 , 𝑠𝑠𝑗𝑗=�� 𝜔𝜔 𝑘𝑘𝑐𝑐𝑇𝑇,𝑗𝑗�2−1�1 2 (5) being the vertical component of the wavenumbers. For the concrete half-space, surface wave behaviour is ensured by enforcing evanescent decay of the wavefield with depth, which is governed by the complex vertical wavenumber components associated with the longitudinal and shear modes: 𝑟𝑟2=𝑖𝑖𝑟𝑟2∗, 𝑠𝑠2=𝑖𝑖𝑠𝑠2 ∗ , where 𝑟𝑟2∗=�1−� 𝜔𝜔 𝑘𝑘𝑐𝑐𝐿𝐿,2�2�1 2, 𝑠𝑠2 ∗=�1−� 𝜔𝜔 𝑘𝑘𝑐𝑐𝑇𝑇,2�2�1 2 (6) From the potentials, the displacement components and corresponding stresses are derived using the linear elastic constitutive relations: 𝜏𝜏𝜕𝜕𝜕𝜕,𝑗𝑗=𝜇𝜇𝑗𝑗�𝜕𝜕𝑤𝑤𝑗𝑗 𝜕𝜕𝜕𝜕 +𝜕𝜕𝑢𝑢𝑗𝑗 𝜕𝜕𝜕𝜕�, 𝜎𝜎𝜕𝜕𝜕𝜕,𝑗𝑗=𝜆𝜆𝑗𝑗∇⋅𝒖𝒖𝑗𝑗+ 2𝜇𝜇𝑗𝑗𝜕𝜕𝑤𝑤𝑗𝑗 𝜕𝜕𝜕𝜕 (7) NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 5 Where 𝜇𝜇𝑗𝑗 and 𝜆𝜆𝑗𝑗 are the Lamé parameters in the layers. The following boundary conditions are applied: 𝜏𝜏𝜕𝜕𝜕𝜕,1=0 ,𝜎𝜎𝜕𝜕𝜕𝜕,1 = 0 at 𝑧𝑧=0 𝑢𝑢1=𝑢𝑢2, 𝑤𝑤1=𝑤𝑤2,𝜏𝜏𝜕𝜕𝜕𝜕,1=𝜏𝜏𝜕𝜕𝜕𝜕,2, 𝜎𝜎𝜕𝜕𝜕𝜕,1 =𝜎𝜎𝜕𝜕𝜕𝜕,2 at 𝑧𝑧=𝐻𝐻 (8) Thus, a system of six homogeneous equations is obtained by applying these boundary conditions. A non-trivial solution exists only when the determinant of the coefficient matrix vanishes, yielding the dispersion relation. This equation must be solved numerically to obtain the wave dispersion curves (e.g. the frequency-wavenumber couples) and characterize the phase velocity and field distribution. To verify the accuracy of the analytical predictions, a two-dimensional Finite Element (FE) model is developed in COMSOL Multiphysics to extract the dispersion characteristics of the layered system. The model simulates a representative section of the plaster-coated concrete domain under plane-strain conditions, with a total depth of 𝐻𝐻𝜔𝜔= 0.36 m, as illustrated in Figure 1. The plaster layer is modeled with a uniform thickness H and perfect bonding is assumed at the interface with the underlying concrete half-space. Bloch boundary conditions (BBCs) are applied at the lateral edges of the domain to enforce periodicity and simulate wave propagation along the surface. In Section 3.2, the analytical and numerical models, informed by experimental data, will be used to predict surface wave behavior and estimate phase velocities. Figure 1. a) Schematic of the 2D plane-strain model consisting of a concrete half-space overlaid by a plaster layer of thickness H. (b) Schematic of the corresponding 2D stripe Finite Element (FE) model of the same system 2.3 Reinforced Concrete Bare Frame Scale Model Ultrasonic indirect measurements were performed on columns of a scale model of a building in the presence of infill walls, before and after the application of a plaster coating. The scale model (Figure 2 and 3) is composed of a reinforced concrete 3D framed structure with infills of hollow bricks and mortar. The structure is a two-storey model and is scaled in NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 6 relation to a real building, with a ratio of 2:3 and an interstorey distance of 2 m. The crosssection of the columns is 20×20 cm and that one of the beams is 30×20 cm. The frame structure design is based on non-seismic old codes; therefore, the reinforcement is not detailed to guarantee large ductility at nodal zones. Figure 2. Views of the scale model with highlighting (red solid hatch and circles) of measured columns and surfaces After the initial series of non-destructive test on the bare columns, a coating of lime/cement plaster was applied both internally and externally. The design compressive strength of the concrete is fck=20 MPa, for which a Young modulus of 29962 MPa is assumed. The plaster compressive strength, as declared in the technical specification of the manufacturer, is fm>3 MPa. Therefore, a plausible value of its Young modulus is 5000 MPa or higher. Experimental tests are in progress and will assess the strength values of concrete and plaster. Standard cube samples of concrete with size 15×15×15 cm produced for the verification of the compressive strength were also subjected to ultrasonic tests. Similarly, ultrasonic tests were also performed on parallelogram samples of the plaster mortar (Figure 4). Table 1 reports the results of the ultrasonic tests on concrete and plaster mortar. NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 7 Figure 3. Lower part of the scale model Figure 4. Ultrasonic test of a parallelogram plaster mortar sample Table 1. Longitudinal velocity and density measurements on samples Sample Dimension [cm] Longitudinal velocity [m/s] Density [kg/m3] Concrete 15×15×15 4317 2225 Plaster 16×4×4 1770 1561 By performing Fourier analysis of the received signal on the plaster sample, the frequency peak is at 44 kHz. Instead, the longitudinal velocity measured on the scale model beam without plaster is 4070 m/s. 3. Results 3.1 Non-destructive Measurements: Indirect Ultrasonic Tests Non-destructive tests were performed both before and after the column plastering to characterize the indirect ultrasonic response and compare the results. Indirect Ultrasonic tests were performed on 3 columns of the ground floor: in the Figure 2, the investigated columns are highlighted by red hatch in the north and south views of the structure and by red circles in the plan view. NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 8 For each series of measurements on a surface of a column, indirect measurements were performed, keeping the position of the emitting probe fixed and positioning the receiving probe at 9 different distances, with a pitch of approximately 15 cm. For each series of measurements, two indirect ultrasonic velocity values were calculated based on two procedures (Table 2): as the mean value of the velocities calculated as the set distance between the probes divided by the measured flight time, procedure PR1, and according to [4], procedure PR2. Table 2. Calculated indirect ultrasonic velocities Identification on Figure 2 Indirect ultrasonic velocity [m/s] Procedure PR1 Procedure PR2 Column Orientation [without plaster] [with plaster] [without plaster] [with plaster] 1 S 1937 1751 1835 1402 1 W 2052 1730 2272 1430 2 S 2346 1921 1888 1768 2 E 2409 1587 2221 1303 3 N 2270 1729 1861 1161 3 E 2273 1684 1671 1595 Mean value on the scale model 2214 1734 1958 1443 3.2 Analytical and Numerical Results In what follows, the dispersive properties of the plastered column, specifically the phase velocity of propagating waves as a function of frequency, are predicted using the analytical and numerical models described in Section 2.2. Two distinct scenarios are considered: • Scenario 1: The plaster and concrete velocities and densities are assumed based on the nominal values reported in Table 1. The missing shear wave velocities are computed by assuming a Poisson ratio of 0.2 for both plaster and concrete. • Scenario 2: The plaster velocities are retained as in Scenario 1, while the shear and longitudinal velocities of concrete are adjusted to match the surface wave velocity estimated for Column 3, orientation N. A Poisson ratio of 0.2 is again assumed. For convenience, data of the two scenarios are collected in Table 3. Table 3. Data of the two scenarios Scenario Concrete Plaster Surface velocity (Rayleigh wave) without plaster 𝑐𝑐 𝐿𝐿[m/s] 𝑐𝑐 𝑇𝑇[m/s] 𝜌𝜌 [kg/m 3 ] 𝑐𝑐 𝐿𝐿[m/s] 𝑐𝑐 𝑇𝑇[m/s] 𝜌𝜌 [kg/m 3 ] 𝑐𝑐 𝑅𝑅[m/s] 1 4317 2644 2225 1770 1084 1561 2408 2 4070 2492 2225 1770 1084 1561 2270 NDT-CE 2025 - The International Symposium on Nondestructive Testing in Civil Engineering Izmir, Türkiye, September 24-26, 2025 9 Figure 5. Dispersive properties of surface waves propagating along a plastered column. In (a), the mechanical properties of Scenario 1 are considered, while in (b), those of Scenario 2 are used For both Scenario 1 and Scenario 2, two different plaster thicknesses are considered, namely H=1 cm and H=0.5 cm, in order to assess the influence of this parameter on the dispersive properties of surface waves in plastered columns. The results for Scenario 1 and Scenario 2 are presented in Figure 5(a) and Figure 5(b), respectively. Each figure shows the fundamental surface wave velocity 𝑐𝑐𝑝𝑝 as a function of frequency, ranging from 0 to 200 kHz, for the two plaster thicknesses. The corresponding curves, plotted in black and gray, exhibit pronounced dispersive behavior. The surface wave velocity transitions from the Rayleigh velocity of bare concrete (indicated by the light blue line) at low frequencies to the Rayleigh velocity of bare plaster (indicated by the blue line) at high frequencies. The analytical results are validated via FE results (blue circles) for Scenario 1 H=1 cm. 4. Discussion The dispersive behaviour of plastered columns, Figure 5, can be interpreted in terms of the frequency-dependent penetration depth of surface waves. At low frequencies, surface waves have longer wavelengths and thus penetrate deeper into the material, sampling a larger portion of the underlying concrete. As a result, their velocity approaches that of surface waves in bare concrete. Conversely, at higher frequencies, the shorter wavelengths lead to shallower penetration depths, which are comparable to the plaster thickness. In this case, the waves are primarily confined within the plaster layer and propagate with a velocity closer to that of surface waves in bare plaster. Considering the measured surface wave velocity values reported in Table 2, it can be seen that in both procedures there is a decrease in surface velocity of about 500 m/s when plaster is present. Considering 44 kHz frequency, which is the one that propagates with higher energy in the presence of plaster, it is possible to note that there is a strong dispersion of the signal due to the presence of the plaster, this may be the reason why the measurements are not uniform, but