6th International Conference on Materials Science and Restoration, MSR-VI Aedificatio Publishers 347358 (2003)

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

1E. Knapen, 2R. Lanoye, 3G. Vermeir, 3W. Lauriks and 1D. Van Gemert 1 Reyntjens Laboratory, Department of Civil Engineering, K.U.Leuven 2 Aspirant van het Fonds voor Wetenschappelijk Onderzoek Vlaanderen 3

Laboratory of Building Physics and Laboratory of Acoustics, K.U.Leuven

Abstract
This research aims at the development of granular, porous materials, to be used in combination with floor or wall panels, as sound absorbing elements. The paper describes the frequency dependent sound absorption of porous mortars, with open porosity up to 40 %. Polymers are used to provide the necessary cohesion and mechanical properties. The polymer/cement ratio is shown to have no influence on the sound absorption quality, due to the small influence on the flow resistivity. This type of sound absorbing materials may contribute to a sustainable construction development and to the improvement of acoustical conditions. This will allow to reduce sound propagation in interior spaces or to improve the control of outdoor noise propagation by use of sound absorbing screens. The analysis is valid for sound absorbing road surfaces as well.

347

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

Introduction

The last few years, the use of sound absorbing porous materials for indoor and outdoor applications is stimulated by legislation. To improve the acoustical performance of noise barriers, the acoustical behaviour of industrial and public areas, many absorbing materials, generally fibrous and foamy, are developed. The polymer-modified mortars, proposed in this paper, can be an alternative for more conventional materials. These granular porous materials offer both a good acoustical absorption and structural strength. The rigidity of the mortars reduces the need for supporting frameworks [1,2]. Porous polymer-modified mortars with various compositions are developed to determine the link between the acoustical behaviour and the polymer/cement ratio (p/c), sand/cement ratio (s/c), grain size and degree of compaction. Usually, sound absorbing materials are developed on an experimental basis. In this paper, the validity of the theoretical model of Biot-Johnson-Allard [3] for predicting the acoustical behaviour is applied to porous granular open mortars [4]. In order to measure the acoustical parameters, such as porosity, flow resistivity and tortuosity of the mortars, the conventional test procedures are adapted. These parameters together with the measured absorption coefficient are used to validate the theoretical model and to determine the corresponding characteristic viscous and thermal length. The prediction of the acoustical behaviour, based on the theoretical model, is shown to deliver a very strong similarity with the measured absorption coefficient. This results in a theoretical background for the development of granular acoustically absorbing surfaces, having high sound absorption values for a broad frequency range.

Test program

To produce mortars with a sufficient sound absorption coefficient, the porosity of the mortars is of great importance. Whereas the porosity of foams measures between 95 and 99 %, the porosity of granular mortars is much smaller. In order to create a high porosity, sand with a discontinuous grain size distribution, in the range of 0.8 and 1.2 mm or in the range of 1.4 and 2 mm, is used. Polymers are added to provide the necessary cohesion and mechanical properties. In the present research, polyvinyl acetate (Compaktuna , a product of PTB ) is added to the fresh mixture. The total solid content is 50 % by mass in this thermoplastic latex. The hydraulic binder is a Portland cement, CEM I 42.5R HSR LA. Table 1 shows the composition of the developed mortars, produced in 2 series to examine the reproducibility of the acoustical properties. The p/c ratio varies between 5 and 15 % and the s/c ratio between 5 and 7. The water/cement ratio

348

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

(w/c) is 0,33 for all samples ratio. The use of a dry mortar mixture prevents the pores getting silted up. The samples 2, 5, 22 and 25 have insufficient cohesion and they are not further examined. The samples are prepared in short plastic tubes with an inside diameter of 103.6 mm and a height of 70 mm. The polymer-modified mortar is put in the molds in two layers. After each layer, the fresh mixture is tamped and shocked on a vibrating table. To examine the influence of the degree of compaction, the number of shocks on the vibrating table is varied between 5 and 15. Only the samples 5, 6, 7, 25, 26 and 27 are shocked 15 times. The curing conditions are important, because of the temperature and humidity dependence of the polymer behaviour [5]. A period of high humidity, stimulating the cement hydratation, is followed by a dry curing period, which is needed for the polymer film formation. The samples are cured at a temperature of 20 C and a relative humidity of 90 % for 7 days in the moist room. Then the specimens are demolded and stored at a temperature of 30 C and a relative humidity of 70 % for 7 days. After sawing, grinding and washing out the pores, the samples have a thickness of about 2 cm. They are dried in an oven at 35 C for 2 days before the acoustical tests are carried out.
Table 1: Composition of the samples

SERIES I Sample
0 1 2 3 4 5 6 7 8 9 10 11

SERIES II Grain Sample size [mm]

1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 0.8/2.0 0.8/2.0 0.8/2.0 0.8/2.0 20 21 22 23 24 25 26 27 28 29 30 31

s/c [kg/kg]
5 5 7 7 7 7 7 7 5 5 7 7

p/c [%]
5 10 5 10 15 5 10 15 10 15 10 15

s/c [kg/kg]
5 5 7 7 7 7 7 7 5 5 7 7

p/c [%]
5 10 5 10 15 5 10 15 10 15 10 15

Grain size [mm]

1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 1.4/2 0.8/2.0 0.8/2.0 0.8/2.0 0.8/2.0

349

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

Acoustical models and parameters

3.1 Introduction In acoustics, several models are developed to predict the acoustical behaviour of porous materials. The difference between the models results from the assumptions that are made with respect to the shape and positions of the pores, the stiffness of the skeleton, ... The model of Biot-Johnson-Allard uses five material parameters to predict the acoustical behaviour, namely porosity, flow resistivity, tortuosity, characteristic viscous length and thermal length [3]. In the next paragraphs, these parameters are defined and the test methods and results are briefly described. 3.2 Porosity The acoustical porosity h is the ratio of the volume of open pores to the total volume. The open porosity is considered because only these pores contribute to the absorption of the incident sound. The porosity can be measured with several techniques, but most of them are not suitable for very porous open mortars. For this type of mortars, the assumption can be made that only open pores are present and therefore the method, based on the absolute density, can be used. The absolute density is determined by use of a pycnometer. The porosity of the different samples is given in Table 2. Only the degree of compaction and the binder amount influence the porosity. 3.3 Flow resistivity Darcys law gives the velocity of volume displacement u , which is equal to the material quantity that flows through an area A in a time period t [6]. To measure the flow resistivity , a known airflow is passed through the sample while measuring the pressure difference, as given in eq. (1)

p A Q d

(1)

p [Pa] is the air pressure difference over the sample when an airflow Q [m/s] flows through the sample. A [m] is the area of the sample and d [m] the thickness. In Table 2, the mean value of the flow resistivity for each sample is given. The flow resistivity is mainly determined by the grain size of the sand. Mortars with small grains have small pores and a high flow resistivity.
3.4 Tortuosity The tortuosity describes the influence of the internal structure on the acoustical behaviour of the material. The tortuosity is a measure for the elongation of the pressing way through the pores, compared to the thickness of the sample. is

350

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

Table 2: Porosity, flow resistivity and tortuosity

SERIES I

SERIES II

Sample Porosity Flow Tortuosity Sample Porosity Flow Tortuosity [%] resistivity [%] resistivity [Ns/m4] [Ns/m4]
0 1 3 4 6 7 8 9 10 11 36 35 38 38 35 36 34 34 39 39 12200 17700 12300 10700 21900 16100 48300 40300 35800 33500 3.4 3.2 2.4 1.9 2.7 2.5 2.3 2.9 2.3 1.6 20 21 23 24 26 27 28 29 30 31 37 34 37 36 36 35 35 35 39 40 9200 15900 12300 17300 13500 15600 41000 55200 30000 28700 3.1 3.0 2.5 3.6 2.2 3.3 3.1 2.9 2.8 2.5

the limit at high frequencies of the dynamic tortuosity ( ) . Johnson et al. [7] associate this dynamic tortuosity to a dynamic phenomenon, namely to the inertia force between fluid and rigid structure. The common measuring technique of the tortuosity is based on ultrasonic sound velocities. Since the wavelength of this ultrasonic sound waves and the length of the grains are the same size, this technique is unusable because of scattering effects. Therefore, Champoux [8] has developed a new technique based on the measurement of two resistivities. This method is rather difficult and inaccurate. Experimentally, a measuring error of 0.3 was estimated for this parameter. The tortuosities of the samples are given in Table 2. There is no relation between the tortuosity and the composition of the mortars.

Characteristic lengths

4.1 Introduction The characteristic lengths are introduced in the model of Biot-Johnson-Allard to characterize the shape of the pores of the sound absorbing material.

351

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

4.2 Characteristic viscous length The characteristic viscous length describes the viscous interaction between the fluid and the stiff frame at high frequencies and is defined by eq. (2) [9].

2 v(r ) dV
2

v(r ) dA
2 A

(2)

v(r ) is the microscopic flow velocity of a non-viscous fluid in a pore, v(r ) is the
microscopic flow velocity of the same non-viscous fluid against the wall of the pores. The numerator is evaluated over the total volume of the pores, the denominator over the area of the pores. For porous mortars, this length can not be measured by the conventional method based on wave attenuation at high frequencies, because of scattering. 4.3 Characteristic thermal length The characteristic thermal length characterizes the heat exchange between the saturated fluid and the stiff skeleton at high frequencies and is defined in eq. (3) [3].

' =

2 dV
V

dA
A

2 V S

(3)

S is the frame area in contact to the air in the porous material, and V is the volume of the pores. The conventional measuring technique for foams, the Brunauer-Emmet-Teller method, is not suitable for porous mortars. This method is based on the adsorption of Krypton by the skeleton at the temperature of liquid nitrogen [10]. Dust in the pores disturbs the measurements.

Experimental determination of the sound absorption coefficient

Sound absorption is the reduction of the sound intensity by transformation of the energy into heat. The absorption coefficient is the ratio of the absorbed intensity to the incident intensity, as given by eq. (4).

= is angle and frequency dependent.

I a ( , f ) . I i ( , f )

(4)

352

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

In our experiments the absorption coefficient is measured under normal incidence of the sound field by use of a so-called impedance tube. The sound field is created by a speaker at the other end of the tube, which is excited by a random noise. A two-microphone technique is used to analyse the sound field. One microphone is fixed and is used as the reference, the other microphone is successively mounted at two specific positions. Microphones are mounted in special holes, drilled through the sidewall of the impedance tube [10].

Fitting of the experimental and the predicted sound absorption coefficient

In this paragraph, the acoustical model of Biot-Johnson-Allard is validated for porous, granular materials. The calculated value is compared to the one that is measured in the impedance tube. The calculation is made with the aid of WinLayers v1.0, developed at the Laboratory for Acoustics and Thermal Physics at the Katholieke Universiteit Leuven. An example of such a fit is given in Figure 1. As mentioned before, the characteristic viscous and thermal length of open porous mortars cannot be measured by conventional techniques. By comparing the experimental sound absorption coefficient with the predicted one for several characteristic lengths, an approximated value for both lengths is deduced for each sample. Table 3 gives the characteristic lengths, found with this technique. The characteristic viscous length shows no relation with the composition of the mortars or the grain size of the sand. The thermal length on the other hand, is higher for mortars with larger grains because of the smaller frame area. In contrast to what was expected, the p/c ratio has no clear influence on the absorption properties: at higher p/c ratio the absorption coefficient does not decrease. The p/c ratio does not change the flow resistivity. Further research is needed to clarify this phenomenon.

Figure 1:

Fit of experimental and predicted sound absorption coefficient

353

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

Table 3:

Characteristic viscous and thermal lengths

SERIES I Sample
0 1 3 4 6 7 8 9 10 11 55 50 55 50 55 55 35 60 50 30

SERIES II [m]
500 500 500 400 500 500 220 200 240 300 20 21 23 24 26 27 28 29 30 31

[m]

Sample
70 70 90 50 75 60 55 80 75

[m]
120

[m]
300 500 500 400 400 500 240 220 200 200

Influence of the parameters

Because of the validity of the model of Biot-Johnson-Allard for granular porous materials, this theoretical model is used to predict the acoustical behaviour of polymer-modified mortars. Some acoustical parameters can not be determined experi-

Figure 2: Influence of the porosity on the sound absorption coefficient

354

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

Figure 3:

Influence of the tortuosity on the sound absorption coefficient

mentally or can hardly be controlled during the production process. By means of the theoretical model, the influence of each parameter on the absorption coefficient can be examined while the other parameters remain unchanged. An increase (or decrease) of the porosity corresponds to an increase (or decrease) of the peak value of the absorption coefficient (Figure 2). At high values of the porosity, the friction surface increases and more energy is dissipated. Variations of the porosity of 2 or 3 % have only a minor influence on the acoustical behaviour. The porosity can be controlled during the production process. By determining the optimal composition with a minimum amount of binder and by limiting the degree of compaction, porous mortars with a high peak value of the absorption coefficient can be developed. Variations of flow resistivity between samples with the same grain size, have no noticeable effect on the absorption coefficient. The samples with a grain size of 0.8/1.2 have a higher flow resistivity than the samples with grains of 1.4/2.0. This increase of the flow resistivity corresponds to a decrease of the peak value of the absorption coefficient. Figure 3 gives the influence of the tortuosity on the absorption coefficient. At high values of the tortuosity, the peak value shifts towards lower frequencies and lower values of the absorption. The thickness of the porous material layer has also a great

355

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

influence on the position of the peak value in the frequency spectrum. The thicker the layer of the porous material, the lower the frequency at which the peak value occurs. The influence of variations of the characteristic thermal length on the absorption coefficient is negligible for the examined polymer-modified mortars. Only the width of the absorption peak shows minor changes. An increase of the characteristic viscous length causes a shift of the peak value towards the higher frequencies and an increase of this peak value. To produce mortars with good sound absorption properties, a high open porosity and a sufficient flow resistivity is needed. By combining multiple layers of porous materials and by varying the thickness of each layer, the desired absorption characteristic over a wide frequency range can be obtained. Optimization of the development of this multiple layer material can be guided by the parameter study presented in this paragraph.

Conclusion

The aim of present research is the verification of the validity of the model of Biot-Johnson-Allard for open porous granular materials e.g., polymer-modified mortars. It has been shown that, using the characteristic viscous and thermal length as fit-parameters, a good match between the predicted and the experimental sound absorption coefficient is found. Since all viscous lengths and all thermal lengths are of the same order of magnitude, they are assumed to be a good approximation of the real material parameters. Because the model is valid for homogeneous materials, small deviations can certainly be explained as due to heterogeneity or limited size of the test samples. Further research will focus on the development of more suitable methods for the determination of some of the parameters. Another goal will be the optimisation of the material properties for a certain absorption goal. The model will allow to do the sensitivity analysis. Finally the performance should be evaluated on larger samples with a random incident sound field.

References
1. K.V. Horoshenkov, D.C. Hughes, A. Cwirzen, The sound speed and attenuation in loose and consolidated granular formulations of high alumina cement, Applied Acoustics 64, 97-212 (2003) N.N. Voronina, K.V. Horoshenkov, A new empirical model for the acoustic properties of loose granular media, Applied Acoustics 64, 415-432 (2003)

2.

356

Acoustic properties of sound absorbing, polymer-modified porous cement mortars

3. 4. 5. 6. 7.

8.

9.

10. 11.

12.

J.F. Allard, Propagation of sound in porous media, Elsevier Applied Science (1993) E. Knapen, R. Lanoye, Geluidabsorberende granulaire materialen, Thesis K.U.Leuven, Leuven (2003) A. Beeldens, Influence of polymer modification on the behaviour of concrete under severe conditions, PhD thesis, K.U.Leuven (2002) S. Teuber, Schallausbreitung in porsen Absorbern, Thesis Carl von Ossietzky Universitt, Oldenburg D.L. Johnson, J. Koplik, R. Dashen, Theory of dynamic permeability and tortuosity in fluid-saturated porous media, J. Fluid Mech. 176, 379-402 (1987) Y. Champoux, Etude exprimentale du comportement acoustique des matriaux poreux structure rigide, Thesis Carleton University, Ottawa (1991) J.F. Allard, C. Depollier, J. Nicolas, W. Lauriks, A. Cops, Proprits acoustiques des matriaux poreux saturs dair et thorie de Biot, Journal dAcoustique 3, 29-38 (1990) S. Brunauer, P.H. Emmet, E. Teller, Adsorption of gases in multimolecular layers, J. Am. Chem. Soc. 11, 309-319 (1938) ISO 10534-1:1996 Acoustics -- Determination of sound absorption coefficient and impedance in impedance tubes -- Part 1: Method using standing wave ratio ISO 10534-2:1998 Acoustics -- Determination of sound absorption coefficient and impedance in impedance tubes -- Part 2: Transfer-function method

357

E. Knapen, R. Lanoye, G. Vermeir, W. Lauriks and D. Van Gemert

358