See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228651416

Sound absorption properties of sustainable fibrous materials in an enhanced

reverberation room

Article · January 2005

CITATIONS READS

38 549

2 authors, including:

Francesco D’Alessandro
Italian National Research Council
70 PUBLICATIONS   955 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Francesco D’Alessandro on 31 May 2014.

The user has requested enhancement of the downloaded file.

 Sound absorption properties of sustainable fibrous
materials in an enhanced reverberation room

Francesco D’Alessandroa, Giulio Pispolab

a,b
Department of Industrial Engineering, University of Perugia,
Via G. Duranti 67, 06125 Perugia, Italy
a
dalessandro.unipg@ciriaf.it; bpispola.unipg@ciriaf.it


Abstract In this paper, the measurement of sound absorption coefficient of novel

sustainable fibrous materials is investigated. Nowadays the use of such materials is
becoming wider for various applications, being ecological, biodegradable and
renewable: they differ from traditional fibrous materials, as rock or glass wool, for their
very low toxicity and polluting effects. These materials can be used in many ways:
noise mitigation and building acoustic correction are surely among the most important.
Sound absorbing layers made of natural fibres and of recycled raw materials have been
tested in the reverberation room of the Acoustics Laboratory of the University of
Perugia according to ISO 354 standard, in order to quantify their sound absorption
properties and to make a comparison with traditional fibrous sound absorbers. An
optimization of the reverberation room characteristics has been also carried out. Good
sound field diffusivity inside the room is a fundamental requirement for the
measurement accuracy. Among the parameters that mainly affect room diffusivity are
the room shape and the sample disposition inside the room. In order to obtain accurate
values of the sound absorption coefficient, specific actions were adopted. Test
specimens were placed on the floor with edges nonparallel to the room walls. A partial
closing of the lower room corners with absorbing and reflecting diffusers and
suspended plane diffusers were also tested, obtaining a significant improvement of the
results. The measured performance of the tested materials seems to be fully comparable
with that of mineral wool fibres: because of their low impacts on the environment and
the human health they can be seen as a valid alternative to conventional materials.

1. INTRODUCTION

Porous materials obtained from synthetic fibres, such as mineral wool or glass wool, are
commonly used for thermal insulation and sound absorption, because of their high
performance and low cost. Their diffuse-field sound absorption coefficient is very high at
mid-high frequencies. On the other hand, they have several cons: they can be harmful for
human health if their fibres are inhaled, since they can lay down in the lung alveoli, and can
cause skin irritation (as stated by the European Council Directive on dangerous substances
67/548/EEC [1] and subsequent amendments). Hence such materials must be adequately
overlaid if directly exposed to the air. Moreover they can pulverize because of vibrations and
are not resistant to water, oil and chemical agents and this makes unwise their application on
absorbing noise barriers.

In recent years, an increasing attention has been turned to natural fibres as alternatives to
synthetic ones, in order to combine high acoustic and thermal performances with a low
impact on the environment and on the human health. Natural fibres have very low toxicity
and their production processes contribute to protect the environment. Recycled materials, as
recycled plastic fibres, can even be regarded as a sustainable alternative. They can be
manipulated without any protection device, not releasing fibres or dust even when subjected
to an extended mechanical stress. The aim of the paper is to investigate the acoustic
performances of such materials through measurements of sound absorption coefficient in
reverberation room and comparisons with conventional fibrous sound absorbers.

2. DESCRIPTION OF THE TESTED SAMPLES

2.1 Sustainable Fibrous Sound Absorbers: Kenaf And Recycled Polyester

Two different novel fibrous materials were tested in reverberation room. The first ones were
sound absorbing blankets of kenaf fibres assembled in semi-rigid panels without using
adhesives. Small percentages of polyester (8-10 %) and of a fireproof additive have been
added and then the components have been thermobonded. Characteristics of the tested
specimens are detailed in Table 1.

Kenaf is the name of a hibiscus plant related to cotton and okra, Hibiscus cannabinus L., and
is member of the Malvaceae family (Fig. 1a). It has been used in Asia and Africa for
thousands of years to obtain raw materials for clothes, paper, oil and medicines. Today it is
used especially for its cordage, canvas and sacking. It can easily be cultivated (even with
biological methods) and grows quickly. The stems produce two types of fibres, a coarser one
in the “bast”, and a finer one in the “core” (Fig. 1b). Bast fibres (about 35 %) are suitable for
paper, textiles and rope; core (about 65 %) is usually used as a biomass or it can be reduced
to particles and bonded into panels similar to particleboard [2].

The second tested samples were sound absorbing blankets of recycled polyester (PET) fibres.
These are made of 100% polyester fibres obtained from a recycling process of PET bottles.
Blankets are assembled in a 3D fibre arrangement through a thermobonding process without
any adhesive. As the natural fibres, polyester blankets respect both the environment and the
human health. Characteristics of the tested samples are detailed in Table 1.

Measured sound absorption properties of the described materials are reported below in Sec.
4, following a brief overview of the actions taken to obtain reliable results in the
reverberation room of the University of Perugia.
 a b
Figure 1: Kenaf: a) rows; b) bast fibres (left) and core fibres (right).

Table 1: Characteristics of the tested samples.

Parameter Unit Kenaf layers PET layers

Structure - Thermobonded panels with Thermobonded panels with
no added adhesives no added adhesives

Raw material - Natural hemp fibres, Recycled polyester fibres

polyester backing fibres

Thickness mm 50 50
Density Kg/m3 30 30
Surface area of the m2 7,56 7,56
test specimens

3. REVERBERATION ROOM IMPROVEMENT

When measuring sound absorption coefficient α s in reverberation room, several causes of

errors can occur. Indeed it is often difficult to obtain high repeatability and reproducibility,
even though the ISO 354 standard [3] specifies the essential requirements to increase
measurement accuracy. The factors that mainly affect the measured absorption coefficients
are the shape of the reverberation room, the placement of absorbing surfaces, the position of
the source, of the microphones and of the samples under test, the control of the microclimatic
conditions (temperature and relative humidity) and above all the diffuseness of the sound
field [4]. The latter subject, being fundamental assumption of this kind of measurements, has
been thoroughly studied by several Authors (for a brief exhaustive review, refer to [5]). Cook
et al. [6] and Balachandran [7] had already suggested in the 1950s to increase the diffuseness
of the sound field through the use of devices as diffusers and rotating vanes.

Especially in a rectangular reverberation room, the sound field is likely to be far from a
perfectly diffuse one. When absorbers are laid on the floor of such a reverberation room, the
sound field characteristics (e.g. sound pressure standard deviation, spatial cross-correlation
and coherence functions) are deeply altered. Considering the geometrical features of the
reverberation room available at the Acoustics Laboratory of the University of Perugia [8]
(see Fig. 2), it was necessary to work mainly on its shape and on the diffuseness of the sound
field inside the room.

Several measurements were performed to investigate the aforementioned factors employing

the interrupted stationary noise method. A two-channel data acquisition system (01dB-
Metravib model Symphonie) was used to record every 20 ms the sound pressure decays
(through ½’’ microphones G.R.A.S. model 40 AR) and to drive an omnidirectional
(dodecahedral) source with random noise. The reverberation time in each third octave band
was evaluated by linear interpolation of the averaged decay curve (starting 5 dB below the
beginning of the decay and over a decay range of 30 dB).

Figure 2: Plan of the coupled reverberation rooms of the Acoustics Laboratory of the University of Perugia.
The aperture between the two chambers was left open during sound absorption measurements.

The influence of the sample position inside the room was firstly examined. ISO 354 standard
suggests to place the test specimens with edges nonparallel to the room walls in order to
reduce the influence of the horizontal axial modes: data for absorbing noise barrier panels can
be seen in Figure 3, where the effect of a different positioning on the absorption coefficient is
clearly shown. A partial closing of the lower corners of the room (Fig. 4 on the left) was then
performed by means of plane absorbing and reflecting diffusers, in order to reduce the
influence of tangential modes [9]. Samples of an absorbing noise barrier were tested in three
different ways. The first configuration (standard) is a bare room one; then rock wool blankets
were placed in each lower corner, while in the third configuration an assembly of four
reflecting gypsum-board panels was employed. The results are shown in Figure 4 on the
right. The use of corner wall reflecting diffusers seems to give the best results, at least for the
tested samples.
 1.00

0.90 parallel edges (a) oblique edges (b)

0.80
a 0.70

0.60

ALPHA
0.50

0.40

0.30

0.20

b 0.10

0.00

00

50

00

00

00

50

00

00
0

0
10

12

16

20

25

31

40

50

63

80
10

12

16

20

25

31

40

50
frequency (Hz)

Figure 3: Effect of the positioning of the sample (sound absorbing panels of a noise barrier) on the absorption
coefficient: edges parallel (a) and not (b) to the room walls.

1.00

0.90

0.80

0.70

0.60
a
ALPHA

0.50

0.40

0.30

0.20

0.10 standard absorbing panels reflecting panels

0.00
b
0

0
00

50

00

00

00

50

00

00
10

12

16

20

25

31

40

50

63

80
10

12

16

20

25

31

40

50

frequency [Hz]

Figure 4: Effect of corner diffusers on the absorption coefficient of sound absorbing panels of a noise barrier:
bare room (“standard”), absorbing panels (a) and reflecting ones (b).

As stated in the annex A of the ISO 354 standard, diffuseness of the sound field can be
enhanced by means of fixed or rotating diffusers. They are especially important in
rectangular rooms; in those cases, the standard suggests that the area of diffusers (both sides)
should be in the range from 15 % to 25 % of the total surface area of the room. Different
numbers of suspended plane gypsum-board diffusers were hung from the ceiling to evaluate
their effect on sound absorption measurements (Fig. 5a); because of the limited available
room, a smaller percentage of diffuser area was used (at maximum 5 % just with suspended
diffusers). The tested samples were layers of polyester fibres (thickness 50 mm, density 50
kg/m3) and were laid on the floor of the reverberation room. Two source positions and four
microphone positions were employed for each single reverberation time measurement, and
the decay recordings were repeated three times at each point (24 acquisitions). Results are
illustrated in Figure 6: it can be seen how the absorption coefficient raises towards reasonable
values increasing the diffuser number, particularly at mid-high frequencies. Similar
behaviours were reported by other Authors for measurements in rooms of regular shape [10].
Such results may be regarded as an indirect proof of the poor diffuseness of the sound field,
as it may be expected considering the shape of the room.

a b

Figure 5: Configuration of the reverberation room with suspended plane diffusers (a) and with a combination of
suspended and corner plane diffusers (b).

Then a combination of 11 suspended and 16 corner wall diffusers was employed (Fig. 5 b)
for an overall area equal to 8 % of the room surface area. Results are shown in Figure 6, as
well. This latter configuration seems to increase the absorption coefficient at the medium
frequencies, while its effect is poorer on the high frequencies range (the peak at 5 kHz has
been regarded as due to measurement errors). Annex A of ISO 354 prescribes to calculate the
mean value of the sound absorption coefficients in the range from 500 Hz to 5000 Hz for
different numbers of diffusers: as reported in Table 2, the mean value calculated in the
mentioned range is next to unity for the configuration with 11 hanging diffusers and 16 wall
diffusers placed at the corners.

1,20

1,00

0,80

αS 0,60

0,40
bare room
6 hanging diffusers
0,20 11 hanging diffusers
11 hanging + 16 corner wall diffusers

0,00
0

0
00

50

00

00

00

50

00

00
10

12

16

20

25

31

40

50

63

80
10

12

16

20

25

31

40

50

frequency [Hz]

Figure 6: Effect of the installation of different numbers of hanging diffusers and of wall diffusers at the room
corners.
 Table 2: Mean sound absorption coefficients of polyester fiber blankets for different room configurations.

Frequency bare room 6 hanging 11 hanging 11 hanging + 16

range [Hz] diffusers diffusers corner wall diffusers
100-5000 0.69 0.70 0.77 0.84
500-5000 0.84 0.85 0.96 1.05

It has to be duly noticed that introducing diffusers in a reverberation room has inevitably the
consequence of increasing its equivalent absorption area or, that is equivalent, of lowering
the reverberation time T, although reflecting are the diffusers. Figure 7 on the left shows this
phenomenon for the empty room in the abovementioned configurations. Estimating the
relative standard deviations S = σ T / T of the reverberation time (here it has been used the
formula provided in [3]), it is possible to calculate the sound absorption coefficient standard
deviation by simply regarding the reverberation times as the only independents variables:

 55.3V   S1   S 2  
2 2 2 2 2
 ∂α   ∂α 
σα =   σ T1 +   σ T2 =     +    (1)
 ∂T1   ∂T2   cS   T1   T2  

where V is the volume of the reverberation room (120 m3), c is the sound speed in air, S is the
test specimen area and the suffixes 1 and 2 denote respectively measurements without and
with the test specimen. It is evident from Figure 7 on the right that the accuracy of sound
absorption measurements becomes poorer as the number of diffusers increases. This means
that a trade-off is needed among the sound field diffuseness and the measurement accuracy,
simply due to the modified room absorption area.

14,00 0,2
bare room bare room
0,18
12,00
6 hanging diffusers 0,16 6 hanging diffusers
10,00 11 hanging diffusers 0,14 11 hanging diffusers

8,00 11 hanging + 16 corner wall diffusers 0,12 11 hanging diffusers + 16

0,1 corner wall diffusers
6,00 σα [-]
T1 [s] 0,08
4,00 0,06
0,04
2,00
0,02
0,00 0
0
5
0
0
0
5
0
0
0

10 0
12 0
16 0
20 0
25 0
00

40 0
00
00

5
0
0
0
5
0
0
0
0
00
50
00
00
00
50
00
00
10
12
16
20
25
31
40
50
63
80
0
5
0
0

12
16
20
25
31
40
50
63
80
31

50

10
12
16
20
25
31
40
50

frequency [Hz] frequency [Hz]

Figure 7: Diffuser effect on the measurement accuracy. Reverberation times of the empty room (on the left) and
estimated standard deviations of the absorption coefficient (on the right) for different room configurations.

Climatic conditions inside the test room should be always kept under control: temperature
and relative humidity have to be kept as similar as possible for the measurements with and
without the sample. In this way, the effect of the correction introduced by the ISO 354
standard (by means of a power attenuation coefficient defined by ISO 9613-1 [11]) can be
minimized, so eliminating one possible cause of uncertainty. Such correction is applied on
the calculation of the equivalent absorption area A through the second term of the following
[3]:

55.3*V
A= − 4*V * m (2)
c *T

where m is the power attenuation coefficient related to the air sound absorption. Its effect is
greater at high frequencies and for low values of temperature (t<15 °C) and relative humidity
(Φ <30 %), as shown in Figure 8. Even when the relative differences of temperature (∆t) and
relative humidity (∆Φ) between the two sets of measurements (with and without the sample)
are the same, the effects on the absorption coefficient is more pronounced when the absolute
values of t and Φ are lower. It was noticed for example that, in the range from 2500 Hz to
5000 Hz, for ∆t = 0.5 °C and ∆Φ = 1 %, the sound absorption coefficient correction was
relatively equal to 6 % with mean temperature and relative humidity respectively of 18 °C
and 26 %, while it was lower than 1 % when t = 20 °C and Φ = 32 %.

Figure 8: Equivalent air absorption areas in m2 for the employed reverberation room calculated according to
[3] and [11] (Aair=4*V*m) for some third-octave band centre frequencies. Different z-axis scales are employed.

4. ABSORPTION PROPERTIES OF SUSTAINABLE FIBROUS SAMPLES

Considering the room optimization described in Sec. 3, a configuration employing 13

diffusers hung from the ceiling and 16 diffusers placed on the room corners was chosen to
evaluate the absorption properties of the samples introduced in Sec. 2. Figure 9 reports the
measured absorption coefficients; a comparison is also performed between the tested samples
and non-specific available data for mineral wool and glass wool of consistent macroscopic
characteristics (thickness 50 mm, density 33 kg/m3). The measured values can also be found
in Table 3.
The kenaf samples (see Table 1) show an averaged absorption coefficient equal to 0.85 in the
500-5000 Hz range and equal to 0.65 in the 100 – 5000 Hz range, while for the recycled
polyester blankets the measured mean values are respectively 0.91 and 0.71. Although their
performances appear to be slightly poorer than those of traditional fibrous absorbers, these
materials can be seen as a valid alternative to the traditional mineral wool blankets for
thermo-acoustic applications having taken into account their beneficial properties of low
impact on the human health and on the environment.

1
kenaf fibers
recycled polyester fibers
0.9 glass wool
mineral wool

0.8

0.7

0.6
αS
0.5

0.4

0.3

0.2

0.1

0
100 500 1000 5000
frequency [Hz]

Figure 9: Third-octave band sound absorption coefficient αS of kenaf and recycled polyester fibrous blankets in
comparison with traditional fibrous absorbers.

Table 3: Third-octave band sound absorption coefficients of the tested samples.

Frequency [Hz] 100 125 160 200 250 315 400 500 630
αS kenaf 0.14 0.19 0.18 0.29 0.48 0.53 0.64 0.74 0.82
αS PET 0.14 0.20 0.19 0.32 0.58 0.66 0.70 0.82 0.93

Frequency [Hz] 800 1000 1250 1600 2000 2500 3150 4000 5000
αS kenaf 0.85 0.91 0.90 0.90 0.86 0.82 0.83 0.83 0.84
αS PET 0.94 0.97 0.99 0.91 0.95 0.88 0.88 0.88 0.85

5. CONCLUSIONS

Aim of this work was the evaluation of the sound absorption performances of innovative
sustainable fibrous materials. Samples made of vegetal fibres (kenaf) and of recycled
polyester fibres were tested in the reverberation room of the University of Perugia. In order to
increase the accuracy, the measurement methodology was deeply analyzed and an
optimization of the reverberation room characteristics was carried out. Attention has been
focused mainly on the diffuseness of the sound field inside the room and on the correction for
air sound absorption. A remarkable influence on the measured absorption coefficients was
observed installing different numbers of diffusers, both in terms of mean values and of
 standard deviations; suspended gypsum-board plane diffusers were employed, even in
combination with wall diffusers placed at the lower room corners. The dependence of the
final data on the climatic conditions of the room was analysed; it has been noticed that, when
temperature and relative humidity gets values close to the limits prescribed by the ISO 354
standard (15 °C and 30 %), the correction for air absorption can strongly influence the
results, mainly at high frequencies. Measured absorption properties of the mentioned fibrous
materials, after the room optimization, seem slightly lower but comparable with those of
traditional synthetic fibres: considering their sustainability, in particular their non toxicity,
they can be properly seen as an alternative, especially for thermo-acoustic applications and
when there is any risk of fibre dispersion in the environment (e.g. sound absorbing noise
barriers).

ACKNOWLEDGMENTS

The Authors are indebted with Prof. Francesco Asdrubali for his suggestions and with Miss
Silvia Dominici for the collaboration during the experimental campaign. The authors have
also to acknowledge Mr. Alberto Giorgiantoni of Euchora s.r.l. (Italy) for having provided
the samples of kenaf and recycled polyester fibrous layers.

REFERENCES
[1] European Council Directive 67/548/EEC of 27 June 1967, On the approximation of laws, regulations and
administrative provisions relating to the classification, packaging and labelling of dangerous substances,
Official Journal of the European Union, 196, 1-98, 16th August 1967.
[2] T. Sellers Jr. and N. A. Reichert, Kenaf Properties, Processing and Products, Mississippi State
University, 1999.
[3] ISO 354 Acoustics—Measurement of sound absorption in a reverberation room, 2003.
[4] A. C. C. Warnock, Some practical aspects of absorption measurements in reverberation rooms, The
Journal of the Acoustical Society of America 74(5), 1422-1432, 1983.
[5] F. Jacobsen and T. Roisin, The Coherence of Reverberant Sound Fields, The Journal of the Acoustical
Society of America 108(1), 204-210, 2000.
[6] R. K. Cook, R. V. Waterhouse, R.D. Berendt, S. Edelman and M. C. Thompson Jr., Measurement of
Correlation Coefficients in Reverberant Sound Fields, The Journal of the Acoustical Society of America
27(6), 1072-1077, 1955.
[7] C. G. Balachandran, Random sound field in reverberation chambers, The Journal of the Acoustical
Society of America 31(10), 1319–1321, 1959.
[8] F. Asdrubali and K. V. Horoshenkov, The Acoustic Properties of Expanded Clay Granulates, Journal of
Building Acoustics 9(2), 85-98, 2002.
[9] A. Cops, J. Vanhaecht and K. Leppens, Sound Absorption in a Reverberation Room: Causes of
Discrepancies on Measurement Results, Applied Acoustics 46(3), 215-232, 1995.
[10] E. Toyoda, S. Sakamoto and H. Tachibana, Effects of room shape and diffusing treatment on the
measurement of sound absorption coefficient in a reverberation room, Acoustical Science and
Technology 25(4), 255-265, 2004.
[11] ISO 9613-1 Acoustics— Attenuation of sound during propagation outdoors—Part 1: Calculation of the
absorption of sound by the atmosphere, 1993.

View publication stats