ARTIFICIAL VEGETATION: A NOVEL PAPER-BASED SOUND ABSORBER by Amirhossein Salimian B.Sc., Industrial Engineering, Tehran Azad University, 2010 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Mechanical Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) January 2018 © Amirhossein Salimian, 2018 ii Abstract Demand for acoustic barriers in urban buildings grown in recent times fed by levels of increasing urban noise. Extraneous urban noise in buildings was commonly controlled using noise barriers like absorptive panels. The efficiency and sustainability of such noise absorbers is suspect. Our research in acoustics led to the development of a novel paper-based sound absorbers as a viable alternative for those currently available in the market. We named this product Artificial Vegetation (AV). We tested AV and found it an efficient and sustainable alternative that can be mass produced with relatively comparable low overheads. In this research paper we explored literature in acoustic barriers, both paper- and foam-based. We also identified gaps in the literature. The paper elaborates measurement of sound absorption coefficient using impedance-tube method, as a basic procedure for producing AV. We developed an analytical model that optimizes and improves sound absorptivity of AV. Data gathered from experiments were used to validate our model and test as a computational platform in more AV designs. iii Lay summary The outcome of our research project was a product for urban sound insulation in rooms and buildings. This product, Artificial Vegetation (AV), utilized a porous paper-based material and modified Origami method. We conducted a series of experiments to obtain optimal acoustical characteristics. These experiments consisted of altering the physical characteristics of the material such as thickness and porosity and studying the impact on its acoustical characteristics. We developed a model to predict the behaviour of AV. This model was validated and optimized from the experiments’ data that led to the final prototype. AV can be used as a sustainable and effective acoustic barrier to urban noise pollution in buildings. iv Preface This thesis is an original, unpublished, and independent work of the author. All the experiments, analysis and writing are performed under the direct supervision of Dr. James Olson and Dr. Murray Hodgson in Mechanical Engineering department in UBC. The MATLAB code of the prediction model, is a modification of Mr. Banda Logawa’s original work. v Table of contents Abstract ........................................................................................................................................................ ii Lay Summary ............................................................................................................................................. iii Preface ........................................................................................................................................................ iv Table of contents ........................................................................................................................................ v List of tables ............................................................................................................................................... vi List of figures ............................................................................................................................................. vii Acknowledgments ..................................................................................................................................... ix Dedication .................................................................................................................................................... x 1. Introduction .......................................................................................................................................... 1 2 Background ......................................................................................................................................... 7 2.1 Sound absorption ..................................................................................................................... 10 2.1.1 Definition ............................................................................................................................ 10 2.2 Transmission loss ..................................................................................................................... 12 2.3 Prediction models for a Helmholtz resonator ....................................................................... 12 2.3.1 Transfer matrix method ................................................................................................... 13 2.4 Origami....................................................................................................................................... 18 2.4.1 Formation ........................................................................................................................... 18 2.5 Foam-paper ............................................................................................................................... 21 3 Experimental Procedure and Results ............................................................................................ 24 3.1 Impedance tube test ................................................................................................................ 24 3.2 Results ....................................................................................................................................... 29 4 Modeling and Optimization ............................................................................................................. 40 4.1 Modeling .................................................................................................................................... 40 4.2 Results ....................................................................................................................................... 41 4.3 Fitting the model to experimental data .................................................................................. 43 4.4 Sensitivity analysis ................................................................................................................... 45 5 Summary, Conclusions, and Final Remarks ................................................................................ 51 References ................................................................................................................................................ 53 vi List of tables Table 4-1 Parameters for the best fit to the experimental results ...............................................45 Table 4-2-Test matrix for sensitivity analysis. ............................................................................45 vii List of figures Figure 1-1 Total paper consumption by printing grade in North America between years 2000-2009 ........................................................................................................................................... 1 Figure 1-2 -NBSK foam-paper at 50% air content. (Photographed by Anna Jamroz, UBC Pulp and Paper Centre) (Ahmadi, 2016) ............................................................................................ 3 Figure 1-3 Folded tessellation (Sina, 2015). ............................................................................... 4 Figure 1-4 module-based green walls; natural vegetation. Adapted from (Azkorra et al., 2015) . 5 Figure 1-5 two self-folding Origami designs ............................................................................... 6 Figure 2-1 Sound behavior facing a surface. .............................................................................11 Figure 2-2 Transfer Matrix Model (TMM) representation for a multi-layer medium. ...................14 Figure 2-3 “Pyramid” designed with AutoCAD software. ............................................................19 Figure 2-4 Graphtec FC2250-180 .............................................................................................20 Figure 2-5 Graphtec 2250-180 dual heads ................................................................................20 Figure 2-6 Polymer attachment to the paper (adapted form (Sina, 2015)) .................................21 Figure 2-7 Drying foam-papers; exposed to air and a heater. ....................................................23 Figure 3-1 Impedance tube setup .............................................................................................25 Figure 3-2 two ¼ inches B&K 4178 phase matched microphones .............................................26 Figure 3-3 impedance tube with two microphones ....................................................................26 Figure 3-4 Sound absorption coefficient of foam-paper; consistency:0.5; Thickness 1 cm in low frequency ..................................................................................................................................30 Figure 3-5 Sound absorption coefficient of foam-paper; consistency:0.5; Thickness 1 cm in high frequency ..................................................................................................................................31 Figure 3-6 Artificial Vegetation ..................................................................................................33 Figure 3-7 Sound absorption coefficient of 1 cm thick foam-paper; consistency:0.5. (no grass) 34 Figure 3-8 Sound absorption coefficient of 1 cm thick foam-paper and 1 cm grass; consistency:0.5. ........................................................................................................................34 Figure 3-9 Sound absorption coefficient of 1 cm thick foam-paper; consistency: 1. ...................35 Figure 3-10 Sound absorption coefficient of 1 cm thick foam-paper and 1 cm grass; consistency:1. ...........................................................................................................................35 Figure 3-11 Sound absorption coefficient of 1 cm thick foam-paper and 3 cm grass; consistency:1. ...........................................................................................................................36 Figure 3-12 Sound absorption coefficient of 1 cm thick foam-paper and 6 cm grass; consistency:1. ...........................................................................................................................36 Figure 3-13 Q-dP graphs of a 1.5 cm thick foam-paper and a 1.5 cm thick artificial grass product. .....................................................................................................................................37 Figure 3-14 Comparison between the sound absorption result of an AV product and a commercial fibreglass absorber (results of the commercial absorber are provided by Kinetics Noise Inc.) .................................................................................................................................38 Figure 4-1 n cm and two n/2 cm material ..................................................................................42 Figure 4-2 Model test: difference between n cm material and two n/2 cm the same material ....42 Figure 4-3 changing the thickness of the layer 1 from 1cm to 5 cm Vs doing the same for the second layer .............................................................................................................................43 Figure 4-4 Model Vs Experiment result 1cm foam and 6cm grass ............................................44 viii Figure 4-5 Tortuosity change of the foam from 2 to 4 ................................................................44 Figure 4-6 Porosity change of the foam from .9 to.8 ..................................................................44 Figure 4-7 Changing of shape factor from 3.4 to 8 ....................................................................44 Figure 4-8 Test A1: Changing the foam-paper porosity from 0.4 to 0.9 for the thicknesses of 1,2 and 6 cm. ..................................................................................................................................46 Figure 4-9 Test A2: Sensitivity analysis of air flow resistivity from 10000 to 1000 and 20000 (porosity = 0.9) ..........................................................................................................................47 Figure 4-10 Test A3: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 (thickness=1cm; porosity = 0.6) ................................................................................................48 Figure 4-11 Test B: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 ( porosity = 0.6 and thickness = 3 cm) .........................................................................................48 Figure 4-12 Test C: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 ( porosity =0.6 and thickness = 9 cm) ..........................................................................................49 Figure 4-13 Test D: Sensitivity analysis of the tortuosity of layer 2, from 0.5 to 5 ......................50 Figure 4-14 Test E: Sensitivity analysis of the shape factor of layer 2, from 1 to 10 ..................50 ix Acknowledgments Firstly, I would like to express my sincerest gratitude to my supervisor, Dr. James Olson, for the continuous support of my study and research since the start of my master’s degree program. I would like to offer my special thanks to Dr. Murray Hodgson who although no longer with us, continues to inspire me by his dedication to his research and work. I was privileged Dr. Hodgson as my supervisory committee member until August 2017 I would especially like to thank Pouyan Jahangiri for his invaluable help in the past years and for the critical reviews of the thesis. I would also like to thank Dr. Ali Elahimehr and my colleagues Hatef Rahmani, Mohammad Shanbghazani, Chitra Arcot and Banda Logawa for their suggestions and discussions. My everlasting gratitude goes to my parents and my beloved brother, Sina, for the endless support on this progressive road. x Dedication To my dear parents, who have always been there for me. 1. Introduction The Pulp and Paper sector has influenced the core of Canadian economy for more than two centuries. Over time, pulp and paper manufacturing enterprises have undergone revolutionary improvements, and diversified from paper grades to smart cellulose-based membranes. The turn of the century saw the demand for the consumption of printing and writing paper grade products dramatically diminish in North America. A large portion of their market declined, substituted by electronic media. Figure 1-1 Total paper consumption by printing grade in North America between years 2000-2009 The State of the Paper Industry in 2011 reported that the market for the printing grades - 66% of the total North American market – diminished by 25 percent (see Figure 1-1). The strategic forest network in Canada realised that the industry was struggling for survival, and that a major part of the Canadian traditional pulp and paper industry was dwindling. It was unlikely to regain profitability in competition with the electronic media market. The 2 industry policies changed, and the industry started focusing on the production of wood based eco-friendly biodegradable materials for a wide range of applications to diversify its depleted markets (Green Fibre Network conference, 2015). The recyclability and bio-compatibility as well as the low net cost of biodegradable materials have drawn interest and investment of scientists and businesses attentions. An abundant and cost-efficient use of these resources is cellulose which has been vastly used in production of many novel green products. For instance, Jahangiri et. al (Jahangiri, 2013) introduced a novel methodology to make low-density porous material using cellulose fibres in a foam-laid medium called “Foam-Paper”. This material is renewable, biodegradable and it is simple and cost-efficient to produce. The honey-comb structure of foam is being utilized to locate the cellulose fibres to make a porous 3D structure. In foam-paper making procedure, the fibre slurry is prepared by adding a percentage of standard surfactant under strong shear force resulting in uniform bubbles in suspension used as a 3D template for the fibre. This idea is expanded based on variations of the foam air-content to obtain different structures with various mechanical, acoustic and thermal insulation features. Jahangiri et al. (Jahangiri, 2013) then studied the overall mechanical and thermal characteristics of various foam papers based on the foam’s air-contents and pulp types. Based on the report presented by Jahangiri et al. (Jahangiri, 2013) and Ahmadi et al. (Ahmadi 2016) foam papers can cover a variety of applications in many industries such as air and water filtration, heat insulation, and acoustic insulation. Jahangiri et. al (Jahangiri, 2013) briefly investigated acoustic properties of foam-paper and found that foam-papers have useful sound absorptive properties. They used dried Northern Bleached Softwood Kraft (NBSK)-(Figure 1-2), and Nano fibrillated Lyocell Fibre 3 (NLF) to make foam-papers for acoustic tests, and studied the effect of air-content, porosity, sample thickness, and pulp consistency (the proportion of mass of pulp to the whole suspension mass). Figure 1-2 -NBSK foam-paper at 50% air content. (Photographed by Anna Jamroz, UBC Pulp and Paper Centre) (Ahmadi, 2016) Samples were made in two different air contents: 50 percent and 70 percent, and the acoustic features were studied in low-frequency range (between 250 Hz and 1750 Hz). The researchers used the three-microphone impedance-tube method (explained in Chapter 3). The experiments showed that higher consistency results in higher sound absorption properties. The same trend was captured for changing the thickness at the certain frequencies. They also showed that increasing the air content had a negative effect on the absorption since less energy could be damped in relatively high air content 4 products. They concluded that the similarity index for the absorption coefficient was more pronounced in relatively lower frequencies of commercial acoustic products. Interesting applications of paper have found their way in various fields in both art and fundamental science. For example, Sina et al (Sina, 2015) introduced a new methodology to make paper-based products with self-folding capability subjected to heat. The science of Origami assists with providing a systematic approach to make a vast range of paper structures with different applications (see Figure 1-3). Sina et al (Sina, 2015) considered acoustical applications for the self-folding paper products, Figure 1-3 Folded tessellation (Sina, 2015). By incorporating wide range of structures, shapes, and designs to optimize the acoustical features of the method. Acoustical results of the self-folding and foam papers, established that two unique wood-based, renewable and biodegradable products had potential to substitute commercial sound absorbers with lower mass-production costs. We summarize that a systematic study on the acoustical behaviour of two novel products, mentioned above, is relatively unexplored; from review of existing literature our research focused on: 5 • Measurement of sound absorption coefficient of self-folding papers. • increasing sound absorption properties of the foam-papers. • Mathematical modeling of combined foam-paper and self-folding paper material to create a new sound absorptive material. We form a porous and complex shapes with sound absorption capability (Taherzadeh et al., 2012) by utilizing Origami. The desired absorption can be the result of thermal and viscous losses (Horoshenkov, 2013). Sound scattering, another method of noise attenuation can be obtained using scattered surfaces. However, experiments in sound scattering are beyond the scope of our research project. Most recently, the structures of the vegetated surfaces were employed to gain the desired absorption. Problems with maintenance, shipping and installation have made this technique unattractive for commercial use. Figure 1-4 module-based green walls; natural vegetation. Adapted from (Azkorra et al., 2015) “Artificial Vegetation” was developed as a unique and effective technology developed to acquire desired absorption based on the idea of vegetated surfaces. This methodology 6 was based upon the combination of foam-paper and Origami technologies elaborated in the following chapter (see figure 2-2). Figure 1-5 two self-folding Origami designs In this study, we investigated the effect of certain origami shapes on the sound absorption of artificial vegetated (AV) surfaces. Chapter 2 reviews knowledge gaps in the literature of foam-paper, self-folding paper, and artificial vegetation areas. Chapter 3 describes the research findings of experiments conducted and, the procedure of making a self-folding paper product as the top layer of the AV. A mathematical model is developed to predict the absorption of the experimental prototypes in Chapter 4, then, we found the best acoustical properties of AV using sensitivity analysis and optimization techniques. (a) (b) 7 2 Background We explicated the research studies conducted in building noise-control so far, as well as introductory and background information concerning Origami and foam-papers. Attention is focused on how these two methodologies can improve the acoustical properties of the Artificial Vegetation. Noise pollution and troubleshooting in houses and commercial buildings have always been a major issue for this industry. Noise pollution is transmitted in buildings either through airborne or mechanically-transmitted sounds, regardless of their sources. Airborne sound is referred to the sound waves travelling in air and reflecting in the spaces between buildings, whereas, mechanically transmitted sound is mostly referred to the sound waves travelling between objects inside a building which is technically in conjunction with the impact-insulation class transmission. In many cases the sources of the noise pollution are detectable, therefore immediate strategies could be established to reduce the noise level and adjust it to an acceptable level. Noise mitigation or noise control is a protocol or set of strategies to diminish the impact of the noise and reduce the noise pollution in the building industry. Scientists and engineers developed multiple techniques and carried out research studies to address the noise difficulties, many of which are encouraged by the building codes; The most popular technologies in this category are sound insulation, sound absorption, and vibration damping. 8 We refer to sound insulation as a series of methodologies to prevent noise transmission by the introduction of a mass barrier or an obstacle. In this method, the density of the obstacle plays vital role in insulation procedure. To overcome this problem, sound absorption utilizes the porosity of a material rather than its density. In this method, a highly porous structure converts the noise energy into heat within the material through multiple energy loss inside the meandering structure. Vibration damping method is mostly common for large vibrant surfaces. In this method, the vibration energy is extracted from a thin sheet and is dissipated as heat. In building industry, sound absorption became more popular due to light weight and inexpensive price of the sound absorbers, and their high efficiency in noise control. In the field of noise absorption, researches developed a green technology called “vegetation” to combine highly porous plant region with less porous soil regions. Vegetation is a useful sound barrier particularly at mid and high frequencies (Aylor, 1972). Vegetation and growing plants can absorb up to 50% of the propagating sound (Horoshenkov, 2013). Many scientific studies were carried out concerning the sound attenuation using vegetation. For example, Taherzadeh et al (Taherzadeh et al., 2012) conducted a comprehensive experimental study on the influence of a planted area on noise reduction and control. In their study, they concluded that noise reduction is mostly related to the air-flow resistivity and filling fraction of the amount of plant occupation. Upon studying plants, it is observed that the low frequency sound attenuation occurs mostly due to the ground surface, stems and their foliage. The scientific study done by Taherzadeh et al (Taherzadeh et al., 2012), however, showed that viscous and thermal losses and sound scattering was the main reason for the sound attenuation in high 9 frequencies ( > 2000 Hz ). Burns (Burns, 1979) argued that Aylor (Aylor, 1972) statement concerning the reason of sound attenuation in high frequencies could not be rejected nor accepted. Burns (Burns, 1979) mentioned that “trunks scattering” was the main reason for this argument. Bullen & Fricke (Bullen & Fricke, 1982) showed that if ground cover was stiff, noise reduction in high frequencies was much more pronounced. They, however, mentioned that if the stiffness factor was relatively low, noise attenuation could even be negative. Vegetation width is an important element in noise attenuation; It is observed that with relatively larger width, a greater amount of noise absorption was achieved (Cook & Van Haverbeke, 1971). Cook and Van Haverbeke also mentioned that width, height, length, and density of tree belts were more important in noise reduction than the size of the leaves. The four above-mentioned characteristics are entangled with the sound diffusivity and the leave size influencing the resonant absorption. Fricke (Fricke, 1984) studied the main factors affecting the sound transmission in vegetation in pine forests and came to the conclusion that vegetation attenuate the sound in high frequencies and excess attenuation rate (decibel per meter) which depends on the space between trees and their size. The noticeable point in the literature of vegetated absorbers is that every type of plant, tree, or vegetated surface has specified inherited sound absorption characteristics. In this research, to overcome this limit, we developed an innovative technology at Pulp and Paper Centre at UBC called “Artificial Vegetation (AV)” which is a sophisticated imitation from the nature. In this method, attempt was to produce a novel sound absorber based on the structures similar to the vegetated absorbers. Using the knowledge of Origami and 10 the novel foam-paper products, we diversified all the morphological features associated with the leaves/plants artificially. This technology provides the capability to produce various shapes in assorted sizes which are inspired by the natural patterns to ultimately come up with a sound absorber. Due to the complex structure of the AV products, insight into their characteristics is difficult, if not impossible. Therefore, building up a brief background knowledge on the dominant mechanisms of the absorbers is highly necessary. This will be discussed in Sections 2.1 to 2.6. 2.1 Sound absorption Noise - defined as unwanted sound - can make serious trouble for most of the daily human activities. Some high frequency noises and vibrations can even lead to failure of the massive structures. Therefore, reducing the noise level in the buildings is one of the very topics that government pay a serious attention. One way to reduce the noise level is using efficient sound absorber materials. In this section, attempt is to shed light on this topic: 2.1.1 Definition In general, when sound collides with a surface, three scenarios may occur; it is either transmitted, reflected, or absorbed. The share of each depends on the acoustical properties of the surface material (see Figure 2-1). 11 Figure 2-1 Sound behavior facing a surface. Sound absorption of a material is usually determined by sound absorption coefficient (𝛼), which is the fraction of the incident sound energy which is absorbed by a medium. A material with absorption coefficient of 1 is fully absorptive which means that all incident sound wave energies are absorbed in the material, whereas, a material with absorption coefficient of 0 does not absorb the incident sound wave i.e. the sound is either reflected or transmitted through the material. The sound absorption of a medium depends on the acoustic impedance of the medium and is a function of the wave frequency and the incident angle. Sound absorption also depends on the structural and morphological characteristics (parameters like porosity, composition, density, and thickness) of the medium. 12 2.2 Transmission loss Another important acoustical concept is transmission loss (TL) which is defined as intensity decrease of a sound wave when it propagates from a source or when it transmits through a material or a structure. Mathematically, it is defined as follows and is measured in “dB” scale: 𝑇𝐿 = 10 𝑙𝑜𝑔 |𝑇𝑖𝑇𝑡| Eq. (2-1) where 𝑇𝑖 and 𝑇𝑡 represent the incident wave power and transmitted wave power respectively. Transmission coefficient is also defined as the fraction of the sound energy which is transmitted through a wall. 2.3 Prediction models for a Helmholtz resonator Helmholtz resonator (HR) is a type of resonant absorber that attenuates the noise due to viscous loss of the system especially in vicinity of its natural frequency. HR was first introduced in 1862 in a book by Helmholtz, as a device with ability of taking specific frequency out of a larger range. When a gas resonates in a cavity, Helmholtz resonance occurs. This phenomenon is widely used to study acoustical sound absorption. To obtain a higher absorption usually arrays of HRs which consist of certain number of cavities are utilized. For further studies and the mechanism of HRs we refer to (Logawa, 2017). The purpose of the prediction models is finally to understand and calculate the absorption coefficient from impedance. To do so, we need to know the governing equations of a HR to understand how a pressure difference and velocity is related in a simple resonator. 13 Upon applying a control volume on a resonator, the force balance on the control volume can be derived and a second order differential equation is derived to describe the relationship between the input pressure (𝑃) and the volume of the gas in the neck of HR (𝑣) as follows. The details of the derivation of Eq.(2-2) is described by (Logawa, 2017). 𝑃 =𝜌𝑐2𝑉𝑣(𝑡) + 𝜌𝑙𝑆𝑑2𝑣(𝑡)𝑑𝑡2 Eq. (2-2) Where 𝜌 is mass density, 𝑙 is length of the neck, 𝑐 is speed of sound in air, 𝑉 is volume inside the cavity and S is surface area. Here, 𝜌𝑐2𝑉 refers to the compliance of the resonator and 𝜌𝑙𝑆 is generally referred to acoustic inertance (Logawa, 2017). Eq. (2-2) is a classical model and is usually accurate to predict the impedance of a simple resonator. The application of this equation will be considerably limited if our medium consists of arrays of various configurations of many HRs. To analyse a complex system with a complex configuration of resonators and to overcome this problem, we applied Transfer Matrix method. 2.3.1 Transfer matrix method Transfer Matrix Method (TMM) is used when a system can be divided into multiple subsystems. TMM is used to connect the layers using matrices which define the acoustical properties of each layer near the boundary (see Figure 2-2). 14 Figure 2-2 Transfer Matrix Model (TMM) representation for a multi-layer medium. For example, the Transfer Matrix for a simple solid and/or fluid layer is defined as: [𝑇] =[ cos (𝑘𝑧𝑑𝑖) 𝑗𝜔𝜌𝑘𝑧𝑠𝑖𝑛(𝑘𝑧𝑑𝑖)𝑗𝑘𝑧𝜔𝜌𝑠𝑖𝑛(𝑘𝑧𝑑𝑖) cos (𝑘𝑧𝑑𝑖) ] Eq. (2-3) Where 𝜔𝜌𝑘𝑧 equals to 𝑍𝑐; characteristic impedance (Rayl or 𝑃𝑎. 𝑠.𝑚−1), 𝜔 is the frequency (𝐻𝑧), 𝑘𝑧 is the wave number is Z direction and 𝑑𝑖 is the thickness of laye i. The definition of this matrix model, however, is much more complex for the porous structures. Here, we elaborate some of the proposed models to describe the transfer matrix for a porous material: 2.3.1.1 Porous models The empirical models are used to obtain 𝑍𝑐 and 𝑘 for transfer matrix. Here we explain the most commonly used models in porous media. 15 2.3.1.1.1 Delany – Bazley model Delany and Bazley (Delany & Bazley, 1969) fulfilled a large set of experiments in a specific frequency range (250-4000 Hz) for 200 measurements. This model is valid under two different conditions; first, porosity should be greater than 95%, second, the model is valid for frequencies lower than 1% of the flow resistivity. The frequency range for their experiments was from 2000 to 80000 N.s/m3. Considering 𝜔 as angular velocity, 𝜌0 as air density (1.213 kg/m3), 𝑐0 as speed of sound in air medium (342.2 m/s), 𝑓 frequency, k as propagation constant, 𝜌0 as ambient air density and 𝜎 flow resistivity then: 𝑍𝑐 = 𝜌0𝑐0 (1 + 0.495 (𝑓𝜎)−0.754− 𝑗0.0754 (𝑓𝜎)−0.732) Eq. (2-4) 𝑘 =𝜔𝑐0(1 + 0.848 (𝑓𝜎)−0.7− 𝑗0.164 (𝑓𝜎)−0.595) Eq. (2-5) 2.3.1.1.2 Miki model Improvement to the model proposed by Delany and Bazley (Delany & Bazley, 1969) were made by Miki (Miki, 1990) to increase the model accuracy in range of low frequencies (frequencies lower than 0.01𝜎). In this model, 𝑍𝑐 and 𝑘 are defined as follows: 𝑍𝑐 = 𝜌0𝑐0 (1 + 0.070 (𝑓𝜎)−0.632− 𝑗0.107 (𝑓𝜎)−0.632) Eq. (2-6) 𝑘 =𝜔𝑐0(1 + 0.109 (𝑓𝜎)−0.618− 𝑗0.164 (𝑓𝜎)−0.618) Eq. (2-7) 16 2.3.1.1.3 Allard - Champoux model Allard and Champoux (1975) proposed a model in which other morphological parameters such as viscous and thermal characteristic lengths (𝛬 and 𝛬′ respectively), porosity (𝜑), and tortuosity (𝛼∞) were included. The viscous and thermal characteristic lengths are average dimensions of the cells related to viscous and thermal losses, respectively. Thermal characteristic length is twice the ratio of volume to surface area in connected pores and viscous characteristic length is similar to this ratio weighted by the local velocity of fluid in the volume and on the surface. Moreover, porosity of a material is defined as the ratio of the volume of the connected pores to the total volume of the material and tortuosity is defined as the measure of propagation path complexity or streamline curvature. The characteristic impedance (𝑍𝑐) and propagation constant (𝑘) can be derived from the effective density and bulk modulus of the medium as follows: 𝑍𝑐 = √𝜌𝑒𝐾𝑒 Eq. (2-8) 𝑘 = 𝜔√𝜌𝑒𝐾𝑒 Eq. (2-9) The effective density and bulk modulus of the medium was also defined as: 𝜌𝑒 = 𝛼∞𝜌0 (1 +𝜎𝜑𝑗𝜔𝛼∞𝜌0√1 +4𝑗𝛼∞2 𝜂𝜌0𝜔𝜎2𝛬2𝑥2) Eq. (2-10) 𝐾𝑒 =𝛾𝑃0𝛾 − (𝛾 − 1) (1 +8𝜂𝑗𝛬′2𝑁𝑃𝑟𝜔𝜌0√1 +𝑗𝛬′2𝑁𝑃𝑟𝜔𝜌016𝜂 )−1 Eq. (2-11) 17 where 𝜂 is viscosity of air, 𝑁𝑃𝑟 as Prandtl number (= 0.72), 𝛾 is specific heat ratio of air (= 1.4), and 𝑃0 is the air pressure. 2.3.1.1.4 Attenborough model To improve the predictions made by Allard and Champoux, Attenborough (1983) proposed a model considering the shape factor of a material. Shape factor is a constant for a given material and associated with the deviation of a pore from a circular cylinder. Here, the characteristic impedance (𝑍𝑐) and propagation constant (𝑘) can be derived from Eq.8 and Eq.9 respectively. The effective density and bulk modulus of the medium in this model is defined as follows: Here, 𝜌𝑥 and 𝐶𝑥 represent complex density and compressibility functions and defined as: where 𝜆 =12𝑆𝑝√8𝜌0𝜔𝛼∞𝜎𝜑 and 𝑇(𝜁) =𝐽1(𝜁)𝐽0(𝜁)⁄ is the ratio between Bessel functions of zero and first order. In this model, 𝑆𝑝 represents the shape factor of the pores. 𝜌𝑒 = (𝛼∞𝜑)𝜌𝑥 Eq. (2-12) 𝐾𝑒 =1𝐶𝑒=1𝜑𝐶𝑥 Eq. (2-13) 𝜌𝑥 = 𝜌0 (1 −2𝑇(𝜆√−𝑗)𝜆√−𝑗)−1 Eq. (2-14) 𝐶𝑥 =1𝛾𝑃0(1 + (𝛾 − 1)2𝑇(𝜆√−𝑗𝑁𝑃𝑟)𝜆√−𝑗𝑁𝑃𝑟) Eq. (2-15) 18 2.4 Origami The word Origami comes from Japanese roots of “Ori” meaning to fold and “kami” meaning paper i.e. the art of folding paper and making shapes without cutting. When cutting is added to this technique, however, it is named Kirigami. Nowadays, this method is applied in many industries from making simple toys to complex shapes in architecture and solar panels (Guest, 1994; Jahedi, 2010; Miura, 1992). Paper is a perfect material for Origami and Kirigami since it is sustainable, light, and inexpensive (Nojima & Saito, 2006). Folding the paper in order to get a desirable shape is, however, a very time taking process which is not ideal for industry. To make this process more effective, (Nisser, 2016) used sensors to fold and make Origami shapes. (Dickey, 2012) innovated self folding polymers by exposing them to the light. The method proposed by (Sina, 2015) is, however, specifically innovated for Origami application in paper industry. (Sina, 2015) developed a novel method to make papers self-fold by exposing them to heat. This method can be used in many industries such as packaging, furniture, acoustics. Light weight, flexibility, and ability to make repetitive patterns are amongst the major reasons to use this method in acoustics. 2.4.1 Formation After making foam-paper samples, we mount origami shapes on top of them to obtain samples for acoustic tests. To reach this goal, we need to design Origami pattern. We used AutoCAD as the most popular and user-friendly software to draw the patterns. Both cutting and creasing lines should be considered carefully in drawings. Moreover, the location of the crease lines - which later will be the pivots – is also very important. We 19 tested many trial and error patterns to gain the desired ones. The outcome of this part is two-dimensional printed papers which are ready to be cut by the cutting machine and eventually, making a three-dimensional self-folding product. Figure 2-3 “Pyramid” designed with AutoCAD software. 2.4.1.1 Cutting For cutting the Origami shapes, we use a Graphtec cutting machine (see Figure 2-4). The table size is 68.5”x 36” with two 500 grams and 1000 grams down force heads (see Figure 2-5). The final designs of AutoCAD are transferred to CorelDraw software, the cutting machine – which is able to run with various speeds and forces - reads the files from the software (Sina, 2015). Upon setting these parameters, the machine cuts and creases the pattern on the paper which is already mounted on the Graphtec plate. 20 Figure 2-4 Graphtec FC2250-180 Figure 2-5 Graphtec 2250-180 dual heads 2.4.1.2 Polymer attachment Polymer attachment is a process to convert the 2D paper patterns to 3D ones. This process provides the capability of self-folding to the 2D papers. The process is initially introduced by (Sina, 2015) by attaching pre-strained polymers in certain points of the patterns designed in the drawings. To do so, (Sina, 2015) modified one of the cutting tips with a soldering-iron tip. The process starts with fixing the pre-strained polymer sheet on 21 the creased-lines of the paper of the same size, soldering iron presses down the polymer to the paper and heats the polymer on the defined points. The polymer is then melted down and is welded to the paper on the defined points. At this point, our polymer-attached paper is ready to be used as a self-folding prototype. To make the 3D structure, we put the polymer-attached paper into an oven of 100oC. The polymer, now reaches the glass transition temperature where starts to shrink. Consequently, stretches the paper and folds it from the crease lines. These folded parts turn the 2D shape into 3D. Figure 2-6 Polymer attachment to the paper (adapted form (Sina, 2015)) 2.5 Foam-paper (Radvan & Gatward, 1972) used foam in production of paper process and also introduced a foam making process named after his name, Radfoam process which is confirmed with several researchers (Smith & Punton, 1975; Tringham, 1974) as a high efficiency process in terms of making the paper more porous and more unified. On the other hand,(M. Deng, 2009), has developed a foam making process resulting in high porosity foam. (Cervin et Heat 22 al., 2013) has done a research on the drying of the foam and how it affects the porosity of the foam and claimed that the porosity in freeze-drying is lower than NFC stabilized foam. (H. Sehaqui, Zhou, & Berglund, 2010) showed that cell structure of foam are the key to improve mechanical specifications of the foam. (Jahangiri, 2013) has developed a cellulose based porous foam called foam-paper. This product is made by adding surfactant to the pulp suspension, mixing air into suspension, and drying in lab environment (air-drying process). He studied the effect of air-content on porosity leading to final product porosity of 99%. He also studied the air flow resistivity of the foam-paper. (Korehei, 2016) investigated the effect of drying methods of foam-paper structure and showed that freeze-drying makes the foam-papers even more porous. In this study, we used the method proposed by (Jahangiri, 2013) to produce foam-paper. A pulp suspension which consists of water and pulp which later turns into suspension of foam and fibre. First, The Caframo Brinkmann RZR1 type mixer applies the shear force to the pulp, meanwhile a certain amount of surfactant is added, both to get higher air content. The velocity of the mixer can be changed to get higher air content that is the same for amount of the surfactant. In second phase, we air dry the sample; to do so, the foam fibre suspension will be dewatered by vacuuming in a Buchner funnel then dried in the room temperature in Constant Temperature and Humidity room for 24 hours. 23 Figure 2-7 Drying foam-papers; exposed to air and a heater. In summary, the research works reviewed in the literature provide insight into the systematic development of the modules required to produce the artificial vegetation. In this thesis, we used and combined all the previous methodologies developed by Jahangiri et al. (Jahangiri, 2013) and Sina et al. (Sina, 2015) to produce a novel and efficient sound absorber based on the idea of Artificial Vegetation. We also developed an automated modeling platform to understand the effect of various parameters on the acoustical behaviour of the new product. 3 Experimental Procedure and Results Production of various samples with different specifications were seen to investigate acoustical behaviour of the AV products. In this chapter, the production procedure and the required experimental apparatuses are described. The results will be discussed at the end of this chapter. 3.1 Impedance tube test Our research used the Impedance tube test as it is the most commonly-used method in the field to evaluate acoustical performance. We used two ¼ inches B&K 4178 phase matched microphones, each connected to a B&K 2269 preamp. A signal analyzer (Soundbook SINUS GmbH) received the polarization voltage provided by preamps. Two tubes measured the samples’ absorption: i) A 101.5 cm long circular steel with a 10-cm diameter for the low frequency zones and ii) a 30 cm long, 2.9 cm diameter in the range of high frequencies. The high and low frequency experiments overlap in the measurements in the frequencies between 1000 Hz and 2000 Hz resulting in approximately 15% discrepancy between the results in this range. This happens since we use different samples of different sizes for these two experiments which can be resolved by taking an arithmetic average of the data in this frequency zone. We measure the low frequency zone between 200 Hz and 2000 Hz and the high frequency zone between 1000 Hz and 6000 Hz respectively. Depending on the application of the sound absorber the 25 range of interest for the frequencies can differ, however, the frequencies below 1000 Hz is paid less attention in resolving building’s noise pollution. (Wolkesson & Wolkesson, 2013) carried out a comprehensive study on the design and the major calculations of an impedance tube and explained the influence of spacing and number of microphones in the measurements. Figure 3-1 Impedance tube setup 26 Figure 3-2 two ¼ inches B&K 4178 phase matched microphones Microphones inside the tube measured sound pressure and Soundbook calculated transfer function using outcoming data resulting in the calculation of absorption coefficient using equation 3-2. We first calculated the reflection coefficient R, derived through the dimensions and geometry of the tube and the transfer function (H). Figure 3-3 impedance tube with two microphones 27 𝑅 = |𝑅|𝑒𝑗𝜙𝑅 = 𝑅𝑟 + 𝑗𝑅𝑖 =𝐻 − 𝑒−𝑗𝑘𝑠𝑒𝑗𝑘𝑠 − 𝐻𝑒𝑗2𝑘(𝑙−𝑠) Eq. (3-1) 𝛼 = 1 − |𝑅|2 Eq. (3-2) Our experiments worked on the assumption that the impedance tube to be a fully sealed. Allowing the total fraction of the incident sound that is not reflected to be absorbed. If the wave-length of the highest frequency was longer than the tube diameter, sound propagation in the impedance tube was defined as plane wave propagation in a duct. The ISO 10537:1998 defined the higher frequency limit for an impedance tube as: 𝑓𝑢 < 0.58𝐶0𝑑 Eq. (3-3) 𝑓𝑢 < 0.45𝐶0𝑠 Eq. (3-4) Another limitation based on ISO 10537 mentioned that upper frequency was dependant on the distance between two microphones. In equation 3-3 and equation 3-4, 𝐶0, 𝑑, and 𝑠 represent speed of sound, internal diameter of the tube, and spacing between two microphones. k0 is the wave number in the air, 𝜑𝐼 and 𝜑𝑅 are phase shift of incident and reflected wave, ?̂?𝐼 and ?̂?𝑅 are the magnitude of incident and reflected wave. A lower band for low-frequency, that is for the wavelength which have the same pressure in the microphones: 𝑓𝑙 > 0.05𝐶0𝑠 Eq. (3-5) So, the sound pressure for incident and reflected waves were defined as follows: 28 𝑝𝐼 = ?̂?𝐼 𝑒(𝑗𝑘0𝑥+𝜑𝐼)𝑒𝑗𝜔𝑡 Eq. (3-6) 𝑝𝑅 = ?̂?𝑅 𝑒(𝑗𝑘0𝑥+𝜑𝑅)𝑒𝑗𝜔𝑡 Eq. (3-7) Where 𝜑𝐼 is the incident wave and 𝜑𝑅 is the reflected waves phase shift and ?̂?𝐼 and ?̂?𝑅 are magnitudes of incident and reflected waves, respectively 𝑥 is distance from surface of the sample and 𝑘0 is the wave number. The sound pressure for microphones 1 and 2 were calculated using: 𝑝1 = 𝑝𝐼(𝑥1) + 𝑝𝑅(𝑥1) = ?̂?𝐼 𝑒−𝑗𝑘𝑥1 + ?̂?𝑅 𝑒𝑗𝑘0𝑥1 Eq. (3-8) 𝑝2 = 𝑝𝐼(𝑥2) + 𝑝𝑅(𝑥2) = ?̂?𝐼 𝑒−𝑗𝑘𝑥2 + ?̂?𝑅 𝑒𝑗𝑘0𝑥2 Eq. (3-9) Here, 𝑥1 and 𝑥2 are defined as explained in Figure 3-3. Moreover, the transfer function was also defined as: 𝐻𝑖𝑗 =𝑝𝑗𝑝𝑖 Eq. (3-10) Considering Equation 3-8, the transfer function between microphone 1 and 2 was computed as: 𝐻12 =𝑝2𝑝1=𝑒−𝑗(𝑘0𝑥2+𝜑) + 𝑅 𝑒𝑗(𝑘0𝑥2+𝜑)𝑒−𝑗(𝑘0𝑥1+𝜑) + 𝑅 𝑒𝑗(𝑘0𝑥1+𝜑) Eq. (3-11) where, 𝑅 is the reflection coefficient which is the ratio of reflected and incident pressures. This equation was re-formatted as: 𝑅 =𝐻12 − 𝐻𝐼𝐻𝑅 − 𝐻12𝑒𝑗𝑘02𝑥1 Eq. (3-12) Where, 𝐻𝐼 and 𝐻𝑅 are the transfer functions for incident and reflected waves between the microphone positions which were defined as: 29 𝐻𝐼 =𝑝𝐼(𝑥2)𝑝𝐼(𝑥1)= 𝑒𝑗𝑘0𝑠 Eq. (3-13) 𝐻𝑅 =𝑝𝑅(𝑥2)𝑝𝑅(𝑥1)= 𝑒−𝑗𝑘0𝑠 Eq. (3-14) The fraction of the incident wave that propagated inside the material to the total energy propagated out was defined as the absorption coefficient. The amount of the incident wave not reflected had to be absorbed. The value of the absorption coefficient was defined in equation 3-2. Finally, equation 3-15 shows the impedance surface of the material consisting of a real part as energy loss and an imaginary part as phase were calculated as: 𝑍𝑠 = 𝑍01 + 𝑅1 − 𝑅 Eq. (3-15) where, acoustical characteristic impedance of air is: 𝑍0 = 𝜌0𝑐0. 3.2 Results The production procedure of the foam-paper samples was described in Section 2.2. The test matrix for the acoustic tests in both low and high frequency zones in the impedance tube was also elucidated earlier in Chapter 2. 30 Figure 3-4 Sound absorption coefficient of foam-paper; consistency:0.5; Thickness 1 cm in low frequency Figure 3-4 shows the absorption coefficient of a 1cm foam-paper with consistency of 0.5 and air content of 80% in the range of 200 – 2000 Hz which defined the range of our low-frequency tests in a three-microphone impedance tube. The graph qualitatively shows that increasing the frequency increased absorption coefficient. The graph also shows that at a reference frequency of 1000 Hz, approximately 20% of the sound was absorbed which justified the work done by (Jahangiri, 2013). The graph also shows a peak at frequency of 200 Hz due to the resonance frequency of the tube and the sample; and it is negligible because it did not show the real absorption properties of the sample and it was part of the experimental errors. 31 Figure 3-5 Sound absorption coefficient of foam-paper; consistency:0.5; Thickness 1 cm in high frequency Figure 3-5 shows the high frequency result for the same sample represented in Figure 3-4. There is an overlap in frequency range between 1000 Hz and 2000 Hz. As shown in Figure 3-4 and Figure 3-5, different results were calculated between the 1000 Hz and 2000 Hz for the absorption coefficient using various low and high frequency methods. This difference is within 15 percent and is computed higher for the high-frequency test. Our findings showed a common phenomenon in impedance measurement that occurred due to a change in the size of the testing sample and the difference between the low and high frequency methods. In our calculations, we used the average of these two tests at each frequency in the overlapped zone. The measured absorption coefficients of the foam-papers made through similar parameters (similar foam air-content, consistency, and 32 drying methodology) showed an uncertainty within 10 percent. This uncertainty was elaborated through the differences in the meandering and tortuous structures of the similar foam-papers. Overall, the results showed that the range of the measured absorption coefficients were relatively lower compared to the one measured in the high-frequency tests. The maximum achievable absorption coefficient in the low frequency tests occurred near 2000 Hz which was approximately 0.3. Moreover, we observed an overshoot in the results close to the frequency of 200 Hz (≈ 250 Hz) that showed the resonance frequency of the tube. Absorption in the range of high frequencies showed an ascending trend as frequency elevates. The effect of production parameters had a significant effect on the results of the absorption coefficient. Foam-papers with different air-contents, thicknesses, and consistencies were tested. We also mounted an Origami design based upon the geometry of the natural “grass” (see Figure 3-6). 33 Figure 3-6 Artificial Vegetation on the foam paper samples to understand the effect of an Origami-based geometry on the overall acoustical behaviour of a foam-paper. A grass structure was produced in different lengths in our measurements. This was the first step to design efficient AV-products. 34 Figure 3-7 Sound absorption coefficient of 1 cm thick foam-paper; consistency:0.5. (no grass) Figure 3-8 Sound absorption coefficient of 1 cm thick foam-paper and 1 cm grass; consistency:0.5. Figure 3-7 and Figure 3-8 show the low and high frequency measurement results for both a foam-paper and a grass-mounted foam-paper respectively. In these measurements, the foam-paper specifications were identical with the only variable being the mounted geometry of artificial grass. The results showed that addition of a layer as tick as 1 cm of grass on top of the foam-paper slightly (within 5%) improved the acoustical behaviour of the final product compared to the original sample of the foam-paper. The successful improvement of the grass structure on the foam-paper built the foundation for our 35 fundamental hypothesis regarding the constructive effect of AV-structures on acoustical behaviour of the foam-papers. Two variables we tested: i) The consistency of the pulp solution to make the foam-papers; ii) The length (thickness) of the grass structure. Figure 3-9 Sound absorption coefficient of 1 cm thick foam-paper; consistency: 1. Figure 3-10 Sound absorption coefficient of 1 cm thick foam-paper and 1 cm grass; consistency:1. Figure 3-9 and Figure 3-10 show the influence of the higher consistency foam-papers (the foam consistency is within 1%) on the acoustical behaviour of the final product. The results clearly showed that the absorption coefficient was improved approximately 10 percent compared to the samples with lower consistency (the foam consistency was 36 within 0.5%). The higher consistency led to more material involving in the structure of the foam-paper causing a more efficient performance in our frequency ranges. High frequency measurements in Figure 3-9 depict a sinusoidal behaviour. This happens due to the presence of the large holes in the structure of the foam-paper as the direct result of the heterogeneity in the foam structure. Figure 3-11 Sound absorption coefficient of 1 cm thick foam-paper and 3 cm grass; consistency:1. Figure 3-12 Sound absorption coefficient of 1 cm thick foam-paper and 6 cm grass; consistency:1. Figure 3-11 and Figure 3-12 show the variation of absorption coefficients of our AV products for a wide range of low and high frequencies for the Origami-grass thickness of 3cm and 6cm respectively. The results show a significant improvement of the absorption 37 coefficient over the range of frequencies by increasing the thickness of the Origami-grass at the same foam-paper properties. The tortuous structure of the AV-product was the main reason for improvement intensified by increasing the thickness of the Origami-grass structure. We also observed that the absorption coefficient of our thickest sample (6cm Origami-grass) showed a large bump close the frequency of 1000 Hz (resonance frequency) depicting that our product behaviour was getting similar to a Helmholtz-resonator. Figure 3-13 Q-dP graphs of a 1.5 cm thick foam-paper and a 1.5 cm thick artificial grass product. 38 Figure 3-14 Comparison between the sound absorption result of an AV product and a commercial fibreglass absorber (results of the commercial absorber are provided by Kinetics Noise Inc.) Figure 3-13 shows the flow rate versus the pressure drop for both a 1.5cm thick foam-paper and a 1.5 cm thick artificial grass product. The slope of the Q-dP curves is an inverse function of the air-flow resistivity of the materials. The results show both products’ resistivity to flow was within the same range, both materials were considered as very low resistive products to the air-flow. This proves that although the artificial grass structure improved the absorption coefficients dramatically, it did not change the flow resistivity by a great factor. Therefore, the AV products could be considered as efficient materials for sound absorption. This was further confirmed by the results provided in Figure 3-14. The graph shows the sound absorption results of an AV product and a commercial fibreglass absorber. The results show that the AV product was compatible to the commercial 39 fibreglass product (an average 35% improvement was achieved within this range of frequencies). In summary, we developed a systematic methodology to test the acoustical behaviour of our Artificially Vegetated products for various range of the sound frequencies. We showed that Origami-grass mounted on a foam-paper significantly improved the acoustical behaviour of the foam-papers in the range of our testing frequencies. Our experiments led this to infer that AV products provide a wide avenue to produce novel and efficient building insulators in the near future. 4 Modeling and Optimization Parametric optimization of acoustical behaviour of porous Origami-based products is relatively unexplored. In this chapter, we attempt to advance a model that was originally developed by Logawa (Logawa, 2017), to follow acoustical behaviour of our AV products to suggest the optimized combination of parameters for efficient production. Based upon the experimental observations (see Chapter 3), we assume our AV products are Helmholtz resonators due to their tortuous structure. We applied the idea of transfer matrix method to solve the governing equations for a multilayer porous media. To simplify the problem, we approximated the characteristic of porous media with their equivalent fluid calculated through the model proposed by (Attenborough, 1983). Here, after the validation process, we analyzed the sensitivity of our model to the input parameters and suggested the physically-reasonable parametric combinations to follow our experimental results. After we examined all aspects of our model and the influence of all parameters, we used our findings to optimize the production procedure to develop more efficient AV absorbers in the future. 4.1 Modeling Our research aimed to understand the acoustical behaviour of a multilayer complex structure in a broad range of sound frequencies. Due to the diverse number of degrees of freedom, the exact numerical simulation of such a complicated problem was difficult. in designing controlled and accurate experimental procedures to study the sound 41 absorption in this multi-dimensional tortuous structure; However, other mathematical methodologies describe. The AV-mounted foam-paper product that we studied was a complex form of a Helmholtz resonator due to multiple cavities in the Origami-grass structure backed with foam-paper. Therefore, we applied the Helmholtz resonator theory to model this system. As we had two layers in our system (foam-paper and Origami-grass) we employed transfer matrix method to transfer the sound energy specification propagating from one layer to another. We studied sound propagation in the system by dividing Helmholtz resonator arrays into multilayer arrays and used transfer matrix method in the boundaries of each layer. We used the formulations provided in Chapter 2 for our numerical computations. 4.2 Results Our developed model was validated through some systematic numerical tests before applying in the practical simulations. The accuracy of the model was tested for some limits. First: theoretically, single layer material with n cm thickness must behave equally as a double layer of the same material but with n/2 cm thicknesses (see Figure 4-1). If all the specifications of the material are the same, two n/2 cm layers of the material should make the exact n cm thick material. Otherwise, the model was not valid for two materials with two different specifications. 42 Figure 4-1 n cm and two n/2 cm material Figure 4-2 shows the absorption coefficients of the above-mentioned test. Both blue and red lines show the exact same trend and values that were intuitive and depict no difference in results for n cm and two layers of n/2 cm samples. Figure 4-2 Model test: difference between n cm material and two n/2 cm the same material Second: the test justified that the result of absorption coefficient was independent of the layer number. we used a material with two identical layers. In each simulation we changed the thickness of each layer. We anticipated that the results of the numerical simulation be identical for the similar overall thicknesses. The blue lines in Figure 4-3 show the variation of the thickness for the layer 1 and the red lines show the same variation for layer 2. The 43 results of both simulations illustrated identical numerical values for the overall thicknesses and consequently the correctness of our hypothesis. Figure 4-3 changing the thickness of the layer 1 from 1cm to 5 cm Vs doing the same for the second layer 4.3 Fitting the model to experimental data We began with a set of parameters used by (Samimi, 2012) to fit our numerical model to the experimental results. (Samimi, 2012) had researched on various materials and obtained the optimized values for her experiment data. We built on this set of parameters as our initial start in of finding the best model to follow our experimental results. 44 Figure 4-4 Model Vs Experiment result 1cm foam and 6cm grass Figure 4-5 Tortuosity change of the foam from 2 to 4 Figure 4-6 Porosity change of the foam from .9 to.8 Figure 4-7 Changing of shape factor from 3.4 to 8 Here, the red curve shows the experimental result, while the blue curve is the primary fitting result form model. Figure 4-4 to Figure 4-7 show the sequence of the attempts based on changing various parameters to find the best fit for our experimental results (the pink curve on Figure 4-7 shows the best fit). The results of the best fit attained were based on the parameters presented in Table 4-1 as follows: 45 Thickness Tortuosity Porosity Shape Factor Flow Restivity Foam-paper 2.0 2.0 0.88 0.1 5000 Origami-grass 2.5 1.0 0.99 3.4 1000 Table 4-1 Parameters for the best fit to the experimental results In summary, we developed a methodology based on the idea proposed by (Attenborough, 1983) to follow the final behaviour of the acoustical properties of a multilayer material such as our AV-products. This modeling is useful to test the acoustical properties of novel multilayer porous material without getting involved in time-consuming experimentation. The model also helps optimize related parameters to produce more efficient sound absorbers. 4.4 Sensitivity analysis The results of the sensitivity analysis of the AV product to various parameters were examined in this section. The test matrix is presented in Table 4-2. In all graphs associated with this section, the red curves show the increase in the evaluation parameter and the blue curves show the reduction. Sensitivity test Test number thickness tortuosity porosity Shape Factor Flow Resistivity A A1 1,2, 6 1 0.4 - 0.9 1 10000 A A2 1 1 0.9 1 1000-20000 A A3 1 1 0.6 1 1000-20000 B 3 1 0.6 1 1000-20000 C 9 1 0.6 1 1000-20000 D 1 0.5 - 5 0.9 1 10000 E 1 1 0.9 0.5-10 10000 Table 4-2-Test matrix for sensitivity analysis. 46 Figure 4-8 Test A1: Changing the foam-paper porosity from 0.4 to 0.9 for the thicknesses of 1,2 and 6 cm. Figure 4-8 shows the sensitivity of the absorption coefficient to the porosity in the range of 0.4 - 0.9 for three various thicknesses of 1, 2, and 6 cm (represented by blue, red, and black family of curves respectively). The results show that increasing the thickness of foam-paper of the AV products led to a significant increase of the absorption coefficients in low frequency zones. The graph also indicates that the variation of the absorption coefficient diminishes in the range of high porosities (> 0.8) compared to the relatively lower porosities. Figure 4-9 shows the sensitivity of the absorption coefficient to the air-flow resistivity. The results showed that increasing the value of the air-flow resistivity generally increases the average value of the absorption coefficient in various frequencies. In practice, the air flow resistivity increased by thickening the foam-paper and increased the foliage of the vegetated region. Figures 4-10, 4-11 and 4-12 show the results of 1cm 2cm 3cm Porosity=0.4 Porosity=0.9 Porosity=0.4 Porosity=0.9 Porosity=0.4 Porosity=0.9 47 porosity of 0.6, thickness of 3cm and 9cm respectively o investigate the similar effect of the air-flow resistivity at different porosities and thicknesses. Figure 4-9 Test A2: Sensitivity analysis of air flow resistivity from 10000 to 1000 and 20000 (porosity = 0.9) Air flow resistivity=1000 Air flow resistivity=20000 48 Figure 4-10 Test A3: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 (thickness=1cm; porosity = 0.6) Figure 4-11 Test B: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 ( porosity = 0.6 and thickness = 3 cm) Air flow resistivity=1000 Air flow resistivity=20000 Air flow resistivity=1000 Air flow resistivity=20000 49 Figure 4-12 Test C: Sensitivity analysis of air-flow resistivity from 10000 to 1000 and 20000 ( porosity =0.6 and thickness = 9 cm) Figure 4-13 and Figure 4-14 show the sensitivity of absorption coefficient on both tortuosity and shape factor of samples. The results generally showed that deviating these parameters from 1.0 resulted in more fluctuations in values of absorption coefficients. We inferred that deviation from 1.0 in shape factor led to more non-spherical average pore shapes and consequently more resonance in the system. Deviation from 1.0 in tortuosity also resulted in increased meandering channels between the pores in the structure resulting in more resonances in the system. Tortuosity was part of characteristic of the materials and it could not be easily changed as a parameter. Air flow resistivity=1000 Air flow resistivity=20000 50 Figure 4-13 Test D: Sensitivity analysis of the tortuosity of layer 2, from 0.5 to 5 Figure 4-14 Test E: Sensitivity analysis of the shape factor of layer 2, from 1 to 10 In summary, we developed a mathematical model based on the idea of Attenborough (1983) and the method of Transfer Matrix to address the acoustical properties of multi-layer sound absorbers. We minimized our model parameters versus our experimental results and validated our model. After the validation process, we used the model implications to understand the acoustical properties of various multi-layer absorbers and to comprehend the effect of each parameter on the absorption coefficient at a particular sound frequency. Once the effect of each parameter is understood, the model can be used to predict a desired absorption coefficient at a particular frequency for a specific multi-layer material. Increasing the tortuosity Increasing the shape factor 51 5 Summary, Conclusions, and Final Remarks In this chapter, the main scientific and engineering contributions of this thesis are highlighted followed by some practical recommendations based on both experimental and modeling results for the future. We attempted to develop a novel green and biodegradable sound absorber product as an alternative to porous absorbers. We developed this product by combining the foam-paper material and a novel technology borrowed from the Origami techniques. This process – which we named Artificial Vegetation (AV) – was inspired by the structure of planted grass in a soil-based platform. We mimicked this structure and produced a novel material with excellent acoustical properties applicable to building construction industry. We reviewed the major methodologies recommended by the foam and Origami techniques. We repeated the work done by (Jahangiri, 2013) and (Sina, 2015) to develop the base products used to produce Artificial Vegetative sound absorbers. Through this procedure, we also validated the results of our tests with the ones already introduced in the literature. In the foam-forming paper-making technology, we reproduced the experimental setup proposed by (Jahangiri, 2013) and developed a series of foam-papers with various foam air-contents as a platform for our novel AV products. We also repeated the Origami method used by (Sina, 2015) to reproduce the self-folding paper technology as a platform to produce the additional porous structure on top of the foam-paper. We proceeded by combining the foam-paper structure with the novel artificial grass geometries developed through the Origami method to produce the Artificial Vegetative 52 sound absorbers. To confirm the positive effect of the grass geometry on the acoustical behavior of the foam-papers, we designed systematic acoustical experiments to examine the sound absorption efficiency of the novel AV products. The experimental results showed that increasing the thickness of the artificial grass layer increased sound absorption of the AV products significantly while the products flow-resistivity remained relatively unaltered. Overall, the experiments showed an eye-catching improvement in the acoustical behavior of the novel AV products. To understand the influence of the different parameters on the acoustical behavior of the AV products and to come up with new efficient designs, we advanced a mathematical model to simulate the absorption coefficients of multi-layer porous structures. We studied the effects of various parameters such as layer-thickness, porosity, shape-factor, flow-resistivity, and tortuosity. Our experimentation assisted in understanding how the absorption coefficient behaved in different frequencies. In conclusion, we proved that the AV product has excellent acoustical properties that can be used as a sound absorber in buildings. Our research findings are the necessary first step to a larger research and development programs to produce novel renewable, biodegradable cellulose based sound absorbers. In future, various foam-paper materials as well as different designs for the artificial grass could help optimize the sound absorptivity of the AV products further. 53 References [1] Ahmadi, H. (2016). Cellulose-Mycelia Foam: Novel Bio-Composite Material. Master’s Thesis, (University of British Columbia). [2] Attenborough, K. (1983). Acoustical characteristics of rigid fibrous absorbents and granular materials Po, 73(March 1983), 785–799. [3] Aylor, D. (1972). Sound Transmission through Vegetation in Relation to Leaf Area Density, Leaf Width, and Breadth of Canopy. The Journal of the Acoustical Society of America, 51(1B), 411. [4] Azkorra, Z., Prez, G., Coma, J., Cabeza, L. F., Bures, S., lvaro, J. E., Urrestarazu, M. (2015). Evaluation of green walls as a passive acoustic insulation system for buildings. Applied Acoustics, 89, 46–56. [5] Bullen, R., and Fricke, F. (1982). Sound propagation through vegetation. Journal of Sound and Vibration, 80(1), 11–23. [6] Burns, S. H. (1979). The absorption of sound by pine trees. Journal of Acoustic Society of America, 65(3), 658–661. [7] Cervin, N. ., Andersson, L., J. B. S, N., Olin, P., Bergstrm, L., and Wgberg, L. (2013). Lightweight and strong cellulose materials made from aqueous foams stabilized by nano brillated cellulose. Biomacromolecules, 14(2), 503–511. [8] Cook, D. I., and Van Haverbeke, D. F. (1971). Trees and shrubs for noise abatement., Res. Bull., 77 p. [9] Delany, and Bazley. (1969). Acoustical properties o f fibrous absorbent materials, (3). [10] Dickey, M. D. (2012). Soft Matter Self-folding of polymer sheets using local light 54 absorption 8(6). [11] Fricke, F. (1984). Sound attenuation in forests. Journal of Sound and Vibration, 92(1), 149–158. [12] Guest, S. . (1994). Folding Triangulated Cylinders. Journal of Applied Mechanic, 61, 773. [13] H. Sehaqui, M. S., Zhou, Q., and Berglund, L. a. (2010). Mechanical performance tailoring of tough ultra-high porosity foams prepared from cellulose suspensions. Soft Matter, 6, 1824. [14] Horoshenkov, K. V. (2013). Acoustic properties of low growing plants. The Journal of the Acoustical Society of America, 133(5), 2554–2565. [15] Jahangiri, P. (2013). Novel cellulose based foam-formed products : applications and numerical studies, PhD thesis(November), University of British Columbia. [16] Jahedi, M. (2010). Folding in Architecture: A Spatial, Structural and Material Investigation, University of British Colombia, Vancouver. [17] Korehei, R. (2016). Effects of drying strategies and micro brillated cellulose ber content on the properties of foam-formed paper. Journal of Wood Chemistry and Technology, 36(4), 235–249. [18] Logawa, B. (2017). Improving The Sound-Absorption of Cross-Laminated Timber Panels Using Resonant Absorbers Layers. Masters' Thesis, (June), University of British Columbia. [19] M. Deng. (2009). Preparation of nanoporous cellulose foams from cellulose-ionic liquid solutions. Materials Letters, 63(21), 1851–1854. [20] Miki, Y. (1990). Aeoustical properties of porous materials Modifications of Delany-Bazley models. Journal of Acoustic Society of Japan, 11(1), 19–24. 55 [21] Miura, K. (1992). Concepts of Deployable Space Structures. Internation Journal of Space Structures, 8(1&2), 3–16. [22] Nisser, M. E. W. (2016). Feedback-Controlled Self-Folding of Autonomous Robot Collectives, 1254–1261. [23] Nojima, T, and Saito, K. (2006). Development of Newly Designed Ultra-Light Core Structures,. JSME Int. J., Ser.A, Vol.49, No.1, 38–42. [24] Radvan, B., & Gatward, A. (1972). Formation of wet-laid webs by a foaming process. Tappi, 55(5), 748. [25] Samimi, M. (2012). Sound Absorbing Porous Materials : Modelling of Normal Absorption Coefficient. Noise Control Lab Course Report. [26] Sina, A. (2015). Origami Engineering: Advanced Converting for Novel Paper Products. Masters’ Thesis. [27] Smith, M., and Punton, V. (1975). The role of the forming process in determining the structure and properties of paper. Pulp Pap Can, 76, 114–117. [28] Taherzadeh, S., Attenborough, K., and Bashir, I. (2012). Sound attenuation by crops and hedges. Soci´ [29] et´ e Fran¸ caise d’Acoustique [30] Tringham, R. (1974). New developments in radfoam process. Paper Technology and Industry 15(5), 288–294. [31] Wolkesson, M., (2013). Evaluation of impedance tube methods - A two microphone in-situ method for road surfaces and the three microphone transfer function method for porous materials Masters' Thesis, Chalmers University.
- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Theses and Dissertations /
- Artificial vegetation : a novel paper-based sound absorber
Open Collections
UBC Theses and Dissertations
Featured Collection
UBC Theses and Dissertations
Artificial vegetation : a novel paper-based sound absorber Salimian, Amirhossein 2018
pdf
Page Metadata
Item Metadata
Title | Artificial vegetation : a novel paper-based sound absorber |
Creator |
Salimian, Amirhossein |
Publisher | University of British Columbia |
Date Issued | 2018 |
Description | Demand for acoustic barriers in urban buildings grown in recent times fed by levels of increasing urban noise. Extraneous urban noise in buildings was commonly controlled using noise barriers like absorptive panels. The efficiency and sustainability of such noise absorbers is suspect. Our research in acoustics led to the development of a novel paper-based sound absorbers as a viable alternative for those currently available in the market. We named this product Artificial Vegetation (AV). We tested AV and found it an efficient and sustainable alternative that can be mass produced with relatively comparable low overheads. In this research paper we explored literature in acoustic barriers, both paper- and foam-based. We also identified gaps in the literature. The paper elaborates measurement of sound absorption coefficient using impedance-tube method, as a basic procedure for producing AV. We developed an analytical model that optimizes and improves sound absorptivity of AV. Data gathered from experiments were used to validate our model and test as a computational platform in more AV designs. |
Genre |
Thesis/Dissertation |
Type |
Text |
Language | eng |
Date Available | 2018-01-15 |
Provider | Vancouver : University of British Columbia Library |
Rights | Attribution-NonCommercial-NoDerivatives 4.0 International |
IsShownAt | 10.14288/1.0363038 |
URI | http://hdl.handle.net/2429/64368 |
Degree |
Master of Applied Science - MASc |
Program |
Mechanical Engineering |
Affiliation |
Applied Science, Faculty of Mechanical Engineering, Department of |
Degree Grantor | University of British Columbia |
GraduationDate | 2018-02 |
Campus |
UBCV |
Scholarly Level | Graduate |
Rights URI | http://creativecommons.org/licenses/by-nc-nd/4.0/ |
AggregatedSourceRepository | DSpace |
Download
- Media
- 24-ubc_2018_february_salimian_amirhossein.pdf [ 2.61MB ]
- Metadata
- JSON: 24-1.0363038.json
- JSON-LD: 24-1.0363038-ld.json
- RDF/XML (Pretty): 24-1.0363038-rdf.xml
- RDF/JSON: 24-1.0363038-rdf.json
- Turtle: 24-1.0363038-turtle.txt
- N-Triples: 24-1.0363038-rdf-ntriples.txt
- Original Record: 24-1.0363038-source.json
- Full Text
- 24-1.0363038-fulltext.txt
- Citation
- 24-1.0363038.ris
Full Text
Cite
Citation Scheme:
Usage Statistics
Share
Embed
Customize your widget with the following options, then copy and paste the code below into the HTML
of your page to embed this item in your website.
<div id="ubcOpenCollectionsWidgetDisplay">
<script id="ubcOpenCollectionsWidget"
src="{[{embed.src}]}"
data-item="{[{embed.item}]}"
data-collection="{[{embed.collection}]}"
data-metadata="{[{embed.showMetadata}]}"
data-width="{[{embed.width}]}"
async >
</script>
</div>
Our image viewer uses the IIIF 2.0 standard.
To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0363038/manifest