Article 1

Acoustic metasurface aided broadband noise reduction in auto- 2

mobile induced by tire-pavement interaction 3

Hyeonu Heo1, Mathew Sofield1, Jaehyung Ju2 and Arup Neogi1,* 4

1 Department of Physics, University of North Texas, P.O. Box 311427, Denton, TX 76203, USA and 5

hyeonu.heo@unt.edu, mathewsofield@my.unt.edu, arup.neogi@unt.edu 6

2 UM-SJTU Joint Institute, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China 7

jaehyung.ju@sjtu.edu.cn 8

* Corresponding authors; J. Ju (jaehyung.ju@sjtu.edu.cn) and A. Neogi (arup.neogi@unt.edu) 9

Abstract: The primary noise sources of the vehicle are engine, exhaust, aeroacoustic noise, and tire- 10
pavement interaction. Noise generated by the first three factors can be essentially reduced by re- 11
placing the combustion engine with an electric motor and optimizing aerodynamic design. Cur- 12
rently, 80-90% of the noise within automobiles occurs from the tire-pavement interaction over a 13
speed of 70-80 km/hr. Most noise suppression efforts aim to use sound absorbers and cavity reso- 14
nators to narrow the bandwidth of acoustic frequencies. A simple technique is demonstrated that 15
utilizes acoustic meta-layers with high reflective characteristics using relatively lightweight material 16
for noise reduction without any change in mechanical strength or weight of the tire. The proposed 17
Citation: Lin, D-F.; Wong, W-J..;
design can significantly reduce the noise arising from tire-pavement interaction over a broadband 18
Chen, C-W; Lin, K-L; Applying Mix-
of acoustic frequencies under 1000 Hz over a wide range of vehicle speeds using a negative effective 19
ture of Municipal Incinerator Bottom
dynamic mass density approach. The proposed approach can be extended to the generalized area 20
Ash and Sewage Sludge Ash for Ce-
ramic Tile Manufacturing . Materials
of acoustic and vibration isolation. 21
2021, 14, x.
https://doi.org/10.3390/xxxxx Keywords: acoustic metasurfaces; tire cavity; noise reduction 22

23
Academic Editor: Firstname Lastname

Received: date
1. Introduction 24
Accepted: date
Noise pollution by traffic is the most widespread environmental problems that cause 25
Published: date sleep disturbance, hearing damage, even cardiovascular disease.[1] Thus, the European 26
Parliament and the Member States have agreed to reduce road noise in 2011 and intro- 27
Publisher’s Note: MDPI stays neu- duced the noise regulations and reduce noise levels by just 3 dB (reducing sound pressure 28
tral with regard to jurisdictional energy to half) in 2013.[2] The primary noise sources of vehicles are engine, exhaust, aer- 29
claims in published maps and institu- oacoustic noise, and tire-pavement interactions.[3] Noise generated by the first three fac- 30
tional affiliations. tors can be essentially reduced by replacing the combustion engine with an electric motor 31
and optimizing aerodynamic design. However, reducing Tire-Pavement Interaction 32
Noise (TPIN) amounts to 80-90% of the total noise generated within a vehicle at a speed 33
of over 70-80 km/h [4,5] is hugely challenging. Any proposed technique should consider 34
Copyright: © 2021 by the authors. Sub-
the tire and wheel's structural and mechanical integrity, which can be problematic due to 35
mitted for possible open access publica-
the tire cavity environment affected by the changes in loading conditions, speed, and tem- 36
tion under the terms and conditions of
perature.[6] Tire and automobile manufacturers are developing soundproofing tech- 37
the Creative Commons Attribution (CC
BY) license (http://creativecom-
niques using resonators and sound-absorbers that meet the new industry standards [2] to 38

mons.org/licenses/by/4.0/).
provide a comfortable feeling for passengers. 39

Materials 2021, 14, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/materials

Materials 2021, 14, x FOR PEER REVIEW 2 of 11

Compressed air in a tire cavity generates resonant noise and vibration at a frequency 40
range below 1,000 Hz. The fundamental frequency is near 230 Hz. Conventional noise 41
reduction techniques of using thick metal plates are not applicable for isolating this reso- 42
nant noise and vibration of compressed air in a tire cavity due to design constraints. Tire 43
manufacturers add a polyurethane absorber glued around the tire's inner liner to reduce 44
noise.[7,8] A resonator attached to the rim has been used to reduce the tire cavity's reso- 45
nance sound.[9,10] The tire tread patterns’ pitch arrangement has also been optimized to 46
minimize the noise.[11] However, the noise reduction methods are mainly focused on the 47
resonator for the tire cavity resonance, i.e., a narrow band or sound absorption capability 48
of porous materials. Therefore, a different approach is necessary for better noise reduction 49
covering the sound generated at low frequency and the wideband noise caused by tire- 50
pavement interaction. 51

Acoustic metasurfaces (AMSes) or metalayers are artificially designed 2D materials 52

of subwavelength thickness that provide a non-trivial local phase shift and alter the direc- 53
tion of propagation of the incident wave. Some recent reports utilize AMSes for extraor- 54
dinary sound absorption [12] using Helmholtz resonators,[13] membranes,[14–16] and 3D 55
space coiling meta-structures.[17] The membranes have shown over 200 times noise re- 56
duction for a specific frequency range [14] and can be optimized for the desired applica- 57
tion by modifying the geometry. A lightweight yet soundproof acoustic metasurfaces 58
were used within an airplane framework to reduce noise.[16] The structure consisted of a 59
perforated stiff periodic pattern and thin, soft materials on a periodic structure. AMSes 60
can be designed with a negative effective dynamic mass density (𝜌𝑒𝑓𝑓 < 0) when the fre- 61
quency is below a thin plate's fundamental frequency. AMSes provide anti-resonance, out 62
−1⁄2
of phase to the incident wave, and exponential decaying wave (∆𝑑 ∝ |𝜌𝑒𝑓𝑓 |), resulting in 63
almost total reflection at the low broad frequency ranges.[14,15] The proposed AMS may 64
be used as an alternative way of noise reduction of tires. 65

The AMSes was designed to maximize the sound transmission loss (STL) over a 66
broad frequency range using lightweight materials without modifying the tire's mechan- 67
ical properties. The noise reduction in the tire was investigated using an AMSes based on 68
hexagonal unit cells attached to the rim's circumference to reflect sound waves arising 69
from the road-pavement interaction aided by the absorption in the radial direction. The 70
AMS was optimized for a particular design parameter with a negative effective dynamic 71
mass density for a practical car tire. Based on the parametric study of the unit cell of AM- 72
Ses, the noise reduction capability of AMSes is demonstrated through the static tests using 73
the tire cavity model and through a dynamic field test. 74

2. Design and fabrication of AMS 75

2.1. Design 76
The noise generated by tire-pavement interaction is the structure-borne noise that 77
occurs at the low-frequency ranges (below 500 Hz), while the air-borne noise occurs at the 78
high-frequency ranges (500-2000 Hz). According to Chang et al.[18], TPIN becomes the 79
primary source of noise occurring at low frequencies below 500 Hz, which is part of the 80
audible frequency range. Besides, the fundamental frequency, f, of tire cavity is a function 81
of the speed of sound of air, c, and wavelength, λ; 𝑓 = 𝑐 ⁄𝜆 = 2𝑐 ⁄𝜋(𝐷𝑜 + 𝐷𝑖 ), where 𝐷𝑜 82
is the outer diameter of the tire cavity toroid, and 𝐷𝑖 is the inner diameter of tire cavity 83
toroid.[19] For the general passenger vehicles, the cavity mode is near 230 Hz, which 84
needs to be reduced. 85

Highly reflective AMSes were explicitly designed for this frequency range, as shown 86
in Figure 1. AMSes are fabricated using silicone rubber and composed of a honeycomb- 87
shaped core panel attached to a tire’s rim. Hexagonal unit cell-shaped metasurfaces have 88
a natural mode of oscillation at high frequencies than squares and triangles with identical 89
Materials 2021, 14, x FOR PEER REVIEW 3 of 11

unit cells with the hydraulic diameters. Moreover, the shape and form of the periodic 90
metasurfaces with hexagonal unit cells are not deformed across a tire’s curved plane. It 91
also offers the best surface filling fraction, which is ideal for noise suppression. The unit 92
cell can be considered as a clamped thin plate because the core panel is relatively rigid. 93
Thus, when the noise occurs, the thin plate only oscillates and propagates acoustic pres- 94
sure while barely passing through the core panel. The unit cell's effective property with 95
hexagonal cross-section was evaluated to predict the AMS's acoustic characteristics. 96
Through Rayleigh’s method of a spring and a mass, the effective dynamic mass density, 97
ρeff, can be determined using the following deceptively simple equation: [15,16] 98

𝑓𝑟2
𝜌𝑒𝑓𝑓 = 𝜌𝑚 (1 − ) (1)
𝑓2
where fr is the lowest eigenfrequency of a circular shaped thin plate, f is the sound 99
frequency, and ρm is the thin plate's density. The given equation originates from Newton’s 100
second law, but the systems’ dynamic inertial mass becomes a function of frequency due 101
to the interactions between internal mass and spring.[15] Thus, the dynamic mass density 102
differs from the conventional gravitational mass density. For the hexagonal clamped thin 103
plate case, fr is calculated using the following equation:[20–22] 104

3
𝜋𝛼 𝐷 𝐸𝑚 ℎ𝑚
𝑓𝑟 = 2 √ , 𝐷= 2)
(2)
6𝑎𝑚 𝜌ℎ𝑚 12(1 − 𝜈𝑚

where hm is the thickness of the thin plate and 𝑎𝑚 is the side length of the thin hex- 105
agonal plate. Em and νm respectively represent Young's modulus and Poisson's ratio of 106
base material for the thin plate. The constant α is a nondimensional frequency parameter 107
calculated by the energy approach and convergence study.[22] For the first mode, α is 108
3.9068. If f < fr, the frequency-dependent effective dynamic mass density becomes nega- 109
tive. It implies that the force and the acceleration have the opposite direction. The clamped 110
thin plate’s local oscillation provides the anti-resonance, which is out-of-phase with the 111
incident wave. Therefore, the acoustic wave through the thin plate ceases to propagate 112
and becomes evanescent since the negative density implies an imaginary wave vector. 113

Figure 1. A concept to reduce tire noise with an acoustic metasurface consisting of a honeycomb
core panel and a soft, thin plate on the tire's rim. For the unit cell, am is the side length of the
hexagon, t is the wall thickness of the core panel, and hm and hc is the thickness of the thin plate
and the core panel's height, respectively.

114
Materials 2021, 14, x FOR PEER REVIEW 4 of 11

Figure 2. The clamped circular plate's dynamic mass density is calculated by the analytical
model (solid black line) and the numerical simulation (solid red line with marker). The fun-
damental resonance of the thin plate is at 2,056 Hz.

The effective dynamic mass density of an AMS including an elastomeric thin plate 115
having an eigenfrequency of 2,056 Hz was obtained from Equation (1) and the numerical 116
simulation using COMSOL Multiphysics when hm = 0.5 mm, Em = 7 MPa, ρm = 1,070 kg/m3, 117
νm = 0.49, and am = 3.65 mm (See Figure 2). The effective dynamic density can be numerically 118
obtained by dividing the out-of-plane surface averaged stress, ̅̅̅̅̅,
𝜎𝑦𝑦 by the product of the 119
surface averaged acceleration, ̅̅̅ 𝑎𝑦 , and the thin plate thickness, ℎ𝑚 , i.e., 𝜌𝑒𝑓𝑓 = 120
𝜎𝑦𝑦 ⁄(𝑎
̅̅̅̅̅ ̅̅̅ℎ
𝑦 𝑚 ) where the y-axis is the direction of wave propagation.[16] 121

2.2. Fabrication 122

To demonstrate the noise reduction of AMSes, we used the tire cavity model with 123
AMSes, conducting a field test with tires covered with AMSes to prove the noise reduction 124
capability. Based on the parametric study, we first fabricated AMSes made up of the com- 125
mercial a honeycomb core panel, aramid – 1/8” cell with 3 (lb/ft3) and 1/4” thickness, of 126
ACP Composites and silicone rubber, dragon skin, of Smooth-On, Inc. The silicone rubber 127
was poured on the clean surface of the wood plate, which was prepared with a 1mm deep 128
channel utilizing a CNC router, as evenly as possible, then smoothed over with an 11" 129
paint shield which rested on the edges of the channel, resulting in the thickness of the 130
silicone rubber about 1 mm. Next, the quarter-inch-thick honeycomb core panel is put on 131
the rubber layer and cured for 2 hours at room temperature, i.e., the silicone rubber cov- 132
ered one side of the panel. 133

The manufactured AMSes were attached to the inner layer of the tire cavity model 134
mimicking a real tire, 235/65R18, for the lab test referring to O’Boy’s design [6] and the 135
rim of each tire (Pirelli Tires), 185/65R15, of the car, Toyota Prius hybrid 2008, for the field 136
test. The tire cavity model consisted of medium density fibreboard (MDF), aluminum 137
metal sheets, and acrylic panel, as shown in Figure 3 (left). The outer and the inner metal 138
sheets represent a tire rubber and a rim of a wheel, respectively. We added a rubber seal 139
to the edge of the aluminum sheets to isolate the cavity. The Rode NT-USB mini micro- 140
phone, which has a sampling rate of 48 kHz and a frequency range of 20 Hz – 20 kHz, was 141
mounted in the center of the inner cavity and connected to the computer with the USB 142
cable. A hole was inserted at the bottom for a speaker emitting white noise generated by 143
the Minirator MR2 audio generator of NTi Audio, which has a resolution of 0.1 Hz. 144
Materials 2021, 14, x FOR PEER REVIEW 5 of 11

Figure 1. (Left) Tire cavity model representing a tire, 235/65R18, where Do = 30” and Di = 18”, re- 145
spectively. The thickness values of Al sheets, MDF, and acrylic panel, are 0.060”, 0.750”, and 146
0.437”, respectively. At the bottom, there is a hole to generate the white noise to represent TPIN by 147
a speaker. The Rode NT-USB mini microphone was mounted in the center of the inner cavity and 148
connected to the computer with the USB cable. (Right) The rim of the tire (Pirelli Tire), 185/65R15, 149
of Toyota Prius hybrid 2008. The foam and the AMS were bonded on the circumference of the rim, 150
3.5” width. The density of foam and AMS are approximately 161 kg/m3 and 233 kg/m3, respec- 151
tively. 152

For the field test, AMSes were bonded on the rim, inside each tire of the vehicle. (See 153
Figure 3) As a comparison, a commercial foam, neoprene sponge foam rubber of Lazy dog 154
warehouse, used for soundproofing with the same thickness of the honeycomb core panel 155
was used. 156

3. Results 157
A parametric study of the unit cell was conducted to evaluate the effect of design 158
parameters, such as side length and thickness of the unit cell’s thin plate, density, and 159
sound transmission loss (STL) (See Figure 4). Then, AMSes were fabricated based on the 160
parametric study to maximize STL yet remain lightweight and attached to the tire cavity 161
model and a real tire for the laboratory and field tests. The AMSes’ performance was com- 162
pared to a commercial foam having the same thickness as that of AMSes. 163

3.1. Design map of the unit cell of AMS 164

The unit cell’s parametric study was carried out using the numerical simulation 165
(COMSOL Multiphysics, acoustic module) to predict design parameters' effect on acoustic 166
properties, such as dynamic mass density and STL. As the design parameters, thickness, 167
hm, and side length, am, of the thin plate were examined. The clamped hexagonal thin plate 168
was considered the unit cell of AMS, and the linear elastic model for the silicone rubber 169
was occupied. Figure 4(a) illustrates the unit cell’s geometry, where the thin plate is placed 170
in the middle of the pipe. On top of the pipe, the acoustic pressure propagates through 171
the structure. Then, the transmitted sound pressure was measured at the bottom. 172
Materials 2021, 14, x FOR PEER REVIEW 6 of 11

Figure 2. The result of the parametric study on design parameters, such as thickness, hm, and side
length, am, of the thin plate of the unit cell of AMSes; (a) The schematic image of the unit cell used
Figure 1. and (2); (c) STL calculated by the numerical simulation.

173

The noise's peak is near 230 Hz, which is the fundamental mode of the tire cavity. 174
The investigation's frequency range was from 100 Hz to 400 Hz to consider the effects on 175
the fundamental mode. As mentioned above, when the frequency is less than the funda- 176
mental mode, the effective density of AMSes becomes negative. Under these conditions, 177
the plate's local oscillation reflects the incident wave results in a substantial noise reduc- 178
tion. Therefore, when the natural frequency is shifted to a higher frequency by modifying 179
the design parameters, the noise reduction effect is enhanced. (See Figure 4(b) and (c)). 180
The AMS shows a significant noise reduction by 23-62 dB at the low-frequency ranges 181
even though the silicone rubber's material properties were simplified using a linear elastic 182
model in this study. A smaller unit cell and a thicker plate have even higher sound losses 183
because the first mode is proportional to the thickness and inversely proportional to the 184
plate's area. (see Equation (2)) 185

3.2. The sound pressure level in the tire cavity model (static test) 186
The AMS feasibility was demonstrated by constructing a tire-cavity that represented 187
an actual tire (235/65R18), as shown in Figure 3. The model consisted of MDF, aluminum 188
metal sheets, and acrylic panels. The outer and the inner metal sheets represent a tire rub- 189
ber and a rim of a tire, respectively. The Rode NT-USB mini microphone is mounted in 190
the inner cavity. A hole at the bottom of the speaker facilitated the generation of white 191
noise. Figure 5 shows the effect of noise reduction due to the AMSes, where Figure 5(a) 192
shows the acoustic spectrum in log scale, and Figure 5(c-d) depict the sound transmission 193
coefficients (STCs) normalized to the maximum sound transmission of the white noise of 194
the cavity mode. Figure 5(b) illustrates the tire cavity with foam and AMS. 195

In Figure 5, the solid black line represents background noise, and it serves as a refer- 196
ence. The speaker generating white noise (W.N.) was turned on to measure the empty 197
inner cavity’s spectra and is depicted by the yellow lines. Due to the circular symmetry of 198
the tire cavity, there are radial and azimuthal eigenmodes. The dominant contribution to 199
automobile noise originates from the fundamental radial mode with a frequency near 200
Materials 2021, 14, x FOR PEER REVIEW 7 of 11

184.7 Hz. There is a peak near 185 Hz in Figure 5. The suppression of this peak noise in 201
the frequency spectrum within the inner cavity strongly reduces the noise transmitted to 202
the car’s cabin. The efficiency of our metasurface based technology is compared with ex- 203
isting sound absorption-based noise reduction technology. The noise in a cavity filled with 204
1/4” thick form (shown by blue dashed lines) is compared with the cavity wrapped using 205
the acoustic metasurface (shown by the red dotted lines). It is evident from the acoustic 206
spectra shown in Figure 5(c-d) that the noise within the inner cavity is reduced in both 207
cases. For the AMS, the reduction is more substantial, especially near the cavity mode, 208
and it remains effective over a broader range of frequencies. The bandwidth of the noise 209
suppression frequency is narrower for the foam as it still transmits sound energy while 210
absorbing due to thermal dissipation. The wavelength at the low frequency is much larger 211
than the porous size of the foam. However, the AMS reflects due to anti-local resonance 212
below the natural frequency of the thin plate. There are several minor peaks at higher 213
frequencies at 350 Hz, and the unit cell design can suppress that. Because the dynamic 214
mass density is a function of design parameters, the modes can be varied or specified to 215
reflect, i.e., noise reduction. 216

Figure 5. Sound pressure level (a) and normalized sound transmission coefficients (c-d) in the tire 217
cavity model measured in the inner cavity. The background noise, black solid dashed lines, is the 218
reference. The yellow lines represent the white noise (W.N.) when the speaker is turned on. The 219
blue and red lines are the cases of attached foam and AMS, respectively. The pictures (b) display 220
the cavity models with foam or AMS utilized in the experiment. 221

3.3. The sound pressure level in the cabin (dynamic test) 222
After the lab-scale test, the field test was performed with the Toyota Prius hybrid 223
2008. This car was chosen to minimize the noise induced by engine and exhaust and focus 224
on tire-pavement interaction. Prius hybrid has E.V. mode up to 60 km/h (~ 37 mph). The 225
tire size is 185/65R15 (Pirelli Tire), and the air pressure in the tire is 44 psi. To investigate 226
the effectiveness of AMS, we consider three cases – i) without any attachment, ii) with 227
Materials 2021, 14, x FOR PEER REVIEW 8 of 11

foam, and iii) with AMS. The test was performed on the local driveway, about 7.2 miles, 228
with relatively new asphalt replaced in 2020. The noise inside the cabin was measured 229
twice from 20 to 60 mph with a ten mph interval. 230

Figure 6. The result of the sound reduction performance of AMS (red) and foam (blue) comparing
to the original tire (yellow) without any attachment through the field test. (a) The sound pressure
spectra from 100 Hz to 1,000 Hz. The solid black line is the case of the stopped state. At 60 mph,
yellow (solid), blue (dashed), and red (dotted) lines represent the cases of without any attach-
ment, foam, and AMS on the rim, respectively. (b)-(f) The normalized sound transmission coeffi-
cients (N-STC) from 200 Hz to 300 Hz depending on the vehicle speed. STC is normalized by the
peak of the cavity mode at 60 mph.

The temporal sound pressure was measured for 19 seconds with a Rode NT-USB 231
mini microphone, and the recorded data were processed with the fast Fourier transform 232
(FFT) to obtain the frequency spectrum of the sound pressure level (SPL) from 50 Hz to 233
Materials 2021, 14, x FOR PEER REVIEW 9 of 11

1,000 Hz as shown in Figure 6(a). The cavity mode occurs near 230 Hz, as expected. Alt- 234
hough both foam and AMS show noise reduction effect, the SPL of AMS is 2-3 dB more 235
than foam and is significantly higher. The frequency range under consideration ranged 236
from 200 to 300 Hz as the tire air cavity mode appears near 230 Hz. The frequency spec- 237
trum of the sound transmission coefficient (STC) at various vehicle speeds is shown in 238
Figures 6(b-f). The maximum peak normalizes STC for the cavity mode at 60 mph. For 239
low speed at E.V. mode, 20-40 mph, the cavity mode’s peak values are similar, but the 240
noise at other frequencies induced by engine noise increases when the vehicle speed is 241
increased gradually. 242

The average sound pressure level depending on the vehicle speed, as shown in Fig- 243
ure 7. The noise level increases with the speed while the slope of noise changes when the 244
mechanical engine kicks on after 40 mph due to the electric to gasoline power modes. It is 245
the average value so that the foam case shows more noise than the reference case. Alt- 246
hough the foam reduces the cavity mode’s noise, more peaks occur near the cavity mode, 247
as seen in Figure 6(b). AMS is a little heavier than foam, 1.45 times. Nevertheless, through 248
all speed ranges, it clearly shows that AMS provides a better noise reduction effect than 249
foam, 2-5 dB near the cavity mode, 200 – 300 Hz. 250

251

Figure 7. The average sound pressure level from 200 Hz to 300 Hz versus the vehicle's speed with 252
error bars. Yellow (solid), blue (dashed), and red (dotted) lines are references, i.e., without any 253
attachment, foam, and AMS on the rim, respectively. 254

4. Further Discussion 255

The foam is generally used to reduce the noise not just for tires but also for stationary 256
structures, e.g., in civil engineering, as an absorber. For the tire application, Mohamed et 257
al. validated the effectiveness of foam[23][23][23] [23]. However, the absorption coefficient 258
is less than 7%, even lower at the low-frequency range. On the other hand, as a reflector, 259
AMS shows better noise reduction capability than foam at the low-frequency range, the 260
region of the cavity mode, and the broadband about 2-3 times. However, the real applica- 261
tion’s effectiveness is much lower than the preliminary results from the unit cell. It is be- 262
cause AMS can only cover partially, 3.5” of the rim having non-flat surface so that noise- 263
induced by TPIN still propagates through the uncovered surface. Besides, the unit cell has 264
a periodic boundary condition, which is hard to implement in a real application. With the 265
modified and optimized design with the rim, the adequate bonding mechanism between 266
the rim and the AMS for durability, and the suitable material selection, the efficiency of 267
noise reduction of AMS can be improved substantially. 268

5. Conclusions 269
Materials 2021, 14, x FOR PEER REVIEW 10 of 11

We propose a noise reduction method for tires using acoustic metasurfaces or meta- 270
layers with high reflective and absorbing characteristics and relatively lightweight for 271
practical applications. The effectiveness of AMS was demonstrated numerically and ex- 272
perimentally. Through the parametric study, the design guideline was provided. A pro- 273
totype of AMSes was fabricated in house and utilized for the experiment both on the la- 274
boratory scale and for actual field tests in a hybrid car. The noise reduction effect of AMSes 275
can be manipulated by tuning the thin plate’s fundamental resonance depending on de- 276
sign parameters. The developed method can reduce the noise substantially near the tire 277
cavity mode at around 230 Hz and extends over a broader range of frequencies under 278
1,000 Hz. 279

Furthermore, the structure of this material used for this approach is lightweight and 280
does not affect the tire’s performance. It can be easily combined with the existing technol- 281
ogies to maximize the sound transmission loss. The design can be applied to other fields 282
for sound and vibration isolation by modifying the design due to its robustness and light- 283
weight to reduce noise issues. 284

Acknowledgment: The authors acknowledge the support from National Science Foundation sup- 285
ported grant entitled "GOALI: EFRI NewLaw: Non-reciprocal effects and Anderson localization of 286
acoustic and elastic waves in periodic structures with broken P-symmetry of the unit cell" Award 287
#1741677. J.J. appreciates the support from Shanghai NSF (Award # 17ZR1414700), the Research 288
Incentive Program of Recruited Non-Chinese Foreign Faculty by Shanghai Jiao Tong University. 289
The authors acknowledge Tracy Lynch, Mark Lanier, Richard Bates, and Lowyn Hendrickson for 290
helps and supports in this study. 291

References 292
[1] WHO, Burden of disease from Burden of disease from, (2011) 126. 293
[2] The European Parliment and the Council of the European Union, REGULATION (EU) No 540/2014 OF THE EUROPEAN 294
PARLIAMENT AND OF THE COUNCIL of 16 April 2014 - On the sound level of motor vehicles and of replacement silencing 295
systems, and amending Directive 2007/46/EC and repealing Directive 70/157/EEC, Off. J. Eur. Union. 158 (2014) 131–195. 296
[3] M.E. Braun, S.J. Walsh, J.L. Horner, R. Chuter, Noise source characteristics in the ISO 362 vehicle pass-by noise test: Literature 297
review, Appl. Acoust. (2013). https://doi.org/10.1016/j.apacoust.2013.04.005. 298
[4] T. Li, J. Feng, R. Burdisso, C. Sandu, The effects of tread pattern on tire pavement interaction noise, in: Proc. INTER-NOISE 2016 299
- 45th Int. Congr. Expo. Noise Control Eng. Towar. a Quieter Futur., 2016. 300
[5] L. Domenichini, A. Fracassa, F. La Torre, G. Loprencipe, A. Ranzo, A. Scalamandrè, Relationship between road surface 301
characteristics and noise emission, First Int. Colloq. Veh. Tyre Road Interact. Rome, Italy, Pap. 99.03. (1999). 302
[6] D.J. O’Boy, S.J. Walsh, Automotive tyre cavity noise modelling and reduction, in: Proc. INTER-NOISE 2016 - 45th Int. Congr. 303
Expo. Noise Control Eng. Towar. a Quieter Futur., 2016. 304
[7] T. Tenzuka, Tire tread for reducing noise, US 2018/0001715 A1, 2018. 305
[8] S.-J. Choi, H.-J. Kim, Polyurethane foam and pneumatic tire, 2016. https://patents.google.com/patent/US9315611B2/en. 306
[9] Y. Kamiyama, K. Ishii, Vehicle wheel having sound-damping structures, US 10131190 B2, 2018. 307
[10] K. Sakakibara, Pneumatic tire, US 10000096 B2, 2018. 308
[11] F.J. Cusimano, Low noise pneumatic tire tread with voids balanced over each half tread region, US 005209793 A, 1993. 309
[12] B. Assouar, B. Liang, Y. Wu, Y. Li, J.C. Cheng, Y. Jing, Acoustic metasurfaces, Nat. Rev. Mater. 3 (2018) 460–472. 310
https://doi.org/10.1038/s41578-018-0061-4. 311
[13] Y. Li, X. Jiang, B. Liang, J.C. Cheng, L. Zhang, Metascreen-Based Acoustic Passive Phased Array, Phys. Rev. Appl. 4 (2015). 312
https://doi.org/10.1103/PhysRevApplied.4.024003. 313
[14] Z. Yang, J. Mei, M. Yang, N.H. Chan, P. Sheng, Membrane-type acoustic metamaterial with negative dynamic mass, Phys. Rev. 314
Lett. 101 (2008). https://doi.org/10.1103/PhysRevLett.101.204301. 315
[15] G. Ma, M. Yang, S. Xiao, Z. Yang, P. Sheng, Acoustic metasurface with hybrid resonances, Nat. Mater. 13 (2014) 873–878. 316
https://doi.org/10.1038/nmat3994. 317
[16] N. Sui, X. Yan, T.Y. Huang, J. Xu, F.G. Yuan, Y. Jing, A lightweight yet sound-proof honeycomb acoustic metamaterial, Appl. 318
Phys. Lett. 106 (2015). https://doi.org/10.1063/1.4919235. 319
[17] Y. Xie, W. Wang, H. Chen, A. Konneker, B.I. Popa, S.A. Cummer, Wavefront modulation and subwavelength diffractive 320
acoustics with an acoustic metasurface, Nat. Commun. 5 (2014). https://doi.org/10.1038/ncomms6553. 321
[18] C. Jin, W.Y. Wang, X.X. Jin, Study on tire noise transfer path identification, in: Int. Conf. Signal Process. Proceedings, ICSP, 2010. 322
https://doi.org/10.1109/ICOSP.2010.5656122. 323
Materials 2021, 14, x FOR PEER REVIEW 11 of 11

[19] X. Wang, Z. Mohamed, H. Ren, X. Liang, H. Shu, A study of tyre, cavity and rim coupling resonance induced noise, Int. J. Veh. 324
Noise Vib. 10 (2014). https://doi.org/10.1504/IJVNV.2014.059628. 325
[20] S. Timoshenko, Vibration problems in engineering, 2nd ed., D. Van Nostrand Company, Inc., New York, 1937. 326
[21] Rayleigh, Theory of Sound, Macmillan & Co., , London, UK, 1877. 327
[22] K.M. Liew, K.Y. Lam, A set of orthogonal plate functions for flexural vibration of regular polygonal plates, J. Vib. Acoust. Trans. 328
ASME. 113 (1991). https://doi.org/10.1115/1.2930167. 329
[23] Z. Mohamed, X. Wang, A study of tyre cavity resonance and noise reduction using inner trim, Mech. Syst. Signal Process. 50– 330
51 (2015) 498–509. https://doi.org/10.1016/j.ymssp.2014.05.044. 331
332

333

334