ArticlePDF Available

3D Printing of Asphalt and its effect on Mechanical Properties

Authors:

Abstract and Figures

The paper describes work to design, build and test an asphalt 3D printer. The main difficulty encountered is that asphalt behaves as a non-Newtonian liquid when moving through the extruder. Thus, the rheology and pressure in relation to set temperature and other operational parameters showed highly non-linear behaviour and made control of the extrusion process difficult. This difficulty was overcome through an innovative extruder design enabling 3D printing of asphalt at a variety of temperatures and process conditions. We demonstrate the ability to extrude asphalt into complex geometries, and to repair cracks. The mechanical properties of 3D printed asphalt are compared with cast asphalt over a range of process conditions. The 3D printed asphalt has different properties from cast, being significantly more ductile under a defined range of process conditions. In particular, the enhanced mechanical properties are a function of process temperature and we believe this is due to microstructural changes in the asphalt resulting in crack-bridging fibres that increase toughness. The advantages and opportunities of using 3D printed asphalt to repair cracks and potholes in roads are discussed.
Content may be subject to copyright.
Materials and Design 160 (2018) 468–474
Contents lists available at ScienceDirect
Materials and Design
journal homepage: www.elsevier.com/locate/matdes
3D printing of asphalt and its effect on mechanical properties
Richard J. Jacksona,b, Adam Wojcika, Mark Miodownika,b,*
aMechanical Engineering Dept., UCL, London, UK
bInstitute of Making, UCL, London, UK
HIGHLIGHTS
We have created a technique to 3D
print asphalt — we believe it is the
first of its kind.
3D printed asphalt is more ductile
than cast asphalt.
The changes in mechanical proper-
ties are related to the microstructural
changes in asphalt that occur during
3D printing.
The mechanical properties of 3D
printed asphalt depend on process
conditions, this can be advantageous
allowing toughness to be tailored to
the repair.
The technique has the potential to
be used on autonomous vehicles or
drones to autonomously repair roads
and complex infrastructure.
GRAPHICAL ABSTRACT
ARTICLE INFO
Article history:
Received 3 May 2018
Received in revised form 13 September 2018
Accepted 14 September 2018
Available online 20 September 2018
Keywords:
Asphalt
Bitumen
3D printing
Additive manufacture
Repair
Toughness
ABSTRACT
The paper describes work to design, build and test an asphalt 3D printer. The main difficulty encountered
is that asphalt behaves as a non-Newtonian liquid when moving through the extruder. Thus, the rheology
and pressure in relation to set temperature and other operational parameters showed highly non-linear
behaviour and made control of the extrusion process difficult. This difficulty was overcome through an inno-
vative extruder design enabling 3D printing of asphalt at a variety of temperatures and process conditions.
We demonstrate the ability to extrude asphalt into complex geometries, and to repair cracks. The mechani-
cal properties of 3D printed asphalt are compared with cast asphalt over a range of process conditions. The
3D printed asphalt has different properties from cast, being significantly more ductile under a defined range
of process conditions. In particular, the enhanced mechanical properties are a function of process tempera-
ture and we believe this is due to microstructural changes in the asphalt resulting in crack-bridging fibres
that increase toughness. The advantages and opportunities of using 3D printed asphalt to repair cracks and
potholes in roads are discussed.
© 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
1. Introduction
Asphalt (bitumen) composites are the most common material
used to surface roads, with 95% of UK roads paved with asphalt
*Corresponding author.
E-mail address: M.miodownik@ucl.ac.uk.
mixtures [1]. Its success is due to a combination of factors that
have been widely studied: it creates a safe and robust road surface
for driving when combined with stone aggregates and appropri-
ate polymer binders [2,3]; road surfacing can be carried out rapidly
and without complex machinery; it has good acoustic properties
and so muffles the sounds of traffic [4]; it is robust, repairable and
indeed self-repairs [5-7]. However asphalt composites do degrade
https://doi.org/10.1016/j.matdes.2018.09.030
0264-1275/© 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
R. Jackson et al. / Materials and Design 160 (2018) 468–474 469
over time due to the effects of road usage, oxidation, loss of volatiles,
moisture damage, and various other factors. This degradation leads
to increased stiffness of the road surface, cracks forming, stripping,
ravelling, loss of aggregate, and development of pot-holes [8,9].
Despite a stipulated minimum lifetime of 40 years, the re-
surfacing of roads is estimated to cost £2 billion per year in the UK
alone [10]. Increasing the life of roads has the potential to reduce
environmental and financial costs associated with road closures and
the congestion they cause. One approach taken to increase the life
of asphalt roads has been to enhance their self-healing properties.
For instance, mixing ferrous fibres into the asphalt composite allows
the material to be heated by induction by the application of an alter-
nating magnetic field [11]. The heating of the fibres locally heats the
asphalt and this has been shown in the laboratory to heal micro-
cracks and restore the strength of the road, as well as de-icing it
[12]. There is currently an on-going trial being carried out in the
Netherlands in which a road section has been surfaced with such a
material and receives the heating through regular applications of a
magnetic field via a specially adapted vehicle [13]. Other approaches
to preventing road surface degradation are the inclusion of micro-
capsules of sunflower oil into the asphalt composite which burst
open in the presence of a crack and increase the fluidity of the asphalt
allowing it to reflow and heal the crack [14].
Environmental considerations have led to much interest in the
use of recycled materials such as rubber and plastic in asphalt
composites [15,16]. These materials are often inexpensive, being cur-
rently considered as waste, but the asphalt mixing phase is energy
intensive [17,18] and the cost of transport is also a factor [19]. So
when looking to increase the environmental sustainability of infras-
tructure maintenance, materials and repair processes should ideally
be optimised together. In the future, many of these demands could
be met by autonomous vehicles which repair locally on demand [20].
In the case of asphalt roads, this optimised preventative approach
to road maintenance would need to focus on the early stages of
road degradation when small cracks form on the road surface. These
cracks allow water ingress and grow rapidly during freeze-thaw
cycles through the de-bonding of aggregates to form potholes. Once
formed it is very hard to stop the growth of these potholes which
cause significant vehicle damage and so lead to shortening of the
usable lifetime of the road.
This paper describes work to produce a 3D printing technology
that could be attached to an autonomous vehicle or drone, and used
to repair small cracks before they turn into potholes. 3D printing is a
method by which objects can be fabricated layer by layer from a CAD
model of the object. By scanning a road surface the negative shape of
the crack can be obtained and processed into a 3D model [21]. This
information can then be processed and passed to a 3D printer, which
can then print exactly the correct amount of material to conform to
the crack shape and volume, thus repairing the crack. 3D printing
technology has previously been used to repair spall damage in con-
crete road surfaces [22]. We show for the first time that it is also
possible to 3D print asphalt into a crack to restore the road surface.
The focus of this paper is a description of the design and operation of
our asphalt 3D printer, a demonstration of its ability to repair cracks,
and an investigation of the mechanical properties of the 3D printed
asphalt.
2. Materials & methods
The 3D printer is designed as a three axis system in which the
extrusion nozzle is moved by individual stepper motors to print
onto a flat bed, see Fig. 1 (a). The printer nozzle consists of an auger
screw, a stepper motor to drive the screw, and a pellet hopper to
take asphalt in the form of pellets. The pellets are softened as they
travel through the auger screw by an increase in temperature due to
the action of heating resistors, this results in a fluid flow of asphalt
out of the nozzle, as illustrated in Fig. 1 (b). The stepper motors,
temperature, temperature gradient, and auger screw rotation rate
are controlled by simple electronics interfaced to a PC, shown
schematically in Fig. 1 (c).
The 3D printer was constructed using an existing frame and
control system from a RepRap Mendel 90 3D printer, see Fig. 2
(a). The extrusion nozzle assembly was 3D printed using a Form 2
Stereolithography 3D printer using the Formlabs proprietary high
temperature acrylic based resin which allowed precise control of
the complex geometry of the extruder housing and auger screw
(attached to the stepper motor shaft via 5 mm grub screw), see
Fig. 2 (b). The extruder assembly had an inner heating sleeve
made from a 1 mm thick, 20 mm outer diameter aluminium pipe,
with the 15 W 20 YCaddock MP915 series TO-126 power resistors
attached radially, spaced 120apart, and connected in parallel. These
were attached to the outer part of the pipe with MG Chemicals
two-part silver epoxy/cold solder. The temperature was measured
using a 100 k EPCOS B57550G1104F thermistor. The thermistor was
attached to the aluminium pipe by drilling a 0.5 mm deep indenta-
tion into the pipe with a 2 mm drill bit approximately 5 mm away
from one of the power resistor contact points, and similarly fixed
with silver epoxy. Heat conduction through the aluminium pipe to
the extrusion tip proved to be insufficient to control the asphalt tem-
perature quickly and accurately enough and so a metal nozzle cap
was employed to improve heat conduction to the asphalt. This was
initially made from CNC machined aluminium but the same results
were obtained by concentrically stacking M8, M4 and M2 stainless
steel washers at the nozzle tip and fixing to each other and the pipe
with silver epoxy, see Fig. 2 (b). Conventional fused deposition mod-
eling (FDM) via another, unmodified Mendel 90 RepRap printer was
used to print the stepper motor housing and PCB and serial port clip
in ABS plastic. Fig. 2 (c) shows a photo of the extrusion nozzle.
By selecting the hardest grade of bitumen, 10/20, we hoped to
match the in-use material properties as much as possible [1]. The
asphalt pellets were formed from larger pieces of asphalt (Bitumen,
CAS 64742-93-4, 10/20 grade, material and data sheet supplied by
IKO PLC, UK [23]) by low temperature casting (below 150 C) into
a machined mould to obtain millimeter scale pellets. Asphalt is a
substance made principally of long-chain hydrocarbon molecules in
a colloidal structure of aphaltenes and maltenes with complex rhe-
ological properties [24]. Above a threshold temperature, typically
between the range of 30–70 C, it behaves as a Newtonian fluid.
Below this threshold, it undergoes shear-thinning. The rheology in
this regime has been studied in detail and shows an Arrhenius-like
behaviour [25]. Much work has been done to study the effect on rhe-
ology of polymeric binders that are added to asphalt, these behave as
viscoelastic components [26]. In the face of this complexity, we chose
not to try to model the regime of rheology under different shear
stresses as the asphalt pellets travelled through our extruder, but
instead we aimed to find the optimum processing variables by carry-
ing out a systematic empirical investigation of the extrusion process.
We identified the important design parameters of the nozzle
through a number of design iterations, see Fig. 3 (a). Using infrared
cameras to identify thermal gradients allowed us to understand the
heat flow within the extruder and so iterate the design towards
optimum parameters. The power limit (P= 45 W) and torque limit
(T= 44N cm) of the three power resistors and stepper motor respec-
tively, were self-imposed design constraints. A number of auger
screw designs were fabricated and systematically tested within the
printer framework. Chamber height, in combination with screw
height and metal insert height was found to be important as an
interim asphalt softening area was needed in between the hop-
per and screw, otherwise the screw would stall, or break, trying
to extrude asphalt which was too viscous. For this reason, the
final design has zero unheated chamber height. We found that for
temperatures above 150 C, the asphalt was so fluid it flowed out of
470 R. Jackson et al. / Materials and Design 160 (2018) 468–474
Fig. 1. System design: (a) 3D printer, (b) the extruder design, (c) the control electronics.
Fig. 2. Experimental: (a) photo of whole system, (b) photo of extruder components, (c) photo of complete extruder.
the extruder aperture under no screw rotation so was not useful for
3D printing.
The temperature range of 100–140 C was explored to print
a range of test objects. Screw length and pitch affect the extru-
sion rate for a given rotations per minute (RPM) but this was not
investigated. Various extrusion diameters from 0.5 to 5 mm were
used with 2.5 mm found to give the best balance between relia-
bility and dimensional accuracy. Print failures were identified by
incomplete or poor first layer adhesion, intermittent layer bonding,
incomplete printing, and partially hollow objects. Our final design of
the extrusion nozzle is shown in Fig. 3 (b) with the following param-
eters: the chamber diameter (dc= 17.5 mm), metal insert height
(hi= 47.5 mm), the tip height (ht=2.5 mm), the unheated chamber
height (hu= 0mm), the extrusion diameter (de= 2.5mm), the auger
screw length (ls= 21 mm), pitch (ps= 7 mm) and the total thermal
input area (At= 270 mm2).
Fig. 3. Design parameters: (a) CAD of extruder, (b) design parameters, (c) process parameters.
R. Jackson et al. / Materials and Design 160 (2018) 468–474 471
Using this design, we found that the optimum operational param-
eters of the extrusion nozzle were the print speed (vp= 1 mm/s), the
z offset (Zo= 3 mm), the layer height (lh= 3mm), the auger screw
rotation speed (Ws= 5–10 RPM), the set temperature (tset = 125–
135 C), see Fig. 3 (c).
We used open access Pronterface control software (version 3) [27]
and Slic3r slicing software [28] to generate g-code and print via stan-
dard STL files designed using the 3D CAD software Sketchup [29]. The
Slic3r software is intended for use with fused deposition modelling
(FDM) printers which heat a filament of defined diameter through
the heated extruder hole. As such, the filament diameter setting was
used as the extruder chamber diameter to represent the width of the
asphalt, however this is not completely accurate as the auger screw
takes up a significant proportion of the chamber volume. Further-
more, the Slic3r “extrusion multiplier” setting is intended to increase
or decrease the filament feed speed to the extruder, whereas in our
case the filament feed motor was used as the auger screw motor
(100×extruder multiplier was found to be equivalent to 4.4 RPM).
An optimisation process was used to set these heat and print speed
settings.
A number of different shapes were printed including standard
mechanical test bars with dimensions 80 ×10 ×6 mm. These were
mechanically tested at room temperature measured as 22 C using
a three point bend test rig on a Hounsfield HK5 universal testing
machine one week after printing or moulding. The tests were carried
out with a 2500 N load cell at constant strain rate of 1 mm/min with
support bars spaced 25 mm apart. For each test condition, six sam-
ples were tested and the data aggregated. A number of cast asphalt
samples were also produced to compare with the 3D printed sam-
ples. These were cast into PDMS moulds with the same dimensions
as the mechanical test bars at 150 C. The time between casting, 3D
printing and mechanical testing was 48 h.
3. Results
Firstly, basic 3D printing experiments were carried out in which
three single lines of asphalt of length 100 mm were printed with
1 mm layer height using an aperture width of 2 mm. These exper-
iments were performed using a range of print-head temperatures
between 100 and 150 C. Temperatures between 125 C and 135 C
were found to be optimal to create a continuous extrusion of asphalt
with a consistent line width. The rotation speed of the auger screw,
Ws, was found to be an important variable that determined the line
thickness. Fig. 4 shows the effect of Wson the line thickness of the
3D printed asphalt at 125 C. Line printing was less successful and
reliable at lower rotation speeds (approximately 2–7 RPM) although
this was not an issue when printing objects, as subsequent layer
deposition and adhesion after the first layer was helped by (and was
reliant on) the initial layer. Nonetheless above 5 RPM we were able
to reliably print single lines.
Fig. 5 (a) shows the ability of the 3D printer to fabricate an object
from a digital CAD file through the layering of lines of asphalt, in this
case the shape is a pyramid, but this technique is general, with the
same object geometry limitations as standard polymer prints. The
print layers are clearly visible and approximately 1 mm in height.
Some flow and bleeding of the asphalt is visible which affects the
feature resolution of the object. Fig. 5 (b) shows the ability of the
system to 3D print a mechanical test sample of known proportions
for mechanical testing. A number of such samples were printed in
which a number of geometry and process variables were varied.
The temperature range of 125–135C at 1 mm/s print speed and
4.4 RPM (100×extrusion multiplier) gave the most accurate and
reliable prints in terms of desired dimensions and the success of
printing a fully formed object. These test samples were then mechan-
ically tested using a 3 point bend test together with cast asphalt test
Fig. 4. Results: Graph of effect of screw rotation speed on printed line thickness.
sample of identical dimensions. Fig. 5 (c) shows the ability of the
system to take as an input the inverted shape of a crack in an asphalt
sample and 3D print asphalt into the crack to fill it.
Fig. 6 shows the stress/strain curves obtained from cast asphalt
and compares them to those from 3D printed asphalt at three dif-
ferent printing temperatures. It can be seen that the mechanical
properties are markedly different for the two fabrication methods.
For the cast samples there was, as expected, anisotropy observed
between those samples tested with their bottom or top surfaces
under compression, see Fig. 6 (a). The former showed a classic brit-
tle fracture while the latter showed some ductile behaviour before
fracture, see Fig. 6 (d). This anisotropy is likely due to the differences
in their surface roughnesses, porosity, and volatile content between
the top and bottom of the sample. There were no differences seen in
testing the 3D printed samples from top or bottom. Both sets of sam-
ples showed similar fracture strengths, see Fig. 6 (c). The 3D printed
specimens showed up to nine times the ductility of cast samples but
had similar fracture strengths of around 2 MPa, see Fig. 6 (b). The
toughness of the moulded and 130 C printed samples were found
to be 10.2 ±7.1 and 24.1 ±7.2 J/cm2respectively.
The effect of extrusion temperature on mechanical properties is
small although there is some evidence that 130 C is optimal, see
Fig. 6 (a). Observation of the fracture surfaces provides some expla-
nation for the difference in ductility between the cast asphalt and
the 3D printed asphalt. When the printed samples were cracked,
a brown substance was revealed which was dotted throughout the
sample cross section, that in many cases stretched out to bridge the
crack, see Fig. 7. This brown phase and crack bridging effect were not
observed in the cast test samples. Using X-ray photoelectron spec-
troscopy (XPS), the elemental composition of the brown phase was
compared to the bulk. No significant differences were found in the
composition, both being hydrocarbons with trace amounts of silicon
and sulphur.
4. Discussion & implications for design
We have successfully managed to design, build and test an asphalt
3D printer capable of printing small objects and repairing cracks
in asphalt. Since 3D printers are a mature technology this might
not seem remarkable, nevertheless it was not an easy task. The
main difficulty we encountered is that asphalt behaves as a rela-
tively low melting point non-Newtonian liquid when the material
is moving through the extruder as it is heated up, and then in
472 R. Jackson et al. / Materials and Design 160 (2018) 468–474
Fig. 5. Results: (a) photo of printed pyramid, (b) photo of three point bend test sample, (c) photo of moulded crack before (left) and after print fill (right).
between the extruder tip and deposition surface, as it cools down.
Although polymers used in filament-fed 3D printers are generally
non-Newtonian too, their simpler extruder system makes flow con-
trol much easier. Flow through our auger screw extruder created a
more complicated regime of rheology and pressure in relation to set
temperature and other operational parameters which showed highly
non-linear behaviour and made control of the extrusion process dif-
ficult. The functional constraints of some of the process variables
affected our ability to print, for instance, the rotation speed of the
auger screw is linked to the print speed (the extrusion multiplier
is programmed to double the rotation speed if the print speed is
doubled in order to deposit material at the same rate). The print
speed was also limited by the materials properties of the auger screw
(we used the high temperature SLA resin), since the low fracture
strength of this resin limited the torque we could apply. The aperture
affects the resolution of the printer, but again, low fracture strength
of SLA resin limited our ability to reduce aperture size since it led
to high pressures and resulted in mechanical failure. It is hoped that
Fig. 6. Mechanical properties: 3 point bend tests. (a) Cast pieces average, tested from top (CT) and bottom (CB) compared to print averages at each temperature. (b) Comparison
of all cast and printed pieces. (c) Stress at fracture comparison. (d) Elongation at break comparison.
R. Jackson et al. / Materials and Design 160 (2018) 468–474 473
Fig. 7. Crack bridging: (a) bridging in printed test samples, (b) closeup, parallel to sample axis shows fibrous crack bridging characteristic of ductile fracture, (c) 20×magnification
showing concentration of oily material throughout print layers. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
future designs with metal parts will allow us to explore a greater
range of extrusion rates and print resolutions.
The impact of 3D printing on mechanical properties is interest-
ing because it allows us to print a more ductile asphalt. There is a
significant increase (up to 900%) in elongation to fracture for the
printed samples. A possible explanation of this increased ductility
lies with the appearance of a crack bridging component in the sam-
ples. It is hypothesised that the brown phase precipitated throughout
the sample is composed of a lighter saturated fraction of the asphalt
that has coalesced due to size dependent mobility conditions during
the heating, screw mixing and/or extrusion process. Small amounts
of softer components coalesce naturally in asphalt, but usually at
scales of around 1–10 lm[30]. Here, the components are around
20–100 lm in diameter. This means that the 3D printing process at
this scale seems to avoid the degradation of its mechanical proper-
ties that can arise from leaving molten asphalt static for a long time
[31]. The 3D printing process seems to create a composite struc-
ture comprising of large concentrations of the brown phase dotted
throughout the asphalt (as seen in Fig. 7 (c)), giving the material
a higher toughness than cast asphalt. This would be advantageous
to any crack repair scenario since sites of cracks on roads are often
areas of increased stress or wear, and so depositing material with
enhanced ductility could prolong the life of the repair.
Printed asphalt extruded at less than 120 C often did not have
sufficient inter-layer bonding to avoid delamination, so were most
likely not fully bonded throughout the bulk of the material. Con-
versely, prints above 135 C gave poor dimensional accuracy due
to the lower viscosity of the hotter deposited asphalt, as well as
showing mechanical properties similar to cast samples. This suggests
that there is a point between delamination and complete intermelt-
ing that gives superior mechanical properties, we found this to be
the optimum print temperature of 130 C. Commonly printed poly-
mers such as ABS and PLA have lower specific heat capacities than
asphalt (1.8, 1.3, and 2.1 kJ/kg/C respectively) which may be the
reason why the initial printing temperature may occasionally be too
high as the print progresses due to heat build up in the printed
asphalt, giving melted edges and/or poor accuracy, and particularly
a “squashed” appearance with too low object height and too high
object width and length. Over the print time (approximately 20 min
for the test bar) this would be enough to deform layers previously
printed due to the excess heat in the upper layers. Although the
printer bed is heated for conventional polymer printing we did not
employ this method nor did we use a cooling fan normally used
with polymer 3D Printing, which may have exacerbated the issue.
With the correct parameters however, we can modulate proper-
ties through changing print temperature fairly rapidly over a small
10–15 C range. Furthermore, the feed and pellet system make
it relatively straightforward to add other materials such as small
microaggregates or nanomaterials (initial test prints with 10%10 nm
diameter titanium dioxide nanoparticles have been successful), and
then vary the composition of the feedstock during printing to cre-
ate more complex, functionally graded infrastructure materials with
a wider range of properties.
We believe these improved and tunable material properties of 3D
printed asphalt, combined with the flexibility and efficiency of the
printing platform, offers a compelling new approach not just to the
maintenance of road infrastructure, but by attaching it to a drone,
opens up a new way to repair hard to access structures such as the
flat roofs of buildings and other complex structures. The advantage
of this is not only in being able to cut costs – other repair methods
often requiring the erection of scaffolding and the closure or shut-
ting down of infrastructure to gain access – but also the repair can
initiated earlier before large scale deterioration has occurred. The
development of such repair drones would have implications both
for the way in which city infrastructure is repaired but also for the
economic model that underpins it. At the moment, much of city
infrastructure is built to fail and then be replaced, with the capital
costs of construction dominating the design parameters. Infrastruc-
ture designed to be continually monitored and repaired by fleet of
drones promises to be a different model which could have economic
benefits to society.
For instance, such an approach has the potential to be used for
roads. If road degradation is continually monitored then small cracks
can be repaired before they turn into potholes. By intervening at this
early stage and repairing the crack autonomously using 3D print-
ing we believe the road surface might be preserved for longer. Such
approaches have been explored for concrete road surfaces for spall
damage repair [22]. We have already demonstrated the 3D printing
of asphalt using a drone. The next stage of developing this technology
involves understanding the effect of environmental variables such
as road temperature, air temperature, the local chemistry, interface
with aggregate, as well as more comprehensive testing such as cyclic
loading of repaired crack roads.
The materials science of repair is not the only consideration in
the application of this technology. Identification and detection of
crack morphology, especially in the case of complex-shaped cracks
will be an important challenge. Automated computer vision systems
are currently being explored to address this issue [20]. The use of
gantry systems versus the employment of 6-axis robotic systems
is another design issue that is pertinent in the area of automated
construction and repair [32,33]. Although 6-axis systems have more
flexibility; for the repair of sub-cm small cracks in a horizontal
road surface, a simple gantry system, such as ours, may well prove
effective enough.
474 R. Jackson et al. / Materials and Design 160 (2018) 468–474
5. Conclusions
We have designed and built 3D printer capable of printing
asphalt. We have shown that this technology can be used to 3D print
asphalt into complex geometries, and to repair cracks. The mechan-
ical properties of 3D printed asphalt are different from cast asphalt,
showing up to nine times the ductility of cast samples with similar
fracture strengths. The increased ductility is due to microstructural
changes in the asphalt which result in crack-bridging fibres that
increase toughness. The material properties of 3D printed asphalt are
tunable, and combined with the flexibility and efficiency of the print-
ing platform, this technique offers a compelling new design approach
to the maintenance of infrastructure.
Acknowledgments
This work was funded by the EPSRC (Balancing the Impact of
City Infrastructure Engineering on Natural systems using Robots —
EP/N010523/1). We would like to thank Professor Quentin Pankhurst
for access to laboratory facilities, and Dr Joseph Bear at Kingston
University for the XPS analysis. We would also like to thank all
members of Self-Repairing Cities network.
References
[1] IAT, Asphalt Professionals in the 21st Century, 2000.
[2] S. Chi, B.R.K. Blackman, A.J. Kinloch, A.C. Taylor, Durability of asphalt mixtures:
effect of aggregate type and adhesion promoters, Int. J. Adhes. Adhes. 54 (2014)
100–111.
[3] The Asphalt Paving Industry: A Global Perspective, 3rd ed. ed., Number ISBN
0-914313-06-1 in Global Series 101, National Asphalt Pavement Association
and European Asphalt Pavement Association, Lanham, Maryland, 2011.
[4] U. Sandberg, Low Noise Road Surfaces a State-of-the-art Review, The Mitigation
of Traffic Noise in Urban Areas, 1992, volume Eurosymposium, May.
[5] A. Garcia, Self-healing of open cracks in asphalt mastic, Fuel 93 (2012)
264–272.
[6] Y. Tan, L. Shan, Y.R. Kim, B. Underwood, Healing characteristics of asphalt
binder, Constr. Build. Mater. 27 (1) (2012) 570–577.
[7] J. Norambuena-Contreras, A. Garcia, Self-healing of asphalt mixture by
microwave and induction heating, Mater. Des. 106 (2016) 404–414.
[8] P.K. Das, H. Baaj, N. Kringos, S. Tighe, Coupling of oxidative ageing and moisture
damage in asphalt mixtures, Road Mater. Pavement Des. 16 (2015) 265–279.
[9] A.A. Butt, B. Birgisson, N. Kringos, Considering the benefits of asphalt modifica-
tion using a new technical life cycle assessment framework, J. Civ. Eng. Manag.
22 (5) (2015) 597–607.
[10] UK Dept. of Transport Highways Agency, Design Manual for Roads and Bridges,
2006.
[11] Q. Liu, A. Garcia, E. Schlangen, V. Martin van de, Induction healing of asphalt
mastic and porous asphalt concrete, Constr. Build. Mater. 25 (2011)
3746–3752.
[12] H. Fang, Y. Sun, Q. Liu, B. Li, S. Wu, J. Tang, Ice melting properties of steel fiber
modified asphalt mixtures with induction heating 2017, IOP Conf. Ser. Mater.
Sci. Eng. 182 (1) (2017).
[13] Q. Liu, H. Schlangen, G. Van Bochove, The first engineered self-healing asphalt
road: how is it performing? ICSHM 2013: Proceedings of the 4th International
Conference on Self-healing Materials, Ghent, Belgium, 16-20 June 2013, 2013.
[14] A. Garcia, C.A. Austin, J. Jelfs, Mechanical properties of asphalt mixture contain-
ing sunflower oil capsules, J. Clean. Prod. 118 (2016) 124–132.
[15] Y. Huang, R.N. Bird, O. Heidrich, A review of the use of recycled solid waste
materials in asphalt pavements, Conserv. Recycl. Resour. 52 (1) (2007) 58–73.
[16] D.L. Presti, Recycled tyre rubber modified bitumens for road asphalt mixtures:
a literature review, Construction and Building Materials 49 (2013) 863–881.
[17] P. Zapata, J. Gambatese, Energy consumption of asphalt and reinforced concrete
pavement materials and construction, J. Infrastruct. Syst. 11 (1) (2005).
[18] Y. Huang, R. Bird, O. Heidrich, Development of a life cycle assessment tool
for construction and maintenance of asphalt pavements, J. Clean. Prod. 17 (2)
(2009) 283–296.
[19] A.A. Butt, I. Mirzadeh, S. Toller, B. Birgisson, Life Cycle Assessment framework
for asphalt pavements; methods to calculate and allocate energy of binder and
additives, Int. J. Pavement Eng. 15 (4) (2014) 290–302.
[20] R. Fuentes, T. Chapman, M. Cook, J. Scalan, Z. Li, R.C. Richardson, Briefing:
UK-RAS white paper in robotics and autonomous systems for resilient infras-
tructure. Volume 170 Issue SC3 72-79, Smart Infrastruct. Construct. 3 (2017)
72–79.
[21] T. Kim, S. Ryu, Review and analysis of pothole detection methods, J. Emerg.
Trends Comput. Inf. Serv. 5 (8) (2014) 603–608.
[22] J. Yeon, J. Kang, W. Yan, Spall damage repair using 3D printing technology,
Autom. Constr. 89 (2018) 266–274.
[23] IKO PLC, Easymelt Bonding Bitumen - CAS 64742-93-4, 2018.
[24] L. Loeber, G. Muller, J. Morel, O. Sutton, Bitumen in colloid science: a chemical,
structural and rheological approach, Fuel 77 (13) (1998) 1443–1450.
[25] A. Rabbani, D.R. Schmitt, Ultrasonic shear wave reflectometry applied to the
determination of the T shear moduli and viscosity of a viscoelastic bitumen,
Fuel 232 (2018) 506–518.
[26] E. Remišová, V. Zatkaliková, F. Schlosser, Study of rheological properties of bitu-
minous binders in middle and high temperatures., Civ. Environ. Eng. 12 (2016)
13–20.
[27] Pronterface, 2012, http://www.pronterface.com/.
[28] Slic3r, 2011, http://slic3r.org/.
[29] Sketchup, 2000, https://www.sketchup.com/.
[30] A. Jager, A. Lackner, C. Eisenmenger-Sittner, R. Blab, Identification of
microstructural components of bitumen by means of atomic force microscopy,
Proc. Appl. Math. Mech. 4 (2004) 400–401.
[31] B. Hofko, A. Falchetto, J. Grenfell, L. Huber, X. Lu, L. Porot, L. Poulikakos, Z. You,
Effect of short-term ageing temperature on bitumen properties., Road Mater.
Pavement Des. 18 (2017) 108–117.
[32] C. Gosselin, R. Duballet, P. Roux, N. Gaudillière, J. Dirrenberger, P. Morel,
Large-scale 3D printing of ultra-high performance concrete—a new processing
route for architects and builders, Mater. Des. 100 (2016) 102–109.
[33] Y.W.D. Tay, B. Panda, S.C. Paul, N.A. Noor Mohamed, M.J. Tan, K.F. Leong, 3D
printing trends in building and construction industry: a review., Virtual Phys.
Prototyping 12 (3) (2017) 261–276.
... The basic design of the printheads developed by Bellini [10] and Reddy et al. [13] was further explored and improved by Silveira et al. [15], Jackson et al. [24], Singamneni et al. [32,33], Tseng et al. [34], Zhou et al. [37], Leng et al. [39], Alexandre et al. [41], and Wang et al. [44]. Except for the printhead developed by Jackson et al. [24], all equipment adopted a screw with a compression profile. ...
... The basic design of the printheads developed by Bellini [10] and Reddy et al. [13] was further explored and improved by Silveira et al. [15], Jackson et al. [24], Singamneni et al. [32,33], Tseng et al. [34], Zhou et al. [37], Leng et al. [39], Alexandre et al. [41], and Wang et al. [44]. Except for the printhead developed by Jackson et al. [24], all equipment adopted a screw with a compression profile. Silveira et al. [15], Jackson et al. [24], and Alexandre et al. [41] integrated their deposition tools into low-cost desktop 3D printers. ...
... Except for the printhead developed by Jackson et al. [24], all equipment adopted a screw with a compression profile. Silveira et al. [15], Jackson et al. [24], and Alexandre et al. [41] integrated their deposition tools into low-cost desktop 3D printers. In contrast, Tseng et al. [34], Singamneni et al. [32], and Wang et al. [44] constructed their own Cartesian positioning platforms. ...
Article
Full-text available
This paper presents a systematic review on extrusion additive manufacturing (EAM), with focus on the technological development of screw-assisted systems that can be fed directly with granulated materials. Screw-assisted EAM has gained importance as an enabling technology to expand the range of 3D printing materials, reduce costs associated with feedstock fabrication, and increase the material deposition rate compared to traditional fused filament fabrication (FFF). Many experimental printheads and commercial systems that use some screw-processing mechanism can be found in the literature, but the design diversity and lack of standard terminology make it difficult to determine the most suitable solutions for a given material or application field. Besides, the few previous reviews have offered only a glimpse into the topic, without an in-depth analysis about the design of the extruders and associated capabilities. A systematic procedure was devised to identify the screw-assisted EAM systems that can print directly from granulated materials, resulting in 61 articles describing different pieces of equipment that were categorized as experimental printheads and commercial systems, for small- and large-scale applications. After describing their main characteristics, the most significant extruder modifications were discussed with reference to the materials processed and performance requirements. In the end, a general workflow for the development of 3D printers based on screw extrusion was proposed. This review intends to provide information about the state-of-the-art screw-assisted EAM and help the academy to identify further research opportunities in the field.
... A consequência deste processo de degradação é o aumento da rugosidade e diminuição da segurança (BUTTLAR et al., 2018). Após o aparecimento das rachaduras, é difícil cessar o crescimento de buracos que causam danos significativos aos veículos e diminuem a vida útil do pavimento (JACKSON et al., 2018). ...
... Tecnologias distintas vêm sendo desenvolvidas no intuito de otimizar a eficiência de autocorreção das misturas asfálticas, como aquecimento por indução de asfalto (processo que é usado para selagem, endurecimento ou amolecimento de metais ou outros materiais condutores) e uso de rejuvenescedores encapsulados (XU et al., 2018a). O aquecimento por indução e rejuvenescedores encapsulados são dois meios tecnológicos importantes no desenvolvimento da autorregeneração de misturas asfálticas (HAGER, 2010;SCHILANGEN, 2016;AYAR et al., 2016;SUN et al., 2018a;XU et al., 2018b). Rejuvenescedores encapsulados são partículas esféricas, ou cápsulas, que contém aproximadamente 70% de óleo ou bio-óleo como aditivo (AL-MANSOORI et al., 2018). ...
Conference Paper
Full-text available
A camada de revestimento asfáltico pode apresentar problemas sob o efeito a longo prazo de cargas e fatores ambientais. A fissura por fadiga é um dos problemas mais comuns encontrados nas rodovias. Técnicas de microencapsulação e aquecimento por indução estão sendo estudadas a fim de minimizar esse feito. O objetivo desta revisão foi destacar os pontos positivos e negativos na utilização de rejuvenescedores encapsulados na autorregeneração de misturas asfálticas por meio de pesquisa da literatura em uma taxonomia coerente e sistemática, apresentando quais materiais foram estudados e as tecnologias de encapsulação implementadas até esse momento. Utilizou-se as bases de dados Scielo e Science Direct na busca dos artigos científicos. Realizou-se uma seleção dos artigos por meio de descritores específicos e priorizou-se por pesquisas publicadas neste ano. Os artigos, em sua maioria, foram encontrados na base de dados Science Direct. A revisão concluiu que o uso de microcápsulas com agentes de autorregeneração é eficiente. As microcápsulas com agentes rejuvenescedores apresentam boa resistência aos procedimentos de usinagem e compactação. Asfaltos porosos demonstraram-se mais eficientes na autorregeneração do que misturas asfálticas densas. A quantidade de octanol utilizado na fabricação de microcápsulas alteram o tamanho e rugosidade destes elementos. Há uma tendência em estudos de autorregeneração com bio-óleos encapsulados.
... Classical methodologies allow the determination of softening, penetration points, brittleness, viscosity, among others, at a single temperature point [9]. However, current methods are based on the total deformation resistance and its relationship between the elastic and viscous parts of the asphalt bitumen [25]. ...
Article
Full-text available
This study focuses on the mechanical behaviour of asphalt mastic composed of filler particles bonded with an as-phalt bitumens. Asphalt mastics are viscoelastic composite materials widely used in the construction of pave-ment layers. The mechanical properties and the influence of the fillers on the filler/bitumen (f/b) matrix is one ofthe main areas of current research. In particular, the elastic determination of fillers for mechanical testing in as-phalt mastic is relevant to understand permanent deformation caused by temperature variations caused by sea-sonal changes and vehicular traffic loads. In this sense, this research proposes a new methodology for rheologicalcharacterization of the elastic properties of the filler ξ2 and elastic-viscous properties of the asphalt bitumen, ξ1and η, respectively, complementing the existing designs of asphalt mixture. The proposed methodology allowsfor identification of the influence of non-conventional fillers in the behavior of the asphalt mastic for the differ-ent recovery cycles of the Multiple Stress Creep Recovery (MSCR) and determination of new rheological parame-ters for the compression of the recovery phenomena and the elastic capacity of the type of filler and weight of thebase bitumen. The results obtained show a greater adjustment to the experimental curves in determining theelastic modulus in each cycle for the hydrated lime and fly ash fillers with different filler/bitumen ratios. In par-ticular, the proposed model for bituminous mastics achieves a strong fit with the experimental curves by empiri-cally reducing the quadratic error (R2 = 0.99) and managing to differentiate the elastic capacity ξ2 of each fillerand its effect with increasing concentration. For example, it establishes that the Hydrated lime filler (HL) ac-quires an average Young's modulus of 0.005 MPa, being 99.31% more elastic than Fly ash filler (FA) for a load of3.2 kPa at a 1.25f/b ratio. In addition, the new model can be used to modify bitumen properties to design opti-mized and stronger asphalt mixtures.
... Especially in 3D concrete printing, more and more studies have turned 3D printing technology from laboratory to industrial application [9][10][11][12][13][14]. However, asphalt pavement-related fields are challenging to use in 3D printing because it behaves as a non-Newtonian liquid during extrusion [15]. ...
Article
In this work, icosahedral 3D printing aggregate (3DPA) was designed, prepared with three polyamide materials and compared with four natural aggregates for surface texture and difference in adhesion to asphalt. Firstly, the morphological and structural characteristics of 3DPAs were analyzed. Then, four kinds of natural aggregates were compared with 3DPAs for the boiling test to quantitatively evaluate the adhesion of 3DPAs and asphalt. Subsequently, the microscopic images of different aggregates were identified by digital image technology. Finally, the adhesion performance of different aggregates to asphalt was predicted by quantifying the number of edge points. The results show that the adhesion of 3DPAs to asphalt under boiling test is similar to basalt, better than diabase and granite, but worse than limestone. In particular, since 3DPAs are regular icosahedrons, the score can be quantified in the boiling test, which is significantly better than irregular natural aggregates. Digital image processing can accurately identify the microscopic morphology of aggregates and calculate their surface structural characteristics, and the results are basically consistent with the boiling test. 3DPA with regular shape is a method to avoid cutting natural aggregate and quantitatively evaluate asphalt adhesion properties accurately. The research results are helpful to promote the application of 3D printing technology in asphalt pavement.
Article
In this paper, a machine process to repair cracks in asphalt pavements by filling them with hot bitumen is demonstrated for the first time. We have characterized the quality of these fillings. Furthermore, the paper examines how the filling speed, temperature, bitumen type, crack width, crack irregularity, and the flow of hot bitumen affect the crack filling quality by the machine. In the laboratory we have used a 3D-printer as the machine to fill the cracks due to its simplicity. We have characterized the quality of the repaired cracks in terms of their porosity and shear and tensile strength. We have demonstrated that the bitumen flow, filling speed, and width and depth of cracks determine the quality of the filling and will need to be carefully controlled in a future autonomous crack filling device. The complex interaction between these parameters will need to be underpinned in future research.
Article
Full-text available
This report identifies a new and potentially transformative class of materials: materials that are created through human agency but emulate the properties of living systems. We call these ‘animate materials’ and they can be defined as those that are sensitive to their environment and able to adapt to it in a number of ways to better fulfil their function. These materials may be understood in relation to three principles of animacy. They are ‘active’, in that they can change their properties or perform actions, often by taking energy, material or nutrients from the environment; ‘adaptive’ in sensing changes in their environment and responding; and ‘autonomous’ in being able to initiate such a response without being controlled. Artificial materials that are fully animate in all these dimensions do not exist at present, but there are many examples of materials with some features that correspond with our definition of animacy, as well as research that indicates potential ways to improve and extend their capabilities. The development of such materials has been identified by the Royal Society as an area of research with potential to deliver major change, most noticeably in the built environment, from roads and buildings to transport and industry, as well as in sectors such as medicine and clothing. Development and implementation of proto-animate materials are currently being pursued in many disciplines, but without any formal co-ordination. The Royal Society is seeking to support interdisciplinary efforts in the field of animate materials, as well as to improve understanding of their potential, while identifying steps needed to accelerate their development in a socially responsible manner.
Article
A 3D printing process involves printing a 3D object repeatedly by sequencing many thin layered of metals in succession. This is a process that produces a 3D model of a finished product. 3D printing allows engineers to create functional prototypes using their 3D models. In this study, the primary materials used are PLA and Alloy Steel. The model of a bolt are then created using 3D printer. Snapmaker software is used for the slicing and printing of the model. The model is then analyzed for shear stress and deformation.
Article
This study investigated the structural mechanical properties and failure modes of large-scale three-dimensional (3D) printed concrete walls under axial compression loads. Eight large-scale 3D printed concrete wall specimens with varying ratios of height to thickness were tested under axial compression, including two specimens without horizontal steel reinforcement and six specimens with horizontal steel reinforcement. It can be concluded that the failure of 3D printed concrete walls under an axial compression load is brittle. The use of horizontal steel reinforcement reduces the ultimate bearing capacity of 3D printed concrete walls under axial compression. The layers in which the horizontal steel reinforcement is set are weak zones. Based on the stress state characteristics, the structural mechanical properties were further studied by modeling the test results of the strains of large-scale 3D printed concrete walls. The failure load was defined accordingly, and a formula for predicting the failure loads was proposed and validated.
Article
3D (three-dimensional) printing was used as a rapid prototyping tool to determine the influence of cell geometry and infill materials on the physical properties of geometrically patterned matrices while subjected to compressive stress. Matrices of comparable patterns but varied scales and densities were fabricated from acrylonitrile butadiene styrene (ABS) plastic using fused deposition modeling (FDM) 3D printing. The test results confirm that some matrices reinforced by infill with sand, gravel, and mixtures of the two show better compressive strength than conventional concrete, and may find application in matting for airfield damage repair. The cell matrix geometry that demonstrated maximum strength (comparable with conventional concrete) was a hexagonal geometry with a relative density to solid plastic of 0.32 infilled with a mixture of sand and gravel. Additional data suggests that at larger scales, maximum strength comparable with conventional concrete could be achieved with even lower relative density.
Article
Full-text available
Three-dimensional (3D) printing (also known as additive manufacturing) is an advanced manufacturing process that can produce complex shape geometries automatically from a 3D computer-aided design model without any tooling, dies and fixtures. This automated manufacturing process has been applied to many diverse fields of industries today due to significant advantages of creating functional prototypes in reasonable build time with less human intervention and minimum material wastage. However, a more recent application of this technology towards the built environment seems to improve our traditional building strategies while reducing the need for human resources, high capital investments and additional formworks. Research interest in employing 3D printing for building and construction has increased exponentially in the past few years. This paper reviews the latest research trends in the discipline by analysing publications from 1997 to 2016. Some recent developments for 3D concrete printing at the Singapore Centre for 3D Printing are also discussed here. Finally, this paper gives a brief description of future work that can be done to improve both the capability and printing quality of the current systems.
Article
Full-text available
Properties of asphalt mixtures after ageing are fundamental parameters in determining long-term performance (e.g. durability) of these materials. With increasing popularity of reduced temperature mixtures, such as warm-mix asphalt, WMA, the question remains how a reduction in short-term ageing affects the properties after long-term ageing of bituminous materials. This paper aims to improve our understanding of the effect of asphalt manufacturing temperature on ageing and the resulting mechanical properties of bituminous binder by studying the effect of short- and long-term ageing of different bitumen samples as a function of short-term ageing temperatures. For this purpose, round robin experiments were conducted within the RILEM technical committee (TC) 252 chemo-mechanical characterisation of bituminous materials by 10 laboratories from 5 countries using four binders of the same grade (70/100 pen) from different crude sources. The short-term ageing was carried out using the standard procedure for rolling thin film oven test (RTFOT), but varying the temperatures. Long-term ageing was carried out using the standard procedure for pressure aging vessel (PAV) in addition to RTFOT. For the mechanical characterisation, rheological data were determined by using the dynamic shear rheometer (DSR) and conventional tests, with needle penetration and softening point using the ring and ball method. The results show that although different short-term ageing temperatures showed a significant difference in the mechanical properties of the binders, these differences vanished after long-term ageing with PAV.
Article
Full-text available
In this paper, the ice melting performance of asphalt concrete with steel fibers was studied. Steel fiber modified asphalt mixtures were prepared, five different fiber amount of steel fiber modified asphalt mixtures were mixed to study their induction heating rate. The samples covered with different thickness of ice were heated with induction heating to study their ice melting efficency. It was proved that the induction heating of steel fiber modified asphalt mixtures could significantly improve their ice melting efficency compared with the natural condition. And it was found that the thickness of the ice had little influence on the induction heating rate of the asphalt concrete.
Article
Full-text available
The bitumen binders in road pavements are exposed traffic loading effect at different climatic conditions. A resistance to these stresses depends on bitumen properties as well. The paper presents rheological properties (G*, δ, ν*) determined and compared for four bituminous binders (unmodified and polymer modified bitumen) at temperature 46 – 60 (80) °C and dynamic viscosity at temperature 130 – 190 °C (Brookfield viscometer). On the basis of viscosity results it is possible to set optimal production and compaction temperatures. Elastic and viscous behavior of binder in the middle temperature is determined in rheometers. The higher value of complex modulus, the stiffer bitumen binder is able to resist deformation. The greater content of elastic components (e.g. polymer in bitumen) varies mainly elastic-viscous properties of primary bitumen.
Article
This paper aims to evaluate the effect of microwave and induction heating on the self-healing of asphalt mixture test samples. With this purpose, dense asphalt mixtures with four different percentages of steel wool fibres have been prepared to build semi-circular asphalt test samples. Asphalt self-healing has been characterised as the three-point bending strength of test samples before and after healing. This process was repeated ten times in every test sample. Moreover, self-healing was induced in the semi-circular test samples by heating them under microwave and induction. Besides, the chemical degradation of asphalt mixture under microwave and induction heating was monitored by measuring the mass of test samples before and after the heating process. It was found that microwave technology is more effective than induction heating to heal cracks in asphalt roads. Furthermore, the healing level of asphalt mixtures reduced with every healing cycle, until the test specimens could not resist more damage-healing cycles. It could be seen that microwave heating degrades bitumen, and increases the porosity of asphalt mixture. Finally, it was hypothesised that air voids in mixture play an important role in asphalt self-healing by increasing the internal pressure and mobility of bitumen during the heating process.
Article
Embedding encapsulated rejuvenators in asphalt mixture is a sustainable way to extend the lifetime of roads, and minimise the negative environmental impacts of current maintenance processes. The aim of this article is to investigate the use of encapsulated oil for use as rejuvenators in asphalt pavements. With this purpose, capsules of 5 different diameters have been built and their mechanical strength and composition characterised. Then, the capsules have been mixed in asphalt mixture, and their resistance to mixing and compacting has been observed. Finally, the effect of capsules in the indirect tensile strength, modulus, and fatigue life of asphalt mixture test samples has been quantified. It was observed that bigger capsules contained more oil and were less strong than smaller capsules. In addition, most of the capsules resisted the mixing and compaction processes. Asphalt mixture with capsules presented reduced indirect tensile strength and modulus, which happened because the capsules had lower strength than required. Finally, the fatigue life of asphalt mixture was not affected by the capsules.
Article
In this paper, a possible way to capture the combined effect of oxidative ageing and moisture damage on mixture performance has been proposed. The formulations that are needed for finite element (FE) modelling of oxygen and moisture diffusion process have been established. The proposed model should be able to link the in-time changes to the mastic as function of mixture morphology, ageing propensity and the moisture diffusion properties to the physical properties of the asphalt mixture due to the loss of adhesive and/or cohesive bonding. Such an FE model can help find the trends and relationships that can assist in the development of predictive pavement performance model. Also, from this, one can figure out the key parameters that are mainly responsible for ageing-moisture-induced premature damage of asphalt pavements.