Beruflich Dokumente
Kultur Dokumente
A review of 4D printing
Farhang Momeni, Seyed M.Mehdi Hassani.N, Xun Liu ⁎, Jun Ni
Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109, USA
H I G H L I G H T S G R A P H I C A L A B S T R A C T
a r t i c l e i n f o a b s t r a c t
Article history: Research into 4D printing has attracted unprecedented interest since 2013 when the idea was first intro-
Received 11 November 2016 duced. It is based on 3D printing technology, but requires additional stimulus and stimulus-responsive
Received in revised form 19 February 2017 materials. Based on certain interaction mechanisms between the stimulus and smart materials, as well
Accepted 21 February 2017
as appropriate design of multi-material structures from mathematical modeling, 4D printed structures
Available online 1 March 2017
evolve as a function of time and exhibit intelligent behavior. Unlike 3D printing, 4D printing is time-
Keywords:
dependent, printer-independent, predictable, and targets shape/property/functionality evolution. This
4D printing allows for self-assembly, multi-functionality, and self-repair. This paper presents a comprehensive re-
Shape-shifting mechanisms view of the 4D printing process and summarizes the practical concepts and related tools that have a
Multi-material structures prominent role in this field. Unsought aspects of 4D printing are also studied and organized for future
Digital materials research.
Voxel © 2017 Elsevier Ltd. All rights reserved.
Inverse and forward problems
http://dx.doi.org/10.1016/j.matdes.2017.02.068
0264-1275/© 2017 Elsevier Ltd. All rights reserved.
F. Momeni et al. / Materials and Design 122 (2017) 42–79 43
3D printer
Static
3D printing Material
structure
3D printer:
To be able to print multi-
material structure
Stimulus
Interaction
Mathematics: mechanism
Mainly an inverse problem to find
the printing paths
4D Printing Bases
which is referred to as the interaction mechanism in this review structure is unloaded at the low temperature and the desired
paper. For example, one of the main interaction mechanisms is shape is achieved; fourth, the original shape can be recovered
constrained-thermo-mechanics. In this mechanism, the stimulus by reheating the structure.
is heat and the smart material has the shape memory effect. It • Mathematical Modeling: Math is necessary for 4D printing in
contains a 4-step cycle. First, the structure is deformed by an ex- order to design the material distribution and structure needed
ternal load at a high temperature; second, the temperature is to achieve the desired change in shape, property, or functionality.
lowered while the external load is maintained; third, the Theoretical and numerical models need to be developed to
Fig. 5. Illustration of the difference between one-way and two-way shape memory materials (Hager et al. [34]).
establish the connections between four core elements: material 1.3. Self-assembly
structure, desired final shape, material properties, and stimulus
properties. These will be discussed in additional detail in the fol- Self-assembly extends from the nanoscale to the planetary scale
lowing sections. [25,26]. Currently, researchers are interested in macroscale appli-
cations (Campbell et al. [26]). One example is the transfer of equip-
ment parts to the inside of a human body through a small hole. The
A 4D printed structure can be regarded as a child born from the
parts can then self-assemble at the desired location for medical
marriage between a 3D printer and smart materials. It can walk by
purposes (Zhou et al. [11]). Another future application of self-
being exposed to the external stimulus through an interaction
assembly will be on a large scale and in a harsh environment.
mechanism, and it learns how to walk properly with the assistance
Individual parts can be printed with small 3D printers and then
of mathematics.
self-assembled into larger structures, such as space antennae and
satellites (Tibbits et al. [5]). This capability paves the way for the
1.2. Motivations
creation of transportation systems to the International Space Sta-
tion (Choi et al. [2]). Further applications include self-assembling
4D printing opens new fields for application in which a structure can
buildings, especially in war zones or in outer space where the ele-
be activated for self-assembly, reconfiguration, and replication through
ments can come together to yield a finished building with mini-
environmental free energies (Tibbits [4]). This brings several advan-
mum human involvement (Campbell et al. [26]). Moreover, some
tages, such as significant volume reduction for storage, and transforma-
limitations in architectural research and experiments can be re-
tions that can be achieved with flat-pack 4D printed structures. The
moved with the capabilities of 4D printing (Čolić-Damjanovic &
latter may include transformations to 3D structures required during ac-
Gadjanski [27]).
tual applications [4]. Another example is that instead of directly creating
a complicated structure using the 3D printing process, simple compo-
nents from smart materials can be 3D printed first and then self- 1.4. Multi-functionality or self-adaptability
assembled to reach that final complex shape (Zhou et al. [11]). In gener-
al, the potential applications of 4D printed structures can be classified Adaptive infrastructures are another application of 4D printing
into the three categories: self-assembly, multi-functionality, and self- (Campbell et al. [26]). 4D printing can integrate sensing and actuation
repair. directly into a material so that external electromechanical systems are
not necessary (Tibbits et al. [5]). This would decrease the number of
parts in a structure, assembly time, material and energy costs, as well
as the number of failure-prone devices, which is usually utilized in cur-
rent electromechanical systems (Tibbits et al. [5]). Multi-functional and
self-adaptive 4D printed tissues (Khademhosseini & Langer [28]; Jung
et al. [29]) and 4D-printed personalized medical devices, such as trache-
al stents (Zarek et al. [30]), are other fascinating applications of 4D
printing.
1.5. Self-repair
Fig. 7. The difference between folding and bending (Liu et al. [41]).
Fig. 8. Types of surface topography: wrinkling, creasing, and buckling (Liu et al. [41]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 47
Shape memory materials belong to the category of stimulus- 2.5. Shape-shifting types and dimensions
responsive materials shown in Fig. 5. Hager et al. [34] described that
in shape memory materials, the permanent shape is “memorized” by 2.5.1. 1D-to-1D expansion/contraction
the material and alterations between a permanent and a temporary Raviv et al. [12] and Yu et al. [16] demonstrated the linear ex-
shape occur. A dual-SME material includes one permanent shape and pansion/contraction from 1D to 1D through a 4D printing process.
one temporary shape, while a triple-SME material has one permanent The research from Raviv et al. is shown in Fig. 9. The stimulus is
shape and two temporary shapes (Fig. 6). Similarly, a multi (n)-SME water and the 4D printed structure contains alternating layers of
material has one permanent shape and (n − 1) temporary shapes [34]. rigid discs and active hydrogels. After the structure is immersed
Multi-SME materials were discussed by Xie [35], Yu et al. [36], Sun in water, the hydrogel parts will swell while the rigid discs remain
et al. [33], Therien-Aubin et al. [37], and Li et al. [38]. the same. By adjusting the ratio of expandable hydrogels to rigid
discs, the structure can show linear shape-shifting behavior with
2.3. Folding vs. bending a desirable final length (Raviv et al. [12]). On the other hand, in
the structure obtained by Yu et al. the stimulus is heat, as shown
The difference between folding and bending is nuanced. Bend- in Fig. 10. Its linear shape-shifting behavior is obtained based on a
ing is a global deformation associated with smoother curvatures, shape memory cycle, which includes the usual programming and
Fig. 10. The illustration of 1D-to-1D shape-shifting by linear expansion/contraction adapted from (Yu et al., 2015) [16].
48 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 12. An illustration of 1D-to-2D sinusoidal shape-shifting by self-bending (Tibbits et al. [5]).
Fig. 14. Two rigid discs to tune the final folding angle for the above structure in Fig. 13
recovery steps for thermoresponsive shape memory polymers (Tibbits [4]).
Fig. 16. 2D-to-2D self-bending in which a rectangular network transforms into a circle.
Fig. 13. The transformation of 1D strand to 3D wireframe cube by self-folding (Tibbits [4]). Scale bar, 200 μm (Villar et al. [44]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 49
Fig. 17. Multi-shape memory effect from 2D to 3D by self-bending in a smart trestle (Wu et al. [17]).
Fig. 18. Multi-shape memory effect from 2D to 3D by self-bending in an active helix shape (Wu et al. [17]).
Fig. 19. Multi-shape memory effect from 2D to 3D by self-bending in an active wave shape (Wu et al. [17]).
50 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 20. Multi-shape memory effect from 2D to 3D by self-bending in a smart insect-like structure (Wu et al. [17]).
swell and the droplets with lower osmolarity shrink. The shape con- down to a low temperature while maintaining the previously applied
tinues to change until the osmolarity gradient becomes zero. force. The shape-shifting behavior is enabled by different thermal expan-
sion coefficients of the active and non-active materials.
2.5.6. 2D-to-3D bending In another experiment performed by Zhang et al. [18], a 3D periodic
Wu et al. [17] applied multi-shape memory effect for 2D-to-3D structure was achieved from a 2D sheet consisting of four periodic cells
transformations in various structures, such as the active trestle in a square arrangement. Each cell included one central region and four
(Fig. 17), active helix shape (Fig. 18), active wave shape (Fig. 19), rectangular neighbors, as shown in Fig. 23. The mechanism of this
smart insect-like structure (Fig. 20), and smart hook (Fig. 21). Heat shape-shifting behavior is similar to that exhibited in the previous
was the stimulus and these shape-shifting behaviors were achieved in experiment.
the usual shape memory cycles with well-known programming and re- Jamal et al. [13] illustrated a shape alteration for tissue engineering
covery steps. purposes in which a 2D bio-origami planar pattern changed to a 3D pat-
Zhang et al. [18] demonstrated 2D-to-3D self-bending behavior by tern based on the self-bending operation (Fig. 24). This shape-shifting is
creating a flower-like structure. This is shown in in Fig. 22(a) and (b). In enabled by the different swelling ratios of hydrogels and rigid materials
Fig. 22(c), a complex structure was created by tearing paper off of the under water.
flower-like 3D structure. Heat is the stimulus in this shape-shifting behav- Villar et al. [44] illustrated 2D-to-3D self-bending behavior in which
ior. However, their experiment is different from the regular shape mem- a flower-shaped network of droplets transforms into a hollow sphere.
ory cycles with programming and recovery steps. In this mechanism, This is shown in Fig. 25. This shape-shifting is enabled by the osmolarity
there is no need to apply external force at a high temperature and cool gradient of the droplets described earlier.
Fig. 21. Multi-shape memory effect from 2D to 3D by self-bending in a smart hook (Wu et al. [17]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 51
Fig. 22. Flower-like 4D structure. (a) The original flat sheet. (b) The final flower-like structure. (c) A complex structure created by tearing paper off of the flower-like 3D structure (Zhang
et al. [18]).
2.5.7. 2D-to-3D folding Mao et al. [15] demonstrated a sequence of self-folding operations
As shown in Fig. 26, Tibbits [4] printed a 2D flat plane that can be for transforming a 2D strip into 3D shape, as shown in Fig. 29. This
folded into a closed-surface cube, which presents 2D-to-3D shape- shape-shifting behavior is also triggered by heat in the regular shape
shifting behavior based on self-folding. This shape-shifting memory cycles with programming and recovery steps.
behavior is enabled by a stress mismatch between rigid and
active materials due to their different swelling behaviors under 2.5.8. 2D-to-3D twisting
water. Zhang et al. [18] illustrated helical structures with different spi-
The same shape-shifting behavior from a 2D pattern to 3D cube ral degree patterns by twisting a 2D sheet and converting it to 3D
based on self-folding was also illustrated by Naficy et al. [45], shapes, as shown in Fig. 30. This shape-shifting behavior has a sim-
where both water and heat are required as stimulus. They [45] ilar mechanism to that shown in Fig. 22 and Fig. 23. In addition,
printed bilayer hinges made of active and non-active gels. The Zhang et al. [18] printed the fibers with certain angles to induce
active layer is a temperature-sensitive hydrogel. In a dry condition twisting, and by adjusting the print angles of active fibers, the
at room temperature, the cube is flat. When submerged in water at final twist angle would be changed.
room temperature, the active hydrogel swells and the flat
structure is folded. However, if the temperature of the water is 2.5.9. 2D-to-3D surface curling
increased above the lower critical solution temperature (LCST) of Tibbits et al. [4] fabricated a surface grid structure with alternating
the active hydrogel, the folded structure unfolds. This occurs upper and lower segments of expandable materials, which could yield
because the active hydrogel collapses above its LCST (which is a mathematical sinusoidal surface under water (Fig. 31). This was a
mainly due to the breaking of the hydrogen bondings). demonstration of 2D to 3D shape-shifting behavior with a surface curl-
As shown in Fig. 27, a truncated octahedron was created by Tibbits ing feature. This shape-shifting behavior is enabled by a stress mismatch
et al. [5] and shows 2D to 3D shape-shifting behavior based on self- between rigid and active materials from their different swelling proper-
folding mechanism. This shape-shifting is enabled by the stress ties under water.
mismatch between rigid and active materials due to their different To further illustrate surface curling behavior, hair-like
swelling ratios. structures were generated from vertical linear strips by
Ge et al. [6] showed self-folding transformations from 2D flat sheets Tibbits et al. [5], as shown in Fig. 32. The mechanism is the same
to 3D shapes, examples of which include an active origami box, pyra- as above.
mid, and airplanes (Fig. 28). These shape-shifting behaviors are trig- Raviv et al. [12] illustrated 2D-to-3D surface curling behavior with
gered by heat in the usual shape memory cycles with the regular various configurations (Fig. 33). The mechanism is the same as that de-
programming and recovery steps. scribed above.
Fig. 23. A 3D periodic structure created from a 2D sheet by self-bending (Zhang et al. [18]).
52 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 24. A bio-origami 2D pattern transforms into a 3D pattern by self-bending: (a) Schematic illustration of the self-folding of PEG bilayer. (b) A fluorescent micrograph of a self-folded
bilayer (Jamal et al. [13]).
Ge et al. [6] demonstrated 2D-to-3D shape-shifting behavior in appropriate stimulus (Fig. 35). This is an illustration of 2D-to-3D change
which a flat sheet transformed into a complex and non-uniform in surface topography. This shape-shifting behavior is enabled by a
curvature shape as a sculpture (Fig. 34). This shape-shifting stress mismatch between rigid and active materials from their different
behavior is triggered by heat in the usual shape memory cycles underwater swelling properties.
with well-known programming and recovery steps. Tibbits et al. [4] showed another 2D-to-3D surface
topographical change in Fig. 36. In this test, three shapes evolved
2.5.10. 2D-to-3D surface topographical change as a function of time (0 min, 30 min, and 24 h in contact with
In one experiment from Tibbits et al. [4], mountain and valley fea- stimulus). These sinusoidal topographies are generated based on
tures can be generated from concentric circles in the presence of an the different swelling ratios between the active and rigid materials.
Fig. 25. (a) The experiment related to 2D-to-3D self-bending in which a flower-shaped network transforms into a hollow sphere. Scale bar, 200 μm (b) Simulation of (a) (Villar et al. [44]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 53
The concentrations of these materials vary from the center to the behavior is triggered by heat in the usual shape memory cycles with
perimeter of the disc. the well-known programming and recovery steps.
2.5.15. Discussions
In many applications, a 4D printing process can be used to avoid
directly printing a complicated 3D shape. Instead, a lower-
dimension shape can be printed first and then the other
dimension(s) can be enabled in the target location with the re-
quired performance. Printing a lower-dimensional object is rela-
tively easy, fast, and has lower manufacturing costs. Furthermore,
the storage and transportation of lower-dimension objects are
more convenient. In the study of 2D-to-3D shape-shifting behavior
from Liu et al. [41], the addition of the third dimension can add
desired value or functions to the 2D substrates, which can be
printed with various ordinary techniques such as inkjet, photoli-
thography, and roll-to-roll printing.
Even though many shape-shifting types and dimensions have
been demonstrated in the literature, hybrid shape-shifting
behavior is largely unexplored. Future studies can focus on serial
and parallel combinations of different shape-shifting types and
Fig. 27. 2D-to-3D self-folding to make a truncated octahedron (Tibbits et al. [5]). dimensions. For example, a complex 3D structure can be generated
54 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 28. An illustration of a 2D-to-3D alteration in which some origami shapes, such as an origami box, pyramid, and airplane can be generated by self-folding (Ge et al. [14]).
3.1. Multi-material structures us to move from analog materials to digital materials is the physical
voxel (Hiller et al. [54]), which is defined as the fundamental and phys-
In 4D printing, multiple materials usually need to be inserted ical bit that occupies 3D physical space. The physical voxel can be of any
into a single and one-time printed structure (Raviv et al. [12]). size and shape [54–56]. In nature, biological structures usually consist of
This multi-material structure can be a mixture of different smart fundamental building blocks that can be considered physical voxels,
materials or a combination of smart materials and conventional such as DNA and proteins (Hiller et al. [54]). In 4D printing and associ-
materials. The single-material structure in 4D printing should al- ated multi-material structures, the physical voxel can be similarly
ways be fabricated with a smart material. In addition, it needs to defined.
be based on the structure with a gradient distribution of materials. Digital material is defined as an assembly of various physical
The gradient distribution of a single material means that the densi- voxels [54,55]. The spatial arrangement of voxels plays a major
ty of the structure is different at various locations. This anisotropy role in determining the features of a 4D-printed structure [12]. In
can generate shape-shifting behaviors such as bending and twist- digital materials, each voxel contains only one material. Adjacent
ing, which is beyond linear expansion and contraction. Most of voxels can be composed of different materials. Each voxel has its
the previous studies on 4D printing focused on multi-material own properties and the collection of different voxels results in
structures. In this review, the concept of digital material is intro- the multi-material structure. According to Hiller et al. [55], a
duced. Based on this concept, all material structures involved in negative Poisson's ratio can be achieved with appropriate voxel
4D printing can be generalized into three categories. arrangement in the digital material structure.
The three most important categories of 4D-printed structures for
3.2. Digital materials digital materials are uniform distribution (Fig. 44(a)), gradient distribu-
tion (Fig. 44(b)), and special patterns (Fig. 44(c)). One main category of
The digital concept was first introduced in the fields of communica- special patterns is the fiber and matrix structure. Each structure in
tion and computation. This digital concept can be similarly expanded Fig. 44 shows only one single layer, but they can be combined to yield
into material structures (Hiller et al. [54]). The element that enables bi-layer or multi-layer structures. In addition, the number of materials
Fig. 30. Helical structures with different degrees of spiral by 2D–to-3D twisting (Zhang et al. [18]).
56 F. Momeni et al. / Materials and Design 122 (2017) 42–79
has limited options for its minimum nozzle size. Zhou et al. inves-
tigated several new techniques for digital material production
using mask-image-projection-based stereolithography. Ge et al.
[60] provided an approach for printing multi-material shape
memory polymers (SMPs) with a high resolution (up to a few
microns). This approach is enabled by a high-resolution
projection microstereolithography (PμSL) additive manufacturing
system with an automated material exchange mechanism
(Fig. 47). In order to enable 4D printing for biomedical applica-
tions, multi-material additive manufacturing systems that can
print from aqueous mediums needs to be developed (Loh [61]). In
this regard, the direct-write (DW) printing technique
(Lewis [62]; Gratson & Lewis [63]; Lebel et al. [64]), which is
suitable for printing polymeric solutions (Guo et al. [65]), can be
engaged.
In some studies on 4D printing, shape-shifting behavior is en-
abled by certain targeted smart hinges embedded inside the
structure. In this case, only the hinges are made from smart
materials and the other parts are made from conventional
materials. A typical example is shown Fig. 48. 4D-printed
structures with hinges are typically used for folding, wherein the
structures can deform through the hinges. In other cases, the
structure itself has shape-shifting capability without dependence
on the hinges. In these hinge-less structures, the spatial
Fig. 31. An illustration of 2D-to-3D sinusoidal shape-shifting by surface curling (Tibbits arrangement of rigid and active materials is extremely crucial to
et al. [5]). precisely yield the desired shape-shifting behavior. (Tibbits et al.
[5]). In general, structures with hinges can achieve local shape-
shifting behavior, while the structures without hinges can have
both global and local shape-shifting behaviors.
can be more than two. One example of a gradient distribution material In summary, structures with hinges vs. structures without
structure is shown in Fig. 45 from Tibbits et al. [4]. In this example, the hinges in 4D printing are categorized in Fig. 49, along with the
concentration of active and rigid materials varies from the center to related studies.
the perimeter within one layer. The disc can yield various sinusoidal
shapes depending on the duration of immersion in water. 4. Materials
In summary, all multi-material structures that have been
studied in the 4D printing process are summarized in Fig. 46, The development of smart material should pursued in parallel
along with the related literature. with the development of printers. Currently, many 4D printing ap-
To model the material structure in various length and time plications are limited because of unsatisfactory material proper-
scales for digital materials, Myres et al. (1999) developed a soft- ties. For example, 4D printing can fabricate artificial muscles;
ware called the Digital Material [57]. Popescu et al. [56] proposed however, the mechanical properties of current materials are insuf-
a new manufacturing process for digital materials that is reversible ficient to yield the desired performance and functions of actual bi-
for disassembly and can reuse the building blocks of the structure. ological muscles (Loh [61]). Therefore, the development of
Huang et al. [58] demonstrated an approach for ultrafast printing of advanced smart materials with desirable properties that are also
shape-shifting materials. Zhou et al. [59] reported that there are compatible with printers is crucial to advance the application of
several limitations to the current 3D additive manufacturing 4D printing. Programmable materials, such as carbon fiber, wood,
process. For example, printers with inkjet nozzles can only print and textiles, have undeniable influence in many applications, in-
materials with certain viscosities and curing temperatures. The cluding aerospace, automotive, clothing, construction, healthcare
fused deposition modeling (FDM) process is relatively slow and and utility (Loh [61]).
Fig. 32. An illustration of 2D-to-3D hair-like shape-shifting by surface curling (Tibbits et al. [5]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 57
Tibbits et al. [5] applied rigid plastic and active expandable monomers that build linear chains during the polymerization pro-
polymer materials in their experiments. They combined these cess with some difunctional acrylate molecules. This kind of
two materials in various spatial arrangements, as shown in crosslink makes the polymer swell under water rather than being
Figs. 11–13. The expandable material was a hydrophilic UV- dissolved (Raviv et al. [12]).
curable polymer, which could expand up to 150% of its original vol- Ge et al. [6] printed glassy shape memory polymer fibers in an
ume under water. Raviv et al. [12] performed a more precise exper- elastomeric matrix. The elastomeric matrix has no shape memory
iment with two base materials similar to those used by Tibbits [4] effect, i.e., the degree of fixity is 0%. The glass transition tempera-
and Tibbits et al. [5]. One of the base materials was rigid plastic ture Tg of the matrix is approximately − 5 °C. The matrix is in a rub-
with an elastic modulus of 2 GPa and a Poisson's ratio of 0.4. The bery state between a modulus of approximately 0.7 MPa at 15 °C.
other base material was an expandable material with an elastic The fiber has a glass transition temperature T g approximately
modulus of 40 MPa in the dry condition and 5 MPa in water. Its 35 °C. Its modulus is 3.3 MPa at the lowest temperature of the
Poisson's ratio is 0.5. This expandable material has a composition thermomechanical cycle (T L = 15 ° C) and 13.3 MPa at the highest
of vinyl caprolactam (50%wt), polyethylene (30%wt), epoxy temperature (T H = 60 ° C).
diacrylate oligomer (18%wt), Irgacure 81 (1.9%wt), and wetting Ge et al. [14] used two base materials: Tangoblack as the elasto-
agent (0.1%w). It could expand up to 200% of its original volume meric matrix with Tg ~ − 5 ° C and Verowhite (Gray 60) as the fiber
under water. Its material structure contains hydrophilic acrylated with T g ~ 47 ° C. Tangoblack is in a rubbery state at room
Fig. 34. An illustration of 2D-to-3D alteration in which a complex, non-uniform curvature sculpture is achieved: (a) Schematic of a flat laminate. (b) The final desired shape after the
thermos-mechanical experiment (Ge et al. [6]).
58 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 35. An illustration of 2D-to-3D surface topographical changes where mountains and valleys are created on a flat surface (Tibbits et al. [5]).
temperature, which can be polymerized by an ink consisting of ure- Bakarich et al. [20] used Alginate/PNIPAAm ionic covalent en-
thane acrylate oligomer, Exo-1, 7, 7-trimethylbicyclo (2.2.1) hept- tanglement (ICE) gel with various concentrations of NIPAAm. In
2-yl acrylate, methacrylate oligomer, polyurethane resin, and their experiments, the thermo-responsive crosslinked network of
photo initiator. Verowhite is a rigid plastic at room temperature poly N-isopropylacrylamide (PNIPAAm) was utilized as the tough-
and can be polymerized by an ink consisting of isobornyl acrylate, ening agent and can also achieve reversible volume changes. The
acrylic monomer, urethane acrylate, epoxy acrylate, acrylic mono- Alginate/PNIPAAm ICE gel contained α-Keto glutaric acid
mer, acrylic oligomer, and photo initiator. Similarly, Bodaghi et al. photoinitiator, acrylamide, alginic acid sodium salt, calcium
[66] used TangoBlackPlus and VeroWhitePlus in a fiber and matrix chloride, ethylene glycol (as a rheology modifier), N-
structure. They also used Sup705a, which is a hydrophilic gel, as a isopropylacrylamide and N, N′-methylenebisacrylamide
sacrificial material for the manufacturing of complex geometries. crosslinker, and a commercial epoxy-based UV-curable adhesive
This auxiliary material can be removed using a compressed water (Emax 904 Gel-SC).
jet during the post-fabrication process, based on the preferential Kokkinis et al. [46] used two cross-linked polymers with different
interactions between the hydrophilic gel and water. swelling properties: a soft, highly swellable polymer and a solid poly-
Jamal et al. [13] (Fig. 24) used photopatterned poly (ethylene mer. The ink for these two polymers consists of PUA oligomers, which
glycol) (PEG)-based hydrogel bilayers. The two PEG bilayers con- act as the base components of all inks. Two of them yield hard polymers,
tain two molecular weights with different swelling ratios and are such as BR 302 and BR 571, and one of them yields a soft polymer, such
crosslinked with conventional photolithography. as BR 3641 AJ. The ink for the two polymers also consists of reactive dil-
Villar et al. [44] printed aqueous droplets in oil, which are connected uents to change the rheological and mechanical properties,
by lipid bilayers and create a cohesive material. photoinitiator (either Irgacure 907 with ultraviolet light or Irgacure
Mao et al. (2015) [15] used the same two base materials as Ge et al. 819 with a longer wavelength blue LED light), rheology modifier, and
[14] did (Tangoblack and Verowhite). They combined these two mate- the alumina platelets. They used various concentrations of components
rials at varying compositions, which is different from the conventional for different objectives.
fiber and matrix structure in Ge et al. [14]. In fact, they fabricated Mutlu et al. [47] (Fig. 40) printed a thermoplastic elastomer (TPE)
seven compositions with various combinations of these two materials material that has viscoelastic behavior and is soft enough for fabrication
for seven hinges, as shown in Fig. 29. of a compliant finger.
Fig. 36. An illustration of 2D-to-3D shape-shifting with surface topographical changes (Tibbits et al. [5]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 59
Fig. 37. 2D-to-3D shape-shifting by the combination of bending and twisting with complex flower morphologies (Gladman et al. [10]).
Gladman et al. [10] (Fig. 37) fabricated a composite hydrogel ink 50 MPa when the temperature was above its Tg. In addition, they as-
that mimics the structure of plant cell walls. It consists of a soft acryl- sumed the coefficient of thermal expansion of sheet paper to be
amide matrix reinforced with the cellulose fibrils that have a high negligible (Fig. 30).
stiffness. The composite is printed using a viscoelastic ink that con- Kuksenok et al. [19] fabricated a composite that consisted of a
tains an aqueous solution of N, N-dimethylacrylamide, Irgacure thermo-responsive polymer gel with poly (N-isopropylacrylamide)
2959 (BASF), nanoclay, glucose oxidase, glucose, and nanofibrillated (PNIPAAm), which is the host gel, and photo-responsive fibers function-
cellulose (NFC). Irgacure 2959 is the ultraviolet photoinitiator. The alized with spirobenzopyran (SP) chromophores. The thermo-
clay particles were used as a modifier for appropriate rheology and responsive gel has a lower critical solution temperature (LCSTand un-
viscoelasticity, which is necessary for desirable ink printing. Larger dergoes contraction at high temperatures. With no light, the
amounts of clay lead to higher crosslink densities and therefore spirobenzopyran chromophores are in open ring form or in an equiva-
lower swelling ratios. Glucose oxidase and glucose scavenge the sur- lent protonated merocyanine McH form. Under the blue light, they are
rounding oxygen, which consequently reduces oxygen inhibition reversibly converted to the closed ring form or the equivalent spiro SP
during the UV curing process. The shape-shifting behavior of the ma- form (Kuksenok et al. [19]).
terial with the composition described above is irreversible. To Wu et al. [17] (Figs. 17–21) used TangoBlack plus and Verowhite,
achieve reversible shape-shifting behavior in hot and cold water, which is similar to what Ge et al. [14] and Mao et al. [15] used. However,
the poly(N, N-dimethylacrylamide) needs to be replaced with their composite contains two types of fibers. They used DM8530 (fiber
thermo-responsive N-isopropylacrylamide. 1) with Tg ~ 57° C and DM9895 (fiber 2) with Tg ~ 38° C. These two fibers
Zhang et al. (2016) [18] printed polylactic acid (PLA) strips as the fi- have shape memory effects in the temperature range between ~ 20 ° C
bers on a fixed sheet paper. PLA strips have a glass transition tempera- and ~70° C. The matrix is TangoBlack with Tg ~ 2 ° C.
ture of Tg ~ 60 ° C and an elastic modulus of 3.5 GPa in its glass state Duigou et al. [69] fabricated hygromorphic biocomposite, which is
[67,68]. Zhang et al. [18] assumed the elastic modulus of PLA to be activated by moisture gradient. Their hygromorphic biocomposite
60 F. Momeni et al. / Materials and Design 122 (2017) 42–79
overall matrix, which is then reinforced with wood fibers. Natural fibers
such as wood, flax, hemp, etc. usually exhibit hydro-elastic behavior.
Biocomposites with natural fibers have a high-porosity microstructure
(approximately 20% volume percent). This structure enables swelling
when subjected to moisture gradient. On the other hand, it brings inher-
ent damage issues.
Ge et al. [60] created a photo-curable SMP network by combin-
ing two categories of materials: benzyl methacrylate (BMA) as the
linear chain builder (LCB) and poly (ethylene glycol)
dimethacrylate (PEGDMA), and bisphenol A ethoxylate
dimethacrylate (BPA) and di(ethylene glycol) dimethacrylate
(DEGDMA) as crosslinkers. This network is useful for a PμSL-
based additive manufacturing system.
Nadgorny et al. [70] tried to achieve high quality, printable pH-
responsive filaments for material extrusion 3D printing (ME3DP),
which is also known as fused deposition modeling (FDM). They
used poly (2-vinylpyridine) (P2VP) core material, and improved
its mechanical properties by adding 12%wt of acrylonitrile–
butadiene–styrene (ABS) as the reinforcement material.
Zarek et al. (2016) [30] produced a thermo-responsive shape
memory tracheal stent based on semi-crystalline methacrylated
polycaprolactone (PCL). PCL has great rheological and viscoelastic
properties and can be easily manufactured into implants and
Fig. 38. An illustration of various 2D-to-3D shape-shifting behavior (Ge et al. [6]). medical devices. PCL has also been used in many drug delivery
and tissue engineering applications (Woodruff & Hutmacher [71]).
consists of polylactic acid (PLA) and polyhydroxyalkanoate (PHA) as the Zarek et al. [30] additionally used 2,4,6-trimethylbenzoyl-diphenyl-
phosphineoxide (TPO) as the photoinitiator, in addition to vitamin E,
to avoid premature cross-linking.
Naficy et al. [45] used poly (N-isopropylacrylamide)
(poly(NIPAM)) as the smart thermo-responsive polymer, along
with a non-active poly (2-hydroxyethyl methacrylate)
(poly(HEMA)) as the bilayer. Long polymer chains of polyether-
based polyurethane (PEO-PU) are used to modify the rheology
and viscosity. α-ketoglutaric acid serves as the UV initiator. N, N′-
methylenebisacrylamide (BIS) is the crosslinking agent.
5. Discussions
From analyzing existing studies, there are two requirements for ma-
terials in the 4D printing process: printability and intelligence (Fig. 50).
If the materials cannot be printed, the 4D structure cannot be
manufactured. Many studies utilized a rheology modifier to provide a
suitable material viscosity for extrusion-based printing processes. Simi-
larly, the photo-initiator and the crosslinking and sacrificial agents are
several other aspects that need to be considered for proper material
printability. If the structure contains only non-active materials, it cannot
achieve any targeted changes over time as a response to the stimulus.
Schweiger et al. [72] studied multilayered anterior teeth and defined
“multi-material-3D-printing” as a 4D printing process. This is not the
4D printing process discussed in this paper because the structure does
not contain any smart material.
Some applications require dual-responsive materials. For example,
the shape-shifting behavior of a material can be triggered by both
water and heat. Triple and other multi-responsive materials have not
been considered in the 4D printing process so far and can be studied
in the future. Another issue is the degree to which the smart materials
can respond to stimulus. Some smart materials can sense stimulus but
only provide minimal actuation or respond after a very long time. The
responsivity of smart materials needs to be further studied as well.
mechanisms can be divided into various categories. Campbell et al. the external stimulus so that the structure can undergo shape-shifting
[26] explained two types of mechanisms in 4D printing: hydro- under water and so that it can return to its original shape after being
mechanics and thermo-mechanics. In this section, other mecha- dried. This mechanism is driven by the different swelling ratios between
nisms from the literature are organized and summarized, including the active and rigid materials. The expansion of the smart material gen-
unconstrained-hydro-mechanics, constrained-thermo-mechanics, erates a force that leads to the shape change. When the expandable ma-
unconstrained-thermo-mechanics, unconstrained-hydro-thermo- terial is appropriately arranged with the rigid material, complex shape-
mechanics, unconstrained-pH-mechanics, unconstrained-thermo- shifting behavior can be achieved. The magnitude and direction of the
photo-mechanics, osmosis-mechanics, and dissolution-mechanics. shape change depends on the spatial arrangements of the two materials
(Tibbits et al. [4]). The mechanism is illustrated in Fig. 51 and the entire
cycle is unconstrained. Tibbits [4], Tibbits et al. [5], Raviv et al. [12], and
6.1. Unconstrained-hydro-mechanics Jamal et al. [13] used this type of mechanism in their experiments.
It should be noted that this mechanism is naturally reversible, i.e.,
In this mechanism, a smart printed structure consists of an expand- the original shape can be recovered by drying the smart structure. How-
able hydrophilic active material and a rigid material. Water is utilized as ever, the shape-shifting behavior can also be forced to be irreversible by
using a special arrangement of rigid and active materials. For example,
Tibbits et al. [4] reported that the structure in Fig. 35 will not unfold
after being dried because of geometrical constraints.
6.2. Constrained-thermo-mechanics
Two levels of temperature and one external load are required in this
mechanism, as illustrated in Fig. 52. One temperature is higher than the
critical temperature of the smart material, such as its glass transition
temperature or crystal-melt transition temperature (TH N TC). The other
temperature should be lower than the critical temperature (TL b TC). In
this mechanism, the printed structure is heated to TH and the cycle starts
at TH in the following order. First, the original structure is stretched at TH
with a certain amount of strain depending on specific applications.
Then, under external stress, the structure is cooled to TL while the strain
Fig. 41. 3D-to-3D shape-shifting by expansion and contraction (Bakarich et al. [20]). remains unchanged. Next, the external stress is removed at TL, and the
62 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 42. Illustration of global and local shrinkage and bending for 3D-to-3D alterations by using two different stimuli (Kuksenok et al. [19]).
desired temporary shape is obtained at the end of this step. Finally, the medium in order to transfer the heat to the printed structure. It should
structure can be reheated to TH in a free stress condition to recover its be noticed that the temperature is still the only stimulus in this cycle to
original shape. In this mechanism, temperature is the external stimulus. trigger the shape-shifting behavior. Mao et al. [15] applied this mecha-
Ge et al. applied this mechanism in their experiments (Fig. 34 and nism in their sequential self-folding multi-shape memory structure
Fig. 28) [6] [14]. Wu et al. [17] also employed this mechanism (Fig. 29). Yu et al. [16] showed the feasibility of controlling the multi-
(Figs. 17–21). They used water with different temperatures as a shape-shifting sequence in a shape memory polymer based on this
b Gradient Distribution
c Special Patterns
Fig. 44. 4D-printed material structures (Digital Materials) (a) Uniform distribution with different concentrations, (b) Gradient distribution structure, and (c) Special patterns.
mechanism. Monzón et al. [73] experimentally studied the correlations original shape. In this mechanism, the external stimulus is
between the amount of recovery force and the mass/thickness/width of temperature. Zhang et al. [18] used this mechanism in their 4D
the printed shape memory polymers. These correlations can be effec- printing studies (Fig. 30). They demonstrated several kinds of
tively applied in the 4D printing process and should be the focus of fu- shape-shifting behaviors in smart, lightweight, and thin composite
ture studies on interaction mechanisms. structures, such as transformation from a planar sheet to flower-
like 3D structures, periodic 3D structures, and adaptive
metamaterials.
6.3. Unconstrained-thermo-mechanics
Unlike the previous mechanism, the external load is not includ- 6.4. Unconstrained-hydro-thermo-mechanics
ed in the cycle of this mechanism. Only the two temperatures are
required. One is higher than critical temperature of the active ma- This mechanism was shown by Bakarich et al. [20], wherein the
terial involved in the structure (TH N TC), and the other one is lower 4D printed structure undergoes two steps in the shape-shifting
than the critical temperature (T L b T C ). As shown in Fig. 53, the cycle. It first swells freely in cold water, and then deswells freely
printed structure is first heated to T H . The cycle then starts at T H in hot water, as shown in Fig. 54. This cycle can be repeated contin-
and proceeds in the following manner. First, the original structure uously. The swelling and deswelling processes are free and there
is cooled to T L , where the desired shape is achieved at the end of are no constraints from external load. In this mechanism, both
this step. Then, the structure can be heated to T H to recover its water and temperature are required as stimuli. Gladman et al.
Fig. 45. The simulation related to gradient distribution of material structure (red indicates the rigid material and purple indicates the active material) and the results of its immersion in
water over time (left to right) (Tibbits et al. [4]).
64 F. Momeni et al. / Materials and Design 122 (2017) 42–79
[10] also employed this mechanism for reversible shape-shifting printing, and enabled shape-shifting behavior with this
behavior (Fig. 37). mechanism.
6.6. Unconstrained-thermo-photo-mechanics
6.5. Unconstrained-pH-mechanics
This mechanism has been demonstrated by Kuksenok et al. [19].
This mechanism was demonstrated by Nadgorny et al. [70]. In In this mechanism, fibers were considered to be photo-responsive
this mechanism, a 4D-printed, pH-responsive hydrogel can linearly materials and the matrix was considered to be a thermo-
swell at a specific pH level and then shrink at another designed pH responsive material. They showed that the application of light,
level (Fig. 55). This cycle is mainly conducted in an aqueous envi- heat, or a combination of both stimuli could yield printed struc-
ronment and is therefore suitable for the shape-shifting of tures with various morphologies. In their study, the gel has a
hydrogels. Many studies on pH-responsive hydrogels have been lower critical solution temperature and shrinks at a high tempera-
conducted in available literature [74–78]. Nadgorny et al. [70] pro- ture. The spiro benzopyran (SP) chromophores functionalize elas-
vided pH-responsive hydrogels with a composition appropriate for tic fibers, which can be converted into a hydrophobic form when
Fig. 47. An illustration of a multi-material additive manufacturing system (Ge et al., 2016).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 65
et al. [19]. The original composite (Fig. 56(a)) shrinks like an accordion
(Fig. 56(b)) when heated freely (unconstrained) and bends like a cater-
pillar (Fig. 56(c)) when subjected to blue light. Fig. 56(d) shows the be-
havior of the composite when both heat and blue light are applied.
6.7. Osmosis-mechanics
Fig. 49. Structures with hinges vs. structures without hinges in 4D printing.
66 F. Momeni et al. / Materials and Design 122 (2017) 42–79
4D printing materials
behavior is reversible by re-immersing the structure in the solvent. In addition, some studies focused specifically on the mathematics of 4D
this mechanism, the appropriate solvent is the stimulus. printing are discussed.
As a summary of this section, all the shape-shifting mechanisms and Development of the 4D printing process relies on appropriate math-
stimuli are categorized in Fig. 58, along with their related studies. ematical modeling (Gladman et al. [10]. In fact, mathematics is neces-
sary in 4D printing for the following reasons:
7. Mathematics • It is needed to predict the shape evolution after printing over time.
• It provides the theoretical models needed to avoid collisions be-
In this section, forward and inverse problems are defined for the 4D tween components of the structure during the self-assembly op-
printing process, and one example is provided for each category. In eration.
• It reduces the number of trial-and-error experiments. Early ex-
periments in 4D printing involve many repetitions for a specific
structure to achieve the desired shape. For example, Tibbits
Unconstrained-Hydro-Mechanics et al. [5] printed and repeated a series of experiments to identify
a Shape 1 the appropriate material structure needed to reach the desired
shape. This is shown in Fig. 59. With mathematics and some the-
oretical models, the final shape can be predicted for a given ma-
terial structure, material properties, and stimulus properties.
Consequently, the number of test experiments can be reduced
dramatically.
Constrained-Thermo-Mechanics
Original Shape at TH >TC
a
Stretch Heat to
TH
TH
Desired Shape at TL < TC
d
Cool to
TL
Unload
TL
c
The Maxwell element contains a serial arrangement of one spring and include desired bending angle, twisting angle, length, volume, etc. The
one dashpot. A previous study conducted on this subject yielded a next is material structure, such as the volume fractions of fibers and
four-element model developed by Tobushi et al. [80], in which the the matrix, filament size, orientation, interfilament spacing, anisotropy.
creep phenomenon was incorporated through the fourth element, Equivalently the material structure can be described by the size, shape
εscalled the creep recovery strain (Fig. 62). and spatial arrangement of the voxels. From the perspective of the
An appropriate theoretical model for 4D printing consists of four printing process, the material structure depends on print paths and noz-
major components. The first is the final desired shape, which may zle sizes. Material properties makeup the third component, and they
Unconstrained-Thermo-Mechanics Unconstrained-Hydro-Thermo-Mechanics
a Original Shape at TH >TC a Shape 1
b Shape 2
Fig. 53. Schematic illustration of the unconstrained-thermo-mechanics mechanism in 4D Fig. 54. Schematic illustration of the unconstrained-hydro-thermo-mechanics mechanism
printing. in 4D printing.
68 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Unconstrained-pH-Mechanics
a Shape 1
pH < x Swelling Deswelling pH ≥ x Fig. 57. (a) Osmosis effect between two droplets, (b) Macroscopic deformation arising
from osmosis effect (Villar et al. [44]).
Fig. 55. Schematic illustration of the unconstrained-pH-mechanics mechanism in 4D The studies on 4D printing related to the inverse problem are
printing. application-oriented, meaning that they are focused on achieving
a desired functionality or shape. Studies related to the forward
include shear modulus, Young's modulus, and the interactive properties problem are mainly fundamental research aimed toward discover-
associated with the stimulus, such as glass transition temperature and ing concepts, theories, and relationships. These are illustrated in
swelling ratio. The final component is the stimulus properties, such as Fig. 63.
the temperature value and light intensity. The example with the spring-mass mathematical model shown
4D printing mathematics can be divided into two categories accord- above can be considered to be a forward problem. Gladman et al.
ing to Gladman et al. [10]: the forward problem and the inverse prob- [10] studied a bilayer structure and utilized the mathematical rela-
lem. The categories are defined below: tionships of coordinate system transformations to identify the
Unconstrained-Thermo-Photo-Mechanics
Thermo-Responsive
a Matrix with Photo-
Responsive Fibers
Heat & Light
Fig. 58. Shape-shifting mechanisms and stimuli which were used in 4D printing.
print paths (Fig. 67); this is an example of an inverse problem. As by the following equations:
shown in Fig. 64(a), the print path of the first layer aligns along
the e x direction (unit vector in x direction) and the print path of α ∥ −α ⊥ sin2 ðθÞ
H ¼ c1
second layer aligns along the direction of cos(θ)e x + sin (θ)e y , h c2 −c3 cosð2θÞ þ m4 cosð4θÞ
where θ is the angle between the two layers. The relationship be-
tween θ and the final desired shapes is then established, as given
ðα ∥ −α ⊥ Þ2 sin2 ðθÞ
K ¼ −c4
h
2 c5 −c6 cosð2θÞ þ m4 cosð4θÞ
Fig. 60. Mathematical modeling of the 4D-printed hinge that was introduced earlier by Tibbits et al. [5] with the spring-mass concept (Raviv et al. [12]).
with the concentric circles. Surfaces with negative Gaussian curva- folded into the same pattern through different methods and can also
ture or saddle-like surfaces can be produced with an orthogonal bi- be folded to different patterns. These studies serve as the mathematical
layer lattice structure. The combination of these two print paths tools for the forward and inverse problems in 4D printing processes.
can then yield a structure with varying Gaussian curvatures.
Kowk et al. [81] conducted design optimization of origami for 8. Applications
freeform surfaces in 4D printing. According to them [81], the 3D struc-
ture needs to be flattenable to achieve self-folding shape-shifting. A The droplets network illustrated by Villar et al. [44] can be used as
flattenable 3D structure can be flattened into a 2D form without tissue engineering substrates or as support for the functionality of fail-
stretching. However, many designed parts are not flattenable. To ad- ing tissues (Fig. 57).
dress this issue, they [81] provided a geometry optimization method, Kokkinis et al. [46] created smart key–lock connectors (Fig. 68)
which can modify a non-flattenable pattern into a flattenable pattern. based on the experiments shown in Fig. 39. The walls are flat before
Wang et al. [82] studied the mathematics of the single-loop foldable swelling. After swelling, convex and concave deformations occur in
8R (revolute joint) with multiple modes, which is also related to folding. the wall, which leads to a reduction in the internal size of the
They mathematically described that a self-folding structure can be cuboid and finally the interlocking can be achieved. The key-lock
Fig. 61. Standard linear solid (SLS) model to explain the mechanism of the shape memory effect in a shape memory polymer (Yu et al. [36]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 71
Fig. 64. Mathematical modeling can make a connection between (a) print paths quantified
by the angle θ between the two layers, and (b) final desired shape quantified by curvature
tensor κ, mean curvature H,and Gaussian curvature K (Gladman et al. [10]).
Fig. 62. Four-element modeling for the shape memory effect (Tobushi et al. [80]).
connector can be used as a physical connection between Using a different stimulus, Nadgorny et al. [70] printed a pH-
biological parts in the body, such as tendons and muscle. In responsive hydrogel valve that demonstrates dynamic and reversible
addition, it can be used as the selective pick-and-place system in swelling and shrinkage behavior that can regulate the flow rate under
soft robots. various pH levels (Fig. 70). Additionally, this structure can serve as an
Bakarich et al. [20] fabricated a skeletal muscle-like actuator with useful platform for pH-responsive membranes (Orlov et al. [83]) and
high response speed and reversibility based on the principle in Fig. 41. photonic gels (Kang et al. [84]).
They incorporated it into a smart valve (Fig. 69) and controlled the Adaptive metamaterials with tunable bandgaps enabled by 4D print-
water flow by opening the valve in cold water and closing the valve in ing were reported by Zhang et al. [18] (Fig. 71) and later by Bodaghi
hot water. This work shows one example of 4D printed soft mechanical et al. [66] (Fig. 72). In both studies, the lattice structure could be
actuators (Loh [61]). reconfigured in a reversible manner, where heat is the external
Material Properties
(such as longitudinal
Stimulus Properties
Young`s modulus,
4D printing (such as magnitude of
transverse Young`s
Mathematics temperature, light
modulus, swelling ratios,
intensity, etc.)
glass transition
tempretaure etc.)
Material Structure
or equivalently “print paths and
nozzle sizes” or equivalently “the
size, shape and spatial
arrangement of the voxels”
(such as volume fractions of fibers
and matrix, filament size, orientation,
and interfilament spacing, anisotropy,
etc.)
Inverse problem: Material structure or equivalently “print paths and nozzle sizes” is unknown
Fig. 63. 4D printing mathematics allows theoretical models to connect the final desired shape, material structure (or equivalently the size, shape and spatial arrangement of the voxels or
equivalently print paths and nozzle sizes), material properties, and stimulus properties.
72 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 65. Longitudinal and transverse swelling strains (α∥ and α⊥), (Gladman et al. [10]).
Fig. 66. Print paths and final shapes (a) positive Gaussian curvature (b) negative Gaussian curvature (c) and varying Gaussian curvature (Gladman et al. [10]).
stimulus. Meta-materials have tunable lattices and their mechanical interaction mechanism, while the one reported by Bodaghi et al. [66]
properties depend on their geometrical architectures instead of their (Fig. 72) works under the constrained-thermo-mechanics interaction
chemical compositions (Bodaghi et al. [66]. The adaptive meta- mechanism.
materials with tunable band gaps are useful for controlling the elastic Jiang and Wang [85] fabricated and analyzed elastomer
waves (Zhang et al. [18]). Adaptive meta-materials can adjust their stiff- metamaterials (Fig. 73) that can be used as reversible shape-shifting
ness based on changes in geometry while keeping their mass constant. connectors. The shape-shifting action is enabled by external mechanical
Therefore, they can be used as a switch between two different dynamic loadings. These materials can be applied through the 4D printing pro-
states (Bodaghi et al. [66]). The fabricated meta-material shown in cess to bridge components for flexible twisting or bending (Jiang &
Fig. 72 can be applied as planar self-expanding/shrinking actuators Wang [85]).
(Bodaghi et al. [66]). The lattice structure demonstrated by Zhang Kuksenok et al. [19] (Fig. 42) developed a 4D-printed structure that
et al. [18] (Fig. 71) works under the unconstrained-thermo-mechanics could mimic the contraction of an accordion. They also developed a
Fig. 67. Using the concepts of the mean curvature, H, and Gaussian curvature, K, generates the print paths by knowing the final desired morphologies (Gladman et al. [10]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 73
Fig. 68. Smart key–lock connectors that can be employed for various purposes (Kokkinis et al. [46]).
structure to mimic the lateral movement of a caterpillar. Their experi- (Fig. 24) proposed and fabricated self-bending, bio-origami structures
ments are suitable to manufacture smart components for new adaptive that can be used in the field of tissue engineering to analyze the behav-
devices, such as soft robots (Mao et al. [86]). ior of cells and self-bending vascularized tissue structures. In addition,
An active origami box, pyramid, and airplanes illustrated by Ge et al. their findings can be applied to the field of microfluidic networks
[14] show that the 4D printing process is capable of developing origami. (Jamal et al. [13]).
Origami has attracted a significant amount of interest lately in an effort The regeneration of Crambin protein structures through self-
to solve engineering issues related to the packing of large structures into folding of a 1D strand in water was demonstrated by Tibbits et al.
small volumes, for the sake of storage and transportation (Ge et al. [14]). [4] to serve as a potential solution for biomedical problems. A line-
In this regard, Kowk et al. [81] performed design optimization of origa- ar stretching joint, ring stretching joint, and folding joint have been
mi for freeform surfaces in 4D printing. In addition, Jamal et al. [13] demonstrated by Raviv et al. [12]. These joints can be used in other
Fig. 69. a) Computer-aided design model of the smart valve, b)Printing of the CAD model of the valve, c) The 4D printed valve in cold water, and d) The 4D-printed valve in hot water
(Bakarich et al. [20]).
74 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 70. pH-responsive flow regulating smart valve (Nadgorny et al. [70]).
Fig. 71. Thermo-responsive adaptive metamaterials with tunable bandgaps (Zhang et al. [18]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 75
Similarly, a 4D-printed thermo-responsive stent (Fig. 75) was creation of organs and tissues. The stimulation of stem cell differ-
demonstrated by Bodaghi et al. [66], in which the stent diameter entiation can provide shape-shifting behavior in a 4D printed scaf-
can be changed reversibly while the height is kept constant. fold (Irvine & Venkatraman [91]; Miao et al. [97]; Wei et al. [98]).
Both stents (Figs. 74 and 75) work under the constrained- Laser-based cell printing (Koch et al. [99]) can pave the way for
thermo-mechanics interaction mechanism. Wei et al. [91] incorpo- 4D bio-printing of cells.
rated Fe3O4 in a thermo-responsive PLA-based ink and 4D printed a Recently, a 4D nano-printing process was demonstrated by
smart stent which can be guided magnetically and actuated Carbonell and Braunschweig [100]. The bottom-up 4D nano-printing
remotely. process is expected to be an effective alternative to the conventional
Zarek et al. [30] 4D printed a thermo-responsive tracheal stent top-down techniques used for fabricating electronics, bioarrays, and
(Fig. 76), which solved two issues related with the current tracheal functional substrates. Future developments in this area depend
stents. First, it fits the arcade pattern and cartilaginous rings better on further developments in surface chemistry and instrumentation
and provides a stable state, which prevents migrations. Second, the [100].
shrunk form during insertion prevents the injurious insertion. The additive construction of buildings can make a significant contri-
After being inserted into the body, the tracheal stent can expand bution to the construction industry (Labonnote et al. [101]). One of the
to the permanent shape with a local increment in temperature primary technologies in this field is contour crafting (CC) (Khoshnevis
(Zarek et al. [30]. [102]). As Khoshnevis mentioned, the purpose of CC is the automated
Ge et al. developed a SMP-based thermo-responsive construction of one house or a colony of houses in a single run using lay-
multimaterial gripper (2016) [60], as shown in Fig. 77, which ered manufacturing techniques (Fig. 79). CC is one of the few feasible
can be applied in drug delivery systems (DDS) (Malachowski methods for construction on planets in outer space, such as Mars
et al. [92]. (Khoshnevis [102]). On the other hand, 4D printing can fabricate
Self-adaptive and multi-functional textiles are some of the smart elements as building blocks that can be self-assembled after
potential applications that can be improved by 4D printing printing. Therefore, as Labonnote et al. [101] indicated, 4D printing
(Truby & Lewis [93]). Self-adaptive smart textile structure can be and contour crafting can effectively complement each other to pave
adapted to a new size without tensile loading, in contrast to the the way for the human colonization of the Mars. One of the main
textiles made of elastic fibers (Fig. 78) (Hu et al. [94]). Multi-
functional smart textiles are capable of managing the moisture or
temperature of the body, monitoring wounds, providing skin
care, protecting against harsh climates, or adaptively changing
color of a dress (Hu et al. [94]).
4D bio-printing (An et al. [95]; Gao et al. [96]) of stem cells di-
rectly into scaffolds is a significant advancement toward the
Fig. 74. A 4D-printed, thermo-responsive stent which is able to reversibly change its Fig. 75. A 4D-printed, thermo-responsive stent which is able to reversibly change its
diameter and height (Ge et al. [60]). diameter (Bodaghi et al. [66]).
76 F. Momeni et al. / Materials and Design 122 (2017) 42–79
Fig. 77. 4D-printed shape memory gripper that can reversibly grab and release the objects by heat (Ge et al. [60]).
Fig. 78. Adaptability of textiles made of SMP vs. the textiles made of elastic fibers (Hu et al. [94]).
F. Momeni et al. / Materials and Design 122 (2017) 42–79 77
9. Conclusion
References [39] J. Ryu, M. D'Amato, X. Cui, K.N. Long, H.J. Qi, M.L. Dunn, Photo-origami—bending
and folding polymers with light, Appl. Phys. Lett. 100 (2012) 161908.
[1] S. Tibbits, The emergence of “4D printing”, TED Conference, 2013. [40] E.A. Peraza-Hernandez, D.J. Hartl, R.J. Malak Jr., D.C. Lagoudas, Origami-inspired ac-
[2] J. Choi, O.-C. Kwon, W. Jo, H.J. Lee, M.-W. Moon, 4D printing technology: a review, tive structures: a synthesis and review, Smart Mater. Struct. 23 (2014) 094001.
3D Printing and Additive Manufacturing, 2, 2015, pp. 159–167. [41] Y. Liu, J. Genzer, M.D. Dickey, “2D or not 2D”: shape-programming polymer sheets,
[3] M. Jacobsen, Clearing the way for pivotal 21st-century innovation, Giftedness and Prog. Polym. Sci. 52 (2016) 79–106.
Talent in the 21st Century, Springer 2016, pp. 163–179. [42] C. Lauff, T.W. Simpson, M. Frecker, Z. Ounaies, S. Ahmed, P. von Lockette, R.
[4] S. Tibbits, 4D printing: multi-material shape change, Archit. Des. 84 (2014) Strzelec, R. Sheridan, J.-M. Lien, Differentiating bending from folding in origami en-
116–121. gineering using active materialsASME 2014 International Design Engineering
[5] S. Tibbits, C. McKnelly, C. Olguin, D. Dikovsky, S. Hirsch, 4D printing and universal Technical Conferences and Computers and Information in Engineering Conference,
transformation, 2014 DOI. American Society of Mechanical Engineers, 2014 pp. V05BT08A040-
[6] Q. Ge, H.J. Qi, M.L. Dunn, Active materials by four-dimension printing, Appl. Phys. V005BT008A040.
Lett. 103 (2013) 131901. [43] Q. Wang, X. Zhao, Phase diagrams of instabilities in compressed film-substrate sys-
[7] E. Pei, 4D printing: dawn of an emerging technology cycle, Assem. Autom. 34 tems, J. Appl. Mech. 81 (2014) 051004.
(2014) 310–314. [44] G. Villar, A.D. Graham, H. Bayley, A tissue-like printed material, Science 340 (2013)
[8] Z.X. Khoo, J.E.M. Teoh, Y. Liu, C.K. Chua, S. Yang, J. An, K.F. Leong, W.Y. Yeong, 3D 48–52.
printing of smart materials: a review on recent progresses in 4D printing, Virtual [45] S. Naficy, R. Gately, R. Gorkin, H. Xin, G.M. Spinks, 4D printing of reversible shape
and Physical Prototyping 10 (2015) 103–122. morphing hydrogel structures, Macromol. Mater. Eng. (2016) DOI.
[9] M. Young, ALL3DP Company, 2016. [46] D. Kokkinis, M. Schaffner, A.R. Studart, Multimaterial magnetically assisted 3D
[10] A.S. Gladman, E.A. Matsumoto, R.G. Nuzzo, L. Mahadevan, J.A. Lewis, Biomimetic 4D printing of composite materials, Nat. Commun. 6 (2015).
printing, Nat. Mater. (2016) DOI. [47] R. Mutlu, G. Alici, M. in het Panhuis, G. Spinks, Effect of flexure hinge type on a 3D
[11] Y. Zhou, W.M. Huang, S.F. Kang, X.L. Wu, H.B. Lu, J. Fu, H. Cui, From 3D to 4D print- printed fully compliant prosthetic finger, 2015 IEEE International Conference on
ing: approaches and typical applications, J. Mech. Sci. Technol. 29 (2015) Advanced Intelligent Mechatronics (AIM)IEEE 2015, pp. 790–795.
4281–4288. [48] Y. Lee, H. Lee, T. Hwang, J.-G. Lee, M. Cho, Sequential folding using light-activated
[12] D. Raviv, W. Zhao, C. McKnelly, A. Papadopoulou, A. Kadambi, B. Shi, S. Hirsch, D. polystyrene sheet, Sci. Rep. 5 (2015).
Dikovsky, M. Zyracki, C. Olguin, Active printed materials for complex self- [49] A.B. Baker, D.F. Wass, R.S. Trask, 4D sequential actuation: combining
evolving deformations, Sci. Rep. 4 (2014). ionoprinting and redox chemistry in hydrogels, Smart Mater. Struct. 25
[13] M. Jamal, S.S. Kadam, R. Xiao, F. Jivan, T.M. Onn, R. Fernandes, T.D. Nguyen, D.H. (2016), 10LT02.
Gracias, Bio-origami hydrogel scaffolds composed of photocrosslinked PEG bilay- [50] K.E. Laflin, C.J. Morris, T. Muqeem, D.H. Gracias, Laser triggered sequential folding
ers, Adv. Healthc. Mater. 2 (2013) 1142–1150. of microstructures, Appl. Phys. Lett. 101 (2012) 131901.
[14] Q. Ge, C.K. Dunn, H.J. Qi, M.L. Dunn, Active origami by 4D printing, Smart Mater. [51] G. Stoychev, S. Turcaud, J.W. Dunlop, L. Ionov, Hierarchical multi-step folding of
Struct. 23 (2014) 094007. polymer bilayers, Adv. Funct. Mater. 23 (2013) 2295–2300.
[15] Y. Mao, K. Yu, M.S. Isakov, J. Wu, M.L. Dunn, H.J. Qi, Sequential self-folding struc- [52] S.M. Felton, M.T. Tolley, B. Shin, C.D. Onal, E.D. Demaine, D. Rus, R.J. Wood, Self-
tures by 3D printed digital shape memory polymers, Sci. Rep. 5 (2015). folding with shape memory composites, Soft Matter 9 (2013) 7688–7694.
[16] K. Yu, M.L. Dunn, H.J. Qi, Digital manufacture of shape changing components, Ex- [53] M. Vaezi, S. Chianrabutra, B. Mellor, S. Yang, Multiple material additive
treme Mechanics Letters 4 (2015) 9–17. manufacturing–part 1: a review: this review paper covers a decade of research
[17] J. Wu, C. Yuan, Z. Ding, M. Isakov, Y. Mao, T. Wang, M.L. Dunn, H.J. Qi, Multi-shape on multiple material additive manufacturing technologies which can produce
active composites by 3D printing of digital shape memory polymers, Sci. Rep. 6 complex geometry parts with different materials, Virtual and Physical Prototyping
(2016). 8 (2013) 19–50.
[18] Q. Zhang, K. Zhang, G. Hu, Smart three-dimensional lightweight structure triggered [54] J. Hiller, H. Lipson, Design and analysis of digital materials for physical 3D voxel
from a thin composite sheet via 3D printing technique, Sci. Rep. 6 (2016). printing, Rapid Prototyp. J. 15 (2009) 137–149.
[19] O. Kuksenok, A.C. Balazs, Stimuli-responsive behavior of composites integrating [55] J. Hiller, H. Lipson, Tunable digital material properties for 3D voxel printers, Rapid
thermo-responsive gels with photo-responsive fibers, Mater. Horiz. 3 (2016) Prototyp. J. 16 (2010) 241–247.
53–62. [56] G.A. Popescu, T. Mahale, N. Gershenfeld, Digital materials for digital printing, NIP &
[20] S.E. Bakarich, R. Gorkin, G.M. Spinks, 4D printing with mechanically robust, ther- Digital Fabrication ConferenceSociety for Imaging Science and Technology 2006,
mally actuating hydrogels, Macromol. Rapid Commun. 36 (2015) 1211–1217. pp. 58–61.
[21] D. Roy, J.N. Cambre, B.S. Sumerlin, Future perspectives and recent advances in [57] C. Myers, S. Arwade, E. Iesulauro, P. Wawrzynek, M. Grigoriu, A. Ingraffea, P.
stimuli-responsive materials, Prog. Polym. Sci. 35 (2010) 278–301. Dawson, M. Miller, J. Sethna, Digital material: a framework for multiscale
[22] M.A.C. Stuart, W.T. Huck, J. Genzer, M. Müller, C. Ober, M. Stamm, G.B. Sukhorukov, modeling of defects in solids, MRS Proceedings, Cambridge Univ Press 1998,
I. Szleifer, V.V. Tsukruk, M. Urban, Emerging applications of stimuli-responsive p. 509.
polymer materials, Nat. Mater. 9 (2010) 101–113. [58] L. Huang, R. Jiang, J. Wu, J. Song, H. Bai, B. Li, Q. Zhao, T. Xie, Ultrafast digital printing
toward 4d shape changing materials, Adv. Mater. (2016) DOI.
[23] L. Sun, W.M. Huang, Z. Ding, Y. Zhao, C.C. Wang, H. Purnawali, C. Tang, Stimulus-
[59] C. Zhou, Y. Chen, Z. Yang, B. Khoshnevis, Digital material fabrication using mask-
responsive shape memory materials: a review, Mater. Des. 33 (2012) 577–640.
image-projection-based stereolithography, Rapid Prototyp. J. 19 (2013) 153–165.
[24] H. Meng, G. Li, A review of stimuli-responsive shape memory polymer composites,
[60] Q. Ge, A.H. Sakhaei, H. Lee, C.K. Dunn, N.X. Fang, M.L. Dunn, Multimaterial 4D print-
Polymer 54 (2013) 2199–2221.
ing with tailorable shape memory polymers, Sci. Rep. 6 (2016).
[25] G.M. Whitesides, B. Grzybowski, Self-assembly at all scales, Science 295 (2002)
[61] X.J. Loh, Four-dimensional (4D) printing in consumer applications, Polymers for
2418–2421.
Personal Care Products and Cosmetics 20 (2016) 108–116.
[26] T.A. Campbell, S. Tibbits, B. Garrett, The programmable world, Sci. Am. 311 (2014) [62] J.A. Lewis, Direct ink writing of 3D functional materials, Adv. Funct. Mater. 16
60–65. (2006) 2193–2204.
[27] V.-M. Čolić-Damjanovic, I. Gadjanski, Potentials of fablabs for biomimetic architec- [63] G.M. Gratson, J.A. Lewis, Phase behavior and rheological properties of polyelectro-
tural research, Multidisciplinary Engineering Design Optimization (MEDO), Inter- lyte inks for direct-write assembly, Langmuir 21 (2005) 457–464.
national Conference, IEEE 2016, pp. 1–6. [64] L.L. Lebel, B. Aissa, M.A.E. Khakani, D. Therriault, Ultraviolet-assisted direct-write
[28] A. Khademhosseini, R. Langer, A decade of progress in tissue engineering, Nat. fabrication of carbon nanotube/polymer nanocomposite microcoils, Adv. Mater.
Protoc. 11 (2016) 1775–1781. 22 (2010) 592–596.
[29] J.P. Jung, D.B. Bhuiyan, B.M. Ogle, Solid organ fabrication: comparison of [65] S.Z. Guo, F. Gosselin, N. Guerin, A.M. Lanouette, M.C. Heuzey, D. Therriault, Solvent-
decellularization to 3D bioprinting, Biomaterials research 20 (2016) 27. cast three-dimensional printing of multifunctional microsystems, Small 9 (2013)
[30] M. Zarek, N. Mansour, S. Shapira, D. Cohn, 4D printing of shape memory-based per- 4118–4122.
sonalized endoluminal medical devices, Macromol. Rapid Commun. (2016) DOI. [66] M. Bodaghi, A. Damanpack, W. Liao, Self-expanding/shrinking structures by 4D
[31] D.L. Taylor, Self-healing hydrogels, Adv. Mater. (2016) DOI. printing, Smart Mater. Struct. 25 (2016) 105034.
[32] J. Zhou, S.S. Sheiko, Reversible shape-shifting in polymeric materials, J. Polym. Sci. B [67] R.E. Drumright, P.R. Gruber, D.E. Henton, Polylactic acid technology, Adv. Mater. 12
Polym. Phys. 54 (2016) 1365–1380. (2000) 1841–1846.
[33] L. Sun, W.M. Huang, Mechanisms of the multi-shape memory effect and [68] F. Cock, A. Cuadri, M. García-Morales, P. Partal, Thermal, rheological and micro-
temperature memory effect in shape memory polymers, Soft Matter 6 (2010) structural characterisation of commercial biodegradable polyesters, Polym. Test.
4403–4406. 32 (2013) 716–723.
[34] M.D. Hager, S. Bode, C. Weber, U.S. Schubert, Shape memory polymers: past, pres- [69] A. Le Duigou, M. Castro, R. Bevan, N. Martin, 3D printing of wood fibre
ent and future developments, Prog. Polym. Sci. 49 (2015) 3–33. biocomposites: from mechanical to actuation functionality, Mater. Des. 96
[35] T. Xie, Tunable polymer multi-shape memory effect, Nature 464 (2010) 267–270. (2016) 106–114.
[36] K. Yu, T. Xie, J. Leng, Y. Ding, H.J. Qi, Mechanisms of multi-shape memory effects [70] M. Nadgorny, Z. Xiao, C. Chen, L.A. Connal, Three-dimensional printing of pH-
and associated energy release in shape memory polymers, Soft Matter 8 (2012) responsive and functional polymers on an affordable desktop printer, ACS Appl.
5687–5695. Mater. Interfaces 8 (2016) 28946–28954.
[37] H.L. Thérien-Aubin, Z.L. Wu, Z. Nie, E. Kumacheva, Multiple shape transformations [71] M.A. Woodruff, D.W. Hutmacher, The return of a forgotten
of composite hydrogel sheets, J. Am. Chem. Soc. 135 (2013) 4834–4839. polymer—polycaprolactone in the 21st century, Prog. Polym. Sci. 35 (2010)
[38] H. Li, X. Gao, Y. Luo, Multi-shape memory polymers achieved by the spatio- 1217–1256.
assembly of 3D printable thermoplastic building blocks, Soft Matter 12 (2016) [72] J. Schweiger, F. Beuer, M. Stimmelmayr, D. Edelhoff, P. Magne, J. Güth, Histo-
3226–3233. anatomic 3D printing of dental structures, Br. Dent. J. 221 (2016) 555–560.
F. Momeni et al. / Materials and Design 122 (2017) 42–79 79
[73] M. Monzón, R. Paz, E. Pei, F. Ortega, L. Suárez, Z. Ortega, M. Alemán, T. Plucinski, N. [89] C.M. Yakacki, R. Shandas, C. Lanning, B. Rech, A. Eckstein, K. Gall, Unconstrained re-
Clow, 4D printing: processability and measurement of recovery force in shape covery characterization of shape-memory polymer networks for cardiovascular
memory polymers, Int. J. Adv. Manuf. Technol. (2016) 1–10 DOI. applications, Biomaterials 28 (2007) 2255–2263.
[74] A.S. Puranik, L.P. Pao, V.M. White, N.A. Peppas, Synthesis and characterization of [90] V. Srivastava, S.A. Chester, L. Anand, Thermally actuated shape-memory polymers:
pH-responsive nanoscale hydrogels for oral delivery of hydrophobic therapeutics, experiments, theory, and numerical simulations, J. Mech. Phys. Solids 58 (2010)
Eur. J. Pharm. Biopharm. 108 (2016) 196–213. 1100–1124.
[75] B. Frohm, J. DeNizio, D. Lee, L. Gentile, U. Olsson, J. Malm, K. Åkerfeldt, S. Linse, A [91] H. Wei, Q. Zhang, Y. Yao, L. Liu, Y. Liu, J. Leng, Direct-write fabrication of 4D active
peptide from human semenogelin I self-assembles into a pH-responsive hydrogel, shape-changing structures based on a shape memory polymer and its nanocom-
Soft Matter 11 (2015) 414–421. posite, ACS Appl. Mater. Interfaces (2016) DOI.
[76] M. Krogsgaard, M.A. Behrens, J.S. Pedersen, H. Birkedal, Self-healing mussel- [92] K. Malachowski, J. Breger, H.R. Kwag, M.O. Wang, J.P. Fisher, F.M. Selaru, D.H.
inspired multi-pH-responsive hydrogels, Biomacromolecules 14 (2013) 297–301. Gracias, Stimuli-responsive theragrippers for chemomechanical controlled release,
[77] J.P. Best, M.P. Neubauer, S. Javed, H.H. Dam, A. Fery, F. Caruso, Mechanics of pH- Angew. Chem. 126 (2014) 8183–8187.
responsive hydrogel capsules, Langmuir 29 (2013) 9814–9823. [93] R.L. Truby, J.A. Lewis, Printing soft matter in three dimensions, Nature 540 (2016)
[78] X.-L. Gong, Y.-Y. Xiao, M. Pan, Y. Kang, B.-J. Li, S. Zhang, pH- and thermal-responsive 371–378.
multishape memory hydrogel, ACS Appl. Mater. Interfaces 8 (2016) 27432–27437. [94] J. Hu, H. Meng, G. Li, S.I. Ibekwe, A review of stimuli-responsive polymers for smart
[79] H.J. Qi, M.L. Dunn, Thermomechanical behavior and modeling approaches, Shape- textile applications, Smart Mater. Struct. 21 (2012) 053001.
Memory Polymers and Multifunctional Composites 2010, p. 65 DOI. [95] J. An, J.E.M. Teoh, R. Suntornnond, C.K. Chua, Design and 3D printing of scaffolds
[80] H. Tobushi, T. Hashimoto, S. Hayashi, E. Yamada, Thermomechanical constitutive and tissues, Engineering 1 (2015) 261–268.
modeling in shape memory polymer of polyurethane series, J. Intell. Mater. Syst. [96] B. Gao, Q. Yang, X. Zhao, G. Jin, Y. Ma, F. Xu, 4D bioprinting for biomedical applica-
Struct. 8 (1997) 711–718. tions, Trends Biotechnol. (2016) DOI.
[81] T.-H. Kwok, C.C. Wang, D. Deng, Y. Zhang, Y. Chen, Four-dimensional printing for [97] S.A. Irvine, S.S. Venkatraman, Bioprinting and differentiation of stem cells, Mole-
freeform surfaces: design optimization of origami and kirigami structures, J. cules 21 (2016) 1188.
Mech. Des. 137 (2015) 111413. [98] S. Miao, W. Zhu, N.J. Castro, M. Nowicki, X. Zhou, H. Cui, J.P. Fisher, L.G. Zhang, 4D
[82] J. Wang, G. Bai, X. Kong, Single-loop foldable 8R mechanisms with multiple modes, printing smart biomedical scaffolds with novel soybean oil epoxidized acrylate, Sci.
New Trends in Mechanism and Machine Science, Springer 2017, pp. 503–510. Rep. 6 (2016) 27226.
[83] M. Orlov, I. Tokarev, A. Scholl, A. Doran, S. Minko, pH-responsive thin film mem- [99] L. Koch, A. Deiwick, B. Chichkov, Laser-Based Cell Printing, 3D Printing and
branes from poly (2-vinylpyridine): water vapor-induced formation of a micropo- Biofabrication, Springer, 2016 1–27.
rous structure, Macromolecules 40 (2007) 2086–2091. [100] C. Carbonell, A.B. Braunschweig, Toward 4D nanoprinting with tip-induced organic
[84] Y. Kang, J.J. Walish, T. Gorishnyy, E.L. Thomas, Broad-wavelength-range chemically surface reactions, Acc. Chem. Res. (2016) DOI.
tunable block-copolymer photonic gels, Nat. Mater. 6 (2007) 957–960. [101] N. Labonnote, A. Rønnquist, B. Manum, P. Rüther, Additive construction: state-of-
[85] Y. Jiang, Q. Wang, Highly-stretchable 3D-architected mechanical metamaterials, the-art, challenges and opportunities, Autom. Constr. 72 (2016) 347–366.
Sci. Rep. 6 (2016). [102] B. Khoshnevis, Automated construction by contour crafting—related robotics and
[86] S. Mao, E. Dong, H. Jin, M. Xu, K. Low, Locomotion and gait analysis of multi-limb information technologies, Autom. Constr. 13 (2004) 5–19.
soft robots driven by smart actuators, 2016 IEEE/RSJ International Conference on [103] F. Momeni, X. Jin, J.A. Ni, Microscopic Approach for Generic Degradation Modeling,
Intelligent Robots and Systems (IROS)IEEE 2016, pp. 2438–2443. International journal of prognostics and health management 7 (2016) Spec Iss on
[87] L. Anand, N.M. Ames, V. Srivastava, S.A. Chester, A thermo-mechanically coupled Smart Manufacturing PHM).
theory for large deformations of amorphous polymers. Part I: formulation, Int. J.
Plast. 25 (2009) 1474–1494.
[88] C.M. Yakacki, R. Shandas, D. Safranski, A.M. Ortega, K. Sassaman, K. Gall, Strong, tai-
lored, biocompatible shape-memory polymer networks, Adv. Funct. Mater. 18
(2008) 2428–2435.