Metallurgy: No more tears for metal 3D printing

Full text

IAIN TODD
A
dditive manufacturing of metals, also
known as metal 3D printing, was not
long ago seen only as a means of mak-
ing prototypes of mechanical components for
industry. But it is now considered to be a poten-
tially transformative manufacturing process for
many sectors. This change has been driven by
diverse factors1, such as the complex geometries
that can be achieved, the relatively low number
of parts required for components made in this
way, and the fact that manufacturing timescales
are much shorter than those of conventional
manufacturing methods. However, the lim-
ited range of industrially useful alloys that can
be processed in the additive manufacturing
of metals is a barrier to its wide adoption. On
page365, Martin etal.
2
report an approach that
expands the range of metallic materials
that can be used, taking lessons from a much
older manufacturing process — casting.
Metal additive manufacturing (AM) often
involves the deposition of layers of an alloy
feedstock in the form of powders or wires,
which are melted together by a rapidly mov-
ing heat source to form a solid mass. Successive
layers are built up to produce a 3D component.
The rate of solidification is often an order of
magnitude higher than that seen during con-
ventional casting techniques, and the process
of building up layers causes non-uniform cool-
ing, which, in turn, leads to large temperature
gradients or thermal stresses in the alloy.
The picture is further complicated by
the solidification processes that occur in
alloys used for engineering applications,
such as high-strength aluminium and nickel
super alloys. These materials tend to have
relatively wide temperature ranges over which
liquid and solid phases can coexist, and often
solidify as either columnar grains or dendrites
(multi-branched crystal formations) under
the conditions imposed by AM. As solidifica-
tion proceeds, the liquid content of the alloy
decreases, and the residual liquid becomes
confined to channels between the cells or
dendrites, where it forms a film. Localized
contraction of the solid can then cause cavi-
ties to form in the liquid film; if these cavities
propagate, they generate cracks known as hot
tears3. The resulting material can therefore
contain both a columnar microstructure and
numerous cracks — neither of which would be
acceptable for engineering applications.
Martin et al. now suggest that a possible
way of suppressing hot tearing in AM would
be to change the dominant solidification mode
from directional columnar growth to non-
directional growth that produces ‘equiaxed’
grains— which have roughly equal width,
length and height. However, once again, the
local solidification conditions that occur
during AM work against this: steep thermal
gradients are generated, which suppress the
Figure 1 | Improving the properties of alloys produced using additive
manufacturing. a,When many industrially useful alloys are processed using
additive manufacturing (3D printing), the resulting materials contain large
cracks and columnar grains (as shown here for the aluminium alloy Al7075).
This makes them unsuitable for engineering applications. b,Martin etal.2
attached nanoparticles to the surfaces of the granules in Al7075 powder. No
cracking was observed when this powder was used as a feedstock for additive
manufacturing, and the microstructure consisted of equiaxed particles (which
have roughly equal widths, lengths and heights). Scale bars, 20micrometres.
(Images from ref. 2.)
a b
METALLURGY
No more tears for
metal 3D printing
3D printing could revolutionize manufacturing processes involving metals, but
few industrially useful alloys are compatible with the technique. A method has
been developed that might open up the 3D printing of all metals. S L .365
Stefan P. Tarnawsky and Mervin C. Yoder
are in the Department of Pediatrics, School of
Medicine, Indiana University, Indianapolis,
Indiana 46202, USA.
e-mail: myoder@iu.edu
1. Palis, J. FEBS Lett. 590, 3965–3974 (2016).
2. Schulz, C. et al. Science 336, 86–90 (2012).
3. Gomez Perdiguero, E. et al. Nature 518, 547–551
(2015).
4. Epelman, S. et al. Immunity 40, 91–104
(2014).
5. Chan, R. J. & Yoder, M. C. Blood 121, 4815–4817
(2013).
6. Li, Z. et al. Nature Genet. 37, 613–619 (2005).
7. Mass, E. et al. Nature 549, 389–393 (2017).
8. Emile, J.-F. et al. Blood 127, 2672–2681
(2016).
9. Tiacci, E. et al. N. Engl. J. Med. 364, 2305–2315
(2011).
10. Trifa, A. P. et al. Leuk. Lymphoma 53, 2496–2497
(2012).
11. Idbaih, A. et al. Neurology 83, 1478–1480
(2014).
12. Hyman, D. M. et al. N. Engl. J. Med. 373, 726–736
(2015).
This article was published online on 30 August 2017.
342 | NATURE | VOL 549 | 21 SEPTEMBER 2017
NEWS & VIEWS
RESEARCH

repeated formation (nucleation) of the tiny
crystal seeds that would allow equiaxed grains
to form. The challenge is to find a way of
allowing this nucleation process to occur
under AM conditions.
Previous attempts to solve this problem have
generally focused on changing the process-
ing parameters of AM, such as the speed, the
power of the laser or electron beam used to
heat the alloy feedstock, or the pattern in which
the printer moves to build up an object, to dis-
rupt the conditions that promote columnar
growth (see refs4 and 5, for example). Unfor-
tunately, it has proved extremely difficult to
exert sufficient control over such process vari-
ables to promote nucleation and hence develop
the desired microstructure.
Luckily, a potential solution to this problem
can be found from casting — in which addi-
tives called inoculants are commonly mixed
into a liquid metal to ‘seed’ nuclei on which
new crystals can grow, even in the presence of
steep thermal gradients and high solidification
velocities. The first reported instance of an
addition being made to deliberately manipulate
microstructure was in 1906, when ferro-
silicon was added to a ladle of cast iron6. Since
then, developments in casting have enabled
the production of strong materials that lack
holes or tears, and which contain equiaxed
microstructures, using high-performance
engineering alloys7.
Inoculants are normally added to an alloy in
its molten state. This poses a problem in AM,
because the melt pool is only tens of micro-
metres long, and exists at any given point for
just tens of microseconds8. Martin and col-
leagues’ solution allows a precise quantity of
inoculant to be delivered to such melt pools
on this timescale.
The authors demonstrate the potential of
their approach using two aluminium alloys
that are well characterized and widely used:
Al7075, a wrought (mechanically worked)
material used in aerospace applications and
which is not well suited to melt processing, and
Al6061, a high-strength alloy used for casting.
Crucially, both are difficult to process by AM.
Martin etal. first modified the surface of the
feedstock alloy powders by decorating them
with nanoparticulate inoculants, which were
tailored to the composition and crystal lattices
of each alloy. These ‘functionalized’ powders
were then used in a standard AM machine,
following manufacturer-recommended
processing conditions. For comparison, the
authors also tested alloys that had not been
surface-modified using the same processing
conditions.
The difference in the microstructures
obtained for the two types of sample was
dramatic. The samples made using unmodi-
fied alloys contained large columnar grains
and a high number density of cracks, as
might be expected (Fig.1a). By contrast,
the functionalized powders produced fine,
equiaxed microstructures that were free of
cracks (Fig.1b). The mechanical properties
of the inoculated Al7075 were also markedly
better than when it was made from the
unmodified powder, and approached those of
the same alloy in the wrought condition.
There is still some way to go, however,
before this becomes the ‘go-to’ manufactur-
ing technology for aerospace applications.
In this context, the resistance of materials to
fatigue—weakening caused by repeatedly
applied loads — is of equal, if not greater,
importance to their strength9. More work is
needed to better understand and control the
fatigue resistance of materials produced using
AM. Another barrier to uptake by industry
is the slow speed of current metal AM pro-
cesses. Methods are emerging that deliver
a step change in the speed of 3D printing of
polymers10, and the race is on to achieve the
same for metals, but this presents a major
technological challenge.
In the meantime, however, Martin and
colleagues have identified an approach that
allows alloys to be made more suitable for
AM. Although they used aluminium alloys,
they note that the method could be readily
extended to other industrially useful alloy
classes, such as non-weldable nickel alloys,
superalloys and intermetallics. This might
take some time to achieve, however, because
inoculants for these materials remain elusive.
But if inoculants can be found to function-
alize the surfaces of powders of these alloys,
then we really would be moving towards the
3D printing of any metal. ■
Iain Todd is in the Department of Materials
Science and Engineering, University of
Sheffield, Sheffield S1 3JD, UK.
e-mail: iain.todd@sheffield.ac.uk
1. Pollock, T. M. Nature Mater. 15, 809–815 (2016).
2. Martin, J. H. et al. Nature 549, 365–369 (2017).
3. Rappaz, M., Drezet, J. & Gremaud, M. Metall. Mater.
Trans. A 30, 449–455 (1999).
4. Raghavan, N. Acta Mater. 112, 303–314 (2016).
5. Herzog, D., Seyda, V., Wycisk, E. & Emmelmann, C.
Acta Mater. 117, 371–392 (2016).
6. Turner, T. The Metallurgy of Cast Iron (Griffin, 1920).
7. McCartney, D. G. Int. Mater. Rev. 34, 247–260
(1989).
8. Khairallah, S. A., Anderson, A. T., Rubenchik, A. &
King, W. E. Acta Mater. 108, 36–45 (2016).
9. Tammas-Williams, S., Withers, P. J., Todd, I. &
Prangnell, P. B. Sci. Rep. 7, 7308 (2017).
10.Tumbleston, J. R. et al. Science 347, 1349–1352
(2015).
KATARZYNA M. KEDZIORA
& JEREMY E. PURVIS
A
fundamental principle of cell theory
is that all cells arise from pre-existing
ones. Every cell, except sperm and
eggs, inherits an essentially identical copy of
its mother’s genome, which it then passes on
to two daughters when it divides. But it can
also inherit a variety of other ‘memories’ from
its mother cell, in the form of proteins, RNA
and other biochemical keepsakes. Identifying
these molecular memories and understanding
how they influence cell behaviour has been a
long-standing puzzle. On page404, Yang etal.
1
tackle the question of how molecular memo-
ries acquired from the previous generation of
cells influence whether daughter cells prolifer-
ate or enter a reversible resting state known as
quiescence.
Proliferation drives both the development
of an organism and the maintenance of its
tissues. In response to growth signals, prolifer-
ating cells proceed through an initial phase of
growth (known as G1), after which they begin
DNA synthesis (Sphase). Following a second
growth phase (G2), the mother cell divides
its contents into two daughter cells through
a process called mitosis. Not all cells proceed
swiftly through these phases, however. Instead,
some temporarily withdraw from the cell cycle
before Sphase, entering quiescence2.
How does a cell ‘de cide’ b etween proliferation
and quiescence? A study in the 1970s suggested
that this decision is made during G1, before a
cell commits to DNA synthesis3. According
to this model, each cell is a clean slate, able to
make an independent decision on the basis of
the signalling molecules to which it is exposed.
However, this idea was challenged in 2013 by
the discovery that some cells are born pre-
disposed to rapidly enter Sphase4. For these
cells, the decision is influenced by the expe-
rience of the mother during its G2. Precisely
CELL BIOLOGY
The persistence
of memory
Live imaging reveals that whether or not a daughter cell proliferates is influenced
by two molecular factors inherited from its mother, providing insight into how
the behaviour of a newly born cell can be predetermined. S L .404
21 SEPTEMBER 2017 | VOL 549 | NATURE | 343
NEWS & VIEWS RESEARCH