stresses and strains during the course of a load cycle are known.
Calculations are used whenever direct measurements are not
possible. To do this, a geometric model of the test specimen or
a section of the coated specimen is created and converted into
a numerical simulation model. This computer model is used to
simulate the effects of thermo-mechanical loads.
Provided that the properties of the different materials of
the coating system are known across the entire temperature
range, the local stresses and strains in the material can be calcu-
lated. But a few unknown variables still remain in this calculation;
a large number of material properties are not available, specifi-
cally for these thin layers with a thickness of just a few microns.
Further, as the material is exposed to high temperatures and a
rising number of load cycles, the properties of the layers change
In principle, this works as follows: atoms in crystalline
systems like those found in the coating systems are arranged
in a three-dimensional regular lattice structure. Given that
the lattice spacing is characteristic to individual materials, the
diffraction patterns emerging from the interactions between
X-rays and the crystalline planes are characteristic as well. The
spacing between the crystalline planes change as thermal or
mechanical stress deforms the material, and these changes
correspond to the altered diffraction patterns. By reverse logic,
the diffraction patterns tell us the extent to which the material
is strained. So in terms of our layer system it means that we
take the strains relating to different thermo-mechanical loading
to identify the mechanical properties of the individual layers.
Approaching the finish line
So, finally, the moment arrived in November 2012. The
teams from Orlando and Cologne packed up the furnace com-
ponents and the specimens coated in Cologne and set off for
the Advanced Photon Source, APS, in Argonne. The APS is a
particle accelerator, used to speed up electrons to almost the
speed of light. Electro-magnets then compel them to follow a
circular path in a ring-shaped, stainless steel tube – the storage
ring – causing them to emit high-energy X-rays. The storage
ring at APS has a circumference of 1104 metres; 35 laboratories,
referred to as ‘beam lines’, are housed in the experiment hall,
arranged around the ring. These beam lines divert some of the
X-rays produced and use them for various test purposes.
Scientists at APS run the laboratories, cooperating with
external users from all over the world to build experiments and
test facilities and offering scientific support during the experi-
ments. Use of the APS as well as the scientific support provided
are almost priceless, but they cost us a thoroughly reasoned
research application. The approved measurement times are
therefore correspondingly valuable, and so initially we assem-
bled the test setup in shifts operating around the clock and then
proceeded to conduct our actual tests with as little interruption
as possible. The yield at the end of four days of measurements
was one hard disk containing a terabyte of raw data. Analysing
it all will take a few months, but we can already confirm that
the measurements were a success.
For instance, the thin aluminium oxide layer that forms
between the ceramic thermal barrier coating and the metallic
oxidation protection layer is visible in astonishing detail. The
diffraction patterns also reveal that this layer is under mecha
nical strain and – something that surprised us – the crystallites
that the layer consists of displays a clearly perceivable, preferred
orientation. We have already presented the initial results at the
37th International Conference of the American Ceramic Society
in Daytona Beach – the exact place where the whole coopera-
tion began two years earlier.
•
About the authors:
Marion Bartsch is Head of the Experimental and Numerical
Methods Department at the DLR Institute of Materials Research
in Cologne. She is also Professor for Materials used in Aero-
space at the Ruhr University in Bochum.
Janine Wischek is Head of the Mechanical Testing of Materials
Group.
The international team:
Janine Wischek, Carla Meid, Marion Bartsch – Institute for Ma-
terials Research in Cologne; Anette Karlsson (now Cleveland
State University); Kevin Knipe, Albert Manero, Sanna Siddiqui,
Prof Seetha Raghavan – University of Central Florida in Orlando;
Jonatan Almer, John Okazinski – Advanced Photon Source in the
Argonne National Laboratory.
with time. Chasing down the unknowns in the system, we
cooperated with a team under Anette Karlsson from the Univer-
sity of Delaware to initially estimate the unknown material data
and then evaluate the calculated results with respect to the
damage pattern observed under the microscope. Varying the
material data, further calculations helped us determine the data
yielding the greatest correspondence between the experiment
and the results of our calculations. But although this provided us
with a plausible set of data concerning the material properties
in the layer materials, we had not yet, by means of measure-
ment, obtained direct verification of the strain in the layer
system when exposed to complex conditions.
An unexpected suggestion
The 35th International Conference of the American
Ceramic Society in January 2011 in Daytona Beach brought an
encounter that would help us move forward. Seetha Raghavan
from the University of Central Florida in Orlando introduced to
the discussion the synchrotron source at the Argonne National
Laboratory and proposed that, there, we build a modified test
facility, similar to the one already developed at DLR. The high-
energy X-rays emitted by the synchrotron would enable us to
penetrate the layers, and hence use the interactions between
the radiation and the material to determine the strain, and thus
the stress, in the individual layers. This was the missing link to
validate our calculations with direct measurements.
After developing a preliminary concept for the new test
facility, and once Seetha Raghavan achieved approval for meas-
urement time for our project on the advanced photon source,
APS, in Argonne, her postgraduate students Kevin Knipe and
Albert Manero travelled to our institute in Cologne for two
months in the summer of 2012, where we jointly designed the
new facility. We essentially needed a compact furnace with high
heating power and grips with ducts for efficient inner cooling
suited to the tubular specimens. The furnace had to fit in an
existing mechanical test machine at the Argonne National Labo-
ratory and needed windows at the right positions to let the
X-rays through. The entire test machine is mounted on electro
motors, which are used to position the specimen to the fine
X-ray beam with a precision of microns. During measurement,
the beam penetrates the layer system and is diffracted. A detector
registers the diffracted radiation, and the resulting pattern can
be used to identify the materials, which are present.
More information:
Damaged
Undamaged
After many long-distance flights, sections of the white ceramic
thermal barrier coating on the left blade have spalled.
The X-rays are diffracted by the atoms of the coating material.
A detector registers the diffracted X-rays. If thermal and/or
mechanical loads strain the coating system, the atomic spacing
changes – and with it the diffraction pattern.
Aerial view of the synchrotron source in Argonne near Chicago
Image: AdvancedPhotonSource
A close-up of the specimen (white in the image)
Kevin Knipe and Albert Manero during data acquisition
Keeping an eye on the special furnace – Albert Manero
from UCF
Janine Wischek sets up the test system
The principle of strain measurement
with synchrotron radiation
X-ray detector
Radiant heater
X-ray
Air cooling
Mechanical
load
Mechanical
load
30
|
DLR
ma
G
azıne
138
·
139
|
MECHANICAL MATERIALS TESTING
MECHANICAL MATERIALS TESTING
|
DLR
ma
G
azıne
138
·
139
|
31
