Investigation on Process Stability and Part Positioning Influence on the Relative Density of Designed Materials via Laser‐Based Powder Bed Fusion of Metals on a Multi‐Laser Machine [original]

Investigation on Process Stability and Part Positioning

Inﬂuence on the Relative Density of Designed Materials

via Laser-Based Powder Bed Fusion of Metals on a Multi-

Laser Machine

Johannes Albert,* Oliver Hermann, Simon Purschke, David Rule, and Claudia Fleck

1. Introduction

Large gas turbines (LGTs) play an impor-

tant role in a challenging energy market,

which is going through a transition from

using mostly fossil fuels to sustainable

and renewable resources.

[1,2]

From one

side, LGT’s main advantage is based on

the high efﬁciencies of up to 65% in a com-

bined cycle.

[2]

The hot exhausts of the gas

turbine run an additional steam turbine.

[2]

As result, signiﬁcantly lower carbon emis-

sions in the exhausts are achieved, if com-

pared with other fossil technologies such as

coal-ﬁred power plants.

[2]

As the LGT efﬁ-

ciency is directly connected to the highest

temperature in the thermodynamic cycle,

the combustion performance has to be

improved.

[3]

In contrast, hydrogen is estab-

lished as an alternative fuel for LGTs

because there is an increasing demand

for a more sustainable and fast-reacting

energy supply.

[1]

Introducing hydrogen as

a fuel leads to rising combustion temperatures.

[1]

In both cases

—enhancing efﬁciency and using hydrogen as an alternative

fuel—the resulting highest temperature within the LGT is dramat-

ically affected.

[1,3]

Current developments point out that the combustion chamber

and its components are submitted to temperatures up to

2000 K.

[4,5]

This highlights the substantial necessity of effective

cooling concepts and designs.

[4,5]

One of the most efﬁcient ways

is via transpiration cooling.

[5,6]

The main characteristic of this

cooling method is a porous wall through which a coolant agent

ﬂows—compressed air in case of LGTs.

[5,6]

By streaming air

from the cold to the hot gas side, a protective ﬁlm cooling the

hot component surface is generated.

[5,6]

Additive manufacturing (AM) and especially laser-based pow-

der bed fusion of metals (PBF-LB/M) can generate complex

geometries out of high temperature alloys.

[1]

This makes it

suitable for manufacturing components in high temperature

applications.

[1]

Next to geometrical deﬁned drillings or lattice

structures, process adaptions form an approach for producing

open-porous media by PBF-LB/M.

[7–18]

The concept of designed materials (DMs) is based on manip-

ulating the standard PBF-LB/M process to create open-porous

J. Albert, O. Hermann, D. Rule

Additive Manufacturing Materials and Technology

Siemens Energy

Huttenstr. 12, 10553 Berlin, Germany

E-mail: albert.johannes@siemens-energy.com

S. Purschke

Design for Additive Manufacturing

Siemens Energy

Huttenstr. 12, 10553 Berlin, Germany

C. Fleck

Materials Science & Engineering

Institute of Technology Berlin

Straße des 17. Juni 135 –Sekr. EB 13, 10623 Berlin, Germany

The ORCID identiﬁcation number(s) for the author(s) of this article

can be found under https://doi.org/10.1002/adem.202100635.

VCH GmbH. This is an open access article under the terms of the Creative

Commons Attribution-NonCommercial-NoDerivs License, which permits

use and distribution in any medium, provided the original work is properly

cited, the use is non-commercial and no modiﬁcations or adaptations are

made.

DOI: 10.1002/adem.202100635

Component production via laser-based powder bed fusion of metals (PBF-LB/M)

is becoming reality for multiple applications. Especially for sophisticated parts, as

in case of gas turbines, assurance of repeatable properties involves control of

process and machine parameters. PBF-LB/M is used to build conﬁgurable open-

porous structures—designed materials (DMs)—out of the high temperature

superalloy Haynes 282 with relative densities from 35% to 65%. The objective is

investigating positioning inﬂuence and process stability based on their relative

densities in discrete scan ﬁelds and build jobs. As DMs made from Haynes 282

are not yet scientiﬁcally investigated, ﬁrst the process boundaries are identiﬁed.

The used PBF-LB/M machine—EOS M 400-4—works with four lasers in four

quadrants—a quadrant benchmarking is carried out. The experiment is repeated

in a second build job. Scatter is identiﬁed in quadrant and build job benchmarking.

A further aspect under investigation is positioning within laser quadrants.

However, keeping laser and parameter combinations constant, sample locations

are assessed. A relative density dependency on the laser deﬂection angle—as a

positional measure—is observed. The results of positioning inﬂuence and scatter

are used to formulate a tolerance equation for relative densities of DMs.

RESEARCH ARTICLE

www.aem-journal.com

structures.

[7–18]

This can be achieved by either adapting powder

stock or process parameters.

[7–18]

Additives can be used to estab-

lish open-porous structures in the PBF-LB/M process.

[15,16]

Chemical compounds use the applied laser energy in the process

for a gas-generating reaction within the melt pool and therein

produce pores.

[15]

Another option are salts that act as place

holders.

[16]

After printing, they can be removed in a subsequently

leach-out step—pores remain.

[16]

Other approaches focus on

standard PBF-LB/M process parameters, which can be set

individually.

[7–14,17,18]

They generate adjustable open-porous

structures by decreasing the introduced line and volume

energy.

[7–14,17,18]

For this purpose, decreasing the laser power,

or increasing the scan speed, hatch distance, or layer thickness

are setting options.

[7–14,17,18]

Some authors report the additional

use of pulsed lasers or changes in scan strategies for achieving

similar results.

[7,11]

The described methods for introducing pores

into the PBF-LB/M components have been showed for different

materials. Titanium alloys, tool steels, stainless steels, and

nickel-based superalloys—IN625, IN718, EP648, and Hastelloy

X—have been analyzed.

[7–14,17,18]

The nickel-based superalloy Haynes 282—UNS N07208

(Ni-20Cr-10Co-8.5Mo-1.5Al-2.1Ti)—is a material with high tem-

perature strength and good processability.

[19]

It is an ideal can-

didate for static components in aircrafts and land-based gas

turbines.

[19]

With precipitation strengthening from 24%

gamma prime, it is considered the successor of choice of the

solid solution strengthened materials used today.

[20]

Bulk

PBF-LB/M processes for Haynes 282 have showed that the

material is additively manufacturable and has promising process

stability.

[21,22]

Haynes 282 as a material for open-porous

structures–intended pores by PBF-LB/M process parameter

adaptions-have not been scientiﬁcally investigated, yet.

Therefore, the process parameter boundaries and the process sta-

bility were not examined, as well.

In addition to the manufacturing challenges in processing

highly porous materials, their functional behavior for applica-

tions like transpiration cooling has been investigated by ﬂow

tests.

[7]

It has been stated that the morphological properties have

a major inﬂuence on the ﬂow behavior.

[7,8,18,23]

In terms of LGT application, DMs are of increasing interest due

to the possibility to manufacture open-porous structures and bulk

material simultaneously.

[7]

They offer the perspective to increase

the efﬁciency and lower the emissions by introducing transpira-

tion cooling.

[7]

For the implementation into commercial products,

however, some essential aspects need to be investigated. Possible

shortcomings, such as the repeatability of morphological proper-

ties, must be mitigated. The necessity for such assessment is

underlined by investigations which have showed positioning inﬂu-

ences-in build layer dimension-on surface roughness and the melt

pool geometry.

[24,25]

Furthermore, nonadjustable parameters such

as the shielding gas ﬂow or plume generation might inﬂuence the

repeatability of parts built with lower energy parameter sets.

[26,27]

As analyzed for different materials in various studies, morpho-

logical DM properties can be manipulated by varying settings for

layer thickness, laser power, scan speed and hatch distance. DM

process development has not yet been carried out for Haynes 282.

An initial experiment provides a range of feasible process param-

eter settings for an EOS M 400-4 PBF-LB/M machine (EOS

GmbH, Krailling/Munich, Germany). For the research standard,

powder feedstock was utilized. The system used for the investiga-

tion is not capable of operating in pulse modus for the four laser

beams, hence this variable has been excluded. Instead, feasible

ranges of laser power, scan speed, hatch distance, and layer thick-

ness are in focus. The aim of this study is to assess the repeatability

of DMs regarding relative density. For this, an inﬂuence study of

part positioning and a benchmarking of the laser quadrants have

been conducted. The laser deﬂection angle (LDA) as a positional

measure might generate an understanding if mechanisms which

affect the melt pool geometry are present for DMs, as well. The

results have shown a dependence of the relative density on the

LDA and non-negligible scatter, which both have been used to for-

mulate a relative density tolerance equation.

2. Experimental Section

2.1. Materials

Commercial Haynes 282 inert gas atomized metal powder

TruForm (Praxair Surface Technologies, Inc., Indianapolis,

USA) has been used as feedstock. Sieve analysis per ASTM

B214 revealed that none of the analyzed powder particles were

bigger than 53 μm in size. The powder particle size distribution

was determined by laser diffraction analysis per ASTM B822.

This investigation revealed that the powder had a D10 of

22.2 μm, D50 of 31.0 μm, and D90 of 45.6 μm.

2.2. Process

2.2.1. Multi-laser PBF-LB/M Process

An EOS M 400-4 multi-laser PBF-LB/M machine (EOS GmbH,

Krailling/Munich, Germany) was utilized for this work. It oper-

ated with four identical laser units (1) in four separate quadrants

(Q1–Q4) with overlapping side areas.

[28,29]

Figure 1 shows a

detailed scheme of its process chamber. The four independent

scan units (2) deﬁned the positions of the laser spots in focal

plane by tilted mirrors.

[28,29]

Four f-θlenses (3) provided a planar

focal level.

[28,30]

For each quadrant, a focused laser beam in an

area of 250 250 mm

(X–Yplane) can be guaranteed.

[29]

The

laser beam focus was measured for each quadrant by the original

equipment manufacturer as per standard service according to its

speciﬁcation. With a build platform (9) of 400 400 mm

(X–Y

plane), each laser will then operate in a quadrant of 200 200 mm

(X–Yplane).

[29]

A sideward intersection of 50 mm to the neighbor

quadrants was utilizable.

[29]

The overlap area and the use of multi-

ple lasers on a single specimen were not considered for this inves-

tigation. Also the restricted area for build platform ﬁxation (4),

located in each corner, was not utilized.

[29]

The focal points of

the four f-θlenses where the laser beams have no deﬂection

(7, 8) were located at 75 mm in X-direction and 75 mm Y-direc-

tion from the build platform center.

[29]

For the operation in all other

locations (5), a deﬂected laser beam (6) was needed. Figure 1

furthermore shows the shielding gas ﬂow in the EOS M 400-4

PBF-LB/M machine. The shielding gas streamed out of a nozzle

in the middle of the build platform on top of the process cham-

ber.

[29]

On the right and the left sides in build layer height, the

shielding gas was sucked off by in Y-direction elongated exhaust

www.advancedsciencenews.com www.aem-journal.com

slots.

[29]

Therefore, symmetrical shielding gas conditions were

achieved in the combinations of Q1 þQ2 and Q3 þQ4.

[29]

As mentioned, the positioning on the build platform could

have an inﬂuence on the morphological properties of DMs.

The LDA can be used as a value which makes the different posi-

tions in build plane dimension comparable. It can be calculated

by Equation (1), where xLDA and yLDA are, respectively, the

X-and Y-distances from the nondeﬂected laser beam spot and

zLDA is the Z-height of the deﬂection device.

LDA ½°¼ arctan ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

LDA þy2

LDA

pzLDA

! (1)

The constant zLDA distance was deﬁned by the span between

the deﬂection device and the build platform, and this then gave a

value of 490 mm for the EOS M 400-4. The combination of this

distance with the sample-speciﬁcxLDA and yLDA offsets can be

used to calculate an individual LDA for each specimen. A range

of values between 1.24° and 14.58° was covered for the four

parameter combinations, which are placed equally in the ﬁve

position blocks. Position block I was in the range from 1.24°

to 4.52°. The second position block II was in the center of the

build platform—7.57° to 10.78°. Symmetrically arranged in the

quadrants the positions III and IV ranged from 11.46° to

14.58°. The ﬁfth position block V had LDAs from 8.18° to 11.38°.

For the layerwise powder spreading, the EOS soft recoating

system for the M400 series with a natural rubber lip (EOS

GmbH, Krailling/Munich, Germany) was used. The part expo-

sure was selected to be against the shielding gas ﬂow by the

according EOS standard setting.

2.2.2. Sample Depowdering

Following manufacturing, samples were submitted to a two

stage depowdering program for ensuring complete removal of

remaining unbound powder. The ﬁrst step, carried out with a

vacuum cleaner, ensured powder removal from larger cavities.

The second one included vibration, rotation, and inversion of

the build platform. It was carried out with a SFM AT800 depow-

dering machine (Solukon Maschinenbau GmbH, Augsburg,

Germany) to guarantee full powder removal even for smallest

cavities.

2.3. Methods

2.3.1. Sample Design

For determination of the relative density, a dedicated sample

design, shown in Figure 2, was developed. It consists of a base

block (3) on which a cylindrical DM specimen (1) with an as-build

diameter dDM of 8 mm is built. The specimen height hDM was

between 9.0 and 12.7 mm depending on the manually set cut-

off plane (2). The base block (3) was used for easier disassembly

off the build platform and for sample identiﬁcation.

2.3.2. Relative Density Measurements and Determination of

Uncertainty Propagation

For determination of relative densities, the specimen height hDM

and diameter dDM were measured manually with a Mitutoyo 293-

661-10 digital dial micrometer (Mitutoyo Corporation, Kawasaki,

Japan). Final values were an average of ten measurements for

each specimen. The specimen mass mDM was measured using

a Sartorius BA 210 S high-resolution balance (Sartorius AG,

Goettingen, Germany).

For simpliﬁcation purposes, it was assumed that the melted

material had a density of 100%. The relative density function

fRD is deﬁned in Equation (2), where the relative density RD

is dependent on dDM as the specimen diameter, hDM as its height,

Figure 1. Schematic side view of the process chamber of an EOS M 400-4 with its shielding gas ﬂow (left), the corresponding top view on its build

platform (middle), and the sample distribution at each position (I–V) with the according parameter sets (PS A–PS D) (right). 1—laser unit; 2—scan unit;

3—f-θlens; 4—restricted ﬁxation area; 5—laser spot (deﬂected); 6—laser beam (deﬂected); 7—laser spots (not deﬂected); 8—laser beam (not

deﬂected); 9—build platform; Q1–Q4—quadrants Q1–Q4; I–V—positions I–V; PS A–PS D—parameter sets PS A–PS D; LDA—laser deﬂection angle.

www.advancedsciencenews.com www.aem-journal.com

and mDM as its mass. These parameters constituted a speciﬁc

DM density ρDM, which results from the ratio of the specimen

mass mDM and the specimen volume VDM. The theoretical bulk

density for Haynes 282 ρH282 was considered as 8.27 g cm3.

[19]

fRDðdDM,hDM,mDMÞ¼RD ½%¼ ρDM

ρH282

mDM

VDM

ρH282

mDM

4⋅d2

DM⋅hDM

ρH282

(2)

The uncertainty calculation was carried out according to DIN

1319-4.

[31]

The relative density uncertainty sRD is given by

Equation (3), where the relative density function fRD is derived

by its variables, specimen diameter dDM, height hDM, and mass

mDM. Another part of this calculation are the measurement accu-

racies sd,sh,smof each variable.

sRD½%¼ ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ

∂

fRD

∂

dDM

⋅sd



∂

fRD

∂

hDM

⋅sh



∂

fRD

∂

mDM

⋅sm



(3)

For this equation, the measurement accuracies of the single

terms are needed. The accuracy of the specimen diameter mea-

surement sdis a combination of the statistical part, which was

determined for each specimen and the device accuracy. The uti-

lized digital dial micrometer, Mitutoyo 293-661-10 (Mitutoyo

Corporation, Kawasaki, Japan), had a device accuracy of

0.01 mm. In average, the cumulated diameter measurement

accuracy sdwas calculated as being 0.026 mm. As the specimen

height hDM was measured also manually with the same tool, the

device accuracy was kept constant. An average height measure-

ment accuracy shof 0.068 mm was present. The measurement

accuracy of the used balance for determination of the specimen

mass mDM is deﬁned with sm¼0.0002 g.

The average relative density uncertainties sRD for the parame-

ter set PS A was about 0.52%. In an equal range parameter set,

PS B is set with 0.42%. For parameter set PS C, an average

relative density uncertainty of 0.31% was calculated.

Parameter set PS D was also comparable with 0.50%. This

resulted in an average relative density uncertainty sRD of about

0.44% for all parameter sets combined.

2.3.3. Microcomputed Tomography Measurements

The DM morphology was analyzed via microcomputed tomogra-

phy (μCT) with a Yxlon FF85CT inspection system (Comet

Technologies USA, Inc., Hudson, USA). Micro focus was used.

Segmentation of the data was done with the software VGStudio

Max 3.3 (Volume Graphics GmbH, Heidelberg, Germany). The

gray value for the surface determination was selected so that the

digitally and physically measured relative density were equivalent.

2.4. Experimental Procedure

The experimental procedure of this study was separated into two

parts: ﬁrst, a preliminary investigation of feasible process

parameter settings and second, the primary analysis of DM

repeatability.

Haynes 282 as material for open-porous PBF-LB/M structures

was not yet scientiﬁcally investigated. Therefore, the process

boundaries were identiﬁed with a grid scanning approach.

It assessed 250 different process parameter settings by visual

inspection. The criterion for the process boundaries was how

good the outer shape of the samples was represented. The visual

assessment was based on four categories—not buildable, build-

able with very poor quality, poor quality, or good quality. This

investigation built the baseline for the parameter selection for

the further analyses.

Out of the feasible parameter range four representative and

widely different process parameter combinations were selected

for the repeatability evaluation. For the main objective of this

study, assessing the relative density repeatability, two subap-

proaches for laser zone benchmarking and part positioning were

pursued. As the used PBF-LB/M machine—EOS M 400-4—works

with four lasers in four quadrants, a benchmarking of these zones

at position I was carried out. This process stability experiment was

repeated in a second build job. The following aspect under inves-

tigation was the positioning of DMs in build layer dimension.

However, keeping laser and process parameter combinations

constant, the sample location was assessed. For this investigation,

the differing LDAs were used as measure.

The identiﬁed scatter and positioning inﬂuence were used to

build a tolerance equation for the relative density of DMs made

from Haynes 282. An untested parameter set within the analyzed

ranged was checked against the formulation.

3. Preliminary Investigation of Feasible Process

Parameter Settings

3.1. Design of Experiment for Identiﬁcation of the Process

Boundary

On one side, processing Haynes 282 as bulk material via PBF-

LB/M was broadly investigated.

[21,22]

On the other side, the

Figure 2. DM sample and its main features after manufacturing (left) and

after specimen separation (right). 1—DM specimen; 2—cut-off plane;

3—base block; hDM—DM height; dDM—DM diameter; mDM—DM mass.

www.advancedsciencenews.com www.aem-journal.com

feasibility of the open-porous DM process for Haynes 282 was

not addressed. In this case, there was the need to ﬁnd parameter

sets between the process boundaries—highly dense and not

consistently buildable at all. For this purpose, a variation of the

process parameters layer thickness tL, hatch distance HD, laser

power LP, and scan speed LShas been carried out and the build-

ability of each combination has been visually investigated.

All parameters are denoted as reference to a standard param-

eter set. Experiments were carried out using both a standard layer

thickness (tStd

L) and twice that standard layer thickness (2 ⋅tStd

L).

Previous studies on such open-porous structures have shown a

high inﬂuence of the layer thickness on the resulting relative

density.

[9,11,17,18]

It was assumed that consistently buildable

DMs of higher layer thicknesses require higher line energies.

Therefore, different settings for laser power and scan speed

depending on the layer thickness were selected. In Table 1,

the multiples of the standard laser power (LStd

P), the standard scan

speed (LStd

S), and standard hatch distance (HStd

D) for the planned

design of experiment (DoE) are given. Furthermore, the volume

energy density EVis provided.

For all varied parameter factors—except layer thickness, ﬁve

levels were used. Two 5·5·5 matrices resulted, one for the stan-

dard layer thickness (tStd

L) and an additional one for the double

standard layer thickness (2 ⋅tStd

L).

The hatch rotation angle (Hrot) has a large inﬂuence on the

resulting morphology including the periodicity, connectiveness,

and the distribution of shapes of porous areas.

[7]

For the study, a

constant value of 67° for the hatch rotation was kept.

3.2. Determination of Process Boundaries

The DoE based on the parameters from Table 1 resulted in a grid

scanning of 250 process parameter combinations (that is, param-

eter sets). The large quantity was required to ensure that enough

data points would be available to examine the factors’contribu-

tions to the key response in this evaluation—buildability.

Figure 3 shows the two grids of the investigated DoE.

Furthermore, the buildability is indicated. Many samples did

not build up (red cross). A further quality assessment was if build

samples showed poor quality indications. Samples built up very

poorly (orange cross) or poorly (yellow cross) showed broken out

parts or rough contours. Exemplary images of samples with such

indications are shown in the lower part of Figure 3. Points that

are not marked with crosses are buildable in a satisfying quality

in order of representing the intended outer geometry.

4. Selection of Process Parameter Combinations

for the Repeatability Study

The use of multiple buildable parameter combinations with dif-

ferent volume energy densities might allow a more general con-

clusion for the investigated range. This should avoid that

ﬁndings can only be stated for a single parameter set. In addition,

this provides information on whether the parameter combina-

tion itself will have an impact on the repeatability. Table 2 shows

the four parameter sets PS A–PS D with widely different volume

energy density levels. As shown in Figure 3, they were chosen

from the standard layer thickness tStd

Lparameter combinations,

as by doubling the layer thickness the printability decreased.

Furthermore, Table 2 shows morphology illustrations taken

via segmented μCT volume data for a representative sample

of each parameter combination. These cross sections show

material (black) and pores (white) in two dimensions of the

DMs—one in build direction (Y–Zplane) and one in build layer

(X–Yplane).

5. Results and Discussion

To investigate the repeatability of DMs, the inﬂuences of part

positioning and process stability through different laser zones

on the relative density were analyzed. First, the part positioning

within the laser zones was investigated. These results were

used to establish two computed positioning compensation

approaches—global compensation approach (GCA) and param-

eter set compensation approach (PSCA). Subsequently, the

repeatability between the laser zones of the PBF-LB/M machine

was assessed for scatter with minimized positioning inﬂuence.

Afterward, the results for scatter and positioning inﬂuence were

combined to formulate tolerance equations, that were checked

against the measurement results.

5.1. Inﬂuence of Part Positioning on the Relative Density

Table 3 shows the inﬂuence of the LDA on the relative density of

the investigated DMs for quadrants Q1 þQ2 (that is, the combi-

nation of the quadrants Q1 and Q2) in positions I–V (compare

Figure 1). The quadrat-speciﬁc points and regressions are shown,

as well. These two quadrants have been assessed to underline the

validity of the ﬁndings even if different laser and scan units are

used. It is assumed that the transferability is also given for

quadrants Q3 and Q4. The elaborated measurement accuracies

of Section 2.3.2.—individual for each parameter set—are indi-

cated in the graphs.

An increase in the LDA results in a linear decrease in the

relative density for all four parameter sets, in both quadrants.

Even if some of the regression have lower coefﬁcients of deter-

mination, the general trends are comparable. The suggested scat-

ter of the data motivates the elaboration of tolerance equations.

Table 3 combines the results of quadrants Q1 þQ2 to gain sta-

tistically more valid linear regression gradients. These then show

that denser DMs show a tendency toward a stronger inﬂuence of

the LDA on the relative density. For an assessment of this state-

ment, the theoretical relative densities at a LDA of 0° for the four

Table 1. Set of parameters used for the ﬁve-level DoE for the process

boundary investigation.

tStd

L2⋅tStd

Maximum EVMinimum EVMaximum EVMinimum EV

HD3.33 ⋅HStd

D7.78 ⋅HStd

D3.33 ⋅HStd

D7.78 ⋅HStd

LP1.06 ⋅LStd

P0.83 ⋅LStd

P1.20 ⋅LStd

P0.94 ⋅LStd

LS0.94 ⋅LStd

S1.20 ⋅LStd

S0.83 ⋅LStd

S1.06 ⋅LStd

EV19.32 J mm35.05 J mm312.40 J mm33.24 J mm3

www.advancedsciencenews.com www.aem-journal.com

Loading more pages...