st

ech

egrs, ps wl.Wmaountionertd pole

gradabpertiesand st-basededic ple for rhydrogatic be, their

ot allowed to containe fabrication process.and a desirable pore-rials. Casting to invert

Materials Science and Engineering C 43 (2014) 494501

Contents lists available at ScienceDirect

Materials Science a

.e lstents cause problems similar to those of permanent stents [1,19].Therefore, other authors have solved this problem by alloying iron-based materials to increase the degradation rate. Alloying with ele-

removable salt forms is problematic because of the high melting pointof iron [6]. Therefore, powder metallurgical techniques seem to bemore suitable for the preparation of porous iron implants because[1,2,15,16]. Although many in vitro studies of iron have shown thatiron may be problematic [17,18], the in vivo studies performed byPeuster et al. [1,19] showed that implanted iron stents cause no signi-cant toxicity issues. However, their degradation rate is too slow, and the

and corrosion behaviors, such materials are nany harmful contaminants originating from thMoreover, an interconnected porous structuresize interval are usually required for such matecandidates for load-bearing applications in orthopedics [3].Iron is a plentiful element in the human body and possesses higher

mechanical properties than magnesium-based materials [3]. Therefore,iron has been studied as a possible material for biodegradable stents

Generally, manymethods exist for the fabrication of porousmetallicmaterials [5,2630], some of which were developed especially for iron[31]; however, not all of these methods are suitable for the fabricationof materials used in implantology. In addition to suitable mechanicalments such as manganese, palladium, silverappears to be a suitable approach to increa

Corresponding author. Tel.: +420 220444055.E-mail address: Jaroslav.Capek@vscht.cz (J. apek).

http://dx.doi.org/10.1016/j.msec.2014.06.0460928-4931/ 2014 Elsevier B.V. All rights reserved.relatively short lifespanle for some applications.investigated as possible

et al. prepared a foamed iron alloyed by phosphorous and implanted itinto a sheep's femur. Even after 12 months, no inammation or toxicitywas observed [5].(approximately 12 months) can be unsuitabTherefore, other metallic materials have beenIn recent decades, research of biodeincreased. Due to the mechanical proapplication has focused on orthopedicshave been performed on magnesiumthat these alloys are suitable for orthopnately, these materials are only suitabbecause their corrosion is connected tofore, larger implants can be problemamounts of hydrogen [3]. Additionallyle metallic materials hasof such materials, theirents [18]. Many studiesalloys that have shownurposes [813]. Unfortu-elatively small implantsen evolution [14]; there-cause they release large

and maintain sufcient mechanical properties and biocompatibility[25,15,16,20]. Another method to increase the degradation rate is thepreparation of metal foams with an interconnected porous structure[19,21,22]. In addition to faster degradation, such materials allow thetransport of bodily uids to healing tissue and the ingrowth of new tis-sue into the implant; moreover, these materials possess lower Youngmoduli than do compact materials, which improves mechanical bio-compatibility by preventing the stress-shielding effect [5,6,2325].These metal foams are very promising candidates for the preparationof bone replacements, also known as scaffolds. For example, Quadbeck1. IntroductionMicrostructural and mechanical characteriby powder metallurgy

Jaroslav apek , Dalibor VojtchDepartment of Metals and Corrosion Engineering, Institute of Chemical Technology in Prague, T

a b s t r a c ta r t i c l e i n f o

Article history:Received 20 February 2014Received in revised form 25 May 2014Accepted 30 June 2014Available online 26 July 2014

Keywords:Porous ironPowder metallurgyScaffolds

The demand for porous biodtions of biodegradable stentprepared porous iron sampleate as a space-holdermateriabehaviors (universal loadingrosity increasedwith the amtion and no other contaminasamples; although, the propnon-metallic biomaterials anallurgy appears to be a suitab

j ourna l homepage: www, phosphorus and siliconse the degradation rateics of porous iron prepared

nick 5, 166 28 Prague 6, Czech Republic

adable load-bearing implants has been increasing recently. Based on investiga-orous iron may be a suitable material for such applications. In this study, weith porosities of 3451 vol.% by powder metallurgy using ammonium bicarbon-e studied samplemicrostructure (SEM-EDX and XRD), exural and compressivechine) and hardness HV5 (hardness tester) of the prepared samples. Sample po-t of spacer in the initial mixtures. Only the pore surfaces had insignicant oxida-was observed. Increasing porosity decreased the mechanical properties of the

ies were still comparable with human bone and higher than those of porousrousmagnesiumprepared in a similarway. Based on these results, powdermet-method for the preparation of porous iron for orthopedic applications.

2014 Elsevier B.V. All rights reserved.

nd Engineering C

sev ie r .com/ locate /msecthey allow the preparation of porous materials that possess the desiredproperties. Quadbeck et al. prepared iron-based foams by the impregna-tion of polyurethane foamwith an iron-based slurry. Subsequently, thepolyurethane foamwas decomposed, and the porous bodywas sintered[5]. Unfortunately, many toxic compounds are generated during

green compacts (diameter 10 mm, length 40 mm) using a compactionpressure of 510MPa at room temperature. The compactswere annealedat 130 C for 4 h in air to thermally decompose the ammonium bicar-bonate particles. Subsequently, the porous bodies were sintered in an

Fig. 1.Morphologies of the initial powders (SEM): a) iron and b) ammonium bicarbonate.

495J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501polyurethane decomposition, which may complicate the process [32].Another powdermetallurgical approach for the preparation of porous ob-jects is a technique using space-holder particles. This approach,which hasbeen successfully used for the preparation of porousmagnesium and tita-nium [21,22,24,29], involves preparing a mixture containing particles ofmetal and some space-holder (such as urea or ammonium bicarbonate).The mixture is compacted, and the spacer particles are removed by low-temperature annealing or leaching. Subsequently, the porous body issintered at higher temperatures [21,22,24,29].

Although many studies have been performed on the preparation ofporous iron and its properties, to the best of our knowledge, none ofthese studies have addressed porous iron prepared by powdermetallur-gy using space-holder particles. Therefore, we prepared porous ironsamples with different porosities using ammonium bicarbonate as aspace-holder and studied the inuence of increasing spacer content onthe microstructural and mechanical characteristics of the materials.

2. Materials and methods

Iron (Alfa Aesar, 99.5 wt.%, b212 m) and ammonium bicarbonate(Penta, p.a., 250500 m) powders were used as initial materials.Their morphologies are shown in Fig. 1, and the microstructure of theiron powder is shown in Fig. 2.

The powders were manually blended into mixtures containing 0, 5,10, 15 and 20 vol.% of ammonium bicarbonate, which was used as thespace-holder. Hexane was added into the mixtures during blending tomake a dough-likemixture for better homogenization and to avoid seg-regation. These mixtures were uniaxially cold-pressed into cylindricalFig. 2.Microstructure of iron powder (LM).evacuated tubular furnace at 1000 C for 4 h. The prepared sampleswere weighed and measured to determine their porosities accordingto Eq. (1), where P is the sample porosity, d is the diameter, l is thelength,m is the weight, and is the density of pure iron (7.87 g/cm3).

P 1 4 ml d2

100 % 1

Portions of the samples were cut both across and longitudinally, andmetallographic sections were prepared in the standard way to observesample microstructure using an Olympus PME3 light metallographicmicroscope (LM) and a TESCAN VEGA-3 LMU scanning electron micro-scope (SEM). The cross-sections were also subjected to EDX and XRDinvestigations to determine the level of contamination during samplepreparation. EDX analyses were performed using an SEM equippedwith an Oxford Instruments INCA 350 EDX analyzer (SEM-EDX). XRDanalyses were performed using a PANalytical X'Pert PRO X-ray diffrac-tometer equipped with a Cu anode. Five samples of each series weresubjected to three-point bending tests, and three other samples with aFig. 3. Porosity versus initial Fe:NH4HCO3 volume ratio.

ies o

Fig. 6. XRD patterns of initial iron powder and porous samples.

496 J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501length of 15 mm were cut and subjected to compression tests. Both ofthese tests were performed at room temperature using a LabTest5.250SP1-VM universal loading machine. Deformation rates were 0.5and 1 mm/min for the exural and compression tests, respectively.Ultimate exural strength (UFS), exural modulus of elasticity (Ef),compressive yield stress at 0.2% plastic strain (CYS) and compressivemodulus of elasticity (Ec) were determined from the stressstraincurves. Average values and standard deviations (shown as error barsin the plots) were calculated from the data. After the exural tests, frac-ture surfaces were observed and documented by SEM. Vickers hardnesswas measured on sample cross-sections using a load of 5 kg. Fifteenindents were performed for each sample, and average values and stan-dard deviations were calculated.

3. Results and discussions

3.1. Microstructure and porosity

Fig. 4.Macrographs of samples with porositThe dependence of porosity on the initial Fe:NH4HCO3 ratio plottedin Fig. 3 shows that porosity increased for initial mixtures containinggreater amounts of NH4HCO3. The difference in porosity betweensamples prepared without spacer particles and those using a mixture

Fig. 5. SEM micrographs (BSE detector) of sampf a) 34, b) 42, c) 46, d) 48 and e) 51 vol.%.containing 5 vol.% of spacer particleswas approximately 8 vol.%. The in-crease in porosity was higher than the amount of spacers added (8 vs.5 vol.% of spacer), most likely due to the expansion of gasses generated

les with porosities of a) 34 and b) 46 vol.%.

by spacer decomposition. However, differences between the porositiesof samples prepared using 520 vol.% of space-holder material wereonly 34%, which suggests that the occurrence of ammonium bicarbon-ate particles slightly improved the compressibility.

Themacrographs of the samples are shown in Fig. 4. Visible pores oc-

While XRD analysis, shown in Fig. 6, did not suggest oxidationduring sintering, Fig. 7 shows that an increased presence of oxygen inthe pores was detected by SEM-EDX. This gure indicates that theinner surface of the pores was most likely covered by an iron oxidelayer. We attribute this oxidation to a reaction with water and enclosed

Fig. 7. SEMmicrograph (BSE detector) and X-ray oxygen map of the sample with porosity of 46 vol.%.

497J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501curred in the structure of samples prepared from mixtures containingspace-holder particles (porosity 4251 vol.%). These macro-poreswere relatively homogeneously distributed, and the amount of macro-pores increased with the amount of added spacer.

Fig. 5 shows SEM micrographs of selected samples. In Fig. 5a, thesamples that were prepared without spacer material contained a largenumber of small pores (34 vol.%), marked as Type I. In samples pre-pared with added spacer material (Fig. 5be), slightly lower numbersof Type I pores appeared, suggesting that the presence of ammoniumbicarbonate improved the compressibility, which is in an agreementwith porosity measurement (Fig. 3). In addition to these small pores,additional large pores, marked as Type II, occurred. The small TypeI pores originated from imperfect compaction and the initial iron pow-der (Fig. 2), while the large Type II pores were formed by spacerdecomposition.Fig. 8. Flexural stressstrain curves of the porous samples.air during NH4HCO3 decomposition. Any contamination by other ele-ments (nitrogen and carbon) originating fromNH4HCO3 decompositionwas not observed by XRD or EDX.

The samples prepared in this study possessed a similar microstruc-ture to differentmetallicmaterials (magnesium and titanium) preparedin other studies using the same method [21,22,24,29]. In the case ofsamples prepared using urea as the spacer material, contamination bycarbon has been observed in some cases [22,23]. When we comparethe microstructure of the iron-based samples prepared in this studywith that of iron-based samples prepared by Quadbeck et al. [5], whoused a method featuring impregnation of a polyurethane foam by aniron-based slurry and its subsequent thermal removal and sintering ofthe obtained porous iron-based body, we nd that their samples weresignicantly more porous, the pores were more spherical, homoge-neously distributed and interconnected, and the cell walls wereFig. 9. Ultimate exural strength versus porosity.

signicantly thinner. The authors did not investigate contamination ofthe products; however, contamination by organic compounds, carbonand oxygen may be expected, based on a study by Font et al. [32].

3.2. Mechanical properties

The exural stressstrain curves of the prepared samples are shownin Fig. 8. An increase in porosity noticeably decreased the mechanicalproperties. While an area of plastic deformation occurred in the stressstrain curve of the sample prepared using iron powder only (porosityof 34 vol.%) (Fig. 8), no signicant plastic deformation occurred duringexural tests of the samples prepared frommixtures containing spacers(porosity of 4251 vol.%). The average UFS and Ef values are plottedin Figs. 9 and 10, respectively. The implementation of Type II pores(porosity above 34 vol.%) signicantly reduced both the UFS and Ef. Asthe porosity increased from 34 to 51 vol.%, the UFS and Ef decreased byapproximately 80 and 50%, respectively. The increasing amount ofpores resulted in decreased load-bearing sections, which resulted in adecrease of the mechanical properties; moreover, the large Type IIpores acted as stronger stress-concentrators than the small Type Ipores. Therefore, the mechanical properties deteriorated strongly whenthe Type II pores were introduced into the microstructure, and noareas of plastic deformationwere observed in their stressstrain exuralcurves as a consequence. The same trend in UFS evolution with increas-ing porosity and macro-pore implementation was observed for porousmagnesium in our previous work [21] as well as by Zhuang et al. [22].

Fracture surfaces were observed after the exural tests to explainexural behavior (Fig. 11). No large Type II pores and relativelyfrequent traces of plastic deformation were observed on the fracturesurfaces of samples prepared without added spacers (Fig. 11a and b).However, large Type II pores were observed, and signicantly fewersigns of plastic deformationwere found on the fracture surfaces of sam-ples prepared using spacer particles (Fig. 11c and d), which explains theobserved exural behavior.

Fig. 10. Flexural modulus of elasticity versus porosity.

498 J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501Fig. 11. Fracture surfaces of samples with porosity of a) and b) 34 vol.% and c) and d) 46 vol.%.

Fig. 14. Compressive yield strength versus porosity.

499J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501The stressstrain curves obtained by compressive tests and the aver-age values of Ec and CYS are plotted in Figs. 1214. The presence of largeType II pores (porosity above 34 vol.%) signicantly decreased com-pressive properties, compared with samples prepared from iron pow-der only (porosity of 34 vol.%). Increasing amounts of these largepores led to an additional decrease of both Ec and CYS. Samples with aporosity of approximately 50 vol.% had an Ec seven times lower andan UCS four times lower than samples with 34 vol.% of Type I pores.The shapes of the stressstrain curves were similar to those of porousiron-based materials reported in the available literature [5,26]. In com-parison with porous magnesium [23,24,29], no peak stresses were

Fig. 12. Compressive stressstrain curves of the porous samples.observed in the stressstrain compressive curves of porous iron, simi-larly to the behavior of porous titanium [29]. Based on our previous

Fig. 13. Compressive modulus of elasticity versus porosity.research of porous magnesium [24], we hypothesize that these peakstresses occurred during compression of porous magnesium due tothe presence of oxidic layers on the surfaces of magnesium particles,which cause brittle fracture. The absence of such oxidic layers in thecase of iron particles improved diffusion connections between the par-ticles, enhanced plastic deformation and consequently led to the disap-pearance of peak stress. Whereas the shapes of the stressstrain curvesdiffered from those of porous magnesium, as mentioned above, the de-pendence of both CYS and Ec on porosity showed the same trend,mean-ing that CYS and Ec deteriorated when the porosity increased [22,23].

Hardness was measured for additional information regarding themechanical properties, and the dependence of hardness on porosity isplotted in Fig. 15. Although the variations of measured values were

very large due to the presence of pores, the average hardness slightly

Fig. 15. HV5 versus porosity.

re s

Natural bone 0500050050525004050504640400

6

500 J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501decreased with increasing porosity. The most signicant decrease wasobserved after the addition of the rst 5 vol.% of NH4HCO3 (porosity of42 vol.%) because the addition created Type II pores. Additional in-creases in the space-holdermaterial caused a smaller hardness decrease(approximately 2 units of HV5) for each 34 vol.% of porosity.

We conducted a review to compare themechanical properties of thesamples prepared in this study with results obtained by other authorsfor the mechanical properties of human bone and other metallic andnon-metallic porous biomaterials. The results of this review are summa-rized in Table 1 and show that the mechanical properties of all theprepared samples were comparable with those of human bone. Theproperties of porous iron were enhanced comparedwith the propertiesof porousmagnesium prepared in a similar way; moreover, the proper-ties were signicantly higher than the properties of non-metallic bio-materials. Therefore, porous iron is a promising candidate for load-bearing orthopedic applications.Porous iron 3451 25Porous Fe0.6P ~80 ~5Porous Ti 78 20Porous Mg 28 17Porous Mg 1444 25Porous Mg 2338 25Porous Mg 5270 ~1Porous Mg 3655 20Porous Mg 50 20Porous Mg 2931 25Porous Mg 3555 10Porous hydroxyapatite 36Porous hydroxyapatite 5077 20Porous composite (poly-L-lactide +2050 wt.% of bioglass) 7788 ~1Porous polycaprolactone 5556 Porous polycaprolactone 4877 Porous polycaprolactone 3755 Porous polylactide-co-glycolide 31 11Porous composite of polylactide-co-glycolide and 45S5bioglass 43 89Table 1Mechanical properties of porous biomaterials.

Material Porosity [vol.%] Po4. Conclusion

Porous iron samples with porosities of 3451 vol.% were successful-ly prepared by powder metallurgy, and their microstructure andmechanical properties were studied in detail. Samples prepared bymixing iron powder and ammonium bicarbonate as a spacer materialcontained two types of pores: 1) small pores originating from imperfectcompaction and 2) large pores formed by spacer decomposition. Imple-mentation of the large pores in the sample microstructure should en-hance biocompatibility and allow for better osseointegration comparedwith as-cast iron. Although the presence of these pores decreased themechanical properties of the prepared samples, the samples still pos-sessed exural and compressive behavior comparable with humanbone. The mechanical properties were signicantly higher than thoseof non-metallic porous biomaterials and even better than those of porousmagnesium prepared in a similar way. Therefore, porous iron preparedby powder metallurgy appears to be a suitable material for load-bearing applications in orthopedics.

Acknowledgments

The authors would like to thank the Czech Science Foundation(project no. P108/12/G043) for supporting this research.References

[1] M. Peuster, P. Wohlsein, M. Brugmann, M. Ehlerding, K. Seidler, C. Fink, H. Brauer, A.Fischer, G. Hausdorf, Heart 86 (2001) 563569.

[2] H. Hermawan, D. Dube, D. Mantovani, J. Biomed. Mater. Res. A 93A (2010) 111.[3] M. Schinhammer, A.C. Hanzi, J.F. Lofer, P.J. Uggowitzer, Acta Biomater. 6 (2010)

17051713.[4] H. Hermawan, A. Purnama, D. Dube, J. Couet, D. Mantovani, Acta Biomater. 6 (2010)

18521860.[5] P. Quadbeck, R. Hauser, K. Kmmel, G. Standke, G. Stephani, B. Nies, S. Rler, B.

Wegener, PM2010 World Congress, Florenz, Italy, 2010.[6] A. Yusop, A. Bakir, N. Shaharom,M. Abdul Kadir, H. Hermawan, Int. J. Biomater. 2012

(2012).[7] F. Witte, H. Ulrich, M. Rudert, E. Willbold, J. Biomed. Mater. Res. A 81A (2007)

748756.[8] S. Virtanen, Mater. Sci. Eng. B 176 (2011) 16001608.[9] H. Waizy, J.M. Seitz, J. Reifenrath, A. Weizbauer, F.W. Bach, A. Meyer-Lindenberg, B.

Denkena, H. Windhagen, J. Mater. Sci. 48 (2013) 3950.[10] M. Hradilova, J. Kubasek, P. Lejcek, L. Tanger, Metal 2011: 20th Anniversary Interna-

tional Conference on Metallurgy and Materials, 2011, pp. 920924.[11] D. Vojtech, V. Knotek, J. Capek, J. Kubasek, Chem. Listy 105 (2011) 678683.

ize [m] UFS [MPa] Young modulus [GPa] CYS [MPa] UCS [MPa] Reference

2150 0.120 2180 [22]00 22.196.6 0.64 16.167.7 This study

10.9 5.5 [5]00 5.3 35 [29]

24 [12]00 25 [33]00 417 [21]

414 [34]00 1427 1.618.1 1531 [22]00 0.35 2 [29]00 315 1353 2070 [24]00 1217 [23]44 30 [35]00 27 0.127 117 [22]

14 0.140.26 ~0.4 [22] 23 [36] 23 [35] 23 [35] 0.5 [37] 0.4 [37][12] X.N. Gu, W.R. Zhou, Y.F. Zheng, Y. Liu, Y.X. Li, Mater. Lett. 64 (2010) 18711874.[13] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit, F. Feyerabend, Curr.

Opin. Solid State Mater. Sci. 12 (2008) 6372.[14] G. Song, A. Atrens, Adv. Eng. Mater. 5 (2003) 837858.[15] H. Hermawan, D. Dube, D. Mantovani, Acta Biomater. 6 (2010) 16931697.[16] M. Moravej, D. Mantovani, Int. J. Mol. Sci. 12 (2011) 42504270.[17] N.J. Hallab, C. Vermes, C. Messina, K.A. Roebuck, T.T. Glant, J.J. Jacobs, J. Biomed.

Mater. Res. 60 (2002) 420433.[18] J. Farack, C.Wolf-Brandstetter, S. Glorius, B. Nies, G. Standke, P. Quadbeck, H.Worch,

D. Scharnweber, Mater. Sci. Eng. B 176 (2011) 17671772.[19] M. Peuster, C. Hesse, T. Schloo, C. Fink, P. Beerbaum, C. von Schnakenburg, Biomate-

rials 27 (2006) 49554962.[20] H. Hermawan, D. Mantovani, Acta Biomater. 9 (2013) 85858592.[21] J. apek, D. Vojtch, Mater. Sci. Eng. C 33 (2013) 564569.[22] H.Y. Zhuang, Y. Han, A.L. Feng, Mater. Sci. Eng. C 28 (2008) 14621466.[23] C.E. Wen, Y. Yamada, K. Shimojima, Y. Chino, H. Hosokawa, M. Mabuchi, Mater. Lett.

58 (2004) 357360.[24] J. apek, D. Vojtch, Mater. Sci. Eng. C 35 (2014) 2128.[25] J.R. Woodard, A.J. Hilldore, S.K. Lan, C.J. Park, A.W. Morgan, J.A.C. Eurell, S.G. Clark, M.

B. Wheeler, R.D. Jamison, A.J. Wagoner Johnson, Biomaterials 28 (2007) 4554.[26] G. Stephani, O. Andersen, H. Ghler, C. Kostmann, K. Kmmel, P. Quadbeck, M.

Reinfried, T. Studnitzky, U. Waag, Adv. Eng. Mater. 8 (2006) 847852.[27] M. Hirschmann, C. Korner, R.F. Singer, Mater. Sci. Forum 539543 (2007)

18271832.[28] G.J. Davies, in: S. Zhen (Ed.), Journal of Materials Science, Springer, Netherlands,

1983, pp. 18991911.[29] C.E. Wen, M. Mabuchi, Y. Yamada, K. Shimojima, Y. Chino, T. Asahina, Scr. Mater. 45

(2001) 11471153.[30] C.S.Y. Jee, N. Ozguven, Z.X. Guo, J.R.G. Evans, Metall. Mater. Trans. B 31 (2000)

13451352.[31] T. Murakami, K. Ohara, T. Narushima, C. Ouchi, Mater. Trans. 48 (2007) 29372944.[32] R. Font, A. Fullana, J.A. Caballero, J. Candela, A. Garcia, J. Anal. Appl. Pyrolysis 58

(2001) 6377.

[33] J. apek, D. Vojtch, L. Tanger, Metal 2011: 20th Anniversary InternationalConference on Metallurgy and Materials, 2011, pp. 903908.

[34] G. Hao, F. Han, W. Li, J. Porous Mater. 16 (2009) 251256.[35] S.J. Hollister, Nat. Mater. 4 (2005) 518524.

[36] D.W. Hutmacher, T. Schantz, I. Zein, K.W. Ng, S.H. Teoh, K.C. Tan, J. Biomed. Mater.Res. 55 (2001) 203216.

[37] H.H. Lu, S.F. El-Amin, K.D. Scott, C.T. Laurencin, J. Biomed. Mater. Res. A 64A (2003)465474.

501J. apek, D. Vojtch / Materials Science and Engineering C 43 (2014) 494501

Microstructural and mechanical characteristics of porous iron prepared by powder metallurgy1. Introduction2. Materials and methods3. Results and discussions3.1. Microstructure and porosity3.2. Mechanical properties

4. ConclusionAcknowledgmentsReferences