Manufacturing Letters 33 (2022) 29–32

Contents lists available at ScienceDirect

Manufacturing Letters
journal homepage: www.elsevier.com/locate/mfglet

Letters

Fused filament fabrication and sintering of 17-4PH stainless steel

Yaozhong Zhang a, Aljoscha Roch a,b,⇑
a
Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI 48823, USA
b
Department of Chemical Engineering and Material Science, Michigan State University, East Lansing, MI 48823, USA

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

Article history: We demonstrate in this paper that Fused Filament Fabrication (FFF) is a reliable and rapid technology for
Received 22 September 2021 manufacturing complex parts made with 17-4PH stainless steel. We printed, sintered, and evaluated the
Received in revised form 12 June 2022 performance of FFF manufactured 17-4PH stainless steel. Our results show that FFF is competitive and
Accepted 28 June 2022
even superior to other manufacturing technologies. It turned out that the mechanical properties are
Available online 9 July 2022
highly dependent on the post-treatment and hardening procedure. The elongation changed from 5 %
9.5 % due to annealing treatments. We found yield strength of 1007 MPa, ultimate strength of
Keywords:
1212 MPa, and density > 99 % after hardening.
Fused filament fabrication
17-4PH stainless steel
Ó 2022 Society of Manufacturing Engineers (SME). Published by Elsevier Ltd. All rights reserved.
Sintering
Mechanical properties
Yield strength
Ultimate strength
Density

1. Introduction such as selective laser melting (SLM) and electron beam melting
(EBM). The FFF belongs to the material extrusion AM techniques
Additive manufacturing (AM) technologies are of increasing and has demonstrated its overall capability in manufacturing
interest to researchers since their flexible design capabilities and sophisticated structures already [3,17–20,21]. The FFF process
broad material selections are especially in line with many indus- itself is straightforward and can be done without certain security
tries’ demands [1,2]. In particular, the successful fabrication of precautions. The metal powder is bound in a polymer matrix. Since
metal components printed with Ti-alloys [3,4], Al alloys [5], Ni the FFF process requires little investment costs (a small printer
alloys [6], pure W [7], or stainless steel [8,9] predicts the great costs about $1.000 USD), the FFF technology is an exciting alterna-
practical potential for AM. tive to more expansive AM technologies like SLM or binder jetting.
The main motivations for AM are saving manufacturing costs A further advantage of the process is that the green parts can be
and expanding the capabilities of manufacturing complex parts. polished directly after printing easily. That can reduce postprocess-
The choice of which AM approach is appropriate depends on sev- ing costs.
eral considerations like costs for raw materials, costs for equip- The printing speed (50–80 mm/s), high material throughput (1–
ment, safety considerations or material performance, and so on. 20 g/min), material efficiency (no powder waste), the design free-
Each AM technique has its advantages and disadvantages. dom for printing even hollow structures, and the competitive
We investigate in this paper the manufacturing process of material properties make FFF a highly compatible manufacturing
17-4PH stainless steel parts made by Fused Filament Fabrication technology for industries like automotive, the aviation and
(FFF). 17-4PH is used for heavy load components in aerospace biomedicines [3,20,21]. Compared to powder-bed technologies,
and medical sectors thanks to its high impact strength, fracture FFF is safe and user-friendly, and no powder can be airborne, which
toughness, and corrosion resistance [10–12]. may cause health issues.
In past years, considerable efforts have been devoted to 17-4PH Since there are just a few articles in literature for printing
fabrication through AM technologies [13–16]. Most AM manufac- 17-4PH by FFF [22,23], we printed, sintered, and evaluated the
turing approaches are focused on powder-bed fusion techniques, material performance of an in-house developed elastomer-based
17-4PH filament. The printed parts were solvent-debound in ace-
tone and sintered/hardened under various conditions. The investi-
⇑ Corresponding author at: Department of Electrical and Computer Engineering,
gated precipitation-hardened steel 17-4PH requires besides
Michigan State University, East Lansing, MI 48823, USA.
sintering also annealing of the material for getting desired material
E-mail address: [email protected] (A. Roch).

https://doi.org/10.1016/j.mfglet.2022.06.004
2213-8463/Ó 2022 Society of Manufacturing Engineers (SME). Published by Elsevier Ltd. All rights reserved.
Y. Zhang and A. Roch Manufacturing Letters 33 (2022) 29–32

properties The tensile properties were recorded, and a proper 3. Results and discussion
annealing strategy was discussed. Finally, the material perfor-
mance was compared with literature values and gear examples In Fig. 1 is shown the cross-sectional images of SS1-, SS2- and
were printed. SS3-samples. Pores are usually < 2 lm for all samples. According
to Archimedes’ principle, the density of all SS1, SS2, and SS3
are > 99.4 % (relative to 7.8 g/cm3).
For SS1 samples (Fig. 1a), the white columns in dark areas can
2. Experiment be assigned to bainite (caused by moderate cooling) or even
martensite (caused by rapid quenching) structures, while the sur-
2.1. Sample preparation rounding needle-like structures, emerging across the surface, are
pearlite structures, a mixture of a-Fe and Fe3C [24,25]. The alter-
The in-house developed filament was derived from a feedstock nating distribution of bainite/martensite/pearlite structures was
consisting of an elastomer-based binder and 64 vol% 17-4PH pow- probably derived from the rapid cooling rate (quenching with
ders (purity > 98%, Chemical Store Inc, spherical shape with parti- 30 l/min N2, S11).
cle size D50 28 lm). The constitution of the binder system was While applying the SS2, the prior structure was re-diffused suc-
reported in previous research [3]. The premixing of the feedstock cessively, and large grains with a-ferrite/Fe3C phases were formed
was carried out by a twin-screw extruder at a temperature of (Fig. 1b). The hardness and tensile strength increased.
around 200–220 °C. A Leistritz 27 mm twin-screw extrusion sys- The additional step in SS3 treatment, annealing at 1040 °C fol-
tem was used. The barrel of the extruder had an aspect ratio of lowed by the treatment at 550 °C, improved the transformation
48. After the 17-4PH powder and the binder were mixed the mate- to the higher ductile a-ferrite (Fig. 1c). The grains are refined and
rial was pelletized and the pellets were extruded into filaments the a-ferrite / Fe3C phases are obvious finely distributed.
with a diameter of 2.85 ± 0.01 mm. We used a Renkforce 2000
FFF 3D printer for printing samples.
The printing parameters for printing test bars were: printing
temperature 210 °C, printing speed 50 mm/s, bed temperature
90 °C, layer thickness 200 lm, nozzle size 400 lm, outer line over-
lap 15 %, infill 105 %, and the infill pattern were parallel lines.
The bars were placed in acetone at 50 °C for 5 h to remove a part
of the binder before the thermal debinding and sintering were exe-
cuted. The duration of the solvent treatment depends on the thick-
ness of the parts. A loss of 3.5–4 wt% is desired by solvent
treatment. The thermal debinding profile was derived from ther-
mal gravimetric analysis (TGA), which has been reported before
[3].
The homogeneity of the powder and the complete removal of
the binder without residues, which might affect the carbon house-
hold in 17-4PH steel, is important to achieve proper material prop-
erties and no cracks or deformation during sintering. The used H2
atmosphere during thermal debinding helped to remove the
organic binder efficiently during the thermal treatment.
After sintering, different annealing strategies were explored for
precipitation of martensite and austenite phases in 17-4PH. The
sintered samples were annealed at 1040 °C and 550 °C respec-
tively. The thermal treatments were: SS1 = sintered at 1360 °C of
1 h and quenched to room temperature by N2 gas flow (30 l/ Fig. 2. (a) in-house developed 17-4PH filament. (b) printed and sintered bars 6 mm
thick. (c) tensile test samples. The width of the tensile test samples is in the gage
min), SS2 = SS1 + annealed at 550 °C for 4 h, SS3 = SS1 + annealed 6 mm. (d) tensile test results.
at 1040 °C for 1 h and annealed at 550 °C for 4 h.

Fig. 1. SEM images of (a) SS1-, (b) SS2-, (c) SS3 sample. Scale bar: 50 lm in large view and 5 lm in the inset.

30
Y. Zhang and A. Roch Manufacturing Letters 33 (2022) 29–32

Table 1
Summary of 17-4PH tensile properties of different technologies (the values measured at MSU have a deviation of 5–8%).

Process Source Yield Str. (MPa) Ultimate Str. (MPa) E-Module Elongation Density Hardness
(GPa) (%) (%)
SS1 - MSU as sintered/1360 °C 900 1161 160 5.9 >99.4 22 HRC
FFF SS3 - MSU hardened 753 1048 131 9.5 >99.4 35 HRC
SS2 - MSU hardened 1007 1212 175 5 >99.4 34 HRC
MIM Ref.27 as sintered/—°C 740 900 190 6 98 27 HRC
Ref.27 hardened-H900 1100 1200 190 4 >97 33HRC

Fig. 3. At MSU printed green parts (left side) and sintered parts (right side). Printing time 40 min. layer height 220 lm. Debinding in acetone for 3 h. Sintering about 8 h
(SS1 - profile). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The SS3 samples show reduced tensile stress but enhanced 4. Conclusion
plasticity.
The annealing seems to have no impact on the density, how- In this paper, we demonstrate the competitiveness of FFF in the
ever, the different annealing treatments have of course an impact additive manufacturing of 17-4PH metal parts. The material per-
on steel phases and the resulting mechanical properties. formance after sintering is very similar to other AM technologies
For the tensile tests the developed 17-4PH filament (Fig. 2a) or traditional manufacturing techniques. We measured a density
was printed and SS1-sintered bars with 120  20  6 mm after sintering of > 99 % and yield strength and ultimate strength
(Fig. 2b) were cut into ASTM E8 conform dog bones (Fig. 2c) with of 1007 MPa and 1212 MPa. These mechanical properties are like
a thickness of 1.5 mm by a hot-wire process. Subsequently, 5 dog conventional MIM 17-4PH materials. The main advantages of the
bones remain untreated (SS1), 5 dog bones were SS2-heat- FFF process are the economic aspects, the simplicity of the technol-
treated, and 5 dog bones were SS3-heat-treated. The closest ogy, and the few required safety standards.
stress-elongation curve to the average was chosen to represent Crucial for part performances in FFF processes are printer set-
the SS1-, SS2- and SS3 stress-elongation curve in Fig. 2d. tings and sintering. Inadequate printer settings may leave gaps
As a comparison to the literature, values are summarized in between printed lines and subsequently in sintered parts. Simi-
Table 1. As shown in Table 1, the yield strength and elongation larly, insufficient sintering and heat treatment can compromise
of SS1, SS2, and SS3 parts are in the range of literature values, while parts’ density. But overall, the FFF process has great potential for
their elastic modulus has an obvious difference probably due to the the AM field and can offer a wide range of materials.
discussed phase- and microstructure differences caused by the The growing number of available FFF filaments makes this tech-
heat treatments. Obviously the as-sintered FFF sample SS1 and nology very competitive and highly interesting for future
the metal injection molding (MIM) material, have different applications.
mechanical properties due to different sintering temperatures
and cooling rates. This is because the desired mechanical proper- Declaration of Competing Interest
ties are only set after sintering by thermal post-treatment.
The metal injection molding procedure is like FFF [27]. Both The authors declare that they have no known competing finan-
technologies deal with debinding and sintering and deliver similar cial interests or personal relationships that could have appeared
results (s. Table 1). to influence the work reported in this paper.
We could not determine the carbon content after debinding and
sintering our FFF samples. A variation of the carbon content due to
Acknowledgments
the absorption of remaining binder residues may have also an
impact on the mechanical properties of 17-4PH.
The authors thank Ed Drown, Adam C Marsh, and Per Askeland
Finally, we printed and sintered a prototype structure
from Michigan State University for their help during the experi-
(Fig. 3). The green part density was 4.89 g/cm3. The shrinkage
mental work.
is 14 %. The shrinkage and final parts geometry of FFF printed
parts can be predicted within < 100 lm [26]. The FFF process
References
is however not flexible in line width during the printing pro-
cess. The chosen nozzle diameter (usually 250–800 lm) printer [1] Wohlers Associates. Wohlers report 2017: 3D printing and additive
settings define the line widths. That means parts geometries manufacturing state of the industry : annual worldwide progress report.
should be designed with dimensions of an integer multiple of (Wohlers Associates, 2017).
[2] Herderick, E. Additive Manufacturing of Metals: A Review. 13.
possible line widths. In particular, when a part has a geometry [3] Zhang Y, Bai S, Riede M, Garratt E, Roch A. A comprehensive study on fused
with sharp edges, small nozzle diameters are required for print- filament fabrication of Ti-6Al-4V structures. Addit Manuf 2020;34:101256.
ing 100 % dense parts. [4] Al-Rubaie KS et al. Machinability of SLM-produced Ti6Al4V titanium alloy
parts. J Manuf Process 2020;57:768–86.

31
Y. Zhang and A. Roch Manufacturing Letters 33 (2022) 29–32

[5] Hu, J. et al. Relation between heat treatment processes and microstructural [16] Wang, X., Liu, Y., Shi, T. & Wang, Y. Strain rate dependence of mechanical
characteristics of 7075 Al alloy fabricated by SLM. Vacuum 177, 109404 (2020). property in a selective laser melted 17–4 PH stainless steel with different
[6] Nagahari T, Nagoya T, Kakehi K, Sato N, Nakano S. Microstructure and Creep states. Mater Sci Eng A 792, 139776 (2020).
Properties of Ni-Base Superalloy IN718 Built up by Selective Laser Melting in a [17] Liu B, Wang Y, Lin Z, Zhang T. Creating metal parts by Fused Deposition
Vacuum Environment. Metals 2020;10:362. Modeling and Sintering. Mater Lett 2020;263:127252.
[7] Xiong Z et al. Selective Laser Melting and Remelting of Pure Tungsten. Adv Eng [18] Allahverdi M, Danforth SC, Jafari M, Safari A. Processing of advanced
Mater 2020;22:1901352. electroceramic components by fused deposition technique. J Eur Ceram Soc
[8] Thompson Y, Gonzalez-Gutierrez J, Kukla C, Felfer P. Fused filament 2001;21:1485–90.
fabrication, debinding and sintering as a low cost additive manufacturing [19] Yang H, Yang S, Chi X, Evans JRG. Fine ceramic lattices prepared by extrusion
method of 316L stainless steel. Addit Manuf 2019;30:100861. freeforming. J Biomed Mater Res B Appl Biomater 2006;79B(1):116–21.
[9] Huang Y et al. Microstructure and wear properties of selective laser melting [20] Zhang Y, Poli L, Garratt E, Foster S, Roch A. Utilizing Fused Filament Fabrication
316L. Mater Chem Phys 2020;254:123487. for Printing Iron Cores for Electrical Devices. 3D Print. Addit Manuf
[10] Kc, S. et al. Tribological behavior of 17–4 PH stainless steel fabricated by 2020;7:279–87.
traditional manufacturing and laser-based additive manufacturing methods. [21] Marsh, A. C. et al. 3D printed bioactive and antibacterial silicate glass-ceramic
Wear 440–441, 203100 (2019). scaffold by fused filament fabrication. Mater Sci Eng C 118, 111516 (2021).
[11] Caballero A, Ding J, Ganguly S, Williams S. Wire + Arc Additive Manufacture of [22] Wu, G., A. Langrana, N., Sadanji, R. & Danforth, S. Solid freeform fabrication of
17–4 PH stainless steel: Effect of different processing conditions on metal components using fused deposition of metals. Mater Des 23, 97–105
microstructure, hardness, and tensile strength. J Mater Process Technol (2002).
2019;268:54–62. [23] Godec D, Cano S, Holzer C, Gonzalez-Gutierrez J. Optimization of the 3D
[12] Schaller RF, Taylor JM, Rodelas J, Schindelholz EJ. Corrosion Properties of Printing Parameters for Tensile Properties of Specimens Produced by Fused
Powder Bed Fusion Additively Manufactured 17–4 PH Stainless Steel. Filament Fabrication of 17–4PH Stainless Steel. Materials 2020;13:774.
CORROSION 2017;73(7):796–807. [24] Ozyurek D, Nalcacioglu E, Cetinkaya K. Microstructure and Wear Behaviour of
[13] Lee HJ, Dao VH, Ma YW, Yu JM, Yoon KB. Effects of process parameters on the 17–4 Precipitation Hardening Stainless Steel with Various Ti Content. Powder
high temperature strength of 17–4PH stainless steel produced by selective Metall Met Ceram 2020;59:386–93.
laser melting. J Mech Sci Technol 2020;34:3261–72. [25] Bai Y, Chaudhari A, Wang H. Investigation on the microstructure and
[14] Alnajjar M et al. Influence of microstructure and manganese sulfides on machinability of ASTM A131 steel manufactured by directed energy
corrosion resistance of selective laser melted 17–4 PH stainless steel in acidic deposition. J Mater Process Technol 2020;276:116410.
chloride medium. Corros Sci 2020;168:108585. [26] AM Extrusion. AM Extrusion https://am-extrusion.com (2021).
[15] Alnajjar, M., Christien, F., Bosch, C. & Wolski, K. A comparative study of [27] German, R. M. Metal powder injection molding (MIM): Key trends and
microstructure and hydrogen embrittlement of selective laser melted and markets. in Handbook of Metal Injection Molding 1–21 (Elsevier, 2019).
wrought 17–4 PH stainless steel. Mater Sci Eng A 785, 139363 (2020). doi:10.1016/B978-0-08-102152-1.00001-5.

32