Effect of hot isostatic pressing treatment on porosity reduction and mechanical properties enhancement of 316L stainless steel fabricated by binder jetting

ABSTRACT
 In this work, a hot isostatic pressing (HIP) treatment was used to reduce the pores and improve the mechanical properties of binder jetting (BJ) fabricated 316L stainless steel. The relative density of the HIP treated samples sintered under vacuum, nitrogen, and argon atmospheres increased from 89.20∼90.76% to 94.02∼98.35%. The mean diameter of the internal closed pores decreased from 11.83∼14.70 μm to 0.94∼1.53 μm, and the internal porosity reduced from 5.47∼8.67% to 0.14∼0.35%. A remarkable enhancement of mechanical properties was achieved with the tensile strength by ∼15% and the elongation by ∼100% due to the reduction of porosity. The mechanical properties of the BJ 316L sintered under nitrogen and treated with HIP were even better than those of the direct powder HIP components due to the interstitial solution strengthening. These findings will be a valuable reference for optimising the HIP parameters to produce metal BJ components with superior mechanical properties.


Introduction
Metal binder jetting (BJ) has recently attracted tremendous interest from academia to industry due to its strong potential for high-speed and low-cost manufacturing (Mostafaei et al. 2021).In BJ process, the printhead (often with a width of several centimetres) selectively jets billions of liquid binder droplets per second to glue the particles and form the 2D geometry, which is then built up layer by layer to fabricate green metal parts (Sachs et al. 1993).Metal parts with complicated geometries could be produced in a batch way without moulds when combined with the subsequent curing and sintering process (Ziaee and Crane 2019).Although metal BJ has the advantages of a high production rate, support-free printing, and low cost, one of the main challenges is the high residual porosity of the sintered parts, which deteriorates the end-use performance and limits the application of the metal BJ technique (Li et al. 2020).
There have been many attempts to improve the sintered density of BJ parts, including improving the green density by using bimodal powders (Du et al. 2021;Bai, Wagner, and Williams 2017), optimising printing parameters (mainly layer thickness and binder saturation) (Lecis et al. 2021), ameliorating sintering curves (Huber, Vogel, and Fischer 2021), etc.By using bimodal powders, Bai et al. increased the relative density of BJ copper green parts from 44% to 53.4% (increased by 9.4%) and the corresponding sintered density by 12.3% (Bai, Wagner, and Williams 2017).Lexis et al. optimised layer thickness (100 μm to 50 μm) and binder saturation (70% to 50%) to increase the sintering density of BJ 316L from 95.78% to 98.33% (Lecis et al. 2021).Huber et al. improved the sintering density of BJ 17-4PH stainless steel to 98.5∼98.7% by optimising the sintering temperature (Huber, Vogel, and Fischer 2021).However, these methods are difficult to fundamentally eliminate the layered porosity formed by the lack of compaction during powder spreading (Kumar et al. 2019) and the impact of binder droplets on the powder bed (Parab et al. 2019).Generally, these processes result in sizeable residual pore size and irregular shape, which have a negative impact on the mechanical properties of BJ sintered parts.Another effective way to obtain full-density parts is to infiltrate the sintered skeleton with lower melting metals.However, this method is only suitable for a few material systems (e.g.stainless steel & bronze) due to the strict requirements for the match of melting point and interface wettability between two materials (Huang et al. 2022;Lu et al. 2020).Moreover, the obtained material from this method is not a singlephase alloy, which is not expected in specific engineering applications (German 2019).Therefore, finding a general processing scheme that can obtain full density BJ metal parts without sacrificing the material homogeneity or process scalability is urgent.
Hot isostatic pressing (HIP) is a technique that applies high isostatic pressure (hundreds of MPa) to a closed object through an inert gas at a high temperature (Atkinson and Davies 2000).High temperature and high pressure during the HIP process can accelerate the sintering of powders by overcoming the surface-energy driving force for pore closure.Therefore, HIP can homogenise the microstructure, reduce porosity, and in turn improve strength (Teng et al. 2022;Cai et al. 2021).HIP was initially used to consolidate loose powders and later widely used for post-processing castings (Atkinson and Davies 2000).Recently, the HIP technique has also been widely used in additively manufactured parts, such as electron beam melting (Tammas-Williams et al. 2016), laser powder bed fusion (Sun et al. 2021), and cold spray additive manufacturing (Chen et al. 2019).
In recent years, several researchers have investigated the convenience of HIP treatment on the metal BJ technique.For example, Yegyan et al. explored the effect of HIP treatment on the density and mechanical properties of BJ fabricated pure copper (Kumar et al. 2018).They reported that the density and mechanical properties were greatly improved with increased relative density from 90.52% to 97.32% and enlarged tensile strength from 144 MPa to 176 MPa after HIP treatment (Kumar et al. 2018).316L austenitic stainless steel is one of the most widely used alloys in additive manufacturing.However, there are no detailed reports on microstructure changes and tensile mechanical properties of BJ 316L parts treated by the HIP.Some significant aspects need to be explored further: (I) Metal BJ has a higher porosity and even larger pore size than other AM methods.Whether these pores can be effectively eliminated by HIP is still unknown.(II) The HIP process involves pore reduction (or elimination) and grain size growth, which have opposite effects on the material's mechanical properties.Therefore, mechanical properties change after HIP treatment needs to be studied.
This work aims to evaluate the effect of HIP treatment on the density, pore features, grain size, and mechanical properties of BJ 316L stainless steel components sintered under vacuum, nitrogen, and argon atmospheres.Additionally, the differences in the microstructure and mechanical properties of the HIP treated BJ 316L parts with the direct powder HIP treated 316L were compared.These findings would be a valuable reference for optimising the HIP parameters to produce metal BJ components with superior mechanical properties.

Samples preparation
The process schematic and experimental variables are shown in Figure 1.Firstly, 316L green parts (cubical 76 × 28 × 18 mm 3 and 15 × 12 × 8 mm 3 ) were printed on an Easy-II BJ machine (Wuhan Easy Co. Ltd, Wuhan, China) with the following parameters: layer thickness of 100 μm, inkjet concentration of 30%, and resolution of 360 × 1080.A gas atomised 316L powder (Zhejiang Asia General Soldering & Brazing Material Co. Ltd, Hangzhou, China) and a homemade binder were used to print BJ green parts.More detailed material properties were illustrated in previous work (Mao et al. 2021).
Then, the samples were sintered in a GSL-1600X tube furnace (Hefei Kejing Materials Technology Co., Ltd., Hefei, China) at 1200°C with a soaking time of 2 h under a vacuum atmosphere for de-binding, subsequently sintered at 1400°C for 3h under vacuum, nitrogen, and argon atmospheres, respectively to densify the samples.The as-sintered BJ 316L samples with different atmospheres were designated as Vu-AS, N 2 -AS and Ar-AS, respectively.
The final step was hot isostatic pressing of the sintered part.The HIP treatment was conducted on a QIH-15 HIP machine (ABB Ltd., America) at 1150⍰ with a pressure of 130 MPa for 4h.These HIP parameters were selected based on the proper conditions to eliminate the internal pores of 316L produced by other techniques (Cooper et al. 2016;Lavery et al. 2017).As the flow stress of austenitic stainless steel is about 30 ∼ 80 MPa at 1150°C (Samantaray et al. 2009), it is theoretically enough to fully deform internal pores with 130 MPa.The samples sintered under different atmospheres after capsule-free HIP treatments were referred as Vu-HIP, N 2 -HIP and Ar-HIP, respectively.Note that the sintered samples (Vu-AS, N 2 -AS and Ar-AS) were not sealed inside the hermetic capsules before capsule-free HIP treatment, since most of the pores were isolated and not connected to the surface of the material.To further investigate the microstructure and mechanical properties difference between BJ-HIP and traditional powder-HIP process, the capsule method was used to prepare BT-HIP samples.Firstly, the same 316L powders was filled into a stainless-steel capsule.Then, the capsule was degassed by a vacuum pump to remove air and sealed by closing the evacuation tube.Finally, the sealed capsule was sent for HIP treatment.Fig. S1 displayed a schematic of the capsule and capsule-free method HIP.
The metallographic specimen and tensile samples of sintered and HIP treated 316L parts were cut by a wire electrical discharge machining.The metallographic observation plane was parallel to the build direction (Z-axis).The metallographic samples were prepared on an automatic grinding and polishing machine (Ecomet300/Automet300, Buehler, USA).The electrolytic process was conducted on mechanically polished samples to prepare samples for electron backscatter diffraction (EBSD) characterisation with an A2 solution.

Characterisation
The density of sintered and HIP treated samples was determined by Archimedes' methods (Mao et al. 2021).The dimension of samples before and after HIP treatment was measured by a calliper (0.01 mm accuracy).Then the linear shrinkage was calculated as (L-L 0 )/L 0 , where L 0 is the size of sintered samples and L is the size of HIPed samples.The tensile testing was performed on an AG-100KN machine (Shimadzu, Japan) with a 2 mm/min strain rate.The elongation was measured and calculated by recording the tagged parallel section length of the tensile specimen.At least four samples were used in density, shrinkage, and strength measurement for each parameter set.
The porosity of the sintered samples was observed by optical microscope (OM, Axio lab A1, Carl Zeiss, Germany).Image Pro Plus V6.0 software was used to analyse pores' size, shape, and quantity (Schneider, Rasband, and Eliceiri 2012).The definition of roundness was P 2 /(4π*A), where P is the perimeter, and A is the pore area (Singh et al. 2021).The mean diameter was the average length of diameters measured at 2-degree intervals passing through the pores' centroid (Singh et al. 2021;Mao et al. 2023).The fracture morphologies of the tensile sample were observed by a scanning electron microscope (SEM, FEI Quanta 650, FEI, America).The EBSD measurements were conducted on the Gemini SEM300 (Carl Zeiss, German) equipped with an Aztec Data Collection system.The nitrogen content was tested on a G8 GALILEO elemental analyzer (Bruker, Germany).

Bulk density
The bulk density measurement was conducted to evaluate the effect of HIP treatment on the porosity reduction of sintered BJ 316L samples.As shown in Figure 2(a), the relative densities of all sintered 316L samples increased after the HIP treatment (approximately 90% to 94∼98%).However, the relative densities sintered in different atmospheres and treated with HIP showed certain variations.The N 2 -HIP samples had the highest relative density improvement (from 89.38% to 97.69%, an increment of 8.31%), while Ar-HIP had the lowest enlargement (4.83%).The reasons for this disparity will be analysed in the following sections.Since HIP treatment has brought a distinguishable porosity reduction, it is necessary to examine the dimensional changes after HIP treatment.Figure 2(b) indicates that all samples experienced around 3% linear shrinkage in all three axes, consistent with the pore volume reduction (4∼8 vol.%) after HIP treatment.It is also noteworthy that the dimensional change in the Z direction was the highest in all sintering atmospheres.Several researchers have observed this phenomenon in metal BJ and believe it is related to the low compaction effects of rollers during the powder spreading process and the ballistic ejection of fine powder hit by the binder droplet (Zhu et al. 2020).

Pore features
The pore morphology variations of argon-sintered 316L before and after HIP treatment were observed via OM.As shown in Figure 3, the black areas correspond to the pores.The 'near surface' observation plane was on the surface of the sample, and the 'far surface' observation plane was approximately 5 mm from the surface.The schematics of the 'near surface' and 'far surface' observation planes are shown in Fig. S2.The pores exhibit the following two changes after HIP treatment.(I) Extensive porosity defects near the sample surface were not eliminated after HIP treatment, and these surface-connected pores were distributed along layer interfaces as illustrated in Figure 3 S3.Pores were barely visible due to the low magnification of OM.The above phenomenon was due to the capsule-free HIP used in this work, and the high-pressure gas cannot effectively act on materials with surface-connected pore regions.Kumar et al. found that the relative density of sintered BJ copper after HIP treatment only increased from 83.6% to 85.8% (Kumar et al. 2019).They concluded that the surface-connected pores were impossible to eliminate through the capsulefree HIP.The pore-reduce efficiency of HIP treatment for sintered BJ metal components could be increased by increasing the green density (e.g.bimodal powder mixtures) (Kumar et al. 2018).
To quantitatively evaluate the feature change of internal closed pores of the sintered parts after HIP treatment, the OM images of polished samples were analysed by ImagePro software.The statistical data are shown in Table 1.The results showed that the pore size, roundness and porosity sintered under three atmospheres were significantly reduced after HIP treatment.The pore number was slightly reduced.Taking argon-sintered samples as an example (Figure 4), the porosity was reduced from 8.67% to 0.35% (data obtained by OM analysis), the number of pores was reduced from 2883/mm 2 to 2583/mm 2 , the roundness was decreased from 1.68-1.16,and the mean pore diameter was significantly reduced from 13.18 μm to 1.38 μm after HIP treatment.These results demonstrated that HIP treatment reduced porosity in sintered BJ 316L primarily by reducing the volume of individual pores but not by reducing the pore number.This was because the material surrounding the closed pores began to creep and collapse due to the high hydrostatic pressure generated by inert gas during the HIP process, resulting in decreased pore volume.Furthermore, due to the high HIP processing temperature (1150°C), atoms will be transported to concave regions of small radius curvature from those where the surface is more gently rounded through grain boundaries and surfaces.As a result, irregularly shaped pores spheroid (Atkinson and Davies 2000).
Table 1 also revealed that the Ar-HIP samples have the highest porosity (0.35%), while the Vu-HIP samples have the lowest (0.14%).This can be explained by the solubility difference of the sintering atmosphere in the 316L matrix.The internal pressure of the gas will increase as the pore volume shrinks until reaching equilibrium with the HIP pressure (130 MPa).As argon atoms are too large to be dissolved in the austenitic matrix, the porosity of Ar-HIP samples was the largest when the gas pressure reached 130 MPa.Because nitrogen atoms could be partially dissolved in the austenite (up to 0.31  wt.% at 1150⍰ (Valente et al. 2021)), the porosity of N 2 -HIP was lower.Furthermore, the initial gas pressure in Vu-AS samples was negative and can be easily compressed.
The change in pressure due to the temperature and volume during HIP can be approximated using the ideal gas law (Tammas-Williams et al. 2016): where P, V, n, and T are the pressure (MPa), volume (m 3 ), number of moles, and temperature (K) of pores, respectively.R is the universal gas constant (8.134JK −1 mol −1 ).Since argon atoms cannot be dissolved in 316L, it is assumed that the number of moles remains constant during sintering and HIP treatment, so the value P*V/T is fixed.The internal pressure of the closed pores of the Ar-AS sample at 1673 K was assumed as 0.1 MPa (atmospheric pressure).According to formula (1), the pressure changed to 0.08616 MPa at the HIP treatment temperature (1423 K).The relationship between pressure and pore volume was calculated in Figure 5 with constant T at 1423K.As the HIP pressure increases, the pore diameter decreases exponentially.However, even when the HIP pressure reaches 130 MPa, the pores are only shrunk to a minimal size rather than entirely eliminated.Taking the max pore diameter (28.84 μm) of the argon sintered component as an example, when the internal pressure of the pore reached the equilibrium state of 130 MPa, the calculated pore diameter shrinks to 2.51 μm, the corresponding volume drops sharply from 12560 μm 3 to 8.32 μm 3 , and the volume shrinkage is as high as 99.93%.The experimental data and estimated pore shrinkage indicate that the HIP treatment of BJ 316L can reduce porosity and formula (1) can be used to predict how pore size will change with pressure.

Grain morphology
Since the HIP is a long-term high-temperature (1150°C) treatment process, it can be expected that grains will inevitably grow up.Table 2 summarises the grain size enlargement of BJ 316L sintered under different atmospheres after HIP treatment.Figure 6 displayed Inverse  that grain size grows significantly from 69.74 μm to 287.39 μm after HIP treatment (nearly tripled).Such results could be attributed to porosity reduction by high-temperature HIP treatment, which reduces the hindrance effect on grain boundary migration.Although the temperature of HIP treatment was much lower than the sintering temperature (1150⍰ vs 1400⍰), the grain size after HIP treatment was much larger than that after sintering.Considering only the temperature, atoms were more active in higher temperatures, leading to considerable grain growth at 1400⍰.However, the result was that the grain growth at 1400⍰ was slower because many pores on the grain boundary play a role in the pinning effect and hinder the grain growth.
During HIP treatment, the internal closed pore size quickly decreased and the pinning effect was lowered.Figure 7 depicts the distribution of grain boundary misorientation angles and the corresponding grain boundary (GBs) map of vacuum sintered and HIPtreated samples.In the Vu-AS and Vu-HIP samples, enormous twin boundaries (TBs) with a misorientation angle of 60°were formed.This is due to the low stacking fault energy of 316L austenitic stainless steels, which results from the face-centred cubic lattice structure.Generally, annealing twins easily in 316L alloy during recrystallization decreases the system's free energy.Furthermore, EBSD statistics revealed that the relative frequency of coincidence site lattice (CSL) boundaries (BGs = 60°) was as high as 56.8%-58.7%.Because of their low interfacial energy and ability to prevent crack propagation, these particular GBs were beneficial for mechanical properties (Sinha et al. 2015;Kumar et al. 2007).Moreover, the fraction of high-angle grain boundaries (HAGBs) increased  after HIP treatment (from 87.4% to 97.7%).It is expected that increasing the percentage of HAGBs will improve the strength.When a fracture spreads to the front of a HAGB, more energy must be expended and overcome than when the crack spreads to a low-angle grain boundary (LAGB) (Li et al. 2022;Kumar et al. 2007).

Mechanical properties
As mentioned in the above sections, the significant reduction of the porosity will contribute to the improvement of mechanical properties of BJ 316L after HIP treatment, while the significant grain size coarsening would reduce the strength.To evaluate the combined influence of these two conflicting effects on the mechanical properties, we performed room-temperature tensile tests on the sintered and HIP treated samples.
Figure 8 shows the room temperature tensile properties (ultimate tensile strength (UTS), yield strength (YS), elongation at break(ε ab )) of all the as-sintered and HIP treated samples.The detailed tensile testing results are summarised in Table 3.By conducting the HIP treatment on sintered BJ 316L components, it is noteworthy that the strength and elongation were significantly improved in all sintering atmospheres.Compared to the tensile properties of the Ar-AS sample, the UTS of the Ar-HIP sample increased from 454.83 MPa to 527.70 MPa, coupled with a remarkable ε ab enhancement from 33.19% to 68.71%.This is to say, the ε ab and UTS of the HIP treated sample has increased by 107% and 16% relative to that of the as-sintered sample, respectively.These outcomes can be explained by increased densification and interparticle contact area by removing remaining pores through diffusion and welding-up during the HIP process (Chen et al. 2019).It is well known that the pores inside the material have passive effects on the mechanical properties of BJ metal: The pores may lead to a decrease in loading area, stress concentration, and crack initiation and propagation (Yuan, Zhang, and Ma 2019).The influence of porosity on the mechanical properties (UTS, YS, ε ab , etc.) of porous metal can be calculated with the power law function (Yuan, Zhang, and Ma 2019) In formula (2), S is the mechanical property of sintered porous metal, S 0 is the mechanical property of corresponding dense metal, ρ and ρ 0 is the relative density of sintered metal and dense metal, respectively.χ refers to the porosity of sintered metal.The index m is usually between 3 and 4.5.The specific value of m depends on the sintering state and mechanical property index (Yuan, Zhang, and Ma 2019;German 2014;Gilmas et al. 2016).According to formula (2), the mechanical properties of BJ 316L will increase exponentially (0.9 m VS 0.99 m ) with the decrease of porosity after HIP treatment.The yield strength (σ y ) and ultimate tensile strength (σ UTS ) of dense austenitic stainless steel can also be predicted by the following empirical formula (Cooper, Brayshaw, and Sherry 2018), which mainly includes the influence of content of alloy element and grain size: In formula (3) and formula (4), the strength unit is MPa, (*) is the percentage mass ratio of an element, and d is the average grain size (mm).Based on formula (3), formula (4), and Table 2, it can be seen that the yield strength and ultimate tensile strength of BJ 316L sintered in vacuum were reduced by∼14 MPa and∼27 MPa, respectively, which was due to grain coarsening.The decrease in mechanical properties caused by grain growth was much smaller than the increase in mechanical properties after HIP treatment (YS of 54 MPa, UTS of 90 MPa).Compared to grain coarsening, the results showed that porosity reduction was the most essential factor in determining mechanical properties.
Figure 9 shows the SEM images of the tensile fracture of BJ 316L before and after HIP treatment under vacuum sintering.It can be found that the cross-section area of Vu-AS specimen after tensile test (the cross section size before tension is 3*1.3 mm 2 ) was larger than Vu-HIP, which was caused by insufficient deformation during tension (Figure 9(a) and (c)).A large number of pores in Vu-AS specimens act as additional crack initiation and nucleation points in plastic deformation.During tensile deformation, microcracks were grown rapidly and converged into networks, leading to early failure of the Vu-AS specimens.However, the HIP treatment reduced the initial volume density of the crack initiation point in the matrix and increased the distance between pores.The adjacent pore polymerisation needs to experience more obvious pore growth.Therefore, pore polymerisation of the Vu-HIP sample occurred under higher plastic deformation.Figure 9 shows that before HIP treatment, a large number of pores and a few dimples could be observed on the fracture surface.After HIP treatment, almost no pores were observed on the tensile fracture surface, and dimples became deeper and larger.This indicates that materials experience great deformation and have good plasticity in the tensile process.

Comparison of microstructure and mechanical properties between BJ-HIP and BT-HIP
The differences in tensile properties between BJ-HIP (Vu-HIP, N 2 -HIP, and Ar-HIP) 316L and direct powder HIP (BT-HIP) were also compared.The same batch of 316L powders was used in the experiment, which neglected the influence of the initial alloy composition on the mechanical properties.The HIP parameters were also the same and achieved a similar low density (Fig. S3).As shown in Figure 8(b), the yield strength, ultimate tensile strength and elongation of BT-HIP samples were 305.73 ± 2.03 MPa, 641.90 ± 1.27 MPa and 51.85 ± 0.19%, respectively.The strength of BT-HIP was higher than those of Vu-HIP and Ar-HIP but lower than those of N 2 -HIP.
The high strength of BT-HIP may originate from grainboundary strengthening.According to the EBSD data statistics shown in Figure 10(a) and (b), the average grain size was 112.27 μm (60% diameter was less than 30 μm), which was significantly smaller than the BJ-HIP (287.39 ∼ 329.61 μm).Based on formulas (3) and (4) and  Table 2, it can be seen that the yield strength and ultimate tensile strength of BT-HIP samples were increased by∼8 MPa and∼16 MPa, respectively, which was due to grain refining.Another important phenomenon is the grain boundary characteristics of the BT-HIP samples.As shown in Figure 11, the relative frequency of CSL boundaries (BGs = 60°) was as high as 60.4%.The total number of CSL boundaries in Figure 11(a) clearly outnumbers those in Figure 11(b).Due to the low interfacial energy in these special grain boundaries, crack propagation could be hindered by destroying the connectivity of the original grain boundaries.Therefore, the mechanical properties were obviously improved in BT-HIP samples.
The excellent mechanical properties of N 2 -HIP were most likely derived from the nitrogen-sintering environment.During sintering, the nitrogen element was partially dissolved and existed as interstitial atoms in the austenitic lattice.The mechanical properties of 316L were comprehensively improved through interstitial solid solution strengthening (Hu et al. 2019).In this work, the nitrogen content in N 2 -AS and N 2 -HIP samples was measured as 0.195 wt.%.Fig. S7 demonstrates that the nitrogen element is distributed uniformly.According to formula (3) and formula (4), the yield strength and final tensile strength increment caused by the nitrogen element can reach 96 and 252 MPa, respectively.Valente et al. reported that the solubility of nitrogen element in 316L austenitic matrix could be up to 0.31 wt.% (Valente et al. 2021).
Table 4 summarises the tensile properties of 316L stainless steel formed by different processing methods, which indicates that the mechanical properties of BJ 316L after HIP treatment could be matched with most manufacturing methods.It is worth noting that the tensile properties of the N 2 -HIP samples were superior to those fabricated by metal injection mollding, wire arc additive manufacturing, casting method, etc.

Conclusions
In this work, we demonstrate HIP treatment to effectively reduce internal closed porosity by decreasing pore size.Remarkable enhancement of mechanical properties was achieved due to the significant porosity reduction.The influence of HIP treatment on the density, geometric change, pore morphology, grain coarsening, as well as room temperature tensile properties of BJ 316L sintered under different atmospheres (vacuum, nitrogen, and argon) was systematically evaluated.Three main conclusions are drawn as follows: (1) HIP treatment can effectively reduce the closed internal porosity of different atmosphere sintered BJ 316L parts by compressing single pores volume (up to 99.93 vol.%).However, HIP treatment has little effect on the surface connected pores.
(2) The tensile properties of BJ 316L were significantly improved after HIP treatment by reducing porosity, minimising pore size and rounding pores.Compared to porosity reduction, grain coarsening was less critical in determining mechanical properties.
(3) The yield strength and ultimate tensile strength of the N 2 -HIP sample were higher than those of the direct powder HIP sample, which confirms the ability of BJ to form high-performance austenitic stainless steel parts.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This

Figure 1 .
Figure 1.Schematic of the experimental process and variables of HIP treated BJ 316L.
(b)(f)(h).(II) The closed pores inside the sample were significantly reduced after HIP treatment as shown in Figure 3(c)(e)(g).The schematic diagram of surface-connected pores and closed-pores is shown in Fig.

Figure 2 .
Figure 2. Effect of HIP treatment on (a) density and (b) linear shrinkage of BJ 316L, which sintered under three atmospheres (vacuum, nitrogen, and argon).

Figure 3 .
Figure 3. Pore morphology variation of as-sintered and HIP treated BJ 316L samples under argon.(a)(d) The optical microscope (OM) photograph of the Ar-AS sample showed numerous pores in the cross-section.(b)(e)(f) The OM photograph of the near surface of the Ar-HIP sample showed a significant reduction in pore areas but a large number of pores remaining near the epidermis.(c)(g)(h) The OM photograph of the far surface of the Ar-HIP sample, a few pores were still observed near the epidermis and the internal pores have almost disappeared.

Figure 4 .
Figure 4. OM images (a) (b) before and (c) (d) after HIP treatment sintered under argon atmosphere and corresponding pore size distribution analysis.

Figure 5 .
Figure 5.Comparison of calculated and experimental pore diameter change after HIP treatment.

Figure 6 .
Figure 6.IPF of (a) (b) vacuum sintered and (c) (d) HIP treated BJ 316L and corresponding grain size distribution.

Figure 7 .
Figure 7. Grain boundary misorientation angle distribution of (a) vacuum sintered and (c) HIP treated BJ 316L and corresponding grain boundary map.

Figure 8 .
Figure 8. Improvements in mechanical properties of sintered BJ 316L after HIP treatment.

Figure 10 .
Figure 10.EBSD IPF images and grain size distribution of BT-HIP and N2-HIP samples.

Figure 11 .
Figure 11.Grain boundary misorientation angle distribution of (a) BT-HIP and (b) N2-HIP samples and corresponding grain boundary map.

Table 1 .
Change of pore features of BJ 316L prior to and after HIP treatment with different sintering atmospheres.

Table 2 .
BJ 316L grain size changes after HIP treatment when sintered in various atmospheres.

Table 3 .
Statistics of tensile properties of BJ 316L after HIP treatment.

Table 4 .
Mechanical properties of 316L stainless steel formed by different processing methods.the Analysis and Testing Centre of Huazhong University of Science and Technology for the tests.
work was supported by the Hubei Key Research and Development Programme (2020BAB049), the Wuhan Science and Technology Project (2020010602012037), and the Development of Strategic Emerging Industries Project of Ronggui Street 2020.Yiwei Mao is a Ph.D. candidate in the School of Materials Science and Engineering at the Huazhong University of Science and Technology.He is engaged in the research of metal binder jetting additive manufacturing.Jiaming Yuan is a master's degree candidate in the School of Materials Science and Engineering at the Huazhong University of Science and Technology.He is engaged in binder jetting experimental platform design and assembly.Yuhua Heng is currently a Ph.D. candidate in the School of Materials Science and Engineering at the Huazhong University of Science and Technology.She is engaged in metal binder jetting and post-treat.Kunhao Feng is currently a Ph.D. candidate in the School of Materials Science and Engineering at the Huazhong University of Science and Technology.He is engaged in ceramic and metal binder jetting additive manufacturing.Daosheng Cai is an Associate Professor in the School of Materials Science and Engineering at the Huazhong University of Science and Technology.Moreover, he is also the chairman of Wuhan Easy Co. Ltd.He is committed to developing and promoting the application of binder jetting technic.Qingsong Wei is a professor in the School of Materials Science and Engineering at Huazhong University of Science and Technology.He studies on additive manufacturing and near-net shaping.