Ti Hf Research Articles

Microstructure and Thermoelastic Martensitic Transformation in Ni-Low and -Rich Ni–Ti–Hf–Nb Hightemperature Shape Memory Alloys

Microstructure and Thermoelastic Martensitic Transformation in Ni-Low and -Rich Ni–Ti–Hf–Nb Hightemperature Shape Memory Alloys

Development of an Al–Ti–Hf Composite Alloy Strengthened with High Volume Fraction of In-situ formed Al3(Ti, Hf)-Type Trialuminide Intermetallic Phase

Development of an Al–Ti–Hf Composite Alloy Strengthened with High Volume Fraction of In-situ formed Al3(Ti, Hf)-Type Trialuminide Intermetallic Phase

Functional Properties of a Ni-rich Ni–Ti–Hf Shape Memory Alloy Fabricated via Laser Beam Powder Bed Fusion—Impact of Porosity and Precipitation Characteristics on the Thermal Hysteresis

Additive manufacturing (AM) is very promising for the fabrication of complex parts made from shape memory alloys (SMAs). In the present study, a Ni–Ti–Hf shape memory alloy has been processed by laser beam powder bed fusion of metals (PBF-LB/M). Employing different sets of processing parameters, i.e., a variation of scanning speed, specimens characterized by various microstructures and porosities were obtained. Microstructural analysis revealed that processing of Ni–Ti–Hf SMAs at a low energy level promotes the formation of a fine-grained microstructure with numerous lack of fusion defects. Transmission electron microscopy (TEM) studies revealed that H-phase precipitates are present in any case, leading to an increase in the Ms-temperature and, thus, a phase transformation at room temperature. The thermal stability of the phase transformation behavior in different Ni–Ti–Hf as-built conditions was studied using differential scanning calorimetry (DSC). DSC analysis showed that specimens with a higher defect density are characterized by smaller thermal hysteresis.

Effects of Zr and Hf on superelasticity, shape memory effect and microstructure of β-type (Ti–Zr–Hf)–Nb–Sn multi-principal element alloys with low magnetic susceptibility

(Ti–Zr–Hf)(97−x)–xNb–3Sn (Ti:(Zr + Hf) = 1:1) alloys were designed for biomedical superelastic alloys with magnetic resonance imaging (MRI) compatibility. The effects of Zr and Hf on constituent phases, superelasticity, shape memory effect and magnetic susceptibility were clarified. The specific Nb content for superelasticity and shape memory effect was also explored. It was found that the (Ti–Zr)–6.5Nb–3Sn alloy exhibited a superelastic recovery strain of 5.2 % with lower magnetic susceptibility compared to conventional Ti–Ni and Ti–Nb alloys. The magnetic susceptibility decreased with increasing Hf content. The (Ti–Hf)–9Nb–3Sn alloy showed a superelastic recovery strain of 0.9 % and very low magnetic susceptibility less than half of that of the conventional alloys. Based on the lattice parameters of these alloys, it was found that these alloys exhibited larger transformation lattice strain between a β phase and an α” phase than those of the Ti–Nb alloys, suggesting high potential to exhibit a large superelastic recovery strain. According to orientation analyses of constituting grains, the larger superelastic recovery strain in the (Ti–Zr)–6.5Nb–3Sn alloy than that of the (Ti–Hf)–9Nb–3Sn alloy is considered to be caused by a strong recrystallization texture of {001}β <110>β. By optimization of heat-treatment temperature, superelastic recovery strain of 3.6 % was achieved in the (Ti–0.5Zr–0.5Hf)–7.5Nb–3Sn alloy.

Superconductivity with a high upper critical field in an equiatomic high-entropy alloy Sc–V–Ti–Hf–Nb

High-entropy alloy (HEA) superconductors have attracted significant attention due to their exceptional low-temperature mechanical and superconducting properties. We report the synthesis and thorough characterization of an equiatomic HEA superconductor with the composition Sc0.20V0.20Ti0.20Hf0.20Nb0.20, crystallizing in a body-centered cubic crystal structure (Im3m¯). Our investigation, using magnetization, transport, and heat capacity measurements, reveals the presence of weakly coupled, fully gapped superconductivity with a transition temperature of 4.17(3) K and the upper critical field exceeding the Pauli paramagnetic limit. The metallic nature, combined with a high upper critical field, positions it as a promising candidate for applications in superconducting devices.

In-depth S/TEM observation of Ti–Hf and Ta–Hf-doped Nb3Sn layers

In superconducting Nb3Sn layers with coherence lengths of approximately 3 nm, grain boundaries act as effective pinning sites. Thus, grain refinement is an essential issue that directly affects the superconducting critical characteristics of the Nb3Sn layer. In recent years, Hf addition to Nb3Sn wires co-doped with Ta has attracted notable interest as a method that enables grain refinement down to several tens of nm. In-depth characterization of the Nb3Sn grain morphology in Hf-doping is crucially important to correlate the microstructure with the flux pinning characteristics. In this article, the grain morphologies of Ti–Hf and Ta–Hf-doped Nb3Sn layers were clarified by scanning transmission electron microscopy (STEM) and TEM-based automated crystal orientation mapping (ACOM-TEM). STEM/energy dispersive x-ray spectroscopy (EDS) revealed no significant oxide precipitates in our samples. The grain size distribution was attained by ACOM-TEM. Although Hf-doping attained a grain refinement effect in the Nb3Sn layer in both doping cases, the degree of this effect was relatively small for Ti–Hf. Kernel average misorientation analysis by scanning electron microscopy-electron backscattered diffraction unveiled no appreciable difference between the internal strain states of the Nb-alloy parent phases in Ti–Hf and Ta–Hf. One remarkable new finding through STEM/EDS was the presence of a Cu–Hf compound phase in the Nb3Sn layer. The Cu–Hf compound sounds analogous to the Cu–Ti compounds that form when Nb–47Ti with Cu matrix is heat treated. The STEM/EDS maps revealed a larger amount of Cu flow from the Cu–Sn side along the grain boundaries. The large Cu deposition on the grain boundaries might facilitate grain growth in Nb3Sn. Those findings make a novel contribution to the literature as they provide a deep insight into Nb3Sn phase formation via Hf doping.

Application of CALPHAD Method for Predicting of Concentration Range of Amorphization of Transition Metals Melts

Early, the efficiency of the CALPHAD (Calculation of Phase Diagrams) method to a targeted search for compositions of amorphous alloys has been shown. The method for predicting the ranges of amorphization is based on the calculation of diagrams of metastable phase transformations between supercooled melts and boundary solid solutions on the base of pure elements. In this work, the model parameters for thermodynamic properties of liquid alloys and boundary solid solutions were summarized in a self-consistent database for the multicomponent Cu–Fe–Ni–Ti–Zr–Hf system. Such database for the multicomponent system is based on a common set of model parameters for boundary binary and ternary systems. This database was used to predict the concentration ranges of amorphization for the quinary Cu–Fe–Ni–Ti–Zr, Cu–Fe–Ni–Ti–Hf and boundary ternary and quaternary systems. The results of calculations are presented along sections in quaternary and quinary systems. The ternary and quaternary equiatomic alloys along with high entropy CuFeNiTiZr and CuFeNiTiHf alloys are trapped into prognosed composition ranges of amorphization. Predicted composition space of amorphization for melts of the Fe–Ni–Ti–Zr system is shown on the concentration tetrahedron. Based on the obtained results, a new criterion for predicting the concentration regions of amorphization of multicomponent melts is proposed, according to which the presence of a sufficient content of metals that are electron acceptors and donors is a chemical factor that affects the thermodynamic stability of melts and determines their glass-forming ability. For multicomponent melts of the Cu–Fe–Ni–Ti–Zr–Hf system the concentration ranges of amorphization correspond to the simultaneous fulfillment of the conditions xFe + xNi + xCu &gt; 0.25 and xTi + xZr + xHf &gt; 0.15, where Fe, Ni, and Cu are electron acceptors and Ti, Zr, and Hf are electron donors.

Enhanced quality factor (∼336,800 GHz) of MgTiO3 ceramic by Hf substitution for Ti-site

Mg(Ti1–xHfx)O3 (0 ≤ x ≤ 0.1) microwave dielectric ceramics were synthesized by solid-state reaction method. The effects of Hf substitution for the Ti-site on the phase composition, structure, and microwave dielectric properties of MgTiO3 ceramic were systematically investigated. The study demonstrated that Hf substitution significantly improved the quality factor of MgTiO3 ceramic. A single phase was confirmed at x = 0.002 and 0.005 by XRD and SEM, while secondary phases were observed at x ≥ 0.02. The Rietveld refinement of XRD data implied that the distortion of [Ti(Hf)O6] octahedron started with alleviation and then gradually deteriorated with an increase in Hf content. Raman spectra displayed that the frequencies of the vibrational modes were high when there was low distortion in [Ti(Hf)O6] octahedron, manifesting that the low distortion weakened the effect of non-harmonic vibration. The quality factor of the samples with x ≤ 0.05 was far higher than that of undoped MgTiO3, which may be related to low distortion in [Ti(Hf)O6] octahedron and high relative density. Notably, the excellent microwave dielectric properties of Q × f ∼336,829 GHz, εr ∼17.2 and τf ∼ -49.7 ppm/°C were achieved for the Mg(Ti0.995Hf0.005)O3 ceramic. The greatly improved quality factor of MgTiO3-based microwave ceramics through Hf substitution makes it a suitable candidate material for ceramic filters.

Synthesis of Multilayer Nanostructures Hf–Ti–N from the Plasma Phase

Synthesis of Multilayer Nanostructures Hf–Ti–N from the Plasma Phase

Local lattice distortions and the structural instabilities in bcc Nb–Ta–Ti–Hf high-entropy alloys: An ab initio computational study

Local lattice distortions (LLD) and structural stability of body-centered cubic (bcc) Nb–Ta–Ti–Hf high-entropy alloys (HEAs) are studied as functions of composition employing ab initio density-functional theory calculations, with specific focus on the role of the relative concentrations of group IV (Ti and Hf) versus group V (Nb and Ta) elements. Calculated results are presented as a function of composition x in NbxTa0.25Ti(0.75−x)/2Hf(0.75−x)/2 alloys, for elastic moduli, phonon spectral functions, LLD and structural energy differences for the bcc and competing hexagonal close-packed (hcp) and ω phases. The results highlight the important role of group V elements and LLD in stabilizing the bcc structure. They further reveal how composition x can be tuned to alter both the magnitude of the LLD and structural energy differences. Specifically, the magnitude of the structural energy differences, and elastic and dynamic stability of the bcc phase, are enhanced with increasing x, while the LLD increase in magnitude as this concentration is decreased. The results also show evidence of correlated LLD at lower values of x, reflecting local structural distortions towards the ω phase, but not hcp. The degree of ω-collapse is nevertheless partial i.e., transformation towards this phase is not observed to be complete due to the presence of Ta and Nb. At lower values of x we further find an energy landscape characterized by multiple, nearly degenerate local energy minima for different values of the LLD.

The giant ShangzhuangREE‐rich apatite deposit, western China: Genesis by ultramafic magma differentiation

The origin of giant rare earth element (REE)‐rich phosphorus deposits related to ultramafic rocks is unclear. To characterize the genetic linkage between ultramafic rocks and P–REE mineralization, we investigated the whole‐rock geochemistry, geochronology and Hf isotopic compositions, along with previously published data of barren and ore‐bearing clinopyroxenites from the giant Shangzhuang REE‐rich apatite deposit (containing about 51.1 million tonnes of P2O5and 157,400 tonnes of REEs) in the Lajishan suture, Western China. The P–REE ores and barren biotite clinopyroxenites are composed predominantly of clinopyroxene and biotite, with high levels of apatite, amphibole, allanite, sulfide and oxide minerals in the ore‐bearing samples. Zircon and allanite U–Pb ages for the barren and ore‐bearing clinopyroxenites indicate they were formed between 467 and 465 Ma. The barren and ore‐bearing samples are characterized by high CaO and low SiO2contents, as well as similar Hf isotope ratios (εHf[t]: +3.4 to +7.7), indicating that they represent an ultramafic magma system and have been derived from a similar source. They are enriched in light REEs and depleted in Nb–Ta–Zr–Hf–Ti, similar to arc‐related lavas found worldwide. The weakly evolved barren samples have high Ba/La, Rb/Y and Dy/Yb values, and their ages are coeval with calc‐alkaline granitic plutons in the belt, which postdate the formation of the Lajishan ophiolite by ~30 myr. This suggests that the primary magma originated from garnet peridotite mantle modified by subducted Proto‐Tethys oceanic crust‐derived fluids in an arc setting. The presence of apatite intergrowth with cumulus clinopyroxene without alteration features indicates that the deposit has a magmatic origin. The ore‐bearing clinopyroxenite contains higher contents of P, REEs, and high‐field‐strength elements than the barren clinopyroxenite, which could be attributed to the in situ accumulation of apatite and allanite in the residual magma. Thus, the formation of the P‐dominated orebodies results from phosphate saturation due to the extensive fractionation crystallization of clinopyroxene and biotite during magmatic evolution. REEs were primarily enriched in the late‐stage melt, which was compatible with apatite and allanite. We propose that the formation of the Shangzhuang REE‐rich apatite deposit can be attributed to a simple differentiation of ultramafic melts in an evolved magmatic system.

Thermodynamic modeling of the Ni–Ti–Hf system

Ni–Ti–Hf alloys have received extensive attention owing to their excellent shape-memory behavior. In the present work, the thermodynamic assessment of the Ni–Ti–Hf system was performed using the CALPHAD method after a critical review of the thermodynamic descriptions of boundary binaries and available experimental data in the ternary system. The solubility ranges of the third elements in binary intermetallic compounds are well described. Moreover, the wide homogeneity ranges of the B2 and NiTi2 phases are satisfactorily reproduced. The ternary τ phase was reasonably described using a two-sublattice model (Ni)0.44(Hf,Ti)0.56. Both the thermochemical properties and phase diagrams were well described using the present model parameters.

A novel (Hf1/3Zr1/3Ti1/3)C medium-entropy carbide coating with excellent long-life ablation resistance applied above 2100 °C

Compared to conventional monocarbides, the oxides of medium/high-entropy carbides (M/HECs) expressed excellent antioxidant resistance and high-temperature structural stability, which are great potential to improve the oxidation/ablation resistance of carbon/carbon (C/C) composites. M/HECs ceramics coated C/C composites are ideal long-term ablation-resistant structural materials applied in ultra-high-temperature (above 2100 °C) and oxygen-containing environments. This work reported a novel MEC (Hf1/3Zr1/3Ti1/3)C coating first prepared on SiC-coated C/C composites. The fabricated MEC coating significantly prolonged the anti-ablation time (more than 210 s) above 2100 °C in an oxyacetylene flame ablation environment with a heat flux of 2.4 MW/m2, compared to monocarbide coating (less than 90 s). The ablated surface of MEC coating was composed of amorphous Zr–Hf–Ti–C–O oxycarbide pinned by m-(Hf, Zr, Ti)O2 nanoparticles, forming a unique structure of oxide scale for long-term ablation resistance. This work provides a new way to improve the ablation resistance of monocarbide coatings and develops a great promise for new-generation structural materials at ultra-high temperature environments.

The Middle Paleozoic Magmatism of the Central Tuvinian Trough (Eastern Altai–Sayan Fold Area): Petrogenesis, Tectonics, and Geodynamics

Abstract —The Tuvinian trough is one of the large grabens of the rift system formed in the Devonian–Carboniferous in the eastern part of the Altai–Sayan fold area. Based on the results of comprehensive studies, the age was refined, and the geochemical features of igneous rocks formed during two stages of tectonic and magmatic activity within the Tuvinian trough were studied. In the Early Devonian (397 Ma, Emsian), at the stage of the initiation of the Tuvinian trough in the stretching setting, the volcanic and subvolcanic rocks of the Kendei Formation formed, which make a bimodal series. The Early Devonian igneous rocks of mafic composition have geochemical features of both intraplate (low values of Mg#, high contents of K2O (up to 2.9 wt.%) and TiO2 (up to 2.2 wt.%), and enrichment in LREE relative to HREE) and suprasubductional (enrichment in Pb and Sr and depletion in Ta and Nb) formations and are characterized by high values of εNd(T) (+5.9 to +8.0). They are assumed to have formed from a mixed source including the depleted mantle and components modified by subduction. The Early Devonian felsic volcanic rocks, which are the extreme member of the bimodal sequence, also combine the geochemical features of rocks of intraplate (high Fe, low Sr, P, and Ti contents, Zr and Hf enrichment) and island arc (Ta and Nb depletion) origin. These rocks with εNd(T) values from +4.0 to +7.0 resulted from the melting of a heterogeneous source corresponding in composition to the lower continental crust. In the Middle Devonian–early Carboniferous (390–350 Ma), the Tuvinian rift trough evolved into a mature stage, at which the mafic rocks of the Torgalyk complex were intruded. The Middle Devonian–early Carboniferous mafic rocks are similar in isotope and geochemical characteristics, including the Nd isotopic composition (εNd(T) = +6.7), to the Early Devonian formations. In contrast to the Early Devonian rockes, the magmas for the Middle Devonian–early Carboniferous mafic rocks were generated a relatively homogeneous mantle source without significant metasomatic transformations, the features of which are better manifested in the Kendei rocks.

Magnesium and Calcium Isotopic Geochemistry of Silica‐Undersaturated Alkaline Basalts: Applications for Tracing Recycled Carbon

AbstractCarbon plays important roles in the evolution of the atmosphere and biosphere, and geochemical differentiation in Earth's interior. Most subducted C is recycled to the deep mantle and returns to the surface by degassing from erupted basalts and associated fluids. The Mg–Ca isotopic systems have been widely used in tracing recycled C. However, the idea that these geochemical proxies truly reflect the deep C cycle has been challenged. Here we present whole‐rock geochemical and Mg–Ca isotopic compositions of Miocene silica‐undersaturated alkaline basalts of the Western Qinling orogen, China. These alkaline basalts are associated with carbonatites and are characterized by low SiO2 (39.0–43.2 wt.%) and Al2O3 (6.98–9.15 wt.%) contents, high CaO/Al2O3 ratios (1.4–1.8), and positive Nb–Ta and negative Pb–Zr–Hf–Ti anomalies, suggesting they were derived by partial melting of carbonatite‐metasomatized asthenospheric mantle. The studied samples have δ26Mg values of −0.24‰ to −0.44‰, ranging from mantle‐like values (δ26Mg = −0.25‰ ± 0.07‰) to lower values. This implies that carbonatite metasomatism does not always produce low‐δ26Mg anomalies. The samples have δ44/40Ca values (0.59‰–0.77‰; relative to the standard SRM915a) that are lower than Bulk Silicate Earth (0.94‰ ± 0.05‰), which are attributed to the involvement of low δ44/40Ca recycled carbonate in the mantle source. We suggest that the shift in mantle δ26Mg values during carbonatite metasomatism is controlled by the type and amount of carbonatite involved, while Ca isotope variations depend largely on the δ44/40Ca values of subducted carbonates. Mg or Ca isotopes alone, however, may not be sufficient to track the deep carbon cycle.

Photoluminescence effects and probing phase transition of Sm3+-doped Ba(Ti0.8Hf0.2)O3-(Ba0.7Ca0.3)TiO3 phosphor prepared by hydrothermal method

A series of [(Ba(1-0.3y)Ca0.3y)1-xSmx][Ti(0.8+0.2y)Hf(0.2-0.2y)]O3 (Sm3+-doped (1-y)Ba(Ti0.8Hf0.2)O3-y(Ba0.7Ca0.3)TiO3, abbreviated as Sm3+-doped BCTH, or (1-y)BHT-yBCT:xSm3+, x = 0.0005–0.03, y = 0.2–0.7, y representing the mole fraction of (Ba0.7Ca0.3)TiO3) nano phosphors were prepared by a hydrothermal method. The Sm3+ ions doping amount and triple-point morphotropic phase boundary (TMPB) composition dependent photoluminescence properties were systematically investigated. The hydrothermal synthesized powders present rather pure pseudo-cubic perovskite phase structure at room temperature and locate on the rhombohedral side of MPB. Under the excitation of 409 nm light, the powders exhibit dominant orange-red emission peaks with characteristic emission of the Sm3+ ions at around 564 nm, 596 nm, 644 nm, and 703 nm, relating to the transitions of 4G5/2 → 6H5/2, 6H7/2, 6H9/2 and 6H11/2, respectively. The BHT-0.5BCT:0.005Sm3+ powder exhibits optimal emission in these Sm3+-doped BCTH powders. Furthermore, the phase transition temperature can be characterized by the multi-feature changes of the temperature dependent fluorescence intensity.

Surface tension measurement of Ti–Zr and Ti–Hf alloys at high-temperature by the pendant drop method

High-temperature alloy melts introduced into carbon/carbon (C/C) composites by reactive melt infiltration (RMI) to form ultra-high-temperature ceramics (UHTCs) is currently an important method for improving their ablation resistance. As a fundamental thermo-property, surface tension has an important impact on the infiltration of alloy melts into the C/C composites. However, the surface tension of binary alloy systems, such as Ti–Zr and Ti–Hf, used for RMI has been little studied. In this work, experimental measurements through the pendant drop method and theoretical calculations of the surface tension of Ti–Zr and Ti–Hf alloys with different compositions have been carried out. Meanwhile, inevitably, impurities are introduced during the determination of surface tension at high temperatures, mainly carbon and oxygen. The effect of impurities on the phase composition, distribution, and surface tension of the alloys is investigated. The results demonstrated that the experimental measurements are in good agreement with the theoretical values of Butler's equation. The surface tension of Ti–Zr and Ti–Hf alloys increases with the content of Zr or Hf at the same temperature. Small amounts of oxygen are mainly dissolved in the lattice and have no significant effect on the surface tension. Carbon slightly decreases the surface tension of the alloy.

Crystal-plasticity modeling of phase transformation–viscoplasticity coupling in high-temperature shape memory alloys

A coupling between phase transformation and viscoplasticity is observed in HTSMAs during actuation. The dislocations generated and retained phases accumulated are responsible for the coupling. It results in functional fatigue, which is reflected as an alteration in the functional properties of the alloy, and an increase in the irrecoverable deformations. The objective of the present study was to develop a theoretical framework to account for the interactions between viscoplasticity and phase transformation to simulate the alteration in functional properties and generation of irrecoverable deformations. A crystal-plasticity based multi-scale approach was followed to develop the framework. This approach accounts for the mechanisms of: phase transformation, transformation induced plasticity, accumulation of retained martensite, plasticity (in a rate-independent manner), and viscoplasticity (in a rate-dependent manner). The coupling was accounted by a direct effect of the dislocation densities produced by the irrecoverable mechanisms onto the transformation resistance and retained martensite. This resulted in an indirect effect on the functional properties of phase transformation. On simulating the response of single crystals and polycrystals, anisotropic responses are captured at the grain scale (from single crystals), and a nearly isotropic response is captured at the macro scale (from polycrystals). The results show consistency between the functional property trends (such as TT and hysteresis) and those observed experimentally at the macroscale. The contributions of this study are presenting an effect of: (i) texture, (ii) thermal cycling rate and (iii) functional fatigue, through the response of several randomly oriented single crystals and a polycrystal of Ni–Ti–Hf HTSMA.

Mixing enthalpy of the Co–Ti–Hf liquid alloys and regularities of the function change in the row of the ternary (Co, Ni, Cu)–Ti–Hf glass-forming melts

The partial molar mixing enthalpies of titanium and hafnium in liquid Co–Ti–Hf alloys were studied at 1900 K by high-temperature isoperibolic calorimetry along xCo/xHf = 3 and xCo/xHf = 3 sections. The partial mixing enthalpies of undercooled liquid titanium and hafnium in liquid alloys and the integral mixing enthalpy are negative over the studied composition range. The thermodynamic mixing functions of liquid (Co, Ni, Cu)–Ti–Hf alloys were modeled in the framework of the associate solution model. Regularities in the change of thermodynamic mixing functions in liquid glass-forming (Co, Ni, Cu)–Ti–Hf alloys are discussed considering donor–acceptor interaction of the components in the melts.

Tensile strength improvement of martensitic ODS steels with Zr and Hf additions

Aimed at evaluating the effectiveness of Zr and Hf additions on oxide particle refinement and tensile strength improvement of transformable oxide dispersion strengthened (ODS) steels, 9Cr–Zr and 9Cr–Hf ODS martensitic steels were fabricated by mechanical milling, spark plasma sintering and tempering heat treatment at 800 °C. The as-sintered 9Cr–Zr steel consisted of martensite and certain amount of transformed ferrite with M23C6 (M = Fe, Cr) precipites within ferrite interior. With respect to the as-sintered 9Cr–Hf steel, the microstructure consisted of martensite and small amount of residual ferrite. The residual ferrite is characteristics of higher number density of oxide nanoparticles than those in martensite. The added Zr or Hf cannot transform all Y2O3 into Y–Zr–O or Y-Hf-O complex oxides. Except for Y2O3, ZrO2 and Y4Zr3O12 were identified in 9Cr–Zr steel, and YxHfyOz and Y2H2O7 were found in 9Cr–Hf steel. The orientation relationship and interface structure between Y–Zr–O/Y-Hf-O and the matrix were studied in detail, as well as the size distributions of oxide nanoparticles in both as-sintered and tempered conditions. Tempering is mandatory to restore ductility of martensitic steels. The tensile strength of tempered 9Cr–Hf steel is superior to most reported results, while the tempered 9Cr–Zr steel exhibits a good combination of tensile strength and ductility. The ferrite phase in both tempered 9Cr–Zr and 9Cr–Hf steels has negligible effects on ductility degradation. The inferior strength of high-angle boundaries to tempered martensite itself is suggested to be responsible for limited ductility of steels. Compared to Al and Ti constituents, Zr and Hf can be considered as potential microalloying elements to produce ODS steels with favorable mechanical performance.