Thermodynamics behind oxide mixture reduction
Neuer Datei-Download
Lowering oxide reduction temperatures by alloy thermodynamics
This work addresses a central limitation of hydrogen-based direct reduction: some technologically important oxides, most notably Cr₂O₃, are too stable to be reduced efficiently by H₂ at moderate
temperatures. Pure Cr₂O₃ requires temperatures above about 1300 °C for effective hydrogen reduction. This makes direct production of Cr-containing alloys by carbon-free reduction difficult, despite
the enormous potential of hydrogen metallurgy for decarbonizing alloy production.
The study shows that this limitation can be overcome by reducing oxide mixtures, not isolated oxides. When Cr₂O₃ is mixed with Fe₂O₃ or NiO, the reduced Cr does not have to form
pure metallic Cr. Instead, it dissolves directly into a metallic Fe–Cr or Ni–Cr solid solution. This changes the reduction thermodynamics. The activity of Cr in the alloy phase is lower than that of
pure Cr, and this lowers the chemical potential of the reduced Cr product. The reduction equilibrium is therefore shifted toward oxide reduction at lower temperature.
The essential reaction is thus not:
but rather:
The product phase, Fe(Cr), acts as a thermodynamic sink for Cr. This is the central physical mechanism behind the observed reduction-temperature lowering of about 200 °C.
A multicomponent Ellingham framework
The paper introduces a quantitative thermodynamic framework for predicting the reducibility of mixed oxides. Classical Ellingham diagrams describe reduction equilibria for pure metals and pure
oxides, usually assuming unit activity of the condensed phases. That treatment is insufficient for alloy-forming reduction, where the reduced element dissolves into a metallic solution and its
activity can be far below unity.
The authors therefore calculate the equilibrium oxygen partial pressure,
O2, using CALPHAD for Fe–Cr–O and Ni–Cr–O systems. This allows the reducibility of oxide mixtures to be expressed as a
function of temperature, oxide composition and alloy chemistry. The resulting concept is a multicomponent Ellingham diagram. It extends the classical reduction map from pure oxides
to alloy-forming oxide mixtures.
The calculations show that increasing temperature strongly raises the equilibrium
O2, making reduction more favorable. For a Ni–10Cr oxide mixture,
O2 increases from
×10−30 at 700 °C to
×10−18 at 1100 °C. For Fe–10Cr, it increases from
×10−31 to
×10−19 over the same temperature range. The temperature effect is
especially strong between 700 and 800 °C, where a 100 °C increase shifts
O2 by at least six orders of magnitude.
The calculations also show that the matrix oxide matters. At low Cr₂O₃ contents, up to about 25 at.%, NiO promotes Cr₂O₃ reduction more effectively than Fe₂O₃ because Cr has a lower chemical
activity in Ni–Cr solid solution. As the Cr₂O₃ fraction increases, this matrix effect weakens because less Fe or Ni is available to dilute and stabilize Cr in the alloy phase. Above about 25 at.%
Cr₂O₃, the difference between Fe–Cr and Ni–Cr oxide mixtures becomes small.
This provides a direct design rule: the reduction of a stable oxide is favored when the reduced element can dissolve into a metallic matrix where its activity is low. Alloy thermodynamics, not
only oxide stability, governs reducibility.
Experimental proof in Fe–Cr oxide mixtures
The thermodynamic predictions were tested by hydrogen-based direct reduction of compacted Fe₂O₃ + Cr₂O₃ powders targeting Fe–10Cr and Fe–50Cr alloys. The powders had an average particle size of
about 1 µm and were reduced in 75 vol.% H₂–25 vol.% Ar at 900 or 1100 °C, with a heating rate of 10 °C min⁻¹ and 1 h isothermal holding.
Reduction started at about 250 °C, mainly due to Fe₂O₃ reduction. Pure Cr₂O₃ showed no measurable reduction at 1100 °C under the same conditions once moisture loss was accounted
for. In contrast, Cr₂O₃ in the Fe₂O₃ + Cr₂O₃ mixtures was partially reduced, confirming the alloy-assisted reduction mechanism.
For the Fe–10Cr oxide mixture reduced at 900 °C, the total reduction degree was 91.37%, slightly above the value expected for complete Fe₂O₃-to-Fe reduction alone. This
corresponds to about 1.31 at.% metallic Cr and 12.31 wt.% reduction of the initial Cr₂O₃. X-ray diffraction showed 89.5 ± 0.9 wt.% BCC Fe–Cr and
10.5 ± 0.4 wt.% residual Cr₂O₃.
At 1100 °C, the Fe–10Cr oxide mixture reached 93.93% reduction, corresponding to 3.74 at.% metallic Cr and 35.03 wt.% reduction of the initial
Cr₂O₃. XRD showed 94.1 ± 1.1 wt.% BCC Fe–Cr and 5.9 ± 0.5 wt.% Cr₂O₃. The BCC lattice parameter increased from 2.867 Å at 900 °C to
2.869 Å at 1100 °C, consistent with substitutional Cr dissolution in Fe.
For the Fe–50Cr oxide mixture reduced at 1100 °C, the total reduction degree was only 52.57%, corresponding to 4.40 at.% metallic Cr and 4.63 wt.%
reduction of the initial Cr₂O₃. The final phase constitution was 46.2 ± 0.5 wt.% BCC Fe–Cr, 48.8 ± 0.6 wt.% Cr₂O₃ and 5.0 ± 0.2 wt.% FCC
Fe–Cr. The large Cr₂O₃ fraction remained kinetically difficult to reduce because much less Fe matrix was available to dissolve Cr and because unreduced Cr₂O₃ impeded sintering and
transport.
The composition trend is decisive. Fe-rich mixtures reduce Cr₂O₃ more readily because the Fe matrix dilutes Cr and lowers its activity. Cr-rich mixtures require higher temperature or longer
holding time. For Fe–10Cr, complete Cr₂O₃ reduction was achieved after about 4 h at 1100 °C, yielding a fully metallic Fe–10Cr alloy.
Atomic-scale mechanism at the metal–oxide interface
The reduction of Cr₂O₃ proceeds at the interface between the oxide and the growing Fe-rich metallic phase. Atom probe tomography of the Fe–10Cr sample reduced at 1100 °C reveals the chemical
partitioning across this interface.
In the metallic phase near the interface, the composition is approximately 94.27 at.% Fe, 3.24 at.% Cr, 0.35 at.% O and 2.11 at.%
H. About 1 µm away from the interface, the metallic phase contains 93.67 at.% Fe and 4.24 at.% Cr, confirming homogeneous Cr dissolution in the Fe-rich
metal. These values agree with the reduction degree derived from thermogravimetry.
The oxide side also shows interdiffusion. The Cr₂O₃ region near the interface contains 0.73 ± 0.11 at.% Fe, indicating Fe incorporation into the oxide. XRD supports this
finding: the Cr₂O₃ lattice parameter increases from 4.955 Å to 4.960 Å with increasing reduction temperature and Cr content, consistent with partial Fe dissolution
into the corundum lattice.
The interface is therefore not a sharp, inert boundary. It is an active reaction zone where Cr is reduced and dissolves into Fe, while a small amount of Fe enters the oxide. The authors suggest
that an Fe-rich oxide complexion may mediate this exchange. The decisive step remains the removal of reduced Cr from the interface into the Fe matrix. This prevents reoxidation of Cr and enables
continued Cr₂O₃ reduction.
Scientific novelty and innovation
The main advance is the thermodynamic explanation of why stable oxides can be reduced at lower temperatures in oxide mixtures. The study shows that the reduction temperature is not fixed only by
the stability of the pure oxide. It is shifted by the activity of the reduced element in the alloy product. This changes the way hydrogen-based reduction must be designed: the target alloy phase is
part of the reduction thermodynamics.
The second innovation is the multicomponent Ellingham approach. By calculating equilibrium
O2 for oxide mixtures and alloy products, the work provides a practical design tool for hydrogen-based co-reduction. It
allows the selection of oxide compositions, matrix elements and reduction temperatures for direct alloy synthesis. This moves Ellingham analysis from pure oxide metallurgy into alloy design.
The third contribution is the direct coupling of thermodynamics with interface-scale experimental proof. Thermogravimetry quantifies the reduction degree, XRD tracks phase fractions and lattice
expansion, and atom probe tomography confirms Cr dissolution into Fe and Fe incorporation into Cr₂O₃ at the nanometre scale. The mechanism is therefore established from process scale to atomic
scale.
The implications are broad. Cr₂O₃ is a model case for hard-to-reduce oxides. The same principle can be applied to other alloy systems in which a stable oxide becomes reducible because the
reduced element is stabilized in a metallic solution. This provides a route to produce Fe–Cr, Ni–Cr and more complex Cr-containing alloys directly from oxides, avoiding the conventional sequence of
separate oxide reduction, metal refining, melting and alloying.
The work defines a core rule for one-step metallurgy from oxides: choose oxide mixtures such that the first reduced metal creates a matrix that lowers the activity of the harder-to-reduce
element. Reduction and alloying then become one coupled event. Hydrogen removes oxygen; the alloy phase supplies the thermodynamic driving force.
This is the key conceptual step from hydrogen reduction as extraction to hydrogen reduction as alloy synthesis.
Thermodynamics and mechanism behind the lowering of direct reduction temperatures in oxide mixtures
Hydrogen-based direct reduction offers a sustainable pathway to decarbonize the metal production industry.
However, stable metal oxides, like Cr2O3, are notoriously difficult to reduce, requiring extremely high temper-
atures (above 1300 ◦C). Herein, we explain how reducing mixed oxides can be leveraged to lower hydrogen- based reduction temperatures of stable oxides and produce alloys in a single process. Using a newly devel-oped thermodynamic framework, we predict the precise conditions (oxygen partial pressure, temperature, and oxide composition) needed for co-reduction. We showcase this approach by reducing Cr2O3 mixed with Fe2O3 at 1100 ◦C, significantly lowering reduction temperatures (by ~200 ◦C). Our model and post-reduction structural and chemical analyses elucidate that th
1-s2.0-S136970212500358X-main.pdf
PDF-Dokument [6.5 MB]
Hydrogen-based direct reduction offers a sustainable pathway to decarbonize the metal production industry.
However, stable metal oxides, like Cr2O3, are notoriously difficult to reduce, requiring extremely high temper-
atures (above 1300 ◦C). Herein, we explain how reducing mixed oxides can be leveraged to lower hydrogen- based reduction temperatures of stable oxides and produce alloys in a single process. Using a newly devel-oped thermodynamic framework, we predict the precise conditions (oxygen partial pressure, temperature, and oxide composition) needed for co-reduction. We showcase this approach by reducing Cr2O3 mixed with Fe2O3 at 1100 ◦C, significantly lowering reduction temperatures (by ~200 ◦C). Our model and post-reduction structural and chemical analyses elucidate that th
1-s2.0-S136970212500358X-main.pdf
PDF-Dokument [6.5 MB]
Druckversion 