Technological Aspects of Lithium-Titanium Ferrite Synthesis by Electron-Beam Heating

Solid-phase synthesis of lithium-titanium ferrite by electron-beam heating of a Fe2O3–Li2CO3–TiO2 initial reagents mixture with different history (powder, compact, mechanically activated mixture) was studied using X-ray diffraction, thermomagnetometric and specific saturation magnetization analyses. Ferrite was synthesized using an ILU-6 pulsed electron accelerator; it generated electrons with electron energy of 2.4 MeV to heat samples to temperatures of 600 and 750 °C. The isothermal holding time upon reaching the synthesis temperature was 0–120 min. The efficiency of ferrite synthesis by electron-beam heating was evaluated via comparison of the characteristics of the obtained samples with those synthesized by conventional ceramic technology under similar temperature-time conditions. It was found that the rate of ferrite formation depends on the heating method, temperature, synthesis time, density, and activity of the initial mixture. It was shown that sample compaction provides the preferential formation of unsubstituted lithium ferrite of Li0.5Fe2.5O4 composition with a Curie temperature of at ca. 630 °C in both synthesis methods. High-energy electron-beam heating of the mechanically activated mixture significantly accelerates synthesis of Li0.6Fe2.2Ti0.2O4 substituted ferrite, for which the Curie temperature and specific saturation magnetization were recorded as 534 °C and 50 emu/g, respectively. Therefore, LiTi ferrites can be obtained at a lower temperature (750 °C) and with a shorter synthesis time (120 min) compared to traditional ceramic technology.


Introduction
Ferrites hold a specific place among the many technological magnetic materials. Ferrite materials, including lithium-containing ferrites, are currently the key elements in most modern radio engineering, electronic, and computing devices [1][2][3][4]. Lithium and substituted lithium ferrites are characterized by a sufficiently low cost, yet they exhibit a certain combination of electrical and magnetic properties, which enable their use in microwave technology. They show high values of the Curie temperature and saturation magnetization, and low dielectric and magnetic losses required to design ferrite devices operating at microwave and millimeter wave frequencies [5][6][7][8].
Lithium-containing ferrites used in microwave technology are typically of a complex composition, which is attained by replacing iron ions, for example, with titanium, zinc, or manganese ions. The electrical properties of ferrites, in particular, can be improved via Ti 4+ ions introduced into pure lithium ferrite Li 0.5 Fe 2.5 O 4 to decrease the formation of Fe 3+ ions during ferrite synthesis [9][10][11]. Therefore, the Li 0.5(1+x) Fe 2.5−1.5x Ti x O 4 substituted lithiumtitanium ferrite with a spinel structure and high electrical resistance is formed. Substitution with titanium ions increases the electrical resistance, yet the saturation magnetization decreases slightly. The Curie temperature also decreases depending on the concentration of introduced titanium ions.
The conventional ceramic technology, the key method for ferrite synthesis, employs solid-phase synthesis of ferrites from mixtures of initial oxides that include iron oxide Fe 2 O 3 as the main component. To obtain ferrites of complex compositions, ceramic technology requires high synthesis temperatures and long high temperature exposures to synthesize ferrites with good properties [13,14].
It is known that the properties of ferrites synthesized by this method significantly depend on the redox processes occurring during heating and cooling. In this case, high temperatures stimulate reduction reactions associated with oxygen loss from the ferrite lattice [21]. Volatile elements, such as lithium, in ferrite chemical composition may complicate the processes. As reported in [21], lithium in combination with oxygen can be released from samples at 1000 • C. Therefore, in many cases, high temperatures used for synthesis of lithium-containing ferrites become critical due to the violation of their stoichiometric composition and, as a result, deterioration of the synthesized sample properties.
This problem can be partially solved by mechanical milling of the initial reagents in a ball mill [22][23][24][25][26][27][28][29][30][31]. Such treatment reduces the size of the powder particles and simultaneously increases imperfection of the powders, which enhances the reactive activity of the reagents. As shown in [22,32], grinding of the initial oxides, used for ferrite formation, in a planetary mill reduces the synthesis temperature by 100-200 • C, depending on ferrite composition. The time of mechanical activation of the powder typically ranges from 1 to 5 h, depending on the energy intensity of the treatment, which depends on the rotation speed of the mill cups, the diameter of the balls, etc.
A synthesis method was previously developed to reduce the temperature and time of ferrite synthesis; it employs heating of the initial reagents by an electron beam with energies above 1 MeV [33][34][35][36][37][38][39][40][41]. High electron energies provide the possibility of volumetric heating of the synthesized samples, and the electron penetration depth depends on the density of the samples. The synthesis temperature can be controlled by changing the parameters of the electron beam, including the current density.
The electron-beam heating method was used to synthesize ferrites of various compositions, including spinel lithium, lithium-zinc and manganese-zinc ferrites, and hexagonal ferrites [35,[38][39][40][41][42]. In [43][44][45], it was shown that this heating method increases the rate of solid-phase interaction between the initial reagents, which significantly decreases the ferrite synthesis temperature. The results reported in [43] showed that the initial reagents mainly interact during nonisothermal heating of the samples within a very short time. Formal kinetic analysis of lithium ferrite synthesis showed that ferrite formation under electron-beam heating is accelerated due to the decreased activation energy of synthesis and the decreased pre-exponential factor in the temperature dependence of the reaction rate.
To develop the electron beam technology for ferrite production, the efficiency of ferrite synthesis under these conditions should be increased, which depends on numerous parameters, including treatment modes and sample history. This study focused on solidphase synthesis of lithium-titanium ferrite by electron-beam heating of samples of different density and activity.
The mixture of Li 2 CO 3 -Fe 2 O 3 -TiO 2 initial reagents was divided into several parts to obtain samples of different density and activity. Samples in the form of reagent powder showed a bulk density of 1.7 g/cm 3 . Samples in the form of pellets were compacted using a hydraulic press under pressure of 200 MPa; the pellet density was 2.8 g/cm 3 .
Some of the samples made from the initial powder mixture were mechanically activated in an AGO-2S planetary ball mill at a rotation speed of 1290 (MA1 samples) and 2220 (MA2 samples) rpm. For milling, steel cups and 6 mm balls were used. The ball-to-powder mass ratio was 10:1.
Ferrite synthesis was performed by two methods. The first method was a conventional ceramic technology; samples were synthesized by thermal heating (T-samples) in the laboratory furnace at 600 and 750 • C with an isothermal holding time of up to 120 min. The second method employed radiation-thermal heating of samples (RT-samples); synthesis was performed by high-energy electron-beam heating using an ILU-6 pulsed electron accelerator with electron energy of 2.4 MeV. The time of the nonisothermal heating stage varied from 3 to 5 min with respect to the synthesis temperature, which depended on the beam current density. More details about the RT method of sample treatment are reported in [45].
An ARL X'TRA X-ray diffractometer, the Powder Cell 2.5 software, and the PDF-4+ powder database were used for X-ray diffraction analysis (XRD) of the samples. The specific saturation magnetization (σ S ) was measured using an H-04 induction magnetometer at a magnetizing field of 4.7 kOe. The Curie temperature (T C ) of the samples was studied by the thermomagnetometric method based on thermogravimetric measurements with an applied external magnetic field. The temperature was analyzed using a Netzsch STA 449C Jupiter thermal analyzer and permanent magnets providing an external magnetic field of 5 Oe. The thermomagnetometry technique is described in more detail in [46].

XRD Analysis
Formation of lithium-titanium ferrite proceeds in accordance with Equation (1): During synthesis, at x = 0.2, soft magnetic spinel ferrite of the Li 0.6 Fe 2.2 Ti 0.2 O 4 composition, is formed. In [44], it was shown that lithium-titanium ferrite synthesis can yield lithium ferrite in accordance with the reaction: Li 2 CO 3 + 5Fe 2 O 3 → 4Li 0.5 Fe 2.5 O 4 + CO 2 (2) and substituted spinel phases of lithium-titanium ferrite with x = 0.2. These spinel phases are typically transition products of the reaction; their concentration depends on the temperature-time treatment conditions. At a sufficiently high synthesis temperature and long isothermal exposure, the transition products interact with each other to form Li 0.6 Fe 2.2 Ti 0.2 O 4 ferrite. In [44], it was shown that the lattice parameters of the Li 0.5 Fe 2.5 O 4 and Li 0.5(1+x) Fe 2.5−1.5x Ti x O 4 ferrite phases are similar, which causes the merge of the reflections in the diffraction patterns. Therefore, the total content of spinel ferrite phases formed during ferrite synthesis was analyzed by the XRD method.
The XRD patterns were obtained for all the studied samples. The paper reports data for samples synthesized for 120 min. Figure 1 shows the XRD patterns of samples synthesized from powdered and compacted mixtures of initial reagents. The analysis showed that after synthesis the samples contained not only spinel phases, which may belong to lithium and titanium-substituted lithium ferrites, but also phases of unreacted initial components, mainly Fe 2 O 3 (marked reflections), which quantitative content depends on synthesis conditions. Consequently, the amount of the synthesized spinel ferrite phase grows, and the amount of powder starting reagents decreases at increased temperature and isothermal holding time in both synthesis methods. Tables 1 and 2 present XRD and other data for the spinel phase synthesized within 120 min in different samples. The lattice parameter for this phase synthesized in powdered and compacted samples is about 8.33 Ǻ. Lithium ferrite of the Li0.5Fe2.5O4 composition [47] exhibits this value of the lattice parameter. It is known that at increased titanium content in the spinel phase, the lattice parameter increases and amounts to 8.335 Ǻ for Li0.6Fe2.2Ti0.2O4 ferrite [47]. Therefore, the values obtained for the lattice parameter of the spinel phase in mechanically activated samples most likely evidence the formation of titanium-substituted lithium ferrites.  Tables 1 and 2 present XRD and other data for the spinel phase synthesized within 120 min in different samples. The lattice parameter for this phase synthesized in powdered and compacted samples is about 8.33Ǻ. Lithium ferrite of the Li 0.5 Fe 2.5 O 4 composition [47] exhibits this value of the lattice parameter. It is known that at increased titanium content in the spinel phase, the lattice parameter increases and amounts to 8.335Ǻ for Li 0.6 Fe 2.2 Ti 0.2 O 4 ferrite [47]. Therefore, the values obtained for the lattice parameter of the spinel phase in mechanically activated samples most likely evidence the formation of titanium-substituted lithium ferrites. It was found that the concentration of the spinel phase of ferrite was higher in compacted samples compared to the powder synthesized under similar temperature-time regime by both heating methods. This phenomenon was interpreted in our study [32], which shows that the decomposition of lithium carbonate and its interaction with iron oxide can be accelerated at an increased compact density of mixtures of Fe 2 O 3 -Li 2 CO 3 powder reagents. This accelerates the synthesis process as a whole and intensifies the formation of the ferrite phase. A similar effect can be observed in a synthesis of ferrite from mechanically activated powder, only with a much greater efficiency due to a smaller particle size, high inter-particle contact density, and additional imperfection induced by powder activation.
The analysis also showed that electron-beam heating significantly accelerates the formation of spinel phases in comparison with thermal heating. At a synthesis temperature of 600 • C, the samples contain a large proportion of initial reagents in addition to the formed spinel lithium-titanium ferrite phases. However, the concentration of the initial reagents in the PT samples is significantly lower than that in the samples synthesized by thermal heating. As temperature increased up to 750 • C, formation of lithium-titanium ferrite phases accelerates. As a result, the diffraction patterns for samples synthesized from compacted reagents at 750 • C for 120 min show the content of only the spinel ferrite phase. Figure 2 shows the diffraction patterns of samples synthesized from mechanically activated reagents at 600 • C for 120 min. All diffraction patterns show reflections from the spinel phase and from the initial Fe 2 O 3 reagent. At increased rotation speed during powder milling, the amount of the synthesized spinel phase increases with both types of heating, yet a large amount of unreacted iron oxide remains. A similar pattern was observed for all synthesis times.
shows that the decomposition of lithium carbonate and its interaction with iron oxide can be accelerated at an increased compact density of mixtures of Fe2O3-Li2CO3 powder reagents. This accelerates the synthesis process as a whole and intensifies the formation of the ferrite phase. A similar effect can be observed in a synthesis of ferrite from mechanically activated powder, only with a much greater efficiency due to a smaller particle size, high inter-particle contact density, and additional imperfection induced by powder activation.
The analysis also showed that electron-beam heating significantly accelerates the formation of spinel phases in comparison with thermal heating. At a synthesis temperature of 600 °C, the samples contain a large proportion of initial reagents in addition to the formed spinel lithium-titanium ferrite phases. However, the concentration of the initial reagents in the PT samples is significantly lower than that in the samples synthesized by thermal heating. As temperature increased up to 750 °C, formation of lithium-titanium ferrite phases accelerates. As a result, the diffraction patterns for samples synthesized from compacted reagents at 750 °C for 120 min show the content of only the spinel ferrite phase. Figure 2 shows the diffraction patterns of samples synthesized from mechanically activated reagents at 600 °C for 120 min. All diffraction patterns show reflections from the spinel phase and from the initial Fe2O3 reagent. At increased rotation speed during powder milling, the amount of the synthesized spinel phase increases with both types of heating, yet a large amount of unreacted iron oxide remains. A similar pattern was observed for all synthesis times.
As shown in Figure 3, as synthesis temperature increased to 750 °C, the amount of ferrite formed significantly increases. In this case, samples synthesized for 10 min or longer contain only the spinel phase with both types of heating. At first sight, a similar type of the XRD indicates the similarity in the phase composition of these samples. As will be shown below, in reality it was different. Tables 3 and 4 present XRD and other data for the spinel phase synthesized for 120 min from mechanically activated reagents using different rotation speeds of cups.  As shown in Figure 3, as synthesis temperature increased to 750 • C, the amount of ferrite formed significantly increases. In this case, samples synthesized for 10 min or longer contain only the spinel phase with both types of heating. At first sight, a similar type of the XRD indicates the similarity in the phase composition of these samples. As will be shown below, in reality it was different. Tables 3 and 4 present XRD and other data for the spinel phase synthesized for 120 min from mechanically activated reagents using different rotation speeds of cups. Figures 4 and 5 show XRD patterns of the dependences of the spinel phase concentration in the synthesized samples at the synthesis time. Under thermal heating, at a synthesis temperature of 600 • C, the content of the spinel ferrite phase in the samples slowly increased with increased synthesis time. In this case, the concentration curves are linear. In all other modes, the concentration curves indicate the region of rapid accumulation of the spinel phase; its time interval depends on the treatment mode and the type of sample preparation; further it showed a slow increase in concentration over time.
Under electron-beam heating of compacted samples at 750 • C, the area of fast accumulation of the spinel phase in the concentration curves was partially shifted to the nonisothermal heating stage; for the given synthesis temperature, it was about 5 min. This effect is more pronounced in mechanically activated samples heated at 750 • C. In the latter case, a large amount of the spinel phase formed at the heating stage could be observed in the samples synthesized using both types of heating.    The behavior of the curves for the ferrite phase formation allows the assumption that synthesis involves diffusion interaction between the initial reagents, which initiates the formation of transition phases. Then the transition phases and phases of the initial reagents interact with each other to form lithium-substituted ferrite phases, namely, lithium-titanium    Under electron-beam heating of compacted samples at 750 °C, the area of fast accumulation of the spinel phase in the concentration curves was partially shifted to the nonisothermal heating stage; for the given synthesis temperature, it was about 5 min. This effect is more pronounced in mechanically activated samples heated at 750 °C. In the latter case, a large amount of the spinel phase formed at the heating stage could be observed in the samples synthesized using both types of heating. The behavior of the curves for the ferrite phase formation allows the assumption that synthesis involves diffusion interaction between the initial reagents, which initiates the formation of transition phases. Then the transition phases and phases of the initial reagents interact with each other to form lithium-substituted ferrite phases, namely, lithium-   Under electron-beam heating of compacted samples at 750 °C, the area of fast accumulation of the spinel phase in the concentration curves was partially shifted to the nonisothermal heating stage; for the given synthesis temperature, it was about 5 min. This effect is more pronounced in mechanically activated samples heated at 750 °C. In the latter case, a large amount of the spinel phase formed at the heating stage could be observed in the samples synthesized using both types of heating. The behavior of the curves for the ferrite phase formation allows the assumption that synthesis involves diffusion interaction between the initial reagents, which initiates the formation of transition phases. Then the transition phases and phases of the initial reagents interact with each other to form lithium-substituted ferrite phases, namely, lithium-  The results obtained in [43] showed that the synthesis of substituted lithium ferrites at the initial heating stage causes formation of unsubstituted lithium ferrite of the Li 0.5 Fe 2.5 O 4 composition. Therefore, the concentration dependences obtained in this study for lithiumtitanium ferrite are like those obtained for lithium ferrite in [43]. Considering that during heating of the Fe 2 O 3 /Li 2 CO 3 mixture by a high-energy electron beam, the initial rate of lithium ferrite formation sharply increases, the initial stage of the RT synthesis of lithium-titanium ferrite involves the formation of unsubstituted lithium ferrite phases. It is also known [10] that stable spinel compounds, such as Li 2 TiO 3 , Li 4 Ti 5 O 12 , Li 4 Ti 5 O 12 and Li 4 Ti 5 O 12 , can be formed during lithium-titanium ferrite synthesis [10]. Apparently, during synthesis of lithium-titanium ferrites from Fe 2 O 3 /Li 2 CO 3 /TiO 2 powder reagents, similar lithium-titanium oxide compounds are also formed in parallel with lithium ferrite at the initial stages of lithium-titanium ferrite formation. At later synthesis stages, lithiumtitanium ferrite is formed due to the interaction between the initially formed phases.

Study of Specific Saturation Magnetization
Figures 6 and 7 present the specific saturation magnetization of the samples versus their synthesis time. The value of σ S for all the samples was observed to grow at increased synthesis time due to the increased magnetic phase. The behavior of the σ S curves was generally like that of the above concentration curves. At a short synthesis time, the σ S value sharply increased; at a synthesis time exceeding 30 min, the sample magnetization varied insignificantly.
Li4Ti5O12, can be formed during lithium-titanium ferrite synthesis [10]. Apparently, during synthesis of lithium-titanium ferrites from Fe2O3/Li2CO3/TiO2 powder reagents, similar lithium-titanium oxide compounds are also formed in parallel with lithium ferrite at the initial stages of lithium-titanium ferrite formation. At later synthesis stages, lithium-titanium ferrite is formed due to the interaction between the initially formed phases. Figures 6 and 7 present the specific saturation magnetization of the samples versus their synthesis time. The value of σS for all the samples was observed to grow at increased synthesis time due to the increased magnetic phase. The behavior of the σS curves was generally like that of the above concentration curves. At a short synthesis time, the σS value sharply increased; at a synthesis time exceeding 30 min, the sample magnetization varied insignificantly.

Study of Specific Saturation Magnetization
The specific saturation magnetization of lithium-titanium ferrite is known to be lower than that of lithium ferrite. Therefore, the fluctuations in the behavior of the kinetic curves are related to competing processes in the formation of both unsubstituted lithium ferrite and substituted lithium-titanium ferrites.
Comparison of the curves revealed a higher degree of magnetization in the samples synthesized from mechanically activated reagents as compared to those obtained from powder or compacted samples. In addition, at increased grinding energy intensity, the content of the magnetic phase in the samples grew. In this case, the samples mechanically activated at 2220 rpm and synthesized by electron-beam heating show a high magnetization value of 50 emu/g, which is like the data reported in [24,45]. This is related to the decreased particle size and increased inter-particle contact area caused by mechanical treatment, which increases the rate of ferrite formation.

Thermomagnetometric Analysis of Samples
A thermomagnetometric (TM) analysis was performed to estimate the Curie temperature of the synthesized samples. As reported in [47], the XRD and TM data can be used to estimate the complete phase composition of ferrites more accurately. Figures 8 and 9 show the results of the TM analysis, including thermogravimetric (TG) data in the external magnetic field, derivative thermogravimetric (DTG) data, and differential scanning calorimetry (DSC) data for the samples synthesized for 120 min. The TG and DSC measurements were performed simultaneously at a high heating rate of 50 K/min. The specific saturation magnetization of lithium-titanium ferrite is known to be lower than that of lithium ferrite. Therefore, the fluctuations in the behavior of the kinetic curves are related to competing processes in the formation of both unsubstituted lithium ferrite and substituted lithium-titanium ferrites.
Comparison of the curves revealed a higher degree of magnetization in the samples synthesized from mechanically activated reagents as compared to those obtained from powder or compacted samples. In addition, at increased grinding energy intensity, the content of the magnetic phase in the samples grew. In this case, the samples mechanically activated at 2220 rpm and synthesized by electron-beam heating show a high magnetization value of 50 emu/g, which is like the data reported in [24,45]. This is related to the decreased particle size and increased inter-particle contact area caused by mechanical treatment, which increases the rate of ferrite formation.

Thermomagnetometric Analysis of Samples
A thermomagnetometric (TM) analysis was performed to estimate the Curie temperature of the synthesized samples. As reported in [47], the XRD and TM data can be used to estimate the complete phase composition of ferrites more accurately. Figures 8 and 9 show the results of the TM analysis, including thermogravimetric (TG) data in the external magnetic field, derivative thermogravimetric (DTG) data, and differential scanning calorimetry (DSC) data for the samples synthesized for 120 min. The TG and DSC measurements were performed simultaneously at a high heating rate of 50 K/min.

Thermomagnetometric Analysis of Samples
A thermomagnetometric (TM) analysis was performed to estimate the Curie temperature of the synthesized samples. As reported in [47], the XRD and TM data can be used to estimate the complete phase composition of ferrites more accurately. Figures 8 and 9 show the results of the TM analysis, including thermogravimetric (TG) data in the external magnetic field, derivative thermogravimetric (DTG) data, and differential scanning calorimetry (DSC) data for the samples synthesized for 120 min. The TG and DSC measurements were performed simultaneously at a high heating rate of 50 K/min. transition is 12 J/g. Therefore, the concentration of the synthesized unsubstituted lithium ferrite estimated from the DSC peak is 21.6 wt.%, which is similar to the spinel phase concentration of 20 wt% according to the XRD analysis of this sample. The presence of Li0.5Fe2.5O4 in this sample is evidenced by the Curie temperature, which corresponds to the magnetic phase transition and equals 629 °C. At this temperature, the TG curve shows a weight jump caused by termination of the sample interaction with the external magnetic field. The obtained value of the Curie point is similar to that reported in [22,49]. For the powder sample synthesized at 600 • C by thermal heating (Figure 8, T_600_ powder sample), the decreased weight in the TG curve and DSC peak at 727 • C is caused by the CO 2 release due to the synthesis reaction between the residues of the Fe 2 O 3 /Li 2 CO 3 /TiO 2 initial reagents according to Equation (1). The calculated weight loss corresponding to the complete reaction equals 6.17% with an enthalpy of 100 J/g [48]. Therefore, the obtained value of 2.54% with an enthalpy of 56 J/g in the TG curve corresponds to approximately 41% completion of the reaction. For the compact sample (Figure 8, T_600_compact), the observed reaction was associated with the presence of initial reagents that did not react during synthesis, but were less numerous. In this case, the peak observed in the DSC curve at 753 • C is due to the "order-disorder" phase transition in the synthesized lithium ferrite α-Li 0.5 Fe 2.5 O 4 . According to [22], the enthalpy of a complete transition is 12 J/g. Therefore, the concentration of the synthesized unsubstituted lithium ferrite estimated from the DSC peak is 21.6 wt.%, which is similar to the spinel phase concentration of 20 wt% according to the XRD analysis of this sample. The presence of Li 0.5 Fe 2.5 O 4 in this sample is evidenced by the Curie temperature, which corresponds to the magnetic phase transition and equals 629 • C. At this temperature, the TG curve shows a weight jump caused by termination of the sample interaction with the external magnetic field. The obtained value of the Curie point is similar to that reported in [22,49]. Figure 8 shows that the height of the weight jump in the magnetic phase transition and the peak area of the DSC α→β transition in the Li 0.5 Fe 2.5 O 4 phase, which characterize the concentration content of lithium ferrite, depend on the synthesis temperature, heating method, and sample activity. When the temperature of thermal synthesis was increased to 750 • C, the content of lithium ferrite increases, whereas the increased RT synthesis temperature reduces the content of the Li 0.5 Fe 2.5 O 4 phase in the samples.
The samples preliminarily mechanically activated and then synthesized do not contain unsubstituted lithium ferrite (Figure 9). The Curie temperature depends on the rotation speed of mechanical grinding. Variation in the Curie temperature corresponds to the formation of lithium-titanium ferrites with different titanium substitutions [47]. Thus, the formation of lithium-titanium ferrite is more intensive during synthesis of samples from mechanically activated powders. In this case, samples mechanically activated at 2220 rpm and synthesized by electron-beam heating show a Curie temperature of 534 • C, which is like that for the T C of Li 0.6 Fe 2.2 Ti 0.2 O 4 ferrite [47].
Apparently, mechanical grinding of the mixture of different powders not only decreases the particle size, but also significantly increases the number of triple contacts Li 2 CO 3 -Fe 2 O 3-TiO 2 , which stimulates a rapid formation of substituted lithiumtitanium ferrites.
Electron-beam heating of the studied material yields many free electrons, which can briefly reduce the charge of cations (Fe 3+ и Ti 4+ ). The change in the charge of cations decreases the energy of their electrostatic interaction with surrounding anions, thereby increasing their mobility. Consequently, an accelerated fusion reaction under RT heating can occur only if the initial stage of solid-phase interaction is limited by the diffusion of multiply charged Fe 3+ and Ti 4+ cations. In addition, local overheating of interfaces can occur due to nonradiative recombination of radiation-generated electronic excitations at the interfaces. In mechanically activated powders, the density of inter-particle contacts increases, which leads to local temperature gradients during RT treatment and changes the diffusion activity of particles.
The increased time of electron-beam heating minimizes the contribution of the radiation component to the diffusion driving force. This is due to the slower interaction of transition phases and subsequent formation of single-phase lithium-titanium ferrite.
Based on the obtained results, technological regimes were developed for efficient synthesis of lithium-titanium ferrites. Figure 10 presents the scheme that includes the mechanical activation of mixtures of ferrite reagents Li 2 CO 3 -Fe 2 O 3 -TiO 2 and their subsequent heating by a pulsed electron beam.

Conclusions
The XRD and TM analyses were employed to study the solid-phase synthesis of lithium-titanium ferrite with the chemical composition of Li0.6Fe2.2Ti0.2O4 under electron-beam heating of a mixture of Fe2O3-Li2CO3-TiO2 initial reagents at 600 and 750 °C, and to analyze the specific saturation magnetization. The initial reagents were used in the form of a powder, a compact, and a powder mechanically activated at different speeds. The

Conclusions
The XRD and TM analyses were employed to study the solid-phase synthesis of lithium-titanium ferrite with the chemical composition of Li 0.6 Fe 2.2 Ti 0.2 O 4 under electronbeam heating of a mixture of Fe 2 O 3 -Li 2 CO 3 -TiO 2 initial reagents at 600 and 750 • C, and to analyze the specific saturation magnetization. The initial reagents were used in the form of a powder, a compact, and a powder mechanically activated at different speeds. The parameters of the samples synthesized by electron-beam heating were compared with those prepared using the conventional ceramic technology.
It was shown that the rate of lithium-titanium ferrite synthesis depends on the history of powder reagents, type of heating, and temperature and time of synthesis. Increased temperature and time of synthesis increased the concentration of the spinel ferrite phase in the samples, which is typical of ferrite synthesis. The compaction of samples increased the rate of ferrite synthesis, but the temperature-time regimes used caused the synthesis of unsubstituted lithium ferrite of Li 0.5 Fe 2.5 O 4 , which is a transition product of the reaction during synthesis of substituted lithium ferrites. The formation of lithium ferrite was confirmed by the registered Curie temperature, equal to 631-634 • C depending on the heating method.
Preliminary mechanical activation of the Fe 2 O 3 -Li 2 CO 3 -TiO 2 initial powder reagents significantly accelerated ferrite formation under both types of heating. When the energy intensity of grinding that depends on the cup rotation speed (1290 and 2220 rpm) grew, the concentration of the synthesized spinel phase increased, which is associated with the formation of lithium substituted phases of variable composition.
Under electron-beam heating mixtures of initial reagents, the rate of lithium-titanium ferrite formation was much higher than that during conventional thermal synthesis. This was evidenced by a faster transformation of the initial reagents into spinel ferrite phases in the entire temperature range. In this case, the samples mechanically activated at a high rotation speed of 2220 rpm and synthesized by electron-beam heating at 750 • C were characterized by a single-phase ferrite composition of Li 0.6 Fe 2.2 Ti 0.2 O 4 , which was confirmed by data on the Curie temperature (534 • C) and specific saturation magnetization (50 emu/g).
The paper proposes a technological scheme for efficient synthesis of substituted lithium-titanium ferrites at a lower temperature (750 • C) and a shorter synthesis time (120 min) compared to traditional ceramic technology, which requires the use of high synthesis temperature of about 1000 • C and long synthesis time over 10 h.

Data Availability Statement:
The data presented in this study are available from the corresponding authors upon reasonable request.