materials

Article Improving the Magnetic Properties of Non-Oriented Electrical Steels by Secondary Recrystallization Using Dynamic Heating Conditions

Ivan Petryshynets 1,*, František Kováˇc 1, Branislav Petrov 1,2,3, Ladislav Falat 1 and Viktor Puchý 1 1 Institute of Materials Research, Slovak Academy of Sciences, Watsonova 47, 04001 Košice, Slovakia; [email protected] (F.K.); [email protected] (B.P.); [email protected] (L.F.); [email protected] (V.P.) 2 Faculty of Materials, Metallurgy and Recycling, Technical University of Košice, Letná 9, 04200 Košice, Slovakia 3 Embraco Slovakia, s.r.o., Odorínska cesta 2, 052 01 Spišská Nová Ves, Slovakia * Correspondence: [email protected]; Tel.: +421-55-792-2442



Received: 23 May 2019; Accepted: 11 June 2019; Published: 13 June 2019


Abstract: In the present work, we have used unconventional short-term secondary recrystallization heat treatment employing extraordinary high heating rate to develop coarse-grained microstructure with enhanced intensity of rotating cube texture {100}<011> in semi-finish vacuum degassed non-oriented electrical steels. The soft magnetic properties were improved through the increase of grains size with favourable cube crystallographic orientation. The appropriate final textural state of the treated experimental steels was achieved by strain-induced grain boundary migration mechanism, activated by gradient of accumulated stored deformation energy between neighbouring grains after the application of soft cold work, combined with steep temperature gradient during subsequent heat treatment under dynamic heating conditions. The materials in our experimentally prepared material states were mounted on the stator and rotor segments of electrical motors and examined for their efficiency in real operational conditions. Moreover, conventionally long-term heat treated materials, prepared in industrial conditions, were also tested for reference. The results show that the electrical motor containing the segments treated by our innovative approach, exhibits more than 1.2% higher efficiency, compared to the motor containing conventionally heat treated materials. The obtained efficiency enhancement can be directly related to the improved microstructural and textural characteristics of our unconventionally heat treated materials, specifically the homogenous coarse grained microstructure and the high intensity of cube and Goss crystallographic texture.

Keywords: electrical steel; crystallographic texture; magnetic losses; efficiency of electrical motor

1. Introduction The non-oriented (NO) electrical steels belong to the group of soft magnetic materials. They represent important materials used for core laminations in the majority of electrical devices, and they are contributing to the efficiency improvement of the equipment [1]. That is why the NO steels have to possess appropriate magnetic properties such as high magnetic induction, high magnetic permeability, low coercive fields, and low core losses in all plane directions [2,3]. Because of magnetic properties isotropy, these steels are typically used in various electromechanical rotating machines, e.g., in electrical motors, generators, and actuator which transform electrical energy to mechanical and vice versa [4]. It is well-known that magnetic properties of NO steels strongly depend on several different factors, mainly the chemical composition (especially, the silicon content), microstructural state (i.e., grain size and grains matrix distribution), crystallographic texture (i.e., rotated cube and

Materials 2019, 12, 1914; doi:10.3390/ma12121914 www.mdpi.com/journal/materials Materials 2019, 12, 1914 2 of 15

Goss texture component) and other factors which also include the strip thickness, impurities, isolation, mechanical stress, sheet thickness, etc. In order to improve the magnetic permeability and reduce total power loss, the coarse-grained microstructure with <100>//ND texture (θ-fibre), which comprises the cube, rotated cube as well as all orientations with the {001} planes parallel to the sheet plane, is desired since it contains the highest number of the easy magnetization axes <001> in the sheet plane. The <111>//ND (γ- fibre) texture has the hard magnetization axes <111> in the sheet plane, and thus needs to be suppressed in the final steel sheets [5–9]. Non-oriented electrical steels are divided into two groups: fully-processed (i.e., fully-finished) and semi-processed (i.e., semi-finished) grade. The fully-finished grade is a final product of conventionally processed NO steel. It has the microstructure, magneto-crystalline texture, and specific magnetic properties which have been adjusted through hot band annealing, cold rolling and final annealing of thin steel strip. This material can be readily used for the final assembly by the equipment manufacturer [10,11]. In contrast, semi-processed electrical steels are finished to their final thickness by the steel producer and are then subjected to the cutting, stacking and final assembly by the customers. In order to eliminate any undesired effects of such operations on final magnetic properties of NO steels, their final annealing is required. After cold rolling, semi-processed steel coils are generally temper rolled (i.e., skin passed) and then they are subjected to final recrystallization annealing to improve their punchability, facilitate strain-induced grain growth during the annealing at final producers of electromechanical machines. It has been estimated that the temper rolling introduces a soft deformation into the material, which accounts for 5–10% reduction in its thickness [12–14]. Thus, the semi-finished steels possess some extent of stored deformation energy, accumulated in dislocation structures, which can provide a sufficient driving force for a selective grain growth process, also known as strain-induced grain boundary migration phenomenon [15]. It is generally accepted, that the stored deformation energy is proportional to the amount of slip activity, which in polycrystalline materials depends on grain orientation plane, according to the sequence E{1 1 1} > E{1 1 2} > E{1 0 0} [16,17], where {hkl} represents the plane parallel to the rolling plane in Miller indices notations. The stored energy of cold work is therefore supposed to change from grain to grain according to the local crystallographic orientation as a function of applied stress. This means that the small rolling strains induce a gradient of local stresses between neighbouring grains, randomly distributed through the sheet thickness. Depending on their crystallographic orientations, the phenomenon of strain-induced grain migration can be used to increase the grains size during the second annealing (i.e., secondary recrystallization) treatment [18,19]. In industrial conditions, the final heat treatment of semi-finished NO electrical steels is carried out according to EN 10 341 standard [20]. In this case, the conventional long-term annealing process of steel laminations is used for the grain size increase by the mechanism of deformation-induced grain growth and the elimination of residual punching stresses, thereby eliminating all the deleterious effects affecting final magnetic properties. However, the main disadvantage of such conventional treatment comes from a limitation of the heating rate during annealing that leads to early recovery processes in the temper rolled NO steel. Consequently, such a premature onset of the recovery processes lowers the driving force for deformation-induced grain boundary motion before achieving the optimal annealing temperature. Moreover, the whole conventional process cycle: heating, annealing and cooling lasts more than 10 h. One important type of electromechanical machines that convert electric energy into mechanical energy is represented by electric motors. Despite differences in size and type, all electric motors in principle work the same way: an electric current flowing through a wire coil in a magnetic field creates a force that rotates the coil, thus creating torque. The main core materials which provide the magnetic flux-carrying member in most electric motors are non-oriented electrical steels. Because of this reason, the final magnetic properties improving these materials allow for an increase of the electrical motor efficiency. In order to understand how the heat treatment conditions examined in this work influence the working characteristics of rotating machines, the operational testing of a electrical motor was Materials 2019, 12, 1914 3 of 15

carriedMaterials 2019 out., 12 It, x isFOR important PEER REVIEW to say that four different kinds of losses occur in a motor: electrical3 of 15 losses, magnetic losses, mechanical losses and stray (i.e., eddy current) losses [21]. These losses can be reducedThese losses by using can qualitybe reduced materials, by using as well quality by optimizing materials, the as design. well by The optimizing magnetic the losses design. occurring The inmagnetic steel laminations losses occurring of stator in and steel rotor laminations segments of can stator be lowered and rotor via improvementsegments can of be their lowered chemical via composition,improvement microstructure, of their chemical substructure, composition, and micr textureostructure, parameters. substructure, and texture parameters. Our previous works [[22–25]22–25] were focused mostly on the investigation of strain-induced selective grain growth mechanism, depending on the valuevalue of accumulatedaccumulated deformation achieved by the temper rolling processes.processes. It has been presented that in NO silicon steels subjected to soft cold rolling deformation (i.e., 2-6% thickness reduction), columnar or coarsened-grains matrix with randomly distributed grains, showing significantsignificant intensity of rotatedrotated cubecube crystallographiccrystallographic orientation,orientation, could only bebe achieved achieved at veryat very high high heating heating rates. rates. Furthermore, Furthermore, the obtained the obtained experimental experimental findings indicatedfindings thatindicated the annealing that the annealing temperature temperature has to be higherhas to thanbe higher 900 ◦C. than After 900 such °C. dynamicAfter such heat dynamic treatments, heat thetreatments, achieved the microstructure achieved microstructure and texture of and the textur treatede of materials the treated were materials found to havewere afound clearly to positive have a aclearlyffect on positive their final affect magnetic on their properties, final magnetic namely properties, the core losses namely and the coercivity. core losses Even and more, coercivity. the proposed Even approachmore, the allowedproposed not approach only for allowed an improvement not only for of an the improvement final magnetic of parametersthe final magnetic of NO steelsparameters but it alsoof NO enabled steels but a reduction it also enabled of the finala reduction production of the costs final for production end customers. costs for end customers. In contrastcontrast to thethe scientificscientific worksworks [[26–28],26–28], addressingaddressing the useuse ofof unconventionalunconventional cold rolling schemes (e.g., with didifferentfferent rolling anglesangles to the rolling direction) and subsequent conventional heat treatment with low heatingheating rate, wewe havehave proposedproposed aa combinationcombination of aa conventionalconventional tempertemper rollingrolling process forfor thethe steelsteel afterafter primaryprimary recrystallizationrecrystallization withwith aa subsequentsubsequent final,final, unconventionalunconventional secondsecond annealing treatmenttreatment atat dynamic dynamic heating heating conditions, conditions, i.e., i.e., using using extraordinary extraordinary high high heating heating rate. rate. On theOn otherthe other hand, hand, other other research research studies studies [29,30] have[29,30] clearly have indicated clearly aindicated beneficial a effbeneect official rapid effect annealing of rapid on theannealing increase on of the average increase grain of size average and significant grain size improvement and significant of crystallographic improvement of texture crystallographic of electrical steeltexture after of strip-castingelectrical steel and after cold strip-casting rolling process. and cold rolling process. In the present study, rotorrotor andand statorstator segments,segments, manufacturedmanufactured by shear cutting of semi-finishedsemi-finished steel, were subjected to the second annealing annealing treatm treatmentent in our our laboratory laboratory dynamic dynamic heating heating conditions. conditions. The mainmain aim aim of presentof present investigation investigation was towas compare to compare and discuss and thediscuss results the obtained results from obtained operational from torqueoperational load teststorque of load electrical tests motorsof electrical assembled motors from assembled NO steel from laminations, NO steel laminations, individually individually heat treated eitherheat treated in laboratory either in or laboratory industrial or heat industrial treatment heat conditions. treatment conditions.

2. Materials and Methods The material investigated in this work was a commercially available semi-finished semi-finished non-oriented electrical steelsteel of of M450-50A M450-50A grade, grade, in thein formthe form of sheets of sheets with 0.5with mm 0.5 in thicknessmm in thickness and following and following chemical compositionchemical composition in wt.%: in Fe wt.%:= 97.95%, Fe = C97.95%,= 0.006%, C = 0. Si006%,= 1.48%, Si = Mn1.48%,= 0.25%, Mn = P0.25%,= 0.040%, P = 0.040%,Al = 0.18%, Al = other0.18%, elements other elements ~0.094%. ~0.094%. Schematic Schematic overview overvi of experimentalew of experimental procedures, procedures, which were which involved were in experimentalinvolved in experimental investigation investigation of the studied of steel,the studied is displayed steel, is in displayed Figure1. in Figure 1.

Figure 1. Scheme of experimental procedures which were used for investigated steel. Figure 1. Scheme of experimental procedures which were used for investigated steel. The experimental samples were in form of stator and rotor segments (see Figure2). These segments wereThe prepared experimental by shear samples cutting inwere industrial in form conditions, of stator inand accordance rotor segments with common (see Figure electro-motors 2). These manufacturingsegments were technology.prepared by More shear than cutting 80 experimentalin industrial segmentsconditions, were in accordance heat treated with in laboratory common conditionselectro-motors and thenmanufacturing used as core technology. material of More rotor than and stator80 experimental in electrical segments motor. The were final heat performance treated in parameterslaboratory conditions of experimentally and then prepared used as rotating core material machine of (i.e., rotor by and using stator laboratory in electrical heat treated motor. segments) The final performance parameters of experimentally prepared rotating machine (i.e., by using laboratory heat treated segments) were compared with those of the electrical motor assembled, according to conventional industrial technology (i.e. by using industrially heat treated segments).

Materials 2019, 12, 1914 4 of 15 were compared with those of the electrical motor assembled, according to conventional industrial Materials 2019, 12, x FOR PEER REVIEW 4 of 15 technologyMaterials 2019, (i.e.12, x byFOR using PEER industriallyREVIEW heat treated segments). 4 of 15

(a) (b) (a) (b) Figure 2. Experimental samples in form of stator (a) and rotor (b) segments of electric motor. FigureFigure 2. Experimental 2. Experimental samples samples in form in form of stator of stator (a) and (a) androtor rotor (b) segments (b) segments of electric of electric motor. motor.

The heat treatments of experimental segments were individually carried out at either The heat treatments of experimental segments were individually carried out at either unconventional dynamic heating or conventional long-term annealing conditions, according to the unconventional dynamic dynamic heating heating or or conventional conventional long-term long-term annealing annealing conditions, conditions, according according to the to regimes schematically shown in Figure 3. The dynamic heat treatment of experimental segments was theregimes regimes schematically schematically shown shown in Figure in Figure 3. The3. The dyna dynamicmic heat heat treatment treatment of experimental of experimental segments segments was performed in laboratory conditions using electric resistance furnace Nabertherm RS 120/1000/13 wasperformed performed in laboratory in laboratory conditions conditions using using electr electricic resistance resistance furnace furnace Nabertherm Nabertherm RS RS 120/1000/13 120/1000/13 (Nabertherm GmbH, Lilienthal, Germany). The stator and rotor laminations were heated up to 950 (Nabertherm GmbH,GmbH, Lilienthal,Lilienthal, Germany). Germany). The The stator stator and and rotor rotor laminations laminations were were heated heated up up to 950to 950◦C °C at a heating rate of about 12 °C/s (see the solid blue line in Figure 3). Then, the segments were kept at°C aat heating a heating rate rate of aboutof about 12 ◦12C /°C/ss (see (see the the solid solid blue blue line line in Figurein Figure3). 3). Then, Then, the the segments segments were were kept kept at at the annealing temperature for 10 min. The annealing atmosphere was pure hydrogen, d.p. (dew theat the annealing annealing temperature temperature for 10for min. 10 min. The The annealing annealing atmosphere atmosphere was purewas hydrogen,pure hydrogen, d.p. (dew d.p. point)(dew point) ~25 °C. Due to dimensional limitations of the work space at the laboratory furnace used, our ~25point)◦C. ~25 Due °C. to dimensionalDue to dimensional limitations limitations of the work of the space work at thespace laboratory at the laboratory furnace used, furnace our laboratoryused, our laboratory heat treatment of large number of electro-motor core segments was carried out heatlaboratory treatment heat of treatment large number of large of electro-motor number of core electro-motor segmentswas core carried segments out sequentially,was carried i.e., out a sequentially, i.e., a series of maximum five segments on a special in-house made sample holder was seriessequentially, of maximum i.e., a fiveseries segments of maximum on a special five segmen in-housets on made a special sample in-house holder was made heat sample treated holder per single was heat treated per single annealing process. This procedure was repeated until full completion the heat annealingheat treated process. per single This annealing procedure process. was repeated This procedure until full was completion repeated until the heat full treatmentcompletion for the all heat the treatment for all the experimental segments. experimentaltreatment for segments.all the experimental segments.

FigureFigure 3.3. SchematicSchematic representationrepresentation ofof individualindividual heatheat treatmenttreatment processesprocesses ofof experimentalexperimental material:material: Figure 3. Schematic representation of individual heat treatment processes of experimental material: solidsolid blueblue line—unconventionalline—unconventional dynamicdynamic heatheat treatmenttreatment ofof statorstator andand rotorrotor laminationslaminations andand dasheddashed solid blue line—unconventional dynamic heat treatment of stator and rotor laminations and dashed yellowyellow line—conventionalline—conventional long-termlong-term annealingannealing treatmenttreatment at at industrial industrial conditions. conditions. yellow line—conventional long-term annealing treatment at industrial conditions. The long-term annealing treatment was carried out at an industrial line of conventional heat The long-term annealing treatment was carried out at an industrial line of conventional heat treatmentThe long-term according annealing to the EN 10treatment 341 standard was carried and it is out schematically at an industrial presented line byof conventional the dashed yellow heat treatment according to the EN 10 341 standard and it is schematically presented by the dashed yellow treatment according to the EN 10 341 standard and it is schematically presented by the dashed yellow line in Figure 3. The duration of conventional industrial heat treatment was about 12.5 h. It is line in Figure 3. The duration of conventional industrial heat treatment was about 12.5 h. It is important to note that the final phase of cooling process included also the treatment at 540 °C/1 h in important to note that the final phase of cooling process included also the treatment at 540 °C/1 h in a special atmosphere, which provided the formation of so-called “surface blue” isolation layer on the a special atmosphere, which provided the formation of so-called “surface blue” isolation layer on the base of complex iron (II, III) oxide Fe3O4, i.e., FeO·Fe2O3. This layer (typical 0.5–1.5 µm) is responsible base of complex iron (II, III) oxide Fe3O4, i.e., FeO·Fe2O3. This layer (typical 0.5–1.5 µm) is responsible

Materials 2019, 12, 1914 5 of 15 line in Figure3. The duration of conventional industrial heat treatment was about 12.5 h. It is important to note that the final phase of cooling process included also the treatment at 540 ◦C/1 h in a special atmosphere, which provided the formation of so-called “surface blue” isolation layer on the base of complex iron (II, III) oxide Fe O , i.e., FeO Fe O . This layer (typical 0.5–1.5 µm) is responsible for the 3 4 · 2 3 isolation between each segment in stator and rotor of electrical motors and thus it reduces eddy-current losses. In order to achieve required isolation layer on the surface of rotor and stator segments heat treated in laboratory conditions, they were additionally processed within the final separated box of industrial heat treatment line. The selected representative samples were used for the microstructure and texture analyses. The texture analysis was carried out by means of electron back-scattered diffraction (EBSD) method in the normal direction plane for each sample of 25 mm 10 mm in size. The scanning electron × microscope (SEM) JEOL JSM 7000F FEG (Jeol Ltd., Tokyo, Japan) with the EBSD detector Nordlys-I (HKL technology A/S, Hobro, Denmark) were used to perform the texture analysis. The obtained EBSD data were processed by the CHANNEL-5, HKL software package (Service pack 7). The magnetic properties of the segments, heat treated according to both individual annealing schemes, were measured on the samples with planar dimensions 45 mm 15 mm. These samples × were prepared by electrical discharge machining using spark erosion machine EIR-EMO 2N (Emotek s.r.o., Nové Mesto nad Váhom, Slovak Republic). In order to evaluate the magnetic properties of the heat treated segments in two main directions, these samples were cut from the segments along the rolling direction (RD) and transverse direction (TD), see Figure2. The magnetic measurements in direct current (DC) and alternating current (AC) magnetic field conditions were carried out using magnetic measuring instrument Brockhaus MPG 100D (Dr. Brockhaus Messtechnik GmbH & Co. KG, Lüdenscheid, Germany). The AC hysteresis loop measurements were performed at a frequency of 50 Hz. The measurements of electrical motors efficiency, which were constructed from the segments heat treated in either short-term (laboratory) dynamic heating or long-term (industrial) static annealing conditions, were carried out by operational torque load testing of electrical rotating machines at the electric motor test station on the manufacturer workplace. For the efficiency evaluation, a so-called direct method was used, which is generally considered to be very accurate. The measurement of the efficiency of electric motors was made directly using the equation:

Mechanical Output Power 100% Efficiency % = · , (1) Electrical Input Power

Thus, it was required to measure both the mechanical output power and the electrical input power. The electric input power was measured with satisfactorily good accuracy. The mechanical power was given by the equation: Mechanical Power = Torque Angular Speed (2) · The speed measurement is a relatively simple procedure which can provide the achievement of quite accurate results ( 1 rpm). The torque measurement was carried out by dynamometer model ± DINA06TRO07 (Kropy industrial Ltd., Joinville, Brazil), which had the possibility of creating a controllable variable load, fitted with an accurate torque transducer.

3. Results and Discussion

3.1. Microstructure Representative samples from the electric motor segments, heat treated using the two different heat treatment conditions, were metalographically prepared for microscopic analysis of their microstructural state using light optical microscope OLYMPUS GX71 (OLYMPUS Europa Holding GmbH, Hamburg, Germany). The evaluation of main microstructural parameters was carried out Materials 2019, 12, 1914 6 of 15

through the sheet plane cross-section parallel to the rolling direction. The initial microstructure of stator and rotor segments produced of semi-finished non-oriented steel before the heat treatment is shown in Figure4. It can be seen that the experimental steel in its initial material state is characterised by quite fine-grained homogenous microstructure with an average grain size of 17 µm 3 µm. Materials 2019, 12, x FOR PEER REVIEW ± 6 of 15 Materials 2019, 12, x FOR PEER REVIEW 6 of 15

FigureFigure 4. The 4. The initial initial microstructure microstructure of stator of stator segment segment obtained obtained from from semi-finished semi-finished non-oriented non-oriented steel steel beforebeforeFigure heat heat 4.treatment. The treatment. initial microstructure of stator segment obtained from semi-finished non-oriented steel before heat treatment. TheThe grain grain matrixmatrix evolutions evolutions of investigatedof investigat segmentsed segments subjected subjected to the secondary to the recrystallization secondary recrystallizationannealingThe ingrain conventional annealing matrix in long-termevolutions conventional andof unconventional investigatlong-termed andsegments dynamic unconventional subjected annealing dynamic conditionsto the secondaryannealing are shown conditionsinrecrystallization Figure are5a,b, shown respectively. annealing in Figure It in5a,b, canconventional respectively. be clearly seenlo ng-termIt can that be bothand clearly samplesunconventional seen that have both the dynamic monophasicsamples annealing have ferrite the conditions are shown in Figure 5a,b, respectively. It can be clearly seen that both samples have the monophasiccoarse-grained ferrite microstructure, coarse-grained which microstructure, comply with which requirements comply with for therequirements final magnetic for the properties final monophasic ferrite coarse-grained microstructure, which comply with requirements for the final magneticof rotating properties equipment of rotating core material. equipment In the core case ofmaterial. the segments In the treated case of according the segments to the ENtreated 10 341 magnetic properties of rotating equipment core material. In the case of the segments treated accordingstandard, to thethe cross-sectionEN 10 341 standard, microstructure the cross-sect is composedion microstructure of inhomogeneous is composed alternation of inhomogeneous of grains with according to the EN 10 341 standard, the cross-section microstructure is composed of inhomogeneous alternationcolumnar of and grains/or equiaxial with columnar symmetric and/or structure. equiaxial The light symmetric optical microscopicstructure. The analysis light shows optical that alternation of grains with columnar and/or equiaxial symmetric structure. The light optical microscopicthe conventionally analysis shows heat treated that the segments conventionally are characterized heat treated by segments the average are graincharacterized size of 120 by µthem microscopic analysis shows that the conventionally heat treated segments are characterized by the ± µ average10averagem, grain see grain Figure size size of5a. 120of The 120 µm microstructural µm ± 10 ± 10µm, µm, see see features Figure Figure of5a. 5a. laminations, The The microstructural microstructural which were featuresfeatures heat treated of laminations, at dynamic whichconditionswhich were were heat at 950heat treated◦ treatedC during at dynamicat dynamic 10 min, conditions isconditions presented at at950 in 950 Figure°C °C during during5b. The 10 10 minutes, lightminutes, optical isis presentedpresented image demonstrates in FigureFigure 5b.that The5b. the Thelight unconventionally light optical optical image image demonstrates heat demonstrates treated steelthat that the is the characterised unconventionally unconventionally by coarse heat heat grains treated treated matrix steelsteel is with characterised a mean grain bysize coarse of 250grainsµm matrix10 µ m.with a mean grain size of 250 µm ± 10 µm. by coarse grains± matrix with a mean grain size of 250 µm ± 10 µm.

(a) (a) ((bb)) FigureFigure 5. 5. The microstructure of experimental stator segments after: conventional long-term heat Figure 5. TheThe microstructure microstructure of ofexperimental experimental stator stator segments segments after: after: conventional conventional long-term long-term heat heat treatmenttreatment (a ()a and) and unconventional unconventional short-termshort-term dynamic dynamic heat heat treatment treatment (b ().b ). treatment (a) and unconventional short-term dynamic heat treatment (b). Evidently,Evidently, most most of of the the grainsgrains havehave a uniaxial shape shape with with warped warped boundaries boundaries and and are are mutually mutually Evidently, most of the grains have a uniaxial shape with warped boundaries and are mutually contiguouscontiguous to to each each other other inin thethe middlemiddle part of of th thee cross-section. cross-section. Such Such grains grains morphology morphology reasonably reasonably contiguous to each other in the middle part of the cross-section. Such grains morphology reasonably leadsleads to to the the conclusion conclusion that that theythey growgrow from the the sheet sheet sub-surface sub-surface region region to to its its central central part. part. It should It should leads to the conclusion that they grow from the sheet sub-surface region to its central part. It should bebe noted noted that that in in the the casecase ofof thethe samplessamples heat tr treatedeated according according to to the the industrial industrial scheme, scheme, a similar a similar be noted that in the case of the samples heat treated according to the industrial scheme, a similar tendencytendency of of thethe growth of of the the grains grains was was not notso obvi soous. obvious. On the On contrary, the contrary, the microstructure the microstructure in Figure in tendency of the growth of the grains was not so obvious. On the contrary, the microstructure in Figure Figure5a contains5a contains a lot of a lotgrains, of grains, with their with size their ranging size ranging from 40 fromµm up 40 toµ 200m up µm, to located 200 µm, mostly located in central mostly in 5a contains a lot of grains, with their size ranging from 40 µm up to 200 µm, located mostly in central centralpart of part the samples. of the samples. This arrangement This arrangement of grains may of grains be due may to the be effect due to of thea simultaneous effect of a simultaneous growth of part of the samples. This arrangement of grains may be due to the effect of a simultaneous growth of growthlarge number of large of number grains ofthrough grains the through thickness the of thickness the sample. of the It is sample. apparent, It isthat apparent, the observed that the largemicrostructural number of grains differences through of indi thevidual thickness experimental of the sample.segments It can is beapparent, directly relatedthat the to observedthe used observed microstructural differences of individual experimental segments can be directly related to microstructuralconditions of differences performed ofsecondary individual heat experimental treatments, whichsegments were can differing be directly from relatedeach other to theby theirused the used conditions of performed secondary heat treatments, which were differing from each other conditionsheating of rate, performed annealing secondary temperature heat treatments and holding, which time. were It differing is well-known from each that other secondary by their by their heating rate, annealing temperature and holding time. It is well-known that secondary heatingrecrystallization rate, annealing [31] leadstemperature to further and migration holding oftime. grain It boundariesis well-known through that the secondary primary recrystallization [31] leads to further migration of grain boundaries through the primary recrystallized recrystallizationrecrystallized structure,[31] leads thereby to further producing migration a struct ureof containinggrain boundaries a small number through of enlarged the primary grains. structure, thereby producing a structure containing a small number of enlarged grains. The driving recrystallizedThe driving structure, force for thereby the growth producing of new a grainsstruct ureis provided containing by a the small removal number of theof enlarged storage energygrains. The associateddriving force with for a theplastically growth deformed of new grains material is providedstate. In principle,by the removal a gradient of the of storage any intensive energy associatedthermodynamic with a plastically variable offers deformed a source material of such astate. driving In force. principle, In the casea gradient of the semi-finished of any intensive non- thermodynamicoriented steel variable material offers obtained a source after ofthe such temper a driving rolling force. process, In the with case a sheetof the thickness semi-finished reduction non- orientedranging steel from material 3% to obtained 10%, the after soft the cold temper work rollingdeformation process, energy with isa sheetstored thickness within dislocation reduction structures. The plastic anisotropy of grains in a polycrystalline microstructure results in a stored ranging from 3% to 10%, the soft cold work deformation energy is stored within dislocation structures. The plastic anisotropy of grains in a polycrystalline microstructure results in a stored

Materials 2019, 12, 1914 7 of 15 force for the growth of new grains is provided by the removal of the storage energy associated with a plastically deformed material state. In principle, a gradient of any intensive thermodynamic variable offers a source of such a driving force. In the case of the semi-finished non-oriented steel material obtained after the temper rolling process, with a sheet thickness reduction ranging from 3% to 10%, the soft cold work deformation energy is stored within dislocation structures. The plastic anisotropy of grains in a polycrystalline microstructure results in a stored energy difference across grain boundaries, which in turn can provide a driving force sufficient for so-called strain-induced grain boundary migration (SIBM). Further annealing of semi-processed steel laminations may result in grain growth, in which the smaller primary recrystallized grains are eliminated, the larger grains grow, and the grain boundaries assume a lower energy configuration. The microstructural analyses of the heat treated segments have shown that the strain-induced grain boundary migration mechanism enabled to achieve the microstructures with a smaller number of enlarged grains which had the desired effect on their magnetic properties. However, the difference in morphology of finally obtained microstructures indicate that in the case of secondary grain growth obtained in our laboratory dynamic heating conditions, the microstructure evolution is affected not only by the stored deformation energy, but also by the driving force related to the heat transfer phenomena during the thermal processing, see Figure5a,b. Special attention is given to the temperature distribution during the heating process. In case of the long-term annealing process, the heating rate is very slow, and the heating of the segments is homogeneous throughout the segments cross-section. These heating conditions enable the progress of recovery processes at lower temperatures and thus decreased the driving force associated with deformation storage energy, which is responsible for the SIBM. Moreover, long-term annealing gives an opportunity to develop a large number of primary recrystallized grains randomly distributed throughout the steel sheet thickness. As a result, the microstructure with the smaller average grain size was obtained after the conventional heat treatment, see Figure5a. On the contrary, the dynamic heat treatment procedure was characterised by the high heating rate. Sidor et al. [32] showed that rapid heating is responsible for the occurrence of temperature gradient along the normal of the sheet thickness. The strong temperature gradient is an additional major driving force, which in combination with the storage deformation energy within the short annealing time, results in an intensive growth of a smaller number of primary recrystallized grains, which mostly start to grow from the sheet surface and then continue in growth towards the sheet central part. It is evident from Figure5b that the similar grain growth was also achieved for the segment heat treated in dynamic heating conditions. However, the obtained microstructure (Figure5b) is characterised not only by a lower number of grains with much larger average size than in microstructure of the segments obtained after long-term annealing (Figure5a), but also by columnar and /or uniaxial grains related to their specific growth features.

3.2. Texture The crystallographic orientation or texture is an important parameter describing the magnetic properties of NO electrical steels. In order to characterize the texture evolution of the investigated segments of semi-finished NO steel in dependence of heat treatment conditions, specific EBSD analyses were performed. The common texture of experimental steel in its initial material state is shown in Figure6. The crystallographic orientation of primary recrystallized grains matrix is represented by an inverse pole figure (IPF) map and the orientation distribution function (ODF) in Figure6a,b, respectively. The EBSD data show that the fine-grained microstructure through the sheet thickness is characterised mostly by grains with deformation texture {111} (blue grains) and rotated cube {001}<110> (red grains) texture, see Figure6a. The most relevant texture components of this sample are α-, γ-, and θ- fibres represented by ODF in Figure6b. It can be clearly seen that the semi-finished state is characterised by the dominance of γ-fibre, which represents the deformation texture <111>//ND with two characteristic maxima {111}<112> and {554}<225>. This crystallographic texture component deteriorates the magnetic properties of silicon steel. However, the θ- fibre of ODF section (Figure6b) as well as the IPF map (Figure6a) clearly demonstrate the presence of weak peaks Materials 2019, 12, 1914 8 of 15

Materials 2019, 12, x FOR PEER REVIEW 8 of 15 of rotated cube {001}<110> grains, which are characterised by <100> directions and are responsible for easyMaterials magnetization 2019, 12, x FOR of PEER Fe-single REVIEW crystal. 8 of 15

(a) (b)

Figure 6. Characterization of the textural state of experimental steel in initial semi-finished state: IPF (a) (b) map (a) and ODF section taken at ϕ2 = 45° (b). The key for the identification of crystallographic orientationFigureFigure 6.6.Characterization Characterization of grain is located ofof theinthe the texturaltextural upper state stateright of ofcorner experimentalexperime of thental IPF steelsteel map. in in initial initial semi-finished semi-finished state: state: IPFIPF

mapmap (a(a)) andand ODF ODF section section taken taken at atφ ϕ2 2= =45 45°◦ ( b(b).). TheThe keykey forfor thethe identificationidentification ofof crystallographiccrystallographic Theorientationorientation IPF maps of of grain grainobtained is is located located on the in inthe segmentsthe upper upper right right treated corner corner in oflong-term of the the IPF IPF map. map. and dynamic annealing conditions are presented in Figure 7a,b, respectively. It can be seen that the secondary recrystallization generally improvedTheTheIPF IPFthemaps maps crystallographic obtainedobtainedon on the thetextures segmentssegments and treatedtreated the maximumin in long-termlong-term intensities andand dynamicdynamic have annealing annealingbeen considerablyconditions conditions modified,arearepresented presented compared in inFigure Figure to 7the a,b,7a,b, initial respectively. respectively. material It state.It can canbe be seen seen that that the the secondary secondary recrystallization recrystallization generally generally improvedimprovedIt is clearly the the crystallographic visiblecrystallographic that the textures recorded textures and EBSD theand maximum IPFthe mapsmaximum intensities (Figure intensities 7) have show been the have considerablysame been microstructural considerably modified, characteristicscomparedmodified, tocompared the as initiallight tooptical material the initial imagines state. material in Figure state. 5. It is clearly visible that the recorded EBSD IPF maps (Figure 7) show the same microstructural characteristics as light optical imagines in Figure 5.

(a) (b)

(a) (b)

(c)

FigureFigure 7. 7. IPFIPF map map ofof the the segments segments treated treated in long-term in long-term (a) and (a) anddynamic dynamic (b) annealing (b) annealing conditions. conditions. The (c) keyThe for key the for identification the identification of crystallographic of crystallographic orientation orientation of ofgrains grains (c). (c ).Here, Here, the the red red colour colour represents represents thetheFigure cube cube 7. crystallographic crystallographic IPF map of the segmentsorientation, orientation, treated the the green greenin long-term colour—Goss colour—Goss (a) and crystallographic crystallographicdynamic (b) annealing orientation orientation conditions. and and blue blue The colour—thecolour—thekey for the deformationidentification deformation texture. texture. of crystallographic orientation of grains (c). Here, the red colour represents the cube crystallographic orientation, the green colour—Goss crystallographic orientation and blue AItcolour—the iscomparison clearly visibledeformation of the that two texture. the IPF recorded maps in EBSD Figure IPF 7 maps clearly (Figure reveals7) showthat the the segments same microstructural annealed in dynamiccharacteristics conditions as light not optical only have imagines a much in Figurebetter 5grains. matrix than those treated in static conditions but theyAA comparison comparisonare also characterised ofof thethe twotwo by IPFIPF a high mapsmaps intensity inin FigureFigure of 7grains 7clearly clearly with reveals reveals appropriate that that the the{100} segments segments easy magnetisation annealed annealed in in axisdynamicdynamic parallelconditions conditions to the normalnot notonly only direction havehave aa much muchto thebetter better sheetgrains grainsplane.matrix matrix In thisthan than case,those those the treated grainstreated in withinstatic static rotatedconditions conditions cube componentbutbutthey theyare are coveralso also characterisedcharacterisedabout 50% of by analysedby aa highhigh intensitycross-sectionintensity ofof grainsgrains microstructure withwith appropriateappropriate (see Figure {100}{100} 7b). easyeasy The magnetisationmagnetisation IPF map in Figureaxisaxis parallelparallel 7a shows toto thethe that normalnormal secondary directiondirection recrystallized toto thethesheet sheet grainsplane. plane. of InInthe thisthis segments, case,case, thethe which grainsgrains were withwith rotatedheatrotated treated cubecube accordingcomponentcomponent to covercoverthe EN aboutabout 10 341 50% standard, of analysed are characterisedcross-section cross-section microstructuremostly microstructure by the deformation (see (see Figure Figure 7b). 7textureb). The The {111}//NDIPF IPF map map in andinFigure Figure weak 7a7 intensitya shows shows that thatof rotated secondary secondary cube recrystallized as well as Goss grains texture of thecomponent.the segments,segments, The whichwhich resulting werewere ODFsheat heat treated fortreated the colouredaccording IPF to themaps EN presented 10 341 standard, in Figure are characterised7a,b are shown mostly in Figureby the deformation8a,b, respectively. texture Here, {111}//ND the evolutionand weak of intensity crystallographic of rotated texture cube ofas thewell heat as Gosstreated texture segments component. is clearly The characterized resulting ODFs in terms for theof maincoloured fibre IPFcomponents. maps presented It is important in Figure to 7a,b note are that shown these inresults Figure represent 8a,b, respectively. the totally differentHere, the evolution of crystallographic texture of the heat treated segments is clearly characterized in terms of main fibre components. It is important to note that these results represent the totally different

Materials 2019, 12, 1914 9 of 15 according to the EN 10 341 standard, are characterised mostly by the deformation texture {111}//ND and weak intensity of rotated cube as well as Goss texture component. The resulting ODFs for the coloured IPF maps presented in Figure7a,b are shown in Figure8a,b, respectively. Here, the evolution ofMaterials crystallographic 2019, 12, x FOR texture PEER REVIEW of the heat treated segments is clearly characterized in terms of main9 of fibre15 components. It is important to note that these results represent the totally different arrangements of the texturearrangements components, of the compared texture components, to those observed compared in the to initialthose observed semi-finished in the material initial semi-finished state (Figure 6). Inmaterial the case ofstate the (Figure sample 6). in In Figure the case8a, it of is the visible sample that in the Figure texture 8a, is it still is visible characterised that the by texture domination is still of γ-fibrecharacterised with high by intensity domination of {111}of γ-fibre<112 >withand high {554} intensity<225> components.of {111}<112> and {554}<225> components.

(a) (b)

FigureFigure 8. 8.ODF ODF sections sections taken taken atatφ ϕ22 == 4545°◦ representing the the through- through-thicknessthickness textures textures evolved evolved after after heatheat treatment treatment in in long-term long-term ( a(a)) and and dynamicdynamic (b) heat treatment treatment conditions. conditions.

OnOn the the other other hand, hand, the the rotated rotated cube cube texture, texture, represented represented by byθ -fibre,θ-fibre, has has partially partially increased increased in in the areathe of area {100} of< 011{100}<011>> and {100} and< {100}<110>110> components. components. The textureThe texture of the of segments the segments heat treated heat treated in dynamic in heatingdynamic conditions heating conditions is represented is represented by ODF section by ODF in section Figure8 inb. Figure A quick 8b. visual A quick inspection visual inspection shows that comparedshows that to compared the primary to the recrystallized primary recrystallized state (Figure state6b), (Figure the three 6b), common the three texturecommon fibres texture are fibres totally changedare totally after changed the secondary after the recrystallization.secondary recrystallization. It is important It is important to note thatto noteγ-fibre that inγ-fibre this casein this was totallycase was reduced totally in reduced comparison in comparison with initial with state initial where state the where deformation the deformation texture had texture the maximumhad the intensity.maximum Moreover, intensity. the Moreover,θ-fibre was the significantly θ-fibre was improved. significantly It is evidentimproved. that It theis maximumevident that intensity the maximum intensity was achieved for the grains with rotated cube components {100}<011> and was achieved for the grains with rotated cube components {100}<011> and {100}<110>. However, {100}<110>. However, there is also a high intensity of {113}<361> component. The {113}<361> there is also a high intensity of {113} <361> component. The {113}<361> orientation belongs to the orientation belongs to the {11h}<12 > fibre [33] and is commonly found in low carbon steels after cold 1 {11h}<12 2 > fibre [33] and is commonly found in low carbon steels after cold rolling and annealing. Therefore,rolling and it isannealing. believed Therefore, that the {113} it is< believed361> component that the {113}<361> might have component resulted frommight the have rotated resulted cube, from the rotated cube, as it has also been pointed out in [34]. as it has also been pointed out in [34]. The EBSD measurements of the experimental segments clearly showed that secondary The EBSD measurements of the experimental segments clearly showed that secondary recrystallization conditions allow for the control of not just the evolution of their microstructure, but recrystallization conditions allow for the control of not just the evolution of their microstructure, but also also have a crucial effect on their resulting crystallographic texture. It has been presented that the have a crucial effect on their resulting crystallographic texture. It has been presented that the high heating high heating rate by dynamic heat treatment promotes the increase of rotated cube texture ratecomponent by dynamic despite heat treatmentof high intensity promotes of the deformation increase of texture rotated <111>//ND, cube texture which component dominated despite in of the high intensityprimary of recrystallized deformation matrix. texture It< can111 >//be ND,concluded which that dominated the strain-induced in the primary grain recrystallized boundary migration matrix. It canmechanism be concluded in combination that the strain-induced with heat flow grain gradient boundary induce migration the mostly mechanism selective grain in combination growth, which with heatis characterised flow gradient by induce the crystal the mostly lattice selective with ease grain magnetisation growth, which axis is <100> characterised parallel to by the the ND crystal of steel lattice withplane. ease The magnetisation authors [18,30] axis concluded<100> parallel that the to thestored ND deformation of steel plane. energy The authorsin polycrystalline [18,30] concluded materials, that thewhich stored were deformation subjected energyto cold in deformation polycrystalline is distributed materials, heterogeneously which were subjected between to coldthe grains deformation and is distributeddepends on heterogeneously their crystallographic between orientation, the grains andat least depends on orientations on their crystallographic families defined orientation, by the at leastcrystallographic on orientations plane families lyingdefined parallel byto the the sheet crystallographic plane. In vacuum plane degassed lying parallel ferrite to steels, the sheet the plane.stored In vacuumdeformation degassed energy ferrite can steels, be categorized the stored as deformation follows: E{1 1 energy1} > E{1 1 can2} > E be{1 0 categorized 0}. As one can as see, follows: the stored E{1 1 1} > E{1energy 1 2} > E {1increase 0 0}. As from one can the see, grains the stored with energy{001}<110> increase to the from grains the grains with with{111}<110>crystallographic {001}<110> to the grains withorientation. {111}<110 It> cancrystallographic be supposed orientation.that between It the can neighbouring be supposed grains that between with various the neighbouring crystallographic grains orientations, a creation of some small gradient of stored deformation energy is related to the lattice defects, such as dislocations, which were induced by the applied mechanical stress. The gradient of stored energy represents a driving force for strain-induced grain boundary migration mechanism and enables promoting the migration of grains with less internal energy [35]. Based on all mentioned information, it can be supposed that semi-finished electrical steels are characterised by a randomly

Materials 2019, 12, 1914 10 of 15 with various crystallographic orientations, a creation of some small gradient of stored deformation energy is related to the lattice defects, such as dislocations, which were induced by the applied mechanical stress. The gradient of stored energy represents a driving force for strain-induced grain Materialsboundary 2019 migration, 12, x FOR PEER mechanism REVIEW and enables promoting the migration of grains with less internal10 of 15 energy [35]. Based on all mentioned information, it can be supposed that semi-finished electrical steels distributedare characterised stored by deformation a randomly distributedenergy gradient stored deformationbetween the energyprimary gradient recrystallized between grains. the primary The presentrecrystallized stresses grains. disparity The on present the grains stresses boundaries disparity in on combination the grains boundarieswith steep temperature in combination gradient with duringsteep temperature the secondary gradient dynamic during annealing the secondary allow to dynamicobtain the annealing growth of allow a small to number obtain the of growthgrains with of a appropriatesmall number rotated of grains cube with crystallographic appropriaterotated orientation. cube crystallographicThis statement is orientation. directly supported This statement by the resultsis directly obtained supported in present by the investigation results obtained showing in present adequate investigation correlation showing between adequate microstructure correlation (see Figurebetween 5b) microstructure and texture (see (see Figures Figure 7b5 andb) and 8b) texturecharacteristics. (see Figures From7 comparisonb and8b) characteristics. of the microstructural From andcomparison textural ofstates the microstructuralof the segments andheat texturaltreated in states long-term of the segmentsand dynamic heat annealing treated in conditions, long-term andit is possibledynamic to annealing conclude conditions, that the ittemperature is possibleto gradient conclude creates that the the temperature conditions gradient in temper creates rolled the polycrystallineconditions in temper materials rolled for polycrystalline the evolution materialsof secondary forthe grains evolution matrix of with secondary preferential grains rotated matrix cube with crystallographicpreferential rotated orientation. cube crystallographic orientation.

3.3. Magnetic Magnetic Measurements Measurements A preciseprecise estimation estimation of theof relationshipthe relationship between between the microstructure the microstructure and texture and in thetexture investigated in the investigatedsteel can be obtained steel can through be obtained the measurements through the measurements of final magnetic of final properties. magnetic It is properties. well-known It thatis well- the knowntexture ofthat NO the electrical texture steelsof NO has electrical a profound steels e ffhasect a on profound the magnetic effect properties, on the magnetic e.g., total properties, power losses, e.g., totalpermeability, power losses, coercivity, permeability, induction, coercivity, etc. of the induction, core laminations etc. of the [36 core,37]. laminations Since these magnetic[36,37]. Since materials these magneticare strongly materials textured are and strongly the texture textured is a crucialand the source texture of anisotropy,is a crucial the source magnetic of anisotropy, properties the are magneticconventionally properties measured are atconventionally various angles measured with respect at to various the rolling angles direction. with Inrespect order to evaluatethe rolling the direction.magnetic anisotropyIn order to of evaluate the investigated the magnetic steel, anisotropy the testing samplesof the investigated were prepared steel, in the rollingtesting directionsamples were(RD) andprepared the transverse in the rolling (TD) direction,direction as(RD) it was and shown the transverse in Figure (TD)2a. The direction, samples as obtained it was shown from the in Figureinitial semi-finished2a. The samples material obtained state asfrom well the as initial from the semi-finished segments annealed material in state different as well conditions as from were the segmentssubjected toannealed the measurements in different inconditions the DC and were AC subjected magnetic to field. the measurements The Figure9 shows in the a dependenceDC and AC magneticof the measured field. The DC Figure coercivity 9 shows of the a dependence investigated of samples the measured on the radialDC coercivity directions. of the Representative investigated samplesmagnetic on measurements the radial directions. were carried Representative out in order magnetic to diff erentiatemeasurements the samples were carried performance out in order after todi fferentdifferentiate heat treatment the samples procedures. performance The measurements after different clearly indicatedheat treatment some anisotropy procedures. and clearThe measurementsdifferences between clearly magnetic indicated properties some anisotropy of the segments and clear heat differences treated under between laboratory magnetic (dynamic) properties and ofindustrial the segments (long-term) heat treated conditions. under laboratory (dynamic) and industrial (long-term) conditions.

FigureFigure 9. Dependence 9. Dependence of the heat of the treated heat segments treated segments coercivity coercivity on the annealing on the conditions annealing and conditions radial direction. and radial direction. From Figure 9, it is evident that coercivity is much higher for the samples in an initial semi- finished material state than for the heat treated ones. Moreover, it should be noted that the initial steel is characterised by its pronounced anisotropy. In this case, the coercivity values in the RD and TD directions were determined to be 230A/m and 265 A/m, respectively. On the other hand, the segments subjected to both performed heat treatments show a significant reduction in their coercivity values, compared to those without heat treatment. The obtained coercivity values in the RD for the samples heat treated in long-term and dynamic conditions are 58 A/m and 20 A/m, respectively. The obtained coercivity values in the TD for the samples heat treated in long-term and dynamic conditions are 63 A/m and 26 A/m, respectively. This indicates that the annealing processes which

Materials 2019, 12, 1914 11 of 15

From Figure9, it is evident that coercivity is much higher for the samples in an initial semi-finished material state than for the heat treated ones. Moreover, it should be noted that the initial steel is characterised by its pronounced anisotropy. In this case, the coercivity values in the RD and TD directions were determined to be 230 A/m and 265 A/m, respectively. On the other hand, the segments subjected to both performed heat treatments show a significant reduction in their coercivity values, compared to those without heat treatment. The obtained coercivity values in the RD for the samples heat treated in long-term and dynamic conditions are 58 A/m and 20 A/m, respectively. The obtained Materialscoercivity 2019 values, 12, x FOR in PEER the TDREVIEW for the samples heat treated in long-term and dynamic conditions11 of are 15 63 A/m and 26 A/m, respectively. This indicates that the annealing processes which are responsible for are responsible for the evolution of microstructural and textural characteristics of the studied steel the evolution of microstructural and textural characteristics of the studied steel clearly improve its clearly improve its magnetic properties as well. Moreover, it has been demonstrated that the regime magnetic properties as well. Moreover, it has been demonstrated that the regime of annealing may of annealing may also have a significant influence on the value of final coercivity. The segments after also have a significant influence on the value of final coercivity. The segments after the heat treatments the heat treatments are characterized by very weak anisotropy which is acceptable for core materials are characterized by very weak anisotropy which is acceptable for core materials of electric-rotated of electric-rotated machines. Comparison of the samples heat treated in industrial and laboratory machines. Comparison of the samples heat treated in industrial and laboratory conditions clearly show conditions clearly show that the lowest coercivity values were obtained for the segments heat treated that the lowest coercivity values were obtained for the segments heat treated in dynamic laboratory in dynamic laboratory conditions. This result perfectly corresponds with an observed evolution of conditions. This result perfectly corresponds with an observed evolution of microstructural and microstructural and textural characteristics of the investigated materials. The evolution of core losses textural characteristics of the investigated materials. The evolution of core losses of experimental of experimental samples prepared in the RD was carried out also in AC magnetic field with a samples prepared in the RD was carried out also in AC magnetic field with a frequency of 50 Hz. frequency of 50 Hz. The results of these measurements are presented by typical B-H loops in Figure The results of these measurements are presented by typical B-H loops in Figure 10. It is well-known 10. It is well-known that the area enclosed by the B-H loop represents the core losses. Here, the that the area enclosed by the B-H loop represents the core losses. Here, the maximum value of core maximum value of core losses was obtained for the segments of the investigated steel before heat losses was obtained for the segments of the investigated steel before heat treatment (i.e., in initial treatment (i.e., in initial semi-finished material state). The AC magnetic measurements results are also semi-finished material state). The AC magnetic measurements results are also presented in Table1. presented in Table 1. In the case of the heat treated samples, it is obvious that the segments annealed In the case of the heat treated samples, it is obvious that the segments annealed in dynamic laboratory in dynamic laboratory conditions have much lower core loss than the segments heat treated in conditions have much lower core loss than the segments heat treated in industrial conditions according industrial conditions according to EN 10 341 standard. to EN 10 341 standard.

FigureFigure 10. 10. TheThe measured measured B-H B-H loops loops at at 50Hz 50Hz with with peak peak flux flux densities densities at at 1.7 T.

ItIt can can be be clearly clearly seen seen that that these these measurements measurements (Figure (Figure 10,10, Table Table 11)) perfectlyperfectly correspondcorrespond withwith thethe resultsresults which were obtained in the DC DC magnetic magnetic fiel fieldd (Figure 99).). Moreover, thethe datadata ofof B-HB-H hysteresis loopsloops showshow thatthat the the steel steel segments segments heat heat treated treate ind dynamic in dynamic conditions conditions show moreshow than more a 20%than decrease a 20% decreaseof core losses of core in comparisonlosses in comparison with the annealing with the annealing process used process in industrial used in conditionsindustrial conditions (see Table1 ).(see Table 1).

Table 1. The magnetic properties of chosen samples of investigated segments.

Watt losses in AC Coercivity in AC Coercivity in DC Sample Type Magnetic Field P Magnetic Field HC Magnetic Field HC (W/kg) (A/m) (A/m) Semi-finished state 13.8 320 230 Long-term annealing treatment 5.45 116 58 Dynamic annealing treatment 4.33 91 20

3.4. Measurement of Efficiency

Materials 2019, 12, 1914 12 of 15

Table 1. The magnetic properties of chosen samples of investigated segments.

Watt losses in AC Coercivity in AC Magnetic Coercivity in DC Magnetic Sample Type Magnetic Field P (W/kg) Field HC (A/m) Field HC (A/m) Semi-finished state 13.8 320 230 Long-term annealing treatment 5.45 116 58 Dynamic annealing treatment 4.33 91 20

Materials 2019, 12, x FOR PEER REVIEW 12 of 15 3.4. Measurement of Efficiency InIn the the present present work, work, the the tested tested electric electric motors motors were were assembled assembled from from rotor rotor and statorand stator segments segments heat treatedheat treated in either in conventionaleither conventional long-term long-term or unconventional or unconventional short-term short-term dynamic annealing dynamic conditions. annealing Theconditions. results of The the results efficiency of the measurements efficiency measurements within the working within torque the working load range torque of theload experimental range of the motorsexperimental are presented motors inare Figure presented 11. in Figure 11.

FigureFigure 11. 11.E Efficiencyfficiency dependence dependence ofof electrical electrical motorsmotors constructedconstructed fromfrom thethe segments segmentsheat heat treated treated in in dynamicdynamic (solid (solid blue blue line) line) and and long-term long-term (dashed (dashed orange orange line) line) conditions conditions on on the the torque torque load. load.

ThisThisgraph graphshows showsthe thee efficiencyfficiency curves curves which which were were measured measured for for electrical electrical motors motors constructed constructed fromfrom corecorematerials materials treatedtreatedby by di differentfferent annealing annealing methods. methods. ItIt isis evidentevident thatthat thethe betterbetter eefficiencyfficiency (solid(solid blue blue line) line) has has the the electrical electrical motor motor which which uses uses the segmentsthe segments treated treated in laboratory in laboratory conditions conditions with thewith high the heating high heating rate. The rate. electrical The electrical motor, which motor, was which completely was completely finished in finished conventional in conventional industrial conditionsindustrial (dashconditions orange (dash line), orange exhibits line), lower exhibits efficiency lower within efficiency the whole within range the of whole torque range load. of It cantorque be seenload. that It can the be highest seen valuethat the of ehighestfficiency value was obtainedof efficiency for the was motor obtained with for laboratory the motor (dynamically) with laboratory heat treated(dynamically) segments heat in thetreated range segments of the torque in the loadrange from of the 0.15 torque Nm upload to from 0.38 Nm.0.15 Nm The up comparison to 0.38 Nm. of theThe obtainedcomparison data of clearly the obtained shows that data the clearly dynamic shows heat that treatment the dynamic of electric heat motor treatment segments, of electric which motor were shearsegments, cut from which semi-finished were shear electrical cut from steels, semi-finished improves notelectrical only theirsteels, microstructure, improves not texture only andtheir coremicrostructure, losses (coercive texture force), and core but losses it also (coercive significantly force), increases but it also the significantly resulting e increasesfficiency ofthe electrical resulting motor.efficiency The of experimental electrical motor. data showThe experimental that our dynamic data heatshow treatment that our dynamic method enables heat treatment for a significant method improvementenables for a significant of the magnetic improvement properties of ofthe electric magnetic motor properties core laminations. of electric motor Specifically, core laminations. more than aSpecifically, 1.2% efficiency more increase than a at1.2% the efficiency load of 0.28 increase Nm was at the achieved load of for 0.28 electric Nm motorwas achieved with dynamically for electric heatmotor treated with laminations,dynamically inheat comparison treated laminations, with the motor in comparison manufactured with fromthe motor laminations manufactured heat treated from accordinglaminations to conventional heat treated industrialaccording conditions.to conventional Thus, industrial the results conditions. of current investigation Thus, the results may representof current promisinginvestigation challenges may represent for the endpromising users of challenges semi-finished for the electrical end users steels, of semi-finished not only in terms electrical of possible steels, improvementsnot only in terms of their of possible products improvements technical parameters, of their butproducts also as technical a potential parameters, decrease of but production also as a costspotential related decrease to the finalof production heat treatment. costs related to the final heat treatment.

4. Summary and Conclusions In the presented paper the effects of the different heat treatment conditions on the evolution of average grain size, texture and magnetic loss components of semi-finished non-oriented electrical steel were studied. The experimental samples in the form of rotor and stator segments after the shear cutting were processed by conventional long-term and unconventional dynamic heat treatment technology. The obtained results have clearly shown that the annealing process with rapid heating promotes the evolution of selective growth of coarse-grains with enhanced intensity rotated cube texture. The main observations and conclusions can be summarized as follows: • The rapid heating at dynamic annealing conditions of semi-finished NO silicon steels leads to a significant increase of average grain size of the obtained microstructure. The distinct evolution

Materials 2019, 12, 1914 13 of 15

4. Summary and Conclusions In the presented paper the effects of the different heat treatment conditions on the evolution of average grain size, texture and magnetic loss components of semi-finished non-oriented electrical steel were studied. The experimental samples in the form of rotor and stator segments after the shear cutting were processed by conventional long-term and unconventional dynamic heat treatment technology. The obtained results have clearly shown that the annealing process with rapid heating promotes the evolution of selective growth of coarse-grains with enhanced intensity rotated cube texture. The main observations and conclusions can be summarized as follows:

The rapid heating at dynamic annealing conditions of semi-finished NO silicon steels leads to a • significant increase of average grain size of the obtained microstructure. The distinct evolution of coarse-grained microstructure is related to the strain-induced grain boundary migration mechanism under the influence of steep temperature gradient through the steel sheet cross-section. The unconventional dynamic heat treatment of the investigated electric motor core segments • resulted in a stronger texture optimization tendency (i.e. the weakening or total absence the γ-fibre and forming strong rotated cube texture) than in the case of using conventional long-term annealing process. The improvement of the texture characteristics illustrated that the used dynamic heat treatment • can optimize soft magnetic properties of the investigated semi-finished steel, namely its magnetic isotropy in combination with magnetic coercivity. The magnetic properties measured at 50 Hz frequency clearly showed that the evolved • microstructures and textures of the segments heat treated by two different procedures are directly responsible for their final magnetic characteristics. The segments which were unconventionally heat treated at higher temperature with rapid heating were characterised by lower value of watt losses (4.33 W/kg), compared to those of the segments which were conventionally heat treated at lower temperature with much slower heating (5.45 W/kg). The measurement of efficiency of electric motors constructed from the segments heat treated • under two different heat treatment conditions have clearly shown, that in comparison to the conventional heat treatment technology, the application of our unconventional dynamic heat treatment leads to a significant efficiency improvement by more than 1.2%.

Author Contributions: Conceptualization, I.P. and F.K.; methodology, I.P., F.K., V.P. and B.P.; validation, I.P. and F.K.; formal analysis, I.P., F.K. and L.F.; investigation, I.P., F.K., L.F., V.P. and B.P.; resources, I.P.; data curation, I.P., L.F. and B.P.; writing—original draft preparation, I.P. and L.F.; writing—review and editing, I.P. and L.F.; visualization, I.P. and V.P.; supervision, F.K and I.P.; project administration, F.K and I.P.; funding acquisition, F.K. Funding: This research was funded by “Agentúra na podporu výskumu a vývoja MŠVVaŠ SR” (contract No. APVV-15-0259). Acknowledgments: This work was carried out within the research project entitled “Unconventional technology development of final processing of isotropic electrical steels”, which is supported by “Agentúra na podporu výskumu a vývoja MŠVVaŠ SR” under the contract No. APVV-15-0259 and partly by “Vedecká Grantová Agentúra MŠVVaŠ SR a SAV” under the projects VEGA 2/0066/18 and VEGA 2/0073/19. The work was also realized within the frame of the projects ITMS 26220220064 and ITMS 26220220061. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Steiner Petroviˇc,D. Non-oriented electrical steel sheets. Mater. Technol. 2010, 44, 317–325. 2. Moses, A.J. Electrical steels: Past, present and future developments. IEE Proc. 1990, 137, 233–254. [CrossRef] 3. You, D.; Park, H. Developmental trajectories in electrical steel technology using patent information. Sustainability 2018, 10, 2728. [CrossRef] 4. Oda, Y.; Kohno, M.; Honda, A. Recent development of non-oriented electrical steel sheet for automobile electrical devices. JMMM 2008, 320, 2430–2435. [CrossRef] Materials 2019, 12, 1914 14 of 15

5. Leuning, N.; Steentjes, S.; Hameyer, K. Effect of grain size and magnetic texture on iron-loss components in NO electrical steel at different frequencies. JMMM 2019, 469, 373–382. [CrossRef] 6. Sidor, J.J.; Verbeken, K.; Gomes, E.; Schneider, J.; Calvillo, P.R.; Kestens, L.A.I. Through process texture evolution and magnetic properties of high Si non-oriented electrical steels. Mater. Charact. 2012, 71, 49–57. [CrossRef] 7. Lee, H.H.; Jung, J.; Yoon, J.I.; Kim, J.K.; Kim, H.S. Modeling the evolution of recrystallization texture for a non-grain oriented electrical steel. Comput. Mater. Sci. 2018, 149, 57–64. [CrossRef] 8. Cheng, L.; Zhang, N.; Yang, P.; Mao, W.M. Retaining {100} texture from initial columnar grains in electrical steels. Scr. Mater. 2012, 67, 899–902. [CrossRef] 9. Sahoo, G.; Singh, C.D.; Deepa, M.; Dhua, S.K.; Saxena, A. Recrystallization behavior and texture of non-oriented electrical steels. Mater. Sci. Eng. A 2018, 734, 229–243. [CrossRef] 10. Leuning, N.; Steentjes, S.; Schulte, M.; Bleck, W.; Hameyer, K. Effect of elastic and plastic tensile mechanical loading on the magnetic properties of NGO electrical steel. JMMM 2016, 417, 42–48. [CrossRef] 11. Steiner Petroviˇc,D.; Mandrino, D. XPS characterization of the oxide scale on fully processed nonoriented electrical steel sheet. Mater. Charact. 2011, 62, 503–508. [CrossRef] 12. Fischer, J.; Schneider, J. Influence of deformation process on the improvement of non-oriented electrical steel. JMMM 2003, 254–255, 302–306. [CrossRef] 13. Barros, J.; Schneider, J.; Verbeken, K.; Houbaert, Y. On the correlation between microstructure and magnetic losses in electrical steel. JMMM 2008, 320, 2490–2493. [CrossRef] 14. Hilinski, E. Recent developments in semiprocessed cold rolled magnetic lamination steel. JMMM 2006, 304, 172–177. [CrossRef] 15. Jafari, M.; Jamshidian, M.; Ziaei-Rad, S.; Raabe, D.; Roters, F. Constitutive modeling of strain induced grain boundary migration via coupling crystal plasticity and phase-field methods. Int. J. Plast. 2017, 99, 19–42. [CrossRef] 16. Dillamore, I.L.; Smith, C.J.E.; Watson, T.W. Oriented nucleation in the formation of annealing texture in iron. Met. Sci. J. 1967, 1, 49–54. [CrossRef] 17. Hatchinson, W.B. Development and control of annealing textures in low-carbon steels. Int. Met. Rev. 1984, 29, 25–42. [CrossRef] 18. Castro, S.F.; Gallego, J.; Landgraf, F.J.G.; Kestenbach, H.J. Orientation dependence of stored energy of cold work in semi-processed electrical steels after temper rolling. Mater. Sci. Eng. A 2006, 427, 301–305. [CrossRef] 19. Kovac, F.; Stoyka, V.; Petryshynets, I. Strain induced grain growth in non-oriented electrical steels. JMMM 2008, 320, e627–e630. [CrossRef] 20. CEN. Cold Rolled Electrical Non-Alloy and Alloy Steel and Strip Delivered in the Semi-Processed State; STN EN 10341; CEN: Brussels, Belgium, 2006. 21. Almeida, A.D.; Bertoldi, P.; Leonhard, W. Energy Efficiency Improvements in Electric Motors and Drives, 1st ed.; Springer: Berlin, Germany, 1997; pp. 102–112. 22. Petryshynets, I.; Kovac, F.; Marcin, J.; Skorvanek, I. Magnetic properties of temper rolled NO FeSi steels with enhanced rotation texture. IEEE Trans. Magn. 2013, 49, 4303–4306. [CrossRef] 23. Petryshynets, I.; Kovac, F.; Marcin, J.; Skorvanek, I. Improved processing technique for preparation of non-oriented electrical steels with low coercivity. Acta Phys. Pol. A 2014, 126, 182–183. [CrossRef] 24. Kováˇc,F.; Petryshynets, I.; Puchy, V.; Šebek, M.; Marcin, J.J. Preparation of NO silicon steels with columnar microstructure and low watt losses. J. Electr. Eng. 2015, 66, 86–89. 25. Petryshynets, I.; Kováˇc,F.; Falat, L.; Puchý, V.; Šebek, M. Magnetic losses evolution of ferric Fe-Si steel subjected to temper rolling at elevated temperature. Acta Phys. Pol. A 2018, 133, 1065–1068. [CrossRef] 26. Mehdi, M.; He, Y.; Hilinski, E.J.; Edrisy, A. Effect of skin pass rolling reduction rate on the texture evolution of a non-oriented electrical steel after inclined cold rolling. JMMM 2017, 429, 148–160. [CrossRef] 27. He, Y.; Hilinski, E.J. Texture and magnetic properties of non-oriented electrical steels processed by an unconventional cold rolling scheme. JMMM 2016, 405, 337–352. [CrossRef] 28. Verbeken, K.; Kestens, L. Strain-induced selective growth in an ultra low carbon steel after a small rolling reduction. Acta Mater. 2003, 51, 1679–1690. [CrossRef] 29. Fang, F.; Xu, Y.-B.; Zhang, Y.-X.; Wang, Y.; Lu, X.; Misra, R.D.K.; Wang, G.-D. Evolution of recrystallization microstructure and texture during rapid annealing in strip-cast non-oriented electrical steels. JMMM 2015, 381, 433–439. [CrossRef] Materials 2019, 12, 1914 15 of 15

30. Wang, J.; Li, X.; Mi, X.; Zhang, S.; Volinsky, A.A. Rapid annealing effects on microstructure, texture, and magnetic properties of non-oriented electrical steel. Met. Mater. Int. 2012, 18, 531–537. [CrossRef] 31. Cotterill, P.; Mould, P.R. Recrystallization and Grain Growth in Metals, 1st ed.; Surrey University Press: London, UK, 1976; pp. 266–300. 32. Sidor, Y.; Kovac, F.; Kvackaj, T. Grain growth phenomena and heat transport in non-oriented electrical steels. Acta Mater. 2007, 55, 1711–1722. [CrossRef] 33. Homma, H.; Nakamura, S.; Yoshinaga, N. On {h,1,1}<1/h,1,2>, the recrystallization texture of heavily cold rolled BCC steel. Mater. Sci. Forum 2004, 470, 269–274. [CrossRef] 34. Gobernado, P.; Petrov, R.H.; Kestens, L.A.I. Recrystallized {311}<136> orientation in ferrite steel. Scr. Mater. 2012, 66, 623–626. [CrossRef] 35. Diligent, S.; Gautier, E.; Lemoine, X.; Berveiller, M. Lattice orientation dependence of the stored energy during cold-rolling of polycrystalline steels. Acta Mater. 2001, 49, 4079–4088. [CrossRef] 36. Campos, M.F.; Teixeira, J.C.; Langraf, F.J.G. The optimum grain size for minimizing energy losses in iron. JMMM 2006, 301, 94–99. [CrossRef] 37. PermKumar, R.; Samajdar, I.; Viswanathan, N.N.; Singal, V.; Seshadri, V. Relative effect(s) of texture and grain size on magnetic properties in a low silicon non-grain oriented electrical steel. JMMM 2003, 264, 75–85. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).