Improved magnetic properties of Zn-substituted strontium W-type hexaferrites

Accepted Manuscript Improved magnetic properties of Zn-substituted strontium W-type hexaferrites Jae-Hyoung You, Sang-Im Yoo PII:

S0925-8388(18)32031-0

DOI:

10.1016/j.jallcom.2018.05.296

Reference:

JALCOM 46272

To appear in:

Journal of Alloys and Compounds

Received Date: 2 March 2018 Revised Date:

17 May 2018

Accepted Date: 24 May 2018

Please cite this article as: J.-H. You, S.-I. Yoo, Improved magnetic properties of Zn-substituted strontium W-type hexaferrites, Journal of Alloys and Compounds (2018), doi: 10.1016/j.jallcom.2018.05.296. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

2 3 4 5 6

ACCEPTED MANUSCRIPT Improved magnetic properties of Zn-substituted strontium W-type hexaferrites Jae-Hyoung You and Sang-Im Yoo* Department of Materials Science and Engineering Seoul National University, Research Institute of Advanced Materials (RIAM), Seoul 151-744, Korea * [email protected]

RI PT

7

Abstract— The Zn2+ substitution for the Fe2+ site was found to be effective for the improvement of the

9

saturation magnetization (Ms) of strontium W-type hexaferrite (SrFe18O27). For this study, polycrystalline

10

SrFe18O27 with various Zn-substituents of SrZnxFe(2-x)Fe16O27 (SrZnxFe(2-x)W, 0.0 ≤ x ≤ 1.5) were sintered at

11

high temperatures in a reduced oxygen atmosphere of 10-3 atm. While pure SrZnxFe(2-x)W-type solid

12

solutions (0.0 ≤ x ≤ 1.0) could be obtained from the samples of x = 0–1.0, the second phase of ZnFe2O4 was

13

obtained from the sample of x = 1.5. With increasing x up to 1.0, Ms values were monotonously increased

14

and the highest Ms value of 87.7 emu/g was achievable from the sample of x = 1.0 sintered at 1250˚C. In

15

addition, post-annealing heat treatments of samples for oxygenation at 300˚C in pure oxygen gas revealed

16

that oxygen non-stoichiometry increased with increasing the sintering temperature, leading to the increase in

17

Ms, unit cell volume, and electrical conductivity.

M AN U

TE D

EP

18

SC

8

Keywords: W-type hexaferrite; low oxygen pressure; Zn2+ substitution; oxygen non-stoichiometry;

20

magnetic property; saturation magnetization

21

22

AC C

19

1

1.

ACCEPTED MANUSCRIPT

Introduction

W-type hexaferrite, a member of hexaferrite family having the hexagonal structure, was first reported in

3

1952 as a phase mixed with M-type and X-type hexaferrites [1]. They have the chemical formula of

4

Sr(Ba)Me2Fe16O27, where Sr or Ba can be substituted by other metal ions having similar ionic radii, and Me

5

is usually divalent transition metal ions such as Mn2+, Fe2+, Co2+, Ni2+, Cu2+, and Zn2+. For Me = Fe2+, W-

6

type hexaferrites form the binary compounds of BaFe18O27 (BaW) and SrFe18O27 (SrW). In 1980, F. K.

7

Lotgering et al. [2] reported that BaW exhibited about 10% higher Ms value (≈ 78 emu/g) and almost equal

8

anisotropy field (Ha ≈ 16 kOe) in comparison with M-type hexaferrites widely used as a ceramic permanent

9

magnet. Thus, in the early stage, W-type hexaferrites had been studied for their application as the permanent

10

magnet [3-10]. Unfortunately, however, their highest coercivity (Hc) value ever reported [6] is 3600 Oe

11

which is much lower than that of M-type hexaferrite (above 5000 Oe) [11-13]. Recently, many research

12

groups have tried to investigate their applicability as antenna and microwave absorber due to their moderate

13

permeability value in the microwave region and high ferromagnetic resonance frequency, which determines

14

the working frequency of device [14-20].

TE D

M AN U

SC

RI PT

2

Most of the studies on the W-type hexaferrites have been focused on the substituted W-type hexaferrites

16

of ternary or quaternary compounds, which is closely related to the phase stability of W-type hexaferrites.

17

Both SrW and BaW, the simplest binary W-type hexaferrites possessing Fe2+ and Fe3+ ions altogether are

18

high-temperature stable phases, i.e., metastable at relatively lower temperatures. According to literature, the

19

SrW and BaW phases are stable only at the temperature regions of 1350–1440˚C in air [21], and 1495–

20

1540˚C in pure oxygen atmosphere [22], respectively. Thus, it is difficult to obtain the single phase of BaW

21

or SrW without a quenching process since the oxidation of Fe2+ into Fe3+ ions, which destabilizes the W-type

22

hexaferrite structure, can easily occur in air during furnace-cooling to room temperature, resulting in the

AC C

EP

15

1

ACCEPTED MANUSCRIPT phase decomposition of W-type hexaferrite. This might be one reason why the W-type hexaferrite was first

2

reported as a mixed phase, not single phase. In order to suppress the phase decomposition of W-type hexaferrites, many researchers tried to substitute

4

Fe2+ ions with other divalent ions which are hardly oxidized. Among potential divalent substituents, the Zn

5

element must be a good candidate since non-magnetic Zn2+ ions have the strongest preference for the

6

tetrahedral site occupation [23]. In addition, a partial substitution of Zn2+ ions for the Fe2+ ions, located at the

7

tetrahedral sites of cubic spinel ferrites, could improve Ms [23], and hence the soft ferrites like Mn-Zn and

8

Ni-Zn ferrites, widely being used for many different magnetic applications, could be developed. Likewise, if

9

the partial substitution of Zn2+ ions for the Fe2+ ions located at the tetrahedral sites of W-type hexaferrite is

10

enabled, an increase in Ms is expected. Referring to the previous papers on Zn-substituted SrW or BaW

11

hexaferrites [3, 6, 7, 9, 24-26], however, studies have been limited to a full substitution of Zn2+ ions for the

12

Fe2+ ion sites, which is most probably due to difficulty in the fabrication of partially Zn-substituted W-type

13

hexaferrites in air. Recently, we have reported that the fatal decomposition problem of SrW during the

14

furnace-cooling in air could be effectively overcome by sintering and subsequent furnace-cooling in the low

15

PO2 of 10-3 atm [27]. By applying this process to this study, it was possible to prepare partially Zn-substituted

16

SrW (SrZnxFe(2-x)Fe16O27; SrZnxFe(2-x)W) polycrystalline samples. Here, we report their magnetic properties

17

for the first time.

EP 2.

AC C

18

19

TE D

M AN U

SC

RI PT

3

Experimental

20

The polycrystalline samples with the nominal compositions of SrZnxFe(2-x)W (x = 0.0, 0.5, 1.0, and 1.5)

21

were prepared by conventional solid-state reaction. The SrCO3, Fe2O3, and ZnO powders of 99.9% purity

22

were used as precursors. The precursors were weighed, ball-milled for 24 h with zirconia balls in ethanol,

23

dried in an oven at 45˚C, and then calcined at 1150˚C for 8 h in air. The calcined powder was ball-milled for

ACCEPTED 24 h and uniaxially pressed into the pellets of 15 mmMANUSCRIPT in diameter and 2 mm in thickness under the pressure

2

of 180 kg/cm2. The pellets were sintered at the temperature region of 1175–1315˚C for 2 h in the PO2 of 10-3

3

atm with a heating rate of 3˚C/min, and then furnace-cooled in an alumina tube furnace. During the sintering

4

process, PO2 was controlled by using O2-N2 mixed gas (1000 ppm O2 in N2). The gas flow was controlled

5

by mass flow controller with a flow rate of 0.5 l/min. For oxygenation of the samples, post-annealing heat

6

treatments were performed at 300˚C for 2–24 h in the O2 atmosphere.

RI PT

1

The phases and crystal structures of samples were analyzed by X-ray diffraction (XRD, BRUKER D8-

8

ADVANCE α1 system) using Cu-Kα1 radiation (λ = 1.5406 Å) with a divergence slit width of 0.2 mm. The

9

lattice parameters and cell volumes were calculated from the refined XRD patterns by using the Pawley

10

whole powder pattern decomposition method (BRUKER AXS TOPAS software version 4.2). Internal

11

calibration with standard Si powder was performed before the refinement process. Magnetic hysteresis loops

12

were measured at room temperature by using physical property measurement system-vibrating sample

13

magnetometer (PPMS-VSM, Quantum design) with the maximum field strength of 5 T. For the

14

measurement of magnetic hysteresis loops, samples were cut into rectangular parallelepiped shape (~1 × 1 ×

15

3 mm3) and the magnetic field was applied parallel to the long axis. The DC electrical resistivity was

16

measured in the temperature region of 100–300K by using standard four-point probe method. The samples

17

were cut into a thin rectangular plate shape (~0.4 × 2 × 10 mm3) for the resistivity measurement.

18 19

3.

AC C

EP

TE D

M AN U

SC

7

Results and discussion

20

In low PO2 of 10-3 atm, the SrZnxFe(2-x)W single phase could be prepared at the temperature region of

21

1300–1315˚C, 1250–1280˚C, 1225–1250˚C for x = 0.0, 0.5, 1.0, respectively. Fig. 1 shows the XRD patterns

22

for the samples of x = 0.0, 0.5, 1.0, 1.5 and 2.0 sintered at 1315, 1280, 1250, 1225 and 1225˚C for 2 h,

23

respectively. For a comparison, the XRD pattern for the sample of x = 0.0 reported in our previous paper [27]

MANUSCRIPT is also presented in Fig. 1. As shown ACCEPTED in Fig. 1, all samples up to x = 1.5 consist of the SrZnxFe(2-x)W single

2

phase. In the case of x = 1.5, a small amount of ZnFe2O4 is detectable as a second phase together with

3

SrZnxFe(2-x)W phase. This implies that the solubility limit of Zn2+ in the SrZnxFe(2-x)W solid solutions exists

4

between x = 1.0 and 1.5 in low PO2 of 10-3 atm, and excessive Zn2+ ions over the solubility limit are

5

considered to form the second phases. As shown in Fig. 1(b), while only ZnFe2O4 phase is detectable for the

6

sample of x = 1.5, both ZnFe2O4 and Sr2Zn2Y phases are detectable for the sample of x = 2.0. The reason why

7

the Sr2Zn2Y phase is undetectable from the sample of x = 1.5 may be due to a small amount of Sr2Zn2Y.

RI PT

1

The calculated lattice parameters a, c and cell volumes (Vcell = a2csin120˚) of the samples are listed in

9

Table 1, and the plots of a, c and Vcell with respect to x are shown in Fig. 2. Although the SrZnxFe(2-x)W solid

10

solution with x = 2.0 could not be obtained by sintering in the PO2 of 10-3 atm due to the lowered solubility

11

limit of Zn2+, its single phase was obtainable by sintering in air as previously reported [7, 9, 24]. Thus the

12

values of x = 2.0 samples sintered at 1300˚C for 2 h in air are also listed for a comparison. First of all, one

13

can see that a, c and Vcell values of the samples having same x increase with increasing sintering temperature

14

for x = 0.0–1.0. In the case of the samples having x = 0.0, Vcell increases from 984.87 to 987.08 Å3 (∆Vcell ≈

15

0.22%) with increasing sintering temperature from 1300 to 1315˚C. This increase is probably attributed to an

16

increase in the oxygen non-stoichiometry (δ) of samples caused by low PO2 sintering. Moreover, with

17

increasing the δ value, the ratio of Fe2+/Fe3+ must be increased to satisfy the charge neutrality by the

18

following reaction.

M AN U

TE D

EP

AC C

19

SC

8

3+ 3+ SrFe 22 + Fe16 O 27 → SrFe 22 ++ 2δ Fe16 − 2δ O 27 −δ +

δ 2

O2

20

Since the ionic radius of Fe2+ (0.61 Å) is larger than that of Fe3+ (0.49 Å), the Vcell value can be increased

21

with increasing δ value for all samples. On the other hand, the increase in a, c and Vcell with increasing x can

22

be understood as follows. For instance, Vcell increases from 984.87 to 991.73 Å3 (∆Vcell ≈ 0.69%) for the

23

sample of x = 0.0 sintered at 1300˚C and the sample of x = 1.0 sintered at 1250˚C, respectively. The ∆Vcell for

ACCEPTED MANUSCRIPT different x (0.69%) is much larger than ∆Vcell for different sintering temperature (0.22%). The ∆Vcell for

2

different x shows a successful substitution of Zn2+ (0.74 Å) ions for smaller Fe2+ (0.61 Å) sites. One can see

3

that the Vcell monotonously increases with increasing x until x = 2.0, which implying successful substitution

4

of Zn2+ for Fe2+. The previously reported Vcell values of 999.75 Å3 [7], 998.16 Å3 [9], and 993.03 Å3 [24] for

5

SrZn2W are also similar to that of our sample.

RI PT

1

The magnetic hysteresis loops of some selected samples are shown in Fig. 3(a), and their Ms and Ha

7

values are listed in Table 1. The Hc values of samples are in the region 90–120 Oe though they are not

8

presented in this table. The Ms and Ha values were estimated from the hysteresis loops by using the law of

9

approach to saturation [28]. The general expression of this law can be written as follows.

SC

6

A B   M = M s 1 − − 2  + χ p H  H H 

M AN U

10

Where M is magnetization induced by applied magnetic field H, Ms is saturation magnetization, A is

12

inhomogeneity parameter related to the inhomogeneity of the material, χp is the high field susceptibility, and

13

B is the anisotropy parameter related to the magneto-crystalline anisotropy. For the hexagonal crystal

14

structure exhibiting uniaxial anisotropy, B can be expressed as

B=

EP

15

TE D

11

H a2 4K12 = 15 15M s2 .

In high field region, where M vs. 1/H2 curve shows a linear relationship, the parameters A and χp become

17

negligibly small. For our samples, the linear relationship between M and 1/H2 was found in the field region

18

of 2–5 T, and thus the law was applied to the data in this field region in order to evaluate their Ms values.

AC C

16

19

The values of Ms and Ha respect to x values in SrZnxFe(2-x)W are plotted in Fig. 3(b). The Ha value

20

decreases from 19468 Oe to 14640 Oe with increasing x from 0.0 to 2.0. This is attributed to the

21

nonmagnetic ion of Zn2+ which can weaken the super-exchange interaction of magnetic ions and spin-orbit

ACCEPTED coupling interaction [23]. The Ha value of 14640 OeMANUSCRIPT for x = 2.0 sample is larger than the reported values of

2

11812 Oe [7] and 12500 Oe [24]. The difference in the values might come from the different evaluation

3

method. While we evaluated the Ha value by fitting the magnetization curve as described above, the reported

4

values were evaluated by measuring two different magnetization curves for aligned sample with applied

5

field parallel and perpendicular to the aligned axis (c-axis) [7], and by measuring the magnetic torque curve

6

of single crystal sample [24], respectively. Contrary to the Ha value, the Ms value increases with increasing x

7

up to x = 1.0 and then decreases with further increasing x. The maximum Ms value of 87.7 emu/g was

8

obtainable from the sample of x = 1.0 sintered at 1250˚C. This Ms value is about 8% larger compared with

9

that of pure SrW (Ms = 81.3 emu/g) and about 13% higher than fully Zn-substituted SrW (Ms = 77.3 emu/g).

10

Here, it should be noted that the Ms value higher than 87.7 emu/g can be obtained if the amount of Zn

11

substitution in the SrZnxFe(2-x)W single phase is optimized. The value of 77.3 emu/g for x = 2.0 sample,

12

sintered at 1300oC in air, is slightly smaller than the value of 79 emu/g reported by H. Graetsch et al. [24],

13

but it is much higher than the value of 57.6 emu/g reported by H. Yamamoto et al. [7]. The large

14

discrepancy with the Ms value reported by H. Yamamoto et al. [7] is attributable not only to their relatively

15

lower sintering temperature of 1200 C in air but also our evaluation method of the Ms value. While we

16

evaluated the Ms value by using the law of approach to saturation as previously mentioned,

17

the magnetization value at the applied field of 1 T as Ms value. According to their evaluation method, the

18

magnetization value of our sample at 1 T is 63.5 emu/g.

they reported

AC C

EP

TE D

M AN U

SC

RI PT

1

19

The improvement of Ms with increasing x, i.e., the amount of Zn substituent, is similar to that of Zn-

20

substituted spinel ferrites. This similarity comes from a similarity between the crystal structure of W-type

21

hexaferrite and spinel ferrite. The unit cell of W-type hexaferrite (SrMe2Fe16O27) can be expressed as

22

stacking of spinel blocks containing two spinel molecular units (S, 2MeFe2O4) and hexagonal closed packed

23

blocks (R, SrFe6O11) in the form of RSSR*S*S*, where the symbol * denotes 180˚ rotation of the blocks

ACCEPTED MANUSCRIPT around the c-axis. Further simplification enables W-type hexaferrite structure to be expressed as M + S

2

where M is M-type hexaferrite (SrFe12O19) and S is spinel block as described above since the crystal

3

structure of SrFe12O19 can be expressed as RSR*S*. As the unit cell of W-type hexaferrite is composed of

4

two molecular units of W-type hexaferrite, 32 Fe3+ ions and 4 Me2+ ions occupy seven different magnetic

5

sublattices in the unit cell. Each sublattice is presented in Table 2 [29]. There are four octahedral sites (12k,

6

4fVI, 6g, 4f), two tetrahedral sites (4e, 4fIV), and one hexahedral site (2d). Each sublattice has different spin

7

alignment along with the c-axis. The direction of the spins in 12k, 6g, 4f, and 2d sites are parallel to the c-

8

axis, and the spins in 4e, 4fIV, and 4fVI sites are antiparallel. If we assume that the Zn2+ ions in SrZnxFe(2-x)W

9

occupy only the tetrahedral sites (4e, 4fIV) in S block which has down spin alignment, then the net magnetic

SC

moment (mB) of SrZnxFe(2-x)W per molecule in Bohr magnetons (µB) at 0K can be expressed as follows,

M AN U

10

RI PT

1

mB = 12 × 5 − (6 − x) × 5 = 30 + 5x µB

12

As a result, the net magnetic moment is increased with increasing Zn2+ content (x). This is exactly same with

13

the case of spinel ferrite that the substitution of magnetic ions (Fe2+) in a ferromagnetic material by

14

nonmagnetic ions (Zn2+) leads to an increase in the net magnetic moment.

TE D

11

There are some reports on the location of divalent metal ions in fully substituted W-type hexaferrites [29-

16

34]. In the case of BaMg2W, the Mg2+ ion is reported to reside in both the octahedral and tetrahedral sites of

17

the S block [30, 34]. The Co2+ ion in BaCo2W is also reported to occupy both the octahedral and tetrahedral

18

sites of S block [31]. The Mn2+ ion is reported to enter the tetrahedral sites of S block with a preference for

19

the 4e site [33]. The Zn2+ ion in BaZn2W is reported to have a tendency to occupy the tetrahedral sites of S-

20

block [29, 32]. According to these reports, one can see that the Me2+ ions in SrMe2W tend to occupy the

21

sublattices in S block, and their distribution in S block is similar to the distribution of the Me2+ ions in spinel

22

ferrites. Thus, substitution of nonmagnetic Zn2+ ions in W-type hexaferrites has the effect on Ms similar to

23

the case of spinel ferrite. This can also explain the reason why the Ms value is first increased and then

AC C

EP

15

ACCEPTED MANUSCRIPT decreased with further increasing x. While the net magnetic moments of Zn-substituted spinel ferrites are

2

increased with increasing the amount of Zn substitution up to around 40–50 mol% [23], for larger Zn2+

3

substitution, the alignments between the magnetic ions are weakened and the anti-parallel alignment of spins

4

is canted so that the net magnetic moment decreases and becomes zero for the full substitution of Zn2+ for

5

the tetrahedral site, i.e., ZnFe2O4. For the same reason, fully Zn-substituted SrW samples (x = 2.0) sintered

6

in air exhibited lower Ms value of 77.3 emu/g as shown in Table 1.

RI PT

1

From the Table 1, it can be seen that Vcell and Ms values of samples vary with the sintering temperature.

8

Fig. 4 shows the variation of a, c, Vcell and Ms values with respect to the sintering temperature. From the Fig.

9

4, one can see that both Vcell and Ms values increase with increasing sintering temperature for all samples. As

10

we discussed in the previous section, the increase in Vcell with increasing sintering temperature may be

11

attributed to an increase in the oxygen non-stoichiometry of the samples, and the increase in Ms values also

12

seem to be closely related to the increase in oxygen non-stoichiometry. Because as the oxygen non-

13

stoichiometry of a sample becomes larger, oxygen vacancy concentration within the sample becomes higher,

14

and more Fe3+ ions can be oxidized into Fe2+ ions. Consequently, the excessive Fe2+ ions formed due to

15

oxygen vacancy cause the increase in Ms since the magnetic moment of Fe2+ (4 µB) is smaller than that of

16

Fe3+ (5 µB). In other words, the valence state change of Fe3+ ions occupying the spin-down sites can cause an

17

increase in the net magnetic moment of SrW as in the case of Zn2+ substitution. A similar behavior of

18

increase in the Ms values due to oxygen deficiency has been reported for the La-Ce-Zn doped SrM sintered in

19

the N2 atmosphere [35].

AC C

EP

TE D

M AN U

SC

7

20

In order to clarify the influence of oxygen non-stoichiometry on the Vcell and Ms values, oxygen annealing

21

was carried out. The samples of x = 0.0 which were sintered at 1300, 1310 and 1315˚C for 2 h in the PO2 of

22

10-3 atm were annealed at 300˚C for 2–24 h in a pure oxygen atmosphere. Since the SrW phase can be

23

decomposed into SrM and Fe2O3 during the oxygenation process [21, 27], the oxygen annealing temperature

24

was carefully determined to prevent the phase decomposition of SrW. The change in the Vcell and Ms values

MANUSCRIPT after the oxygen annealing process is ACCEPTED shown in Fig. 5. As the oxygen annealing period is extended, the Vcell

2

values of three different samples decrease and become almost identical to each other as shown in Fig. 5(a).

3

Moreover, the total decrease in Vcell is the largest for the sample sintered at 1315˚C in the PO2 of 10-3 atm,

4

which is considered to be the most oxygen-deficient, while the total decrease in Vcell for the sample sintered at

5

1300˚C is the smallest. This result indirectly reveals that the difference of Vcell between the as-sintered

6

samples is closely related to the oxygen non-stoichiometry, and the oxygen vacancy has been removed by the

7

oxygen annealing process.

RI PT

1

There is another evidence that oxygen non-stoichiometry of the sample exists. Fig. 6 (a) and (b) show the

9

DC resistivity (ρ) change of the samples at the temperature region of 100–300 K before and after the

10

oxygenation process on the ρ versus T (K) and lnρ versus 1000/T (K) plots, respectively. In Fig. 6(b), the

11

activation energy (Ea) for conduction is also calculated from the lnρ versus 1000/T (K) plots by using the

12

Arrhenius equation as follows,

M AN U

SC

8

13

ρ = ρ0·exp(Ea /kBT)

where ρ0 is the resistivity at room temperature and kB is the Boltzmann constant. Before the oxygenation

15

process, the sample sintered at the higher temperature tends to be more electrically resistive as it can be seen

16

from the Fig.6. In addition, the Ea is smaller for the sample sintered at the higher temperature as well as the ρ

17

value. This is due to the facts that the electron hopping can occur more easily and frequently for the oxygen-

18

deficient sample which can contain a larger amount of Fe2+ ions, and the dominant conduction mechanism in

19

ferrite is the electron hopping between Fe2+ and Fe3+ ions [36]. Similar behavior of decrease in electrical

20

resistivity has been reported for the Z-type hexaferrite sintered in the N2 atmosphere [37]. After the

21

oxygenation process at 300˚C for 24 h in the O2 atmosphere, the values of ρ and Ea both increases for all

22

samples, which indirectly reveals the removal of oxygen vacancy and reduction of Fe2+ ions in samples.

AC C

EP

TE D

14

1

MANUSCRIPT However, further research is requiredACCEPTED for a quantitative analysis of the Fe2+/Fe3+ ratio of samples for better

2

understanding.

3 4

4.

Conclusion With the sintering and subsequent furnace-cooling process in the low PO2 of 10-3 atm, we could

6

successfully prepare SrZnxFe(2-x)W polycrystalline samples for 0.0 ≤ x ≤ 1.0. The lattice parameters and cell

7

volumes of the samples increased with increasing x up to 1.0, which also represents a successful substitution

8

of Zn2+ for the Fe2+ site. The Ms values of samples monotonously increased from x = 0.0 to x = 1.0, and thus

9

the highest Ms value of 87.7 emu/g was obtainable from the sample of x = 1.0 sintered at 1250˚C for 2 h. The

10

increase in Ms value with increasing x is believed to originate from the occupation of Zn2+ ions to the

11

tetrahedral site of S block in W-type hexaferrite structure, which can cause an increase in the net magnetic

12

moment by reducing the magnetic moment of down spin site. Although further quantitative analyses on the

13

oxygen non-stoichiometry and the Fe2+/Fe3+ ratio of samples are required, oxygen annealing experiment

14

clearly suggests that the oxygen non-stoichiometry is larger for the samples sintered at higher temperatures in

15

low PO2 of 10-3 atm, leading to the increase in the Ms, Vcell, and ρ values of samples.

16

18 19 20

Acknowledgement

This research was supported by Future Materials Discovery Program through the National Research

AC C

17

EP

TE D

M AN U

SC

RI PT

5

Foundation of Korea (NRF) funded by the Ministry of Science and ICT (2017M3D1A1013748).

1 2

-3 ACCEPTED MANUSCRIPT Table 1. Magnetic and structural properties of the SrZn atm. The xFe(2-x)W samples sintered in PO2 = 10 sample of x = 2.0 marked with * is sintered at 1300˚C for 2 h in air.

3

0.5

Ms (emu/g)

Ha (Oe)

1300

78.8(8)

16364(7)

5.8906(1) 32.774(1)

1310

79.8(8)

17265(36)

5.8932(9) 32.784(9)

986.02(4)

1315

81.3(8)

19468(44)

5.8953(1) 32.795(1)

987.08(5)

1250

81.9(7)

15414(69)

5.8913(2) 32.793(2)

985.68(9)

1275

84.5(6)

16443(60)

5.8947(3) 32.805(2)

987.19(14)

1280

85.2(8)

16875(52)

5.8949(4) 32.816(3)

988.57(17)

1225

87.1(7)

1250

87.7(5)

1300*

77.3(2)

1.0 2.0

EP

6

AC C

5

3

Vcell (Å ) 984.87(5)

RI PT

c (Å)

15114(71)

5.9002(2) 32.852(2)

990.43(8)

15067(69)

5.9025(2) 32.870(2)

991.73(8)

14604(92)

5.9100(2) 32.874(3)

994.42(11)

TE D

4

a (Å)

SC

0.0

Annealing temperature (˚C)

M AN U

x

MANUSCRIPT Table 2. Magnetic sublattices of W-typeACCEPTED hexaferrite and their coordination, number of ions, and spin alignment

Coordination

Block

12k

octahedral

R-S

6

4e

tetrahedral

S

4fIV

tetrahedral

S

4fVI

octahedral

R

6g

octahedral

S-S

4f

octahedral

2d

hexahedral

EP

6 7

AC C

4 5

up down

2

down

2

down

3

up

S

2

up

R

1

up

SC

2

TE D

3

Spin

RI PT

Sublattice

Number of ions per molecule

M AN U

1 2

1 2 3 4 5 6 7

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 1. XRD patterns for the SrZnxFe(2-x)W samples of x = 0.0, 0.5, 1.0, 1.5 and 2.0, sintered at 1315, 1280, 1250, 1225 and 1225˚C, respectively for 2 h in PO2 = 10-3 atm. XRD patterns were measured by a θ-2θ scan with the step size of 0.02˚, and time per step of 0.6 sec for the 2θ range of 20–70˚ (a) and time per step of 2.4 sec for the 2θ range of 29–38˚ (b). The XRD peaks of SrZnxFe(2-x)W, SrZn2Fe12O22 and ZnFe2O4 are indexed according to the JCPDS 00-051-1882, JCPDS 00-051-1880 and JCPDS 00-022-1012.

3 4 5 6 7 8 9

Fig. 2. The plots of the values of (a) Vcell (a2csin120˚) and (b) a and c for the samples listed in Table 1. The numbers in Fig 2(a) indicate the sintering temperature of each data point. The data point marked with * is that of the sample sintered in air.

AC C

1 2

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

2 3 4 5 6 7 8

AC C

1

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 3. (a) The magnetic hysteresis loops of the selected samples which mainly consist of SrZnxFe(2-x)W phase. The inset figure shows magnetization curve in the field region of 2–5 T. (b) the plot of the values of Ms and Ha listed in Table 1 vs. x in SrZnxFe(2-x)W.

1 2 3 4

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 4. The plots of (a) a and c vs. sintering temperature, (b) Vcell vs. sintering temperature, (c) Ms vs. sintering temperature for the samples listed in Table 1.

2 3 4 5

Fig. 5. The effect of oxygen annealing on the (a) Vcell and (b) Ms of the samples (x = 0.0) sintered at 1300, 1310 and 1315˚C for 2 h in PO2 = 10-3 atm.

AC C

1

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Fig. 6. The plots of (a) ρ-T and (b) lnρ-1000/T in the temperature region of 100–300K of the samples (x = 0.0) sintered at 1300, 1310 and 1315˚C for 2 h in PO2 = 10-3 atm, and oxygen annealed at 300˚C for 24 h in PO2 = 1 atm. The inset figure in (a) is a ρ-T curve in the temperature range of 270–300K.

AC C

1 2 3 4 5 6

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

[5] [6] [7] [8]

[9]

[10] [11] [12] [13] [14]

[15] [16] [17]

[18] [19] [20] [21]

RI PT

[4]

SC

[3]

M AN U

[2]

J. Went, G. Rathenau, E. Gorter, G. Van Oosterhout, Ferroxdure, a class of new permanent magnet materials, Philips Tech. Rev. 13 (1952) 194–208. F. Lotgering, P. Vromans, M. Huyberts, Permanent- magnet material obtained by sintering the hexagonal ferrite W = BaFe18O27, J. Appl. Phys. 51 (1980) 5913–5918. J. Mignot, P. Wolfers, J. Joubert, A new series of materials for permanent magnets: The W ferrites BaZn2(1−x)(LiFe)xFe16O27, J. Magn. Magn. Mater. 51 (1985) 337–341. A. Collomb, J. Mignot, The Ba(Sr)Mn2Fe16O27 W-type hexagonal ferrites as permanent magnets, J. Magn. Magn. Mater. 69 (1987) 330–336. T. Kagotani, N. Abe, M. Okada, M. Homma, Effects of Alkali-Earth and Rare-Earth-Metal Fluorides Additions on Magnetic-Properties of W-Type Sr Ferrite Powders, Mater. T. Jim. 31 (1990) 879–883. S. Ram, J. Joubert, Development of high-quality ceramic powders for Sr0.9Ca0.1Zn2-W type hexagonal ferrite for permanent magnet devices, IEEE T. Magn. 28 (1992) 15–20. H. Yamamoto, T. Mitsuoka, Effect of CaO and SiO2 additives on magnetic properties of SrZn2-W type hexagonal ferrite, IEEE T. Magn. 30 (1994) 5001–5007. F. Leccabue, O. A. Muzio, M. S. E. Kany, G. Calestani, G. Albanese, Magnetic-Properties and Phase Formation of SrMn2Fe16O27 (SrMn2-W) Hexaferrite Prepared by the Coprecipitation Method, J. Magn. Magn. Mater. 68 (1987) 201–212. S. Ram, J. Joubert, Synthesis and magnetic properties of SrZn2-W type hexagonal ferrites using a partial 2Zn2+ → Li+Fe 3+ substitution: a new series of permanent magnets materials, J. Magn. Magn. Mater. 99 (1991) 133–144. H. Yamamoto, T. Nakajima, D. Yamada, Magnetic-Properties of Sr-Zn System W-Type Hexagonal Ferrite Magnets Prepared by New Manufacturing Method, IEEE T. Magn. 28 (1992) 2868–2870. X. S. Liu, P. Hernandez-Gomez, K. Huang, S. Q. Zhou, Y. Wang, X. Cai, Research on La3+-Co2+-substituted strontium ferrite magnets for high intrinsic coercive force, J. Magn. Magn. Mater. 305 (2006) 524–528. Y. M. Kang, K. S. Moon, Magnetic properties of Ce-Mn substituted M-type Sr-hexaferrites, Ceram. Int. 41 (2015) 12828–12834. G. H. An, T. Y. Hwang, J. Kim, J. Kim, N. Kang, K. W. Jeon, Novel method for low temperature sintering of barium hexaferrite with magnetic easy-axis alignment, J. Eur. Ceram. Soc. 34 (2014) 1227–1233. S. P. Ruan, B. K. Xu, H. Suo, F. Q. Wu, S. Q. Xiang, M. Y. Zhao, Microwave absorptive behavior of ZnCosubstituted W-type Ba hexaferrite nanocrystalline composite material, J. Magn. Magn. Mater. 212 (2000) 175–177. Y. J. Kim, S. S. Kim, Microwave absorbing properties of Co-substituted Ni2W hexaferrites in Ka-band frequencies (26.5–40 GHz), IEEE T. Magn. 38 (2002) 3108–3110. U. Ozgur, Y. Alivov, H. Morkoc, Microwave ferrites, part 1: fundamental properties, J. Mater. Sci-Mater. El. 20 (2009) 789–834. M. J. Iqbal, R. A. Khan, S. Takeda, S. Mizukami, T. Miyazaki, W-type hexaferrite nanoparticles: A consideration for microwave attenuation at wide frequency band of 0.5–10 GHz, J. Alloy. Compd. 509 (2011) 7618–7624. C. Stergiou and G. Litsardakis, Electromagnetic properties of Ni and La doped strontium hexaferrites in the microwave region, J. Alloy. Compd. 509 (2011) 6609–6615. V. G. Harris, Modern Microwave Ferrites, IEEE T. Magn. 48 (2012) 1075–1104. R. C. Pullar, Hexagonal ferrites: A review of the synthesis, properties and applications of hexaferrite ceramics, Prog. Mater. Sci. 57 (2012) 1191–1334. N. Langhof, D. Seifert, M. Göbbels, J. Töpfer, Reinvestigation of the Fe-rich part of the pseudo-binary system

TE D

[1]

EP

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

ACCEPTED MANUSCRIPT

References

AC C

1

[29] [30] [31] [32] [33] [34] [35] [36] [37]

RI PT

[28]

SC

[27]

M AN U

[26]

TE D

[25]

EP

33

[22] [23] [24]

MANUSCRIPT SrO-Fe2O3, J. Solid State Chem. ACCEPTED 182 (2009) 2409–2416. H. Van Hook, Thermal stability of barium ferrite (BaFe12O19), J. Am. Ceram. Soc. 47 (1964) 579–581. J. Smit, H. Wijn, Ferrites, Philips Technical Library, Eindhoven, 1959. H. Graetsch, F. Haberey, R. Leckebusch, M. S. Rosenberg, K. Sahl, Crystallographic and magnetic investigation on W-type-hexaferrite single crystals in the solid solution series SrZn2-xCoxFe16O27, IEEE T. Magn. 20 (1984) 495–500. G. Litsardakis, D. Samaras, A. Collomb, Magnetic properties of the Sr(Ba)ZnxMn2-xW hexagonal ferrites, J. Magn. Magn. Mater. 81 (1989) 184–188. X. H. Wang, T. L. Ren, L. Y. Li, L. S. Zhang, Preparation and magnetic properties of BaZn2-xCoxFe16O 27 nanocrystalline powders, J. Magn. Magn. Mater. 184 (1998) 95–100. J. H. You, H. J. Kim, S.I. Yoo, Preparation of strontium W-type hexaferrites in a low oxygen pressure and their magnetic properties, J. Alloy. Compd. 695 (2017) 3011–3017. R. Grössinger, A critical examination of the law of approach to saturation. I. Fit procedure, Phys. Status Solidi A 66 (1981) 665–674. G. Albanese, M. Carbucicchio, G. Asti, Spin-Order and Magnetic-Properties of BaZn2Fe16O27 (Zn2-W) Hexagonal Ferrite, Appl. Phys. 11 (1976) 81–88. A. Collomb, O. Abdelkader, P. Wolfers, J. C. Guitel, D. Samaras, Crystal-Structure and Magnesium Location in the W-Type Hexagonal Ferrite-[Ba]Mg2-W, J. Magn. Magn. Mater. 58 (1986) 247–253. A. Collomb, B. Lambert- Andron, J. Boucherle, D. Samaras, Crystal Structure and Cobalt Location in the W- Type Hexagonal Ferrite [Ba]Co2-W," Phys. Status Solidi A 96 (1986) 385–395. A. Collomb, M. Valletregi, Zinc Location in the W-Type Hexagonal Ferrite-BaZn2Fe16O27, Mater. Res. Bull. 22 (1987) 753–760. A. Collomb, G. Litsardakis, J. Mignot, D. Samaras, Manganese location and magnetic properties of the Sr(Ba)Mn2Fe16O27 W-type hexagonal ferrite, IEEE T. Magn. 24 (1988) 1936–1938. G. Albanese, G. Asti, Sublattice magnetization and cation distribution in BaMg2Fe16O27 (Mg2W) ferrite, IEEE T. Magn. 6 (1970) 158–161. Y. M. Kang, High saturation magnetization in La-Ce-Zn-doped M-type Sr-hexaferrites, Ceram. Int. 41 (2015) 4354–4359. Y. Bai, J. Zhou, Z. Gui, L. Li, Electrical properties of non-stoichiometric Y-type hexagonal ferrite, J. Magn. Magn. Mater. 278 (2004) 208–213. X. Zhang, Z. Yue, S. Meng, B. Peng, L. Yuan, Magnetic and electrical properties of Z-type hexaferrites sintered in different atmospheres, Mater. Res. Bull. 65 (2015) 238–242.

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

ACCEPTED MANUSCRIPT Partially Zn-substituted strontium W-type hexaferrite was prepared for the first time. Zn substitution improved the saturation magnetization (Ms) of the samples. The existence of oxygen non-stoichiometry of the samples was indirectly found.

AC C

EP

TE D

M AN U

SC

RI PT

The oxygen non-stoichiometry seems to increase Ms and conductivity of the samples.