Negative magnetoresistance in nanotwinned NiMnGa epitaxial films

1
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
www.na u e.com/scien i ic epo s
Nega i e Magne o esis ance in
Nano winned NiMnGa Epi axial
Films
V. O. Golub
1, V. A. Che nenko2,3, A. Apolina io
4, I. R. Aseguinolaza2, J. P. A aujo4,
O. Salyuk1, J. M. Ba andia an2 & G. N. Kakazei
4
Magne ic shape memo y alloys a e unde in ensi e in es iga ion due o hei unusual physical
p ope ies, such as magne ic shape memo y e ec , magne ic ield induced supe elas ici y, di ec
and in e se magne ocalo ic e ec e c., p omising o no el applica ions. One o he in iguing
p ope ies o hese ma e ials in a single phase s a e is a gian magne o esis ance. He e we epo
he ema kable esul s abou he magne o esis i e p ope ies o epi axial ilms o Ni52.3Mn26.8Ga20.9
magne ic shape memo y alloy in he empe a u e ange o 100–370 K, well below he ma ensi ic
ans o ma ion empe a u e. I was ound ha he o ma ion o non-collinea magne ic s uc u e due
o a nano winning o he ilm esul s in elec on sca e ing on such a s uc u e and no iceable nega i e
magne o esis ance in he en i e in es iga ed empe a u e ange.
The Heusle ype magne ic shape memo y alloys ha e been in ensi ely s udied du ing he las decades bo h due
o hei qui e unusual physical p ope ies and po en ial applica ions in di e en a eas o enginee ing and echnol-
ogy1. The in e es o hese ma e ials was s imula ed by he disco e y o he magne ic shape memo y (MSM) e ec
consis ing in a d ama ic size change (up o 12%) unde a magne ic ield, as a esul o win bounda y mo ion in
he ma ensi ic s a e2,3. Since ha a numbe o in e es ing physical e ec s ha e been e ealed in hese ma e ials,
such as he magne ic ield induced supe elas ici y, di ec and in e se magne ocalo ic e ec , exchange bias e c.
(see, e.g.4–7, and e e ences he ein). No e, ha acco ding o he heo e ical p edic ions, some magne ic shape
memo y alloys can exhibi up o a hund ed pe cen o spin pola iza ion, which is e y a ac i e o spin onic
applica ions7.
One o he in iguing p ope ies o hese ma e ials is nega i e magne o esis ance (MR) obse ed nea he ma -
ensi ic ans o ma ion. In he he e ogeneous magne ic/non-magne ic ilms, ei he mul ilaye ed8 o g anula 9, he
MR consis s in he esis i i y change due o he educ ion o magne ic diso de caused by a magne ic ield. Fo a
homogeneous e omagne ic ma e ial wi h localized spins, one can expec MR empe a u e dependence exhibi -
ing a peak a Cu ie empe a u e and going o ze o a low empe a u es10. Such beha io is commonly obse ed o
many Heusle alloys, whe eas he o e all empe a u e dependence o magne o esis ance is much mo e complex
(see, e.g.11,12, and e e ences he ein) and does no exhibi a egula dependence on he phase s a e o he alloys11.
A numbe o epo s we e ocused on in es iga ion o he magne o esis ance a he ma ensi ic ans o ma ion,
whe eby he la ges nega i e magne o esis ance alue 60–70% was epo ed o NiMnIn alloys13,14.
The e a e di e en (some imes con o e sial) explana ions in he li e a u e conce ning he magne o esis ance
in he ma ensi ic and aus eni ic phases, as well as a he ma ensi ic ans o ma ion. The ollowing h ee main
mechanisms we e used o explain he expe imen al magne o esis ance da a ob ained in Heusle MSM alloys:
aniso opic magne o esis ance, in insic magne ic sca e ing and in e ace spin sca e ing11,15,16. The in luence o
magne ic ield on he c ys alline, elec onic and magne ic s uc u e in hese ma e ials was discussed in e s11–17.
Pa icula ly, in MSM single c ys als, ei he in a single ma ensi ic o in a single aus eni ic s a e, he nega i e
magne o esis ance is conside ed o be ela ed o he s-elec on sca e ing by he localized d-spins diso de ed
ei he due o he mal luc ua ions o due o some weak an i e omagne ic exchange in such sys ems10,12. In mos
cases, elec on sca e ing om he domain walls can be neglec ed due o hei low concen a ion. Fo some MSM
alloys in ma ensi ic s a e, he win bounda ies ea angemen unde he ac ion o magne ic ield is conside ed
1Ins i u e o Magne ism NASU and MESU, Kyi , 03142, Uk aine. 2BCMa e ials & Uni e si y o Basque Coun y, P.O.
Box 644, Bilbao, 48080, Spain. 3Ike basque, Basque Founda ion o Science, Bilbao, 48013, Spain. 4IFIMUP-IN/
Depa amen o de Fisica e As onomia, Uni e sidade do Po o, 4169-007, Po o, Po ugal. Co espondence and
eques s o ma e ials should be add essed o V.O.G. (email: [email p o ec ed])
Recei ed: 27 June 2018
Accep ed: 1 Oc obe 2018
Published: xx xx xxxx
OPEN
www.na u e.com/scien i ic epo s/
2
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
as a mechanism o he esis i i y change (see e.g.11), bu i has been a gued18 ha he esis i i y o a single c ys al
is no ela ed o he concen a ion o win bounda ies. Ne e heless, i will be shown in he p esen wo k ha he
exis ence o s able ine win s uc u e can lead o he nega i e magne o esis ance in he Heusle MSM alloys in he
en i e empe a u e ange o he ma ensi ic s a e s abili y. This e ec can be obse ed o a lo o di e en MSM
alloys, whe e he alue o magne o esis ance can be much highe , in a wide empe a u e ange.
Resul s
A 300 nm- hick Ni52.3Mn26.8Ga20.9 (a .%) ilm was epi axially g own by magne on spu e ing on o a MgO(001)
subs a e. The composi ion o he ilm and sample p epa a ion p ocedu e was he same as o he samples in es-
iga ed in19. The composi ion and he p epa a ion p ocedu e we e chosen o ob ain epi axial ilms wi h he
ma ensi ic ans o ma ion empe a u e well abo e he Cu ie empe a u e o ge id o any in luence o he ans-
o ma ion e ec s on magne o esis ance. Composi ion o he ilm was de e mined wi h an unce ain y o 0.5 a .%
by ene gy-dispe si e x- ay spec oscopy. The cubic- o-o ho hombic ma ensi ic ans o ma ion was obse ed
by x- ay di ac ion a abou 420 K. The x- ay analysis e ealed o ho hombic c ys al s uc u e wi h he longes ,
a, and sho es , c, axes lying in he ilm plane, while b is pe pendicula o he ilm plane (Fig.1). The a omic o ce
mic oscopy image demons a es a o ma ion o ine s ipe-like win s uc u e wi h a cha ac e is ic win wid h, l,
o abou six y nanome e s (see he Inse o Fig.1). This mo phology is simila o he one obse ed in19. Magne ic
suscep ibili y and e omagne ic esonance (FMR) measu emen s showed ha he ilm exhibi s a e omagne ic
o de ing wi h he Cu ie empe a u e a ound 370 K.
The magne o esis ance was measu ed using a s anda d ou -p obe me hod unde magne ic ields up o 9 kOe
in he empe a u e ange o 100–370 K. The magne ic ield was applied in he ilm plane. The da a a e p esen ed in
Fig.2. A nega i e magne o esis ance wi h simila dependences on he applied magne ic ield, as depic ed in he
Inse o Fig.2, is obse ed in he en i e empe a u e ange. The compa ison wi h he magne iza ion loop shown
in he Inse o Fig.3 e eals ha o high ield alues (abo e he hys e esis a ea, H > 1 kOe) he a ia ion o he
magne o esis ance wi h he magne ic ield ollows he change o magne ic induc ion B = H + 4πM (Fig.3), which
is ypical o s-d sca e ing mechanism10, whe e M is he magne iza ion. No ewo hy, he empe a u e beha io o
MR in Fig.2 is non-mono onic. Two b oad peaks a e obse ed: he i s one is in he icini y o he Cu ie empe -
a u e and he second one is a ound 200 K. The i s peak can be asc ibed o he conduc ion s-elec ons sca e ing
by localized d-spins which a e diso de ed due o he mal luc ua ions, whe eas he appea ance o he second one
will be discussed below.
Discussion
The o igin o he app eciable alues o magne o esis ance in wide empe a u e ange can be deduced om he ol-
lowing lines o easoning. As i has been men ioned abo e, elec on sca e ing om domain walls can be excluded
om he conside a ion due o hei low concen a ion in his ilm. The addi ional con i ma ion o his s a emen
is ha he cha ac e o esis ance beha iou p ac ically does no change o low ields (below 1 kOe, whe e he
hys e esis beha iou o he magne iza ion is obse ed) and o high magne ic ields (up o 9 kOe). The s uc u al
cha ac e iza ion does no e eal any addi ional phase in he ilms. Only a single ma ensi ic phase is obse ed
below he Cu ie empe a u e, which is, in u n, encoun e ed below he ma ensi ic ans o ma ion empe a u e.
Magne ic measu emen s showed no signs o he p esence o an i e omagne ic clus e s in he ilm. The p es-
ence o such clus e s usually esul s in he shi o hys e esis loop a low empe a u es due o he exchange bias
e ec (see, e.g.20). These ac s allow excluding he mechanism o sca e ing om he phase o clus e in e aces.
Figu e 1. XRD pa e n o Ni-Mn-Ga/MgO(001) hin ilm a oom empe a u e showing (020) and (040)
peaks o he o ho hombic uni cell o he ilm and he e lec ions belonging o he subs a e. A omic o ce
mic oscopy image o he ilm su ace is shown in he inse .
www.na u e.com/scien i ic epo s/
3
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
Tempe a u e beha iou o magne o esis ance can no be also asc ibed o he spin luc ua ion esul ing om a
weak an i e omagne ic exchange in he sys em, due o a small manganese ions excess in he alloy12.
Taking in o accoun an i ele ance o he men ioned ac o s, he appea ance o spin diso de in he s udied
ilm can be a ibu ed o he o ma ion o s ipe-like nano win (mesoscale) s uc u e. P e iously we discussed his
issue in e .19 o simila epi axial NiMnGa ilm wi h a hickness o 500 nm. I has been shown he e ha he win
wid h l is compa able wi h he exchange co ela ion leng h (domain wall wid h),
δ
~
aH H(/)/2
EA
1/2
, whe e a is he
la ice pa ame e , HE and HA a e he in e nal magne ic ields ha cha ac e ize he spin exchange in e ac ion and
he magne oc ys alline aniso opy, espec i ely. The ough es ima ion o he exchange co ela ion leng h
(7–19 nm) was pe o med o single a ian ma ensi e using he exis ed expe imen al da a and heo e ical calcu-
la ions. Al hough he men ioned heo y can’ be di ec ly applied o calcula e δ o exchange coupled wins i is sa e
o s a e ha his alue should be highe due o he e ec i e aniso opy ield educ ion19, up o 30 nm. The e o e
he exchange co ela ion leng h in ou 300 nm ilm is also compa able wi h he win wid h (l ≈ 60 nm). This s a e-
men is con i med by he shape o e omagne ic esonance spec a in pe pendicula geome y ( he magne ic
ield pe pendicula o he ilm plane). The ypical esonance spec um is p esen ed in he Inse o Fig.4. This
spec um does no quali a i ely modi y wi h he empe a u e, see Fig.5.
When he exchange co ela ion leng h is smalle han a win wid h (
δl
), he magne iza ion ec o in he
a ian s aligns along he in-plane aniso opy easy axes wi h 90° domain walls con ined wi hin he win bounda-
ies (no coupling be ween wins), see he schema ic in Fig.4. In his case he magne ic esonance condi ions o
pe pendicula geome y a e he same as o single a ian ilms, because he magne isa ion in each a ian p e-
cesses independen ly. The posi ion o he esonance line was calcula ed using Eq. (17) om e .19 wi h he se o
Figu e 2. Tempe a u e dependence o he magne o esis ance, MR, unde magne ic ield, H = 9 kOe. The
dependence o MR on magne ic ield a T = 200 K is shown in he Inse .
Figu e 3. Dependence o MR on magne ic ield and hys e esis loop o he ilm (in he Inse ) a T = 300 K.
www.na u e.com/scien i ic epo s/
4
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
magne ic pa ame e s discussed he e (g een line in he Inse o Fig.4). The schema ic in Fig.4 also shows ha in
case δ > l he magne iza ion ec o in he a ian s aligns along he in-plane aniso opy easy axes wi h 90° domain
walls con ined wi h he win bounda ies (s ong coupling be ween wins). In his case he in-plane o ho hombic
uniaxial aniso opy is pa ially a e aged among he a ian s, wi h a co esponding educ ion (see Eq. (18) in
e .19) o he FMR esonance ield ( ed line in he Inse ). Finally, i δ is compa able o l, a non-collinea magne ic
s uc u e should be o med, esul ing in inhomogeneous b oadening o he esonance line19. Indeed, he inhomo-
geneously b oadened line was obse ed in ou FMR expe imen s (black line in he Inse ). I is wo h o men ion
ha he win wid h o 300 nm hin ilm (60 nm) was ound o be a bi smalle han o 500 nm one (80 nm), lead-
ing o he smalle deg ee o he magne iza ion inhomogenei y.
The highe he deg ee o he magne iza ion inhomogenei y is, he b oade line should be obse ed. A pa am-
e e ha cha ac e izes he magne iza ion inhomogenei y is ΔHp19, see Fig.4. O he ypes o inhomogenei ies
causing he sp ead o magne ic pa ame e s, such as he composi ion, s uc u e and/o esidual s ess, also lead o
Figu e 4. Dependencies o he e omagne ic esonance line pa ame e s ΔHp (□) and ΔH (○) on he
empe a u e. Righ -up co ne a ea shows a schema ic o he magne ic momen s (a ows) dis ibu ion in
nano wined ilms o di e en ela ions be ween he exchange co ela ion leng h δ and he win a ian wid h l:
δl
(a), δ > l (b) and l ~ δ (c). Double side a ows deno e he magne oc ys alline aniso opy easy axis
di ec ions in win a ian s. The Inse p esen s a ypical FMR spec um obse ed in he ilm a T = 290 K unde
magne ic ield pe pendicula - o- he- ilm-plane (solid line). Calcula ed spec a o he cases o s ong exchange
coupling be ween win a ian s (dash-do line) and no coupling (do line) a e shown, o compa ison.
Figu e 5. The FMR spec a measu ed a a ious empe a u es.
www.na u e.com/scien i ic epo s/
5
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
he esonance line b oadening which can be desc ibed by he pa ame e , ΔH (Fig.4). Figu e4 shows ha
ΔΔHH
p
in he en i e in es iga ed empe a u e ange, excep o he Cu ie empe a u e egion.
ΔHp g adually inc eases when he empe a u e dec eases, whe eas he dependence o ΔH on he empe a u e
is nonmono onous (Fig.4). An inc ease o ΔH nea he Cu ie empe a u e is ela ed o he he mal luc ua ion o
he magne iza ion when he empe a u e app oaches he magne ic o de -diso de egion. A low empe a u es,
he inc ease o ΔH is ela ed o he inc ease o he magne iza ion, magne ic aniso opy ields and s esses due
o di e en coe icien s o he mal expansion o he ilm and he subs a e, which enhance he inhomogenei ies.
The inc ease o ΔHp du ing cooling is ela ed o he inc ease o magne iza ion a ia ion in he nano wins, due o
he dec ease o he exchange co ela ion leng h. The la e is p oduced by he inc ease o he magne ic aniso opy
ield21,22 a low empe a u es19.
Appa en ly, he inc ease o magne ic inhomogenei y en ails an inc ease o elec on sca e ing in he nano -
win s uc u e. This should esul in he inc ease o magne o esis ance alue a low empe a u es. Howe e , he
highe he magne iza ion inhomogenei y is, he highe magne ic ields should be applied o align he magne ic
momen s leading o a dec ease o he magne o esis ance in a ini e magne ic ield. We canno also disca d he
dec ease o he nega i e magne o esis ance as a esul o he inc ease o no mal (posi i e) magne o esis ance in
he low empe a u e egime, due o he well-known Lo en z o ce ac ion on cha ge ca ie s. The con ibu ion o
his mechanism g ows when he empe a u e dec eases, due o he empe a u e-induced educ ion o he elec on
mean ee pa h. The in e play o hese ac o s esul s in he appea ance o he b oad maximum on he empe a u e
dependence o magne o esis ance obse ed expe imen ally (Fig.2).
In conclusion, i is ound ha he nano win s uc u e in he ma ensi ic s a e o epi axial e omagne ic shape
memo y ilms esul s in he o ma ion o a non-collinea magne ic s uc u e, which leads o he nega i e mag-
ne o esis ance obse ed expe imen ally in a wide empe a u e ange. This mechanism is he eason o a high
alue o magne o esis ance in he Heusle alloys wi h la ge elec on ee pa h. A simila e ec is also expec ed
o me amagne ic shape memo y alloys wi h ine win s uc u es exhibi ing e omagne ic exchange inside he
win a ian s and an i e omagne ic exchange ac oss win bounda ies, as ecen ly epo ed in e .23. A ema kable
applied aspec o he s udied magne o esis ance e ec is he possibili y o gene a ing spin wa es wi h de ined
equencies by elec ic cu en pulses on pe iodic nano win s uc u es o med in he MSM epi axial ilms. Such
an e ec can be used in magnonic de ices.
Me hods
A 300 nm- hick Ni52.3Mn26.8Ga20.9 (a .%) ilm was epi axially g own by magne on spu e ing on o a MgO(001)
subs a e hea ed a 500 °C unde 2.6 × 10−2 mba p essu e and 150 W powe . Composi ion o he ilm was de e -
mined wi h an unce ain y o 0.5 a .% by ene gy-dispe si e x- ay spec oscopy (EDX), using a scanning elec on
mic oscope Jeol JSM-6400. Philips X’Pe PRO x- ay di ac ome e (CuKα adia ion) was used o he s uc-
u e in es iga ion in 100–450 K empe a u e ange. The a omic o ce mic oscopy (Veeco Nanoscope IVA) was
used o he isualiza ion o he win s uc u e. Magne ic pa ame e s o he ilms we e de e mined om mag-
ne ic (Quan um Design SQUID MPMS-5 magne ome e ) and e omagne ic esonance measu emen s (B uke
ELEXSYS E500 elec on spin esonance spec ome e ). The magne o esis ance was measu ed using a s anda d
ou -p obe me hod unde magne ic ields up o 9 kOe in he empe a u e ange o 100–370 K.
Re e ences
1. Heusle Alloys. P ope ies, G ows, Applica ions (ed. Felse , C. & Hi oha a, A.) 222 (Sp inge Se ies in Ma e ials Science 2016).
2. Heczko, O., Sozino , A. & Ullakko, K. Gian ield-induced e e sible s ain in magne ic shape memo y NiMnGa alloy. IEEE T ans.
Mag. 36, 3266–3268 (2000).
3. Mu ay, S. J., Ma ioni, M., Allen, S. M., O’Handley, R. C. & Log asso, T. A. 6% magne ic- ield-induced s ain by win-bounda y
mo ion in e omagne ic Ni–Mn–Ga. Appl. Phys. Le . 77, 886–888 (2000).
4. Ace , M., Manosa, L., & Planes, A. Magne ic- ield-induced e ec s in ma ensi ic Heusle -based magne ic shape-memo y alloys in
Handbook o Magne ic Ma e ials (ed. Buschow, K. H. J.) 19, 231–289 (Else ie 2011).
5. Planes, A., Manosa, L. & Ace , M. Magne ocalo ic e ec and i s ela ion o shape-memo y p ope ies in e omagne ic Heusle
alloys. J. Phys.: Condens. Ma . 21, 233201 (2009).
6. Diso de and s ain-induced complexi y in unc ional ma e ials (ed. Kakeshi a, T., Fukuda, T., Saxena, A. & Planes, A.) 148 (Sp inge
Se ies in Ma e ial Science, Sp inge -Ve lag (2012)).
7. Magne ic Nanos uc u es. Spin Dynamics and Spin T anspo (ed. Zabel, H. & Fa le, M.) (Sp inge -Ve lag, (2013).
8. Baibich, M. N. e al. Gian Magne o esis ance o (001)Fe/(001)C Magne ic Supe la ices. Phys. Re . Le . 61, 2472–2475 (1989).
9. Kakazei, G. N. e al. In luence o co-e apo a ion echnique on he s uc u al and magne ic p ope ies o CoCu g anula ilms.
J. Magn. Magn. Ma e . 196–197, 29–30 (1999).
10. Ka aoka, M. Resis i i y and magne o esis ance o e omagne ic me als wi h localized spins. Phys. Re . B 63, 134435 (2003).
11. Ba andia án, J. M., Che nenko, V. A., Lázpi a, P., Gu ié ez, J. & Feuch wange , J. E ec o ma ensi ic ans o ma ion and magne ic
ield on anspo p ope ies o Ni-Mn-Ga and Ni-Fe-Ga Heusle alloys. Phys. Re . B 80, 104404 (2009).
12. Banik, S. e al. Magne o esis ance beha io o e omagne ic shape memo y alloy Ni1.75Mn1.25Ga. Phys. Re . B 77, 224417 (2008).
13. Sha ma, V. K., Cha opadhyay, M. K., Shaeb, K. H. B., Chouhan, A. & Roy, S. B. La ge magne o esis ance in Ni50Mn34In16 alloy. Appl.
Phys. Le . 89, 222509 (2006).
14. Yu, S. Y. e al. La ge magne o esis ance in single-c ys alline Ni50Mn50−xInx alloys (x = 14–16) upon ma ensi ic ans o ma ion. Appl.
Phys. Le . 89, 162503 (2006).
15. Tolea, F. e al. Speci ic Changes in he Magne o esis ance o Ni–Fe–Ga Heusle Alloys Induced by Cu, Co, and Al Subs i u ions. IEEE
T ans. Mag. 53, 4400307 (2017).
16. Golub, V. O. e al. Anomalous magne o esis ance in NiMnGa hin ilms. J. Appl. Phys. 96, 3865 (2004).
17. Auge, A. e al. Thickness dependence o he ma ensi ic ans o ma ion, magne ism, and magne o esis ance in epi axial Ni-Mn-Sn
ul a hin ilms. Phys. Re . B 85, 214118 (2012).
18. S i as a a, V. K., Cha e jee, R. & O’Handley, R. C. E ec o win bounda ies on elec ical anspo in a Ni–Mn–Ga single c ys al.
Appl. Phys. Le . 89, 222107 (2006).
19. Che nenko, V. A., L o , V. A., Golub, V., Aseguinolaza, I. R. & Ba andia án, J. M. Magne ic aniso opy o mesoscale- winned Ni-Mn-
Ga hin ilms. Phys. Re . B 84, 054450 (2011).

www.na u e.com/scien i ic epo s/
6
Scien i ic REPOR S | (2018) 8:15730 | DOI:10.1038/s41598-018-34057-8
20. Ace , M. & Wasse mann, E. F. Magne ic in e ac ions in Ni-Mn-based magne ic shape-memo y heusle alloys. Ad . Eng. Ma e . 14,
523–529 (2012).
21. Sakon, T., Adachi, Y. & Kanoma a, T. Magne o-S uc u al P ope ies o Ni2MnGa Fe omagne ic Shape Memo y Alloy in Magne ic
Fields. Me als 3, 202–224 (2013).
22. L’ o , V. A. e al. The ole o aniso opic he mal expansion o shape memo y alloys in hei unc ional p ope ies. Ac a Ma e ialia
57, 5605–5612 (2009).
23. Golub, V. O. e al. An i e omagne ic coupling be ween ma ensi ic win a ian s obse ed by magne ic esonance in Ni-Mn-Sn-Co
ilms. Phys. Re . B 95, 024422 (2017).
Acknowledgemen s
The Uk ainian eam is g a e ul o he suppo om Nanoscience and Nano echnology g an 1/18-H, p o ided
by NAS o Uk aine. The Po uguese eam acknowledges he Ne wo k o Ex eme Condi ions Labo a o ies-
NECL and Po uguese Founda ion o Science and Technology (FCT) suppo h ough he p ojec s NORTE-01–
0145-FEDER-022096,POCI-01-0145-FEDER-031302, and EXPL/IF/00981/2013. Spanish eam acknowledges
suppo om Spanish Minis y o Economy, Indus y and Compe i i eness (MINECO) h ough he P ojec s No.
MAT2014–56116-C04-02 and No. MAT2014–56116-C04-03).
Au ho Con ibu ions
V.O.G., G.N.K., J.P.A., J.M.B. and V.A.C. o mula ed he p oblem and planned he wo k, I.R.A., J.M.B. and
V.A.C. designed and ab ica ed he samples and pe o med hei s uc u e and magne ic cha ac e iza ion, A.A.,
G.N.K. and J.P.A. pe o med a omic o ce mic oscopy and magne o esis ance measu emen s, V.O.G. and O.S.
pe o med e omagne ic esonance in es iga ions. All he au ho s con ibu ed o he discussion and w i ing o
he manusc ip ex .
Addi ional In o ma ion
Compe ing In e es s: The au ho s decla e no compe ing in e es s.
Publishe ’s no e: Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and
ins i u ional a ilia ions.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional
License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o
o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C e-
a i e Commons license, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his
a icle a e included in he a icle’s C ea i e Commons license, unless indica ed o he wise in a c edi line o he
ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons license and you in ended use is no pe -
mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he
copy igh holde . To iew a copy o his license, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
© The Au ho (s) 2018