1
Chemical and elec ical p ope ies o LSM ca hodes p epa ed by
mechanosyn hesis
R. Mo iche*
1,2
, D. Ma e o-López
3
, F. J. Go o
1
and M. J. Sayagués
1
1
Ins i u o de Ciencia de Ma e iales de Se illa, Cen o Mix o CSIC-US, Amé ico Vespucio 49, 41092
Se illa, Spain.
2
Depa amen o de Ciencia e Ingenie ía de Ma e iales, Uni e sidad Rey Juan Ca los, C/Tulipán s/n,
28933 Mós oles, Mad id, Spain.
3
Depa amen o de Física Aplicada I, Facul ad de Ciencias, Uni e sidad de Málaga, Campus Tea inos s/n,
29071 Málaga, Spain.
*Co esponding au ho : ocio.mo iche@u jc.es; phone: 0034 914888083
Abs ac
Mechanosyn hesis o La
1-x
S
x
MnO
3
(x = 0, 0.25, 0.5, 0.75 and 1) was ca ied ou a
oom empe a u e om s oichiome ic mix u es o La
2
O
3
, Mn
2
O
3
and S O, ob aining
monophasic powde s wi h he pe o ski e s uc u e. Physical p ope ies o hese
ma e ials and hei chemical compa ibili y wi h he elec oly e y ia s abilized zi conia
(YSZ), which depend s ongly on he La/S a io, we e e alua ed o co obo a e
a ailabili y o be implemen ed as ca hode ma e ial in solid oxide uel cells (SOFCs).
Elec ical conduc i i y alues in ai anged be ween 100–400 Scm
-1
in he empe a u e
ange o 25–850 °C. Samples p esen ed low eac i i y wi h YSZ in he wo king
empe a u e ange (600–1000 ºC) main aining he g ain size small enough o p ese e
he ca aly ic ac i i y o oxygen educ ion.
Keywo ds: Solid Oxide Fuel Cells, LSM, mechanochemis y, pe o ski e s uc u e
2
1. In oduc ion
Solid Oxide Fuel Cells (SOFCs) ha e been widely s udied in he las decade in o de
o each a blue economy concep : “a ze o emission wo ld” [1]. SOFCs a e specially
in e es ing because o hei ad an ages espec o he o he ypes o uel cells:
heo e ical e iciency o 80–90%, lowe cos s, uel lexibili y and be e s abili y wi h
ime, al hough he ope a ion empe a u e is s ill oo high be ween 800–1000 °C o
p ac ical applica ion [2,3].
Di e en mixed ionic-elec onic conduc o s wi h pe o ski e s uc u e ha e been
p oposed in he las ew yea s as po en ial ca hode ma e ials o SOFC, such as
cobal i es, e i es, nickela es and double pe o ski es [3]; howe e , hey usually exhibi
chemical and he mal expansion incompa ibili ies wi h y ia s abilized zi conia (YSZ)
elec oly e. S –doped lan hanum mangani es, La
1-x
S
x
MnO
3
(LSM), a e he mos
common ca hode ma e ial used in SOFC sys ems due o hei high s abili y unde
oxidan a mosphe es and high empe a u es compa ed o o he al e na i e ma e ials [4].
Syn hesis o hese ma e ials has been ex ensi ely s udied using di e en syn he ic
ou es, bu he mechanochemical me hod i sel (no as ac i a ion) is no ye a common
one [5]. Mechanochemis y is a ela i e simple p ocess ha uses high-ene gy ball mills
and pe mi s p oduc ion among o he s o nanos uc u ed mixed oxides [6]. The
mechanical ene gy om impac and shea o ces by applica ion o a high equency
mo emen is ans e ed o he powde inducing solid s a e chemical eac ions. One o
he mos impo an ad an ages o mechanochemis y is i s capabili y o p oduce la ge
ma e ial quan i ies a oom empe a u e and in a e y sho ime.
In he p esen wo k, a s udy o he chemical and physical p ope ies o La
1-x
S
x
MnO
3
(LSM) sys em ob ained by mechanosyn hesis was ca ied ou as small di e ences in he
La/S a io cause signi ican changes in he ca hode pe o mance. A s uc u al s udy o
3
his sys em, also p epa ed by mechanochemis y, was ecen ly published by Sayagués e
al. [7], al hough smalle amoun s o samples we e syn hesized. A summa y o he
ob ained s uc u al esul s is p esen ed he e also o demons a e he scalabili y o
mechanochemis y. Chemical compa ibili y be ween ca hode and elec oly e and
elec ical p ope ies o he ca hode we e measu ed in a cell ha was buil using he
syn hesized LSM as ca hode and YSZ as elec oly e.
2. Expe imen al
La
1-x
S
x
MnO
3
powde samples wi h di e en S con en (x = 0, 0.25, 0.5, 0.75 and 1)
we e syn hesized by a mechanochemical me hod using a plane a y ball mill (model
Mic o-Mill Pul e ise e 7, F i sch) om s oichiome ic mix u es o La
2
O
3
(Ald ich
99.98 %), Mn
2
O
3
(Ald ich 99 %) and S O. The las one was ob ained om calcina ion
o S CO
3
(Ald ich 98%) a 1200 °C o 12 h. This oxide mix u e was placed in o a
ha dened ch omium s eel ja along wi h 7 WC balls (26.4 g; Ø = 15 mm) and was
milled a 600 pm (disc and ial) in ai o ob ain 6 g o each sample. The di e en
composi ions and he equi ed milling ime o ob ain single phase powde s (P samples)
a e p esen ed in Table I. P samples we e hen uniaxially p essed in o pelle s o 12×5
mm (mic os uc u al cha ac e iza ion) o 12×1 mm ( o al conduc i i y) and sin e ed in
ai a 1300 °C o 8 h wi h a hea ing a e o 10 °Cmin
-1
and ee cooling (H samples).
S uc u al cha ac e iza ion and phase iden i ica ion we e ca ied ou by a
PANaly ical X’Pe Di ac ome e . X-Ray Di ac ion (XRD) pa e ns we e scanned
be ween 10–80° in 2θ and s ep–scan mode, using a s ep o 0.05° and an acquisi ion ime
o 320 s. Peaks we e indexed using X’Pe HighSco e Plus and FullP o and WinPlo
so wa es [8,9] we e used o do he i ing and calcula e he cell pa ame e s and he
4
di ac ion domain size (D). Scanning elec on mic oscopy (SEM) images we e
ob ained on Hi achi S5800 SEM-FEG (H samples) and Hi achi S-2400N ( he elec ode-
elec oly e in e ace) mic oscopes. The composi ional a ia ion h ough he in e ace
was analyzed using Ene gy Dispe si e X- ay Spec oscopy (EDX) in Hi achi S-2400N,
which was equipped wi h a B uke de ec o .
T ansmission Elec on Mic oscopy (TEM
and HRTEM) images and elec on di ac ion (ED) pa e ns we e pe o med on a 200
kV Philips CM200 mic oscope equipped wi h a supe win objec i e lens and a LaB
6
ilamen (poin esolu ion=0.25 nm) and a 300 kV TECNAI G2 F30 mic oscope wi h a
ield emission sys em (poin esolu ion=0.2 nm). The analysis o he HRTEM images
was done wi h he Digi al Mic og aph so wa e (Ga an Inc.). The samples we e
p epa ed by dispe sion o he powde in ace one and d ople s o he suspension we e
deposi ed on o a coa ed ca bon coppe g id.
The o al conduc i i y o he pelle s (H samples) was de e mined by he ou -poin
Van de Pauw me hod be ween 100 and 900 ºC du ing he cooling p ocess [10]. The
a ea-speci ic pola isa ion esis ance (ASR) alues we e ob ained unde symme ical
a mosphe es in a wo elec ode con igu a ion. Dense YSZ pelle s (8% Y
2
O
3
Tosoh), o
10 mm o diame e and 1 mm o hickness we e ob ained a 1400 ºC o 5 h. The
symme ical cell o LSM ca hode and YSZ elec oly e was p epa ed by sc een p in ing
using a slu y o 50 w % LSM (P samples) and 50 w % Te pineol; and hen sin e ed a
1100 ºC o 1h [10]. Impedance spec a o he cells was pe o med using a Sola on
1260 FRA, a open ci cui ol age (OCV), in he 0.01–10
6
Hz equency ange wi h an
ac signal ampli ude o 50 mV. The spec a was analysed by using he ZView so wa e
[11].
5
3. Resul s and Discussion
3.1. Mic os uc u al and s uc u al analysis
XRD esul s o he La
1-x
S
x
MnO
3
samples ob ained by mechanochemis y a e
p esen ed in Figu e 1a. In he powde samples P1–P4, all he maxima we e indexed in a
pseudo-cubic symme y (Pm-3m, 221) because he small size o cohe en di ac ion
domains ( he peaks a e wide) did no allow esol ing he eal c ys al symme y.
Howe e , sample P5 (x = 1) exhibi ed some ex a XRD peaks ha indica ed he
o ma ion o he hexagonal symme y (P6
3/
mmc, 194). Simila esul s we e ound in ou
p e ious wo k [7]; al hough in ha case all milled samples p esen ed he pseudo-cubic
symme y (Fig. 2 in e . [7]). This di e ence could be due o he highe milling in ensi y
egime used in [7] (ball- o-powde a io (BPR) o 92 ins ead o 31 as in p esen wo k),
which p obably induced a u he dec easing o he cohe en di ac ion domains. A e
he sin e ing ea men (Fig. 1b), samples P1–P4 de eloped he hombohed al s uc u e
(R-3c, 167; H1–H4 samples), whe eas sample P5 main ained he same hexagonal
symme y (H5). Uni cell dimensions o each phase and he a e age c ys alli e sizes
(D) a e p esen ed in Table II. As can be deduced om he la ge b oadening o
e lec ions in Fig. 1a he c ys alline domain o he as-p epa ed samples is e y small,
a ound 15 nm. Compa ing H and P samples, he size o he di ac ion domains o
sin e ed samples has signi ican ly inc eased (abou 200 nm) as can be in e ed om he
na owe XRD peaks obse ed (Fig. 1b).
When La
3+
is subs i u ed o S
2+
in LaMnO
3
la ice, an amoun o Mn
3+
con e s in o
Mn
4+
o main ain he c ys al elec oneu ali y and he aniso opic de o ma ion due o he
Jahn-Telle e ec is educed. I he la ice pa ame e s o he samples a e analyzed
(Table II; P5 pa ame e s could no be calcula ed app op ia ely since all he hexagonal
peaks we e no su icien ly esol ed), i can be obse ed ha when x alue inc eases, a

6
pa ame e dec eases ( o bo h P and H samples) while he solid solu ion emains.
Howe e , when he o al La is subs i u ed by S (H5), a pa ame e shows an inc ease as
a esul o he change in he symme y g oup (P6
3/
mmc, 194). Simul aneously, an
inc ease o c pa ame e ( o H samples) occu s om x = 0 o x = 0.5 and hen i sligh ly
dec eases o x = 0.75. This a ia ion in c pa ame e can be caused by a p e e en ial
occupa ion o a om posi ions in he cell as i was p oposed by Jang e al. [12]. Finally, c
achie es a alue o 9.0762 Å o H5 sample gi en place o he hexagonal s uc u e. All
o hese e ec s a e e lec ed in a olume cell a ia ion esul ing in a dec ease om
353.05 Å
3
o 338.67 Å
3
in he solid solu ion as S con en inc eases. Fo H5 wi h a
100% in S con en he cell olume u he diminishes eaching a alue o 233.24 Å
3
;
again his change is due o he o ma ion o he new hexagonal uni cell wi h smalle c
pa ame e .
Rela i e densi y measu emen o he pelle s (H samples) has shown alues close o
80% (Table II), showing a sligh inc ease wi h inc easing he S con en while
main aining he solid solu ion ( he hombohed al s uc u e). Howe e , o x = 1 (H5),
whe e he e is a s uc u al change, he ela i e densi y sligh ly dec eases.
SEM and TEM ep esen a i e esul s o he powde and sin e ed samples (P1, P3,
H1 and H3) a e p esen ed in Fig 2 and 3 espec i ely. The mic os uc u e o P samples
obse ed by SEM is e y simila o all o hem, being o med by agglome a ions o
small pa icles. Rep esen a i e mic og aphs co esponding o samples P1 and P3 a e
p esen ed in ig 2(a-b). Howe e he pa icle size o H samples a ies as a unc ion o
he La/S a io and ep esen a i e mic og aphs a e p esen ed in Figu e 2 (c-d) o H1
and H3 samples. H1 (LaMnO
3
) shows an a e age g ain size o 2 µm (Fig. 2c) and H3
sample (Fig. 2d), wi h a highe S con en , shows an a e age g ain size o 0.5 µm. Bo h
images show ace ed shape c ys als, p esen ing se e al unca ed p isms. I is wo h
7
no ing ha he p esence o ace ed g ow h can imp o e he ca aly ic p ope ies o
samples due o an inc ease in he speci ic su ace a ea.
Figu e 3 shows he TEM, ED and HRTEM esul s o he P1, P3, H1 and H3
samples. The powde samples P1 and P3 a e o med by agglome a ed pa icles as seen
in he TEM images (le ). The c ys al di ac ion domain is e y small as can be
deduced om he analysis o he co esponding elec on di ac ion pa e ns (inse ) and
he HRTEM mic og aphs ( igh ). Some c ys al domains a e ma ked in he HRTEM
images and he size is be ween 5 and 20 nm. In all o he domains (110) in e plana
spacing was measu ed and his dis ance could be indexed in bo h, he cubic and he
hombohed al sys em; i was indexed in he cubic pe o ski e s uc u e (P) o be
cohe en wi h he X- ay esul s. The c ys al size o he sin e ed samples (H1 and H3)
has inc eased conside ably as obse ed by SEM. A wide dispe sion o sizes om 100
nm o 2 mic ons was ound. The H1image (le ) shows an o ien ed c ys al along he
[122] zone axis o he hombohed al s uc u e, as indica ed by he EDP (inse ) and he
HRTEM mic og aphs is p esen ed on he igh whe e he in e plana spacing and he
(hkl) planes a e ma ked. TEM image shown o H3 sample p esen s an elonga ed shape
pa icle wi h di e en con as in he ip; ha co espond o wo di e en o ien ed
c ys alline domains as can be deduced om he co esponding ED pa e ns (inse ). One
o hem is o ien ed along he [001] hombohed al s uc u e; i is e y easy o dis inguish
he hexagonal symme y in he con as o he HRTEM mic og aph p esen ed on he le
( he (110) planes and spacing a e depic ed). HRTEM image o he second c ys alline
domain is p esen ed o he igh and belongs o he cubic pe o ski e s uc u e o ien ed
along [001], he (100) planes ha e been ma ked in he image.
All o hese esul s a e in acco dance wi h ou p e ious wo k [7], which
demons a es he scalabili y o mechanosyn hesis as i is possible o ob ain 6 g ins ead
8
o 2 g (in 1 h o milling) o LSM ma e ial only by inc emen ing sligh ly he milling ime
(see able I). This makes e iden some o he ad an ages o mechanochemis y o e
o he syn he ic me hods, since he adi ional ce amic me hod equi es a empe a u e o
1000-1200 ºC [13]; o i s scalabili y con as ed wi h Pechini, ci a e [14] o plasma-
sp aying me hods [15-17].
3.2. Ca hode-Elec oly e In e ace
Chemical compa ibili y be ween he elec odes and he elec oly e is an impo an
aspec because di e en ma e ials a e in con ac o a p olonged ime a high
empe a u e wha can p omo es eac i i y and ca ion di usion be ween he componen s,
a ec ing nega i ely he cell pe o mance. A mix u e o powde s 50:50 w .% composed
o La
1-x
S
x
MnO
3
and YSZ elec oly e has been p epa ed and annealed in a u nace o
24 h in a ange o empe a u es om 800 o 1300 °C. The mix u e a e hea ing was
analyzed by XRD and he esul s o x=0 and x= 0.25 a e shown in Fig. 4, he p esence
o absence o new p oduc s as a consequence o he eac ion be ween LSM and YSZ
was s udied.
Phases wi h low S con en (x=0.25) showed low chemical eac ion wi h YSZ in all
he empe a u e ange s udied. O he wise, o a S con en o x = 0.5, a eac ion be ween
S – and Z – con aining phases akes place leading o he o ma ion o S Z O
3
as
eac ion p oduc a 1300 °C, howe e he e is no eac ion below 1000 ºC, his ac is
impo an as he SOFCs usually wo ks below his empe a u e .
The LSM/YSZ in e ace o symme ical cells p epa ed by sc een p in ing and
sin e ed a 1100 ºC o 1 h was also s udied by SEM and EDX o analyze he chemical
composi ion h ough he ca hode-elec oly e in e ace and elucida e he possible
in e di usion o he di e en elemen s. SEM mic og aphs o he c oss sec ion o
9
in e aces wi h di e en S composi ions a e shown in Figu e 5. All ca hodes p esen
adequa e po osi y and appa en ly good con ac wi h he elec oly e. As could be
expec ed, he LSM g ains ha e g own less han o H samples sin e ed a 1300 °C. LSM
samples wi h x = 0 and x = 0.25 ha e g ain sizes o ~ 1 µm, sligh ly smalle han a
comme cial sample o P axai (La
0.8
S
0.2
MnO
3
), also shown in Figu e 5 o a
compa ison pu pose. Fo la ge x alue samples, he g ain size dec eases wi h
inc easing he S con en , eaching e y small alues o less han 0.5µm o S MnO
3
. In
his case he a ia ion end is e e sed when compa ed wi h H samples, p obably due
o he di e en sin e ing empe a u e and way o p epa ing he bulk sample.
EDX line p o iles o La, S , Mn, Y and Z we e ca ied ou h ough he in e ace and
ep esen a i e esul s ( o x=0.25 and 0.5) a e p esen ed in Figu e 6. I could be deduced
ha o highe S con en s, Z di usion in o he ca hode laye is mo e p onounced,
eaching a pene a ion dis ance o 1 µm o La
0.25
S
0.75
MnO
3
. These esul s a e in
ag eemen wi h he o ma ion o S Z O
3
p e iously obse ed by XRD analysis as
p oduc o eac ion be ween componen s. The appa en di usion o La, S and Mn
elemen s is due o ca hode pa icles de achmen , deposi ed on he analyzed su aces.
3.3. Elec ical p ope ies
The subs i u ion o S
2+
o La
3+
inc eases he con en o Mn
4+
in he s uc u e as i
has been men ioned abo e, c ea ing elec onic holes o keeping cha ge neu ali y [18].
Fo his eason, LSM is p- ype elec onic conduc o unde oxidan a mosphe e [19, 20]:















 













Bu no only elec onic conduc i i y is p esen , also ionic conduc i i y due o he ac
ha acancies a e in oduced o non-s oichiome ic oxygen con en and alio alen










10 20 30 40 50 60 70 80
P1
2
θ
*
••
•
•
•
•
••
•
•
••
•
•
•
••
•
•
•
•
*
•
•
*
*
•
*
**
*
••
•
•
••••
P2
P3
P5
P4

10 20 30 40 50 60 70 80
2
θ
H1
*
♦♦♦
H2
♦
♦
♦
H3
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦♦♦♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
*
H4
*
**
*
*
*
****
*
*
*
**
*
H5



Figu e 1






(a) (b)


























Figu e 2












































Figu e 3

P1
P3
H1
5 nm
[001]p
H3




Figu e 4









(a) (b)
La
0.75
S
0.25
MnO
3
- YSZ
LaMnO
3
- YSZ
*
*
*
••
•
•
*
••
•
1100
o
C
900
o
C
*
*
**
*
*
•
•
•
•
*
••
•
*
*
**
*
*
••
•
•
*
•
•
•
20 30 40 50 60 70 80
*
*
**
*
*
••
•
•
*
••
•
25
o
C
*
*
**
*
*
•
•
•
•
*
••
•
20 30 40 50 60 70 80
2
θ
•
*
*
**
*
*
•
•
•
•
*
•
•
2
θ













Figu e 5









Figu e 6
(a)
(b)
0
20
40
60
80
100
3
2
1
0
1
23
La
0.75
S
0.25
MnO
3
- YSZ
Rela i e In ensi y (%)
Posi ion (µm)
La
Mn
S
Y
Z
4
0
20
40
60
80
100
La
0.50
S
0.50
MnO
3
- YSZ
Rela i e In ensi y (%)
Posi ion (µm)
La
Mn
S
Y
Z
43 2 101 2 3
(b)

Figu e 7



0.8
1.0

1.2 1.4
10

-
1
10

0
10

1
10

2
10

3900

800 700 600 500
LaMnO

3

La

0.75

S

0.25

MnO

3

La

0.5

S

0.5

MnO
3

La

0.25

S

0.75

MnO

3
La

0.8

S

0.2

MnO

3

/ P axai

T(ºC)

(b)
0.8 1.2 1.6 2.0 2.4 2.8 3.2
10-2
10-1
100
101
102
103
800 600 400 200
LaMnO
3
La
0.75
S
0.25
MnO
3
La
0.5
S
0.5
MnO
3
La
0.25
S
0.75
MnO
3
S MnO
3
σ(
Scm
-1
)
10
3
/T (K
-1
)
T(ºC)
phase
ansi ion
(a)

La

S

3

MnO

LaMnO

3

0.75

0.25

MnO
3

La

0.5

S

0.5

La

S

0.75

MnO

3

S

0.25

MnO

3

200

400

600

10

3

0

10

10

-2

10

-1

10

2

10

1

800

T(ºC)

0.8

1.6

1.2

2.0

2.4

2.8

3.2

10
3
/T (K
-
1
)

10
3
/T (K
-
1
)





















Figu e 8
