ORIGINAL RESEARCH
published: 12 Sep embe 2018
doi: 10.3389/ chem.2018.00420
F on ie s in Chemis y | www. on ie sin.o g 1Sep embe 2018 | Volume 6 | A icle 420
Edi ed by:
Ca los Lodei o,
Uni e sidade No a de Lisboa,
Po ugal
Re iewed by:
Guo-Hong Tao,
Sichuan Uni e si y, China
Sukhendu Mandal,
Indian Ins i u e o Science Educa ion
and Resea ch, Thi u anan hapu am,
India
*Co espondence:
Ma ilde Fondo
[email p o ec ed]
Special y sec ion:
This a icle was submi ed o
Ino ganic Chemis y,
a sec ion o he jou nal
F on ie s in Chemis y
Recei ed: 29 June 2018
Accep ed: 24 Augus 2018
Published: 12 Sep embe 2018
Ci a ion:
Fondo M, Co edoi a-Vázquez J,
Ga cía-Deibe AM,
Sanma ín-Ma alobos J, He e a JM
and Colacio E (2018) Field-Induced
Single Molecule Magne s o
Phosphine- and A sine-Oxides.
F on . Chem. 6:420.
doi: 10.3389/ chem.2018.00420
Field-Induced Single Molecule
Magne s o Phosphine- and
A sine-Oxides
Ma ilde Fondo1*, Julio Co edoi a-Vázquez1, Ana M. Ga cía-Deibe1,
Jesús Sanma ín-Ma alobos1, Juan Manuel He e a2and En ique Colacio2
1Depa amen o de Química Ino gánica, Facul ade de Química, Uni e sidade de San iago de Compos ela, San iago de
Compos ela, Spain, 2Depa amen o de Química Ino gánica, Facul ad de Ciencias, Uni e sidad de G anada, G anada, Spain
The coo dina ion chemis y o dysp osium and e bium owa d phosphine and a sine
oxides was u he explo ed. Thus, he new ni a e [M(NO3)3(Ph3PO)3] (M=Tb,
1; Dy, 2), [Dy(NO3)3(E OH)(Ph3XO)2] (X=P, 3; As, 4), chlo ide [DyCl2(Ph3AsO)4]Cl
(5), i la e [Dy(OT )2(MePh2PO)4]OT (6; OT = i la e) and hexa luo oace ylace ona e
[M(h a)3(Ph3PO)2] (h a =hexa luo oace ylace ona e; M=Tb, 7; Dy, 8) complexes
we e isola ed and ully cha ac e ized. The c ys al s uc u es o 1·CH3CN, 2·CH3CN, 4,
5·2.75E OH·1.25H2O, 6,7, and 8show MO9co es in 1, 2, and 4, wi h highly dis o ed
geome y, be ween sphe ical capped squa e an ip ism and mu in-like, hexacoo dina ed
en i onmen s o he dysp osium ions in 5and 6, wi h oc ahed al geome y, and oc a-
coo dina ion o he lan hanoid me als in 7and 8, wi h geome y close o squa e
an ip ism. Compa ison o he magne ic beha io o all he complexes allows analyzing
which me al ion (Tb o Dy), phosphine o a sine oxide, o anionic ligand a o mo e he
slow elaxa ion o he magne iza ion. Al e na ing cu en magne ic measu emen s show
ha only 2,4, and 8p esen slow elaxa ion o he magne iza ion in he p esence o
an ex e nal magne ic ield, 8being he complex wi h he highes Ue (44.85 K) o hose
desc ibed he ein.
Keywo ds: e bium, dysp osium, iphenyla sine oxide, phosphine oxide, single ion magne
INTRODUCTION
The obse a ion o he i s ime o slow magne ic elaxa ion in mononuclea lan hanoid
complexes (TBA)[Pc2Ln] (TBA =Bu
4N+; Pc =ph halocyanide; LnIII =Tb o Dy) (Ishikawa
e al., 2003) p o ided a eal b eak h ough in molecula magne ism, opening he ield o single
ion magne s (SIMs) in 2003. This ield has ecei ed g owing a en ion since i s o igins, gi en
ha he non- i ial memo y e ec and quan um cha ac e in SIMs ende s hem as po en ial
ul a-high densi y da a s o age medium and spin onic de ices (Wood u e al., 2013; Shiddiq
e al., 2016; Lu e al., 2017). Acco dingly, he huge amoun o wo k de o ed o his esea ch
du ing he las 15 yea s has led o many ema kable ad ances. Thus, Rineha and Long
published in 2011 (Rineha and Long, 2011) a benchma k s udy whe e hey p o ided a
clea explana ion o how he elec onic s uc u e o -elemen s can in heo y be manipula ed
o c ea e new single molecule magne s (SMMs). In his s udy, hey gi e he ela ionship
be ween he coo dina ion en i onmen o lan hanoid ions and he magne ic aniso opy o
he complex and, he e o e, hey sugges ha one can ma ch an app op ia e ligand ield o
maximize magne ic aniso opy on he basis o he shapes (obla e o p ola e) o he 4 -shell
elec on densi y dis ibu ions. Fo he obla e Tb(III) and Dy(III) ions, heo e ical calcula ions
Fondo e al. SMMs o Phosphine- and A sine-Oxides
es ablished ha he op imum en i onmen o maximize he
aniso opy is he axial one (Ungu and Chibo a u, 2011). Thus,
he op imum geome y will be lineal bu he coo dina ion
numbe wo seems oo low o be s abilized by he lan hanoid ions.
Acco dingly, in he absence o his possibili y, he coo dina ion
numbe 7, wi h pen agonal bipy amidal geome y, was he mos
explo ed one.
On he basis o hese s a ing heo e ical s udies, much
expe imen al wo k was done and con inuous a ainmen s
ela ing o aniso opic ene gy ba ie s (Ue ) and blocking
empe a u es (TB) we e achie ed in he ield o SIMs. Hence, he
highes Ue desc ibed up o now is 1815 K (Ding e al., 2016), a
alue signi ican ly highe han 331 K epo ed o he i s SIM
(Bu
4N)[Tb(Pc)2] (Ishikawa e al., 2003). Besides, he blocking
empe a u es ha e been con inuously inc easing, om he ini ial
1.7 K (Ishikawa e al., 2005) ho ough he 20 K achie ed in
2016 o an-ai s able [Dy(Cy3PO)2(H2O)5]B 3complex (Chen
e al., 2016) o he as onishing 60 K ecen ly epo ed o he
me allocene complex [(Cp )2Dy][B(C6F5)4] (Goodwin e al.,
2017; Guo e al., 2017).
In spi e o hese ad ances, i should be no ed ha mos o
he SIMs wi h high Ue and/o TBa e ai -uns able, and his is a
handicap ha mus be su passed. I mus be no ed ha a se ies o
simple, easy o ob ain, and ai s able phosphine oxide complexes
[Dy(R3PO)2(H2O)5]X3(R =Cy3o CyPh2, X =Cl, B o I)
show blocking empe a u es be ween 19 and 20K (Chen e al.,
2016, 2017), he la ges ones among SMMs i he 60K epo ed
o he uns able [(Cp )2Dy][B(C6F5)4] is excluded. Acco dingly,
i seems ha he coo dina ion chemis y o lan hanoids wi h
phosphine oxides is eally in e es ing om he magne ic poin o
iew. This coo dina ion chemis y has been ex ensi ely s udied,
mainly by Pla (2017), bu , in spi e o his, he magne ic beha io
o hese compounds is s ill poo ly in es iga ed. The e o e, aking
in o accoun all he abo e conside a ions, we ha e decided o
e ise and ex en he s udy o he coo dina ion chemis y o Dy
and Tb wi h phosphine and a sine oxides, and o in es iga e he
magne ic beha io o he ob ained compounds. The compa ison
o he magne ic p ope ies o he isola ed complexes be ween
hem allows es ablishing some pa e ns.
MATERIALS AND METHODS
Gene al
All chemical eagen s and sol en s we e pu chased om
comme cial sou ces and used as ecei ed wi hou u he
pu i ica ion. Elemen al analyses o C, H and N we e eco ded
on a Ca lo E ba EA 1108 analyze . In a ed spec a we e
pe o med in he ange 4000–500 cm−1on a Va ian 670 FT/IR
spec opho ome e in he ATR mode.
Syn heses o he Complexes
[Tb(NO3)3(Ph3PO)3] (1): To a solu ion o Tb(NO3)3·5H2O
(0.130g, 0.30 mmol) in ace oni ile (10 mL), iphenylphosphine
oxide, (0.167g, 0.60 mmol) and ace oni ile (10mL) we e added.
The mix u e was s i ed o 4 h a oom empe a u e and he
esul an colo less solu ion was le o slowly e apo a e, un il
single c ys als o 1·CH3CN p ecipi a ed. The single c ys als we e
il e ed, and hey lose he ace oni ile sol a e on d ying o yield
1. Yield (based on Ph3PO): 0.22g (93%). M.W.: 1179.80. Anal.
calcd. o C54H45TbN3O12P3: C 54.97, H 3.84, N 3.56%. Found:
C 54.89, H 3.59, N 3.95%. IR (ATR, ˜ν/cm−1): 1120, 1153 (P=O),
1305 (NO−
3).
The same p oduc is isola ed when Tb(NO3)3·5H2O and
Ph3PO a e mixed in 1:3 mola a io in ace oni ile.
[Dy(NO3)3(Ph3PO)3] (2) was ob ained in a simila way o
1: amoun s o Dy(NO3)3·6H2O (0.23 g, 0.66 mmol) and Ph3PO
(0.360g, 1.32 mmol). Single c ys als o 2·CH3CN we e isola ed in
he same way as hose o 1·CH3CN, which lose he ace oni ile
sol a e on d ying o yield 2. Yield (based on Ph3PO): 0.403g
(77%). M.W.: 1183.37. Anal. calcd. o C54H45DyN3O12P3: C
54.81, H 3.83, N 3.55%. Found: C 54.84, H 3.40, N 3.53%. IR
(ATR, ˜ν/cm−1): 1120, 1153 (P=O), 1305 (NO−
3).
The same p oduc is isola ed when Dy(NO3)3·6H2O and
Ph3PO a e mixed in 1:3 mola a io in ace oni ile.
[Dy(NO3)3(E OH)(Ph3PO)2] (3): A ho solu ion o
Dy(NO3)3·6H2O (0.137 g, 0.30 mmol) in e hanol (5 mL) was
added o a ho solu ion o iphenylphosphine oxide (0.170 g,
0.60 mmol) in e hanol (5 mL). The mix u e was s i ed a oom
empe a u e o 24 h, when a colo less powde p ecipi a ed. The
solid was il e ed and d ied in ai . Yield: 0.250 g (88%). M.W.:
951.15. Anal. calcd. o C38H36DyN3O12P2: C 47.98, H 3.81, N
4.42%. Found: C 47.90, H 3.81, N 4.29%. IR (ATR, ˜ν/cm−1):
1119, 1150 (P=O), 1308 (NO−
3), 3346 (OH).
[Dy(NO3)3(E OH)(Ph3AsO)2] (4) was ob ained in he same
way as 3: amoun s o Dy(NO3)3·6H2O (0.23 g, 0.66 mmol) and
Ph3AsO (0.425 g, 1.32 mmol). The mix u e yields a solu ion ha
by slow e apo a ion gi es ise o single c ys als o 4. Yield: 0.313g
(46%). M.W.: 1039.05. Anal. calcd. o C38H36DyN3O12As2: C
43.92, H 3.49, N 4.04%. Found: C 44.44, H 3.35, N 3.92%. IR
(ATR, ˜ν/cm−1): 902, 924 (As=O), 1313 (NO−
3), 3334 (OH).
[DyCl2(Ph3AsO)4]Cl·1.25H2O (5·1.25H2O): To an e hanol
(10 mL) solu ion o DyCl3·6H2O (0.460 g, 1.22 mmol), Ph3AsO
(0.77g, 2.395 mmol) was added. The mix u e was s i ed unde
e lux o 1 h and he colo less solu ion was concen a ed in a
o ae apo a o o ca. 2 mL, yielding an oil. A e s anding o
24 h, single c ys als o 5·2.75E OH·1.25H2O, sui able o X- ay
di ac ion s udies, we e isola ed. The c ys als we e il e ed and
d ied in ai , losing he e hanol sol a e o gi e ise o 5·1.25H2O.
Yield (based on Ph3AsO): 0.74 g (78%). M.W.: 1579.93. Anal.
calcd. o C72H62.5AsCl3DyO5.25: C 54.68, H 3.95%. Found: C
54.27, H 3.33%. IR (ATR, ˜ν/cm−1): 886 (As=O), 3264 (OH).
The same p oduc is isola ed when DyCl3·6H2O and Ph3AsO
a e mixed in 1:4 mola a io in e hanol.
[Dy(OT )2(MePh2PO)4](OT )·THF (6·THF): Dy(OT )3
(0.423g, 0.694 mmol) was dissol ed in THF (7 mL), and
me hyldiphenylphosphine oxide (0.600 g, 1.387 mmol) was
added. The mix u e was s i ed a oom empe a u e o 24 h,
and he esul an colo less solu ion was concen a ed in a
o ae apo a o up o ca. 2 mL, yielding an oil. The oil was le
o s and o 10 days a oom empe a u e un il single c ys als o
6·THFwe e ob ained. Yield: 0.250 (47%). M.W.: 1546.63. Anal.
calcd. o C59H60DyF9O14P4S3: C 45.78, H 3.88, S 6.21%. Found:
C 45.79, H 3.65, S 5.91%. IR (ATR, ˜ν/cm−1): 633 (δSO3), 1020,
1262 (SO3), 1156, 1220 (CF3), 1132 (P=O).
F on ie s in Chemis y | www. on ie sin.o g 2Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
The same p oduc is isola ed when Dy(OT )3and MePh2PO
a e mixed in 1:4 mola a io in THF.
[Tb(h a)3(Ph3PO)2] (7): Te bium chlo ide hexahyd a e
(0.14 g, 0.36 mmol) was dissol ed in dis illed wa e (5 mL).
An e hanol solu ion (20 mL) o hexa luo oace ylace one (0.22g,
1.08 mmol) was added o he aqueous solu ion. An e hanolic
e ame hylammonium hyd oxide solu ion (0.1 M) was added
d opwise un il pH 7 was eached. A e s i ing he mix u e
a oom empe a u e o 6 h, he sol en was elimina ed in a
o ae apo a o . The whi e solid ha p ecipi a ed was dissol ed
in me hanol (30 mL), and iphenylphosphine oxide (0.10 g,
0.36 mmol) was added. The mix u e was hea ed unde e lux
while s i ing o 6 h. Slow e apo a ion o he esul an solu ion
yields single c ys als o 7, sui able o X- ay di ac ion s udies.
The c ys als we e il e ed and d ied in ai . Yield (based
on Ph3PO): 0.070 (29%). M.W.: 1336.65. Anal. calcd. o
C51H33TbF18O8P2: C 45.83, H 2.49%. Found: C 45.79, H 2.42%.
IR (ATR, ˜ν/cm−1): 1138 (P=O), 1650 (C=O), 1250, 1160 (CF3).
The same p oduc is isola ed when TbCl3·6H2O, h ah and
Ph3PO a e mixed in 1:3:2 mola a io in wa e /e hanol.
[Dy(h a)3(Ph3PO)2] (8) was ob ained in a simila way o 7:
amoun s o DyCl3·6H2O (0.400 g, 1.07 mmol), Ph3PO (0.298,
1.07 mmol) and h ah (0.664 g, 3.21 mmol). Single c ys als o 8a e
isola ed in he same way as hose o 7. Yield 0.210 (29%). M.W.:
1340.22. Anal. calcd. o C51H33DyF18O8P2: C 45.70, H 2.48%.
Found: C 45.38, H 2.86%. IR (ATR, ˜ν/cm−1): 1136 (P=O), 1652
(C=O), 1250, 1161(CF3).
The same p oduc is isola ed when DyCl3·6H2O, h ah and
Ph3PO a e mixed in 1:3:2 mola a io in wa e /e hanol.
Magne ic Measu emen s
Magne ic suscep ibili y di ec cu en (dc) and al e na ing
cu en (ac) measu emen s o 1–8we e pe o med wi h a
Quan um Design SQUID MPMS-XL-5 suscep ome e . The
magne ic suscep ibili y dc da a we e eco ded a empe a u es
anging om 2 o 300 K, unde a magne ic ield o 1000
Oe. Magne iza ion measu emen s unde magne ic ields o 0–
50000 Oe a 2.0 K we e also eco ded. Diamagne ic co ec ions
we e es ima ed om Pascal’s Tables. ac suscep ibili y da a we e
egis e ed wi h an oscilla ing ac ield o 3.5 Oe and ac equency
o 1400 Hz, unde di e se applied s a ic ields (Hdc =0 o 1,000)
o all he epo ed compounds. In he case o 2and 8,ac
suscep ibili y measu emen s we e mo eo e eco ded unde a dc
ield o 1000 Oe a ac equencies in he ange 50–1400 Hz.
C ys allog aphic Re inemen and S uc u e
Solu ion
C ys al da a and de ails o e inemen a e gi en in Table S1.
Single c ys als o 1·CH3CN, 2·CH3CN, 4,5·2.75E OH·1.25H2O,
6·THF, 7and 8we e ob ained as de ailed abo e. Da a
we e collec ed a 100 K on a B uke Kappa APEXII CCD
di ac ome e , employing g aphi e monoch oma ed Mo-kα(λ
=0.71073 Å) adia ion. Mul i-scan abso p ion co ec ions we e
applied using SADABS (Blessing, 1995; K ause e al., 2015).
The s uc u es we e sol ed by s anda d di ec me hods, using
SHELXT (Sheld ick, 2015a), and hen e ined by ull-ma ix leas -
squa es echniques on F2, using he p og am package SHELXL
(Sheld ick, 2015b) om he p og am package SHELX (Sheld ick,
2008). All non-hyd ogen a oms we e e ined aniso opically, wi h
he excep ion o some a oms wi h low occupa ion si es. Mos
o he hyd ogen a oms we e included in he s uc u e ac o
calcula ions in geome ically idealized posi ions. Howe e , non-
diso de ed hyd ogen a oms, which could be po en ially in ol ed
in H-bonding schemes, we e mos ly loca ed in he co esponding
Fou ie maps. These H a oms we e eely e ined, o wi h he mal
pa ame e s de i ed om hei pa en a oms.
CCDC 1850933–1850938 and 1851725 con ain he
supplemen a y c ys allog aphic da a o his pape . These
da a can be ob ained ee o cha ge om The Camb idge
C ys allog aphic Da a Cen e ia www.ccdc.cam.ac.uk/
s uc u es.
RESULTS AND DISCUSSION
Syn hesis and Spec oscopic
Cha ac e iza ion
As p e iously s a ed, he chemis y o lan hanoids wi h
phosphine oxides is qui e well s udied and many complexes
wi h di e en me al ion:phosphine oxide a ios we e isola ed
(Pla , 2017). These p e ious s udies show ha he s oichiome y
o he complexes seems o be dependen on he na u e o he
me al sal employed, and on he mola quan i ies o sal and
phosphine oxide in he eac ion mix u e, bu also on he cone
angle o he phosphine (Tolman, 1977). Recen ly, i has been
epo ed ha complexes o o mulas [Ln(Cy3PO)2(H2O)5]X3
(Ln =La, Dy, E , Yb, Lu; X =Cl o B ) (Lees and Pla , 2014,
2015) can be easily ob ained. These complexes ha e pen agonal
bipy amidal geome y, and some o hem a e ai s able single
ion magne s, wi h high blocking empe a u es (Chen e al.,
2016). Ne e heless, simila pen agonal bipy amidal complexes
we e no epo ed o Ph3PO, wha would lead o e alua e
he in luence o he a oma ic s. he alipha ic ing in he
magne ic beha io o he complexes. Wi h hese conside a ions
in mind, di e en dysp osium and/o e bium sal s we e ini ially
mixed wi h Ph3XO (X =P o As) o MePh2PO in 1:2 mola
a io, wi h he in en ion o isola ing complexes o s oichiome y
Ln(R3XO)2Y5(R3=Ph3o MePh2; X =P o As; Y =OH2
and/o monoden a e anion).
The ob ained esul s indica e ha he sal :R3XO mola a io
plays a seconda y ole in he s oichiome y o he isola ed
complexes (Figu e 1), and ha in his case he ob aining
o compounds wi h coo dina ion numbe 7 and pen agonal
bipy amidal geome y was no possible. Thus, mixing o
Ln(NO3)2·6H2O (Ln =Dy o Tb) wi h Ph3XO (X =P o
As) in 1:2 mola a io leads o isola e [Ln(NO3)3(Ph3PO)3] o
[Dy(NO3)3(E OH)(Ph3XO)2] compounds, as a unc ion o he
sol en employed in he eac ion, as p e iously desc ibed o
Ph3PO (Cousins and Ha , 1967). Acco dingly, i also seems
ha he na u e o he X a om (P o As) in he Ph3XO dono
does no play any ole in he s oichiome y o he isola ed
complexes. Besides, i should be no ed ha he compounds
[Ln(NO3)3(Ph3PO)3]1and 2a e also ob ained when he sal s
and oxides a e mixed in 1:3 mola a io, and ha hese complexes
F on ie s in Chemis y | www. on ie sin.o g 3Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
FIGURE 1 | Reac ion scheme o isola ion o he me al complexes.
p esen he same s oichiome y as hose epo ed o Ln =La,
Nd, Eu, E , Tm, and Yb wi h Cy3PO (Hun e e al., 2007).
When he ni a e sal was changed by dysp osium chlo ide,
and Ph3AsO was added in 1:2 mola a io, he hyd a ed
complex [DyCl2(Ph3AsO)4]Cl·1.25H2O was ob ained
(Figu e 1). I s s oichiome y is he same desc ibed o
[DyCl2(Ph3PO)4]Cl·2H2O (Glazie e al. 2004), ob ained
by mixing o DyCl3and Ph3PO in 1:4 mola a io, bu clea ly
di e s om ha o [Dy(Cy3PO)2(H2O)5]Cl3(Lees and Pla ,
2014), which also has been ob ained by mixing o DyCl3and
Cy3PO in 1:4 mola a io in e hanol. Acco dingly, compa ison
o his expe imen wi h he eac ions wi h dysp osium ni a e
and wi h p e ious esul s (Glazie e al., 2004; Hun e e al.,
2007;Lees and Pla , 2014) clea ly show ha he cone angle o
he phosphine plays a undamen al ole in he s oichiome y
o phosphine o a sine oxide complexes, bu ha he simila i y
be ween he chemis y o Cy3PO and Ph3XO also depends on
he na u e o he anion o he sal employed.
Since h ee ni a e dono s o wo o he chlo ide anions a e
coo dina ed o he lan hanoid me al cen e in 1-4and ela ed
complexes (Glazie e al., 2004; Hun e e al., 2007; Bowden e al.,
2014), a sal o a non-coo dina ing anion, such as i la e, was
chosen, in o de o see he in luence o he coo dina ing abili y
o he anion o he sal in he inal s oichiome y, and in he hope
F on ie s in Chemis y | www. on ie sin.o g 4Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
o isola ing complexes o s oichiome y [Dy(R3XO)2(H2O)5]Y3.
In his way, mixing o dysp osium i la e wi h MePh2PO in 1:2
mola a io, and in a non-d ied non-coo dina ing sol en such as
THF, yields [Dy(OT )2(MePh2PO)4]OT , which also has a mola
a io me al:oxide o 1:4. Thus, his expe imen shows ha , in spi e
o he quan i y o wa e p esen in he non-d ied THF sol en ,
he dysp osium a om links o wo poo ly coo dina ing i la e
ions, p e en ing he coo dina ion o wa e o he me al cen e .
Acco dingly, his esul esembles in pa hose ob ained o Ce,
Nd and Lu i la e complexes o Ph3PO (Fawce e al., 2002;
Be he e al., 2003;), whe e he s oichiome y is he same, bu
o he Ce and Nd de i a i es he coo dina ion numbe is 7, wi h
one i la e ac ing as a biden a e chela e ligand and he o he one
as a monoden a e dono .
Finally, he eac i i y o Dy and Tb owa d Ph3PO, in 1:1
mola a io and in he p esence o hexa luo oace ylace one,
was s udied, gi en ha a [Eu( mh)3(RPh3PO)] ( mh =
2,2,6,6- e ame hylhep ane-3,5-dione, R =H, m-Me o
p-Me) complex, wi h coo dina ion numbe se en, was
p e iously epo ed (Yanagisawa e al., 2017). As Eu3+is
signi ican ly bigge han Tb3+and Dy3+, he isola ion o
Tb and Dy complexes wi h coo dina ion numbe 7 was
expec ed e en in he p esence o he less s e ically hinde ed
hexa luo oace ylace one dono . Ne e heless, his syn hesis
leads o complexes [M(h a)3(Ph3PO)2] (M =Tb, Dy) whe e he
me al:oxide a io is 1:2, as i occu s in he p e iously desc ibed
[Eu(h a)3(Ph3PO)2] and ela ed compounds (Hasegawa e al.,
2013), sugges ing ha o his kind o complex he size o he
lan hanoid ion has a poo in luence in he s oichiome y o he
isola ed compound.
As a summa y, compa ison o he esul s desc ibed he ein
wi h hose p e iously desc ibed o Cy3PO (Chen e al.,
2016, 2017) seems o indica e ha he lowe cone angle
o iphenylphosphine compa ed wi h icyclohexylphosphine
p e en s he isola ion o pen agonal bipy amidal complexes
o iphenylphosphine oxide wi h dysp osium o e bium. In
he same way, hexa luo oace ylace one allows coo dina ion
o wo phosphine oxide ligands, leading o coo dina ion
numbe o 8 o he lan hanoid a om while he mo e
s e ically hinde ed 2,2,6,6- e ame hylhep ane-3,5-dione only
allows linking one phosphine ligand, leading o coo dina ion
numbe 7 (Yanagisawa e al., 2017). The e o e, his s udy
clea ly shows ha no only he cone angle o he phosphine
oxides bu also he olume o he auxilia y ligands a e
undamen al ac o s in con olling he coo dina ion numbe
in his ype o complex, he sal :oxide me al a io, and
e en he size o he lan hanoid ion, playing seconda y
oles.
All he compounds desc ibed he ein we e ully cha ac e ized
by analy ical echniques, IR spec oscopy and by single c ys al
X- ay di ac ion s udies, excep 3. In addi ion, di ec and
al e na ing cu en magne ic measu emen s we e eco ded o all
he compounds.
The IR spec a o all he phosphine oxide complexes show
s ong bands in he ange 1119–1153 cm−1, which can be
assigned o he P=O s e ching ib a ion (Bowden e al., 2012).
Fo complexes 1,2, and 3 he e a e wo bands in his ange,
indica ing he exis ence o wo ypes o non-equi alen phosphine
oxide. Ne e heless, complexes 6-8p esen jus one band a ca.
1135 cm−1, which sugges ha all he phosphine oxide ligands
p esen in he compounds a e equi alen .
Complexes 4and 5·1.25H2O show s ong bands abou 900
cm−1, which can be assigned o As=O ib a ions (Le ason e al.,
2001). In 4 wo bands a e obse able, also sugges ing wo non-
equi alen a sine oxide ligands, while 5·1.25H2O only shows an
As=O band, in ag eemen wi h he equi alence o he ou a sine
oxide dono s p esen in he complex.
In addi ion, he syn he ized complexes also p esen
bands ela ed o he auxilia y ni a e, i la e o
hexa luo oace ylace ona e ligands. Thus, complexes 1-4
show a s ong band a ca. 1305 cm−1, assigned o he ν1(N-O)
ib a ion o he ni a e ligand, ac ing as biden a e (Bowden e al.,
2012). Complex 6show many bands (633, 1020, 1262, 1156,
and 1220 cm−1) ela ed o S-O o C-F ib a ions, which sugges
he p esence o i la e (Johns on and Shi e , 1993). Finally,
he IR spec a o 7and 8p esen s ong bands a 1652 cm−1,
assigned o he s e ching equency o he C=O g oup, and a
1250 and 1161, in ag eemen wi h ib a ions o he CF3and
C-H moie ies, all o which clea ly sugges he p esence o he
hexa luo oace ylace ona e ligand (Richa dson e al., 1968).
X-Ray Di ac ion S udies
[Tb(NO3)3(Ph3PO)3]·CH3CN (1·CH3CN) and
[Dy(NO3)3(Ph3PO)3]·CH3CN (2·CH3CN)
The c ys al s uc u es o 1·CH3CN and 2·CH3CN a e e y
simila and hey will be discussed oge he . The uni cell
o each complex con ains neu al [M(NO3)3(Ph3PO)3] (M =
Tb o Dy) molecules, in addi ion o ace oni ile as sol a e.
Ellipsoid diag ams o 1and 2a e shown in Figu e 2,Figu e S1,
espec i ely, and main dis ances and angles in Table S2.
Bo h complexes c ys allize in he P21/c g oup, wi h no
symme y elemen s ela ing he di e en ligands.
In he [M(NO3)3(Ph3PO)3] molecules, h ee phosphine oxide
monoden a e ligands and h ee biden a e chela e ni a e dono s
comple e he coo dina ion sphe e o he me al ions. Thus, in
bo h complexes, he lan hanoid ion is in an O9en i onmen .
Calcula ions o he deg ee o dis o ion o he LnO9co e
wi h espec o a pe ec nine- e ex polyhed on using he
SHAPE so wa e (Llunell e al., 2005, 2010; Ruiz-Ma ínez e al.,
2008) lead o shape measu emen s be ween sphe ical capped
squa e an ip ism and mu in-like, bu close o sphe ical capped
squa e an ip ism (Figu e 3,Table S3). In hese polyhed a, all he
dis ances and angles a e wi hin hei usual ange o lan hanoid
complexes wi h phosphine oxide and ni a e dono s (Bowden
e al., 2011), showing M-OPh3PO dis ances no ably sho e han
he M-Oni a e ones (Table S2).
[Dy(NO3)3(E OH)(Ph3AsO)2] (4)
An ellipsoid diag am o 4is shown in Figu e 4 and main bond
dis ances and angles in Table S4.
The c ys al s uc u e o 4 esembles ha o 2. Besides, i should
be no ed ha 4is isomo phous wi h he p e iously desc ibed and
c ys allog aphically cha ac e ized [La(NO3)3(E OH)(Ph3AsO)2]
complex (Le ason e al., 2001). Thus, h ee biden a e chela e
ni a e ligands, wo monoden a e Ph3AsO dono s and an e hanol
molecule ill he O9coo dina ion sphe e o he dysp osium
F on ie s in Chemis y | www. on ie sin.o g 5Sep embe 2018 | Volume 6 | A icle 420

Fondo e al. SMMs o Phosphine- and A sine-Oxides
FIGURE 2 | Ellipsoids diag am (50% p obabili y) o 1. Only he Tb, P, and dono O-a oms ha e been labeled, o cla i y. Colo code: Tb, ligh blue; C, g ay; H, ligh
g ay; N, da k blue; P, o ange; O, ed.
FIGURE 3 | Coo dina ion polyhed on o Dy in 2, illus a ing he dis o ed
sphe ical capped squa e an ip ism geome y shown by he me al ion in
1and 2.
ion. SHAPE calcula ions indica e ha he geome y abou he
me al ion is also be ween sphe ical capped squa e an ip ism
and mu in-like, as in 1and 2, bu in his case he mu in-like
disposi ion seems o be a bi mo e s able (Table S2). All he
dis ances and angles in his complex a e wi hin he usual ange,
bu i should be no ed ha he Dy-Oa sine oxide dis ances a e a
bi sho e han he co esponding Dy-Ophospine oxide ones in 2,
FIGURE 4 | Ellipsoids diag am (50% p obabili y) o 4. Only he Dy, As, and
dono O-a oms ha e been labeled, o cla i y. Colo code: Dy, ligh blue; As,
iole ; C, g ay; H, ligh g ay; N, da k blue; O, ed.
and his can be a consequence o he smalle elec onega i i y o
a senic wi h espec o phospho ous, and, he e o e, o he g ea e
dono s eng h o he a sine oxide, gi en ha he geome y o
bo h complexes is e y simila .
Finally, i is wo h o men ion ha his complex shows a
hyd ogen bond be ween one oxygen a om (O6) o one ni a e
F on ie s in Chemis y | www. on ie sin.o g 6Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
dono and he e hanol ligand o a neighbo ing uni . This
hyd ogen bond is ecip ocal and gi es ise o a pseudodime , wi h
a Dy···Dy dis ance o 7.1870(9) Å (Figu e 5).
[DyCl2(Ph3AsO)4]Cl·2.75CH3CH2OH·1.25H2O (5·2.75
CH3CH2OH·1.25H2O)
An ellipsoid diag am o [Dy(Ph3AsO)4(Cl)2]Cl is shown in
Figu e 6 and main dis ances and angles in Table S5.
The asymme ic uni o 5·2.75CH3CH2OH·1.25H2O con ains
a [DyCl2(Ph3AsO)4]Cl molecule, and e hanol and wa e as
sol a es. The [DyCl2(Ph3AsO)4]+ca ion shows he Dy(III) ion
FIGURE 5 | Hyd ogen bond o 4, showing he pseudodime .
FIGURE 6 | Ellipsoids diag am (50% p obabili y) o [DyCl2(Ph3AsO)4]+in 5.
Only he Dy, As, Cl and dono O-a oms ha e been labeled, o cla i y. Colo
code: Dy, ligh blue; As, iole ; C, g ay; Cl, g een; H, ligh g ay; O, ed.
joined o ou Ph3AsO ligands h ough hei O-a om, and o
wo chlo ide ions. Thus, he dysp osium a om is in an O4Cl2
en i onmen , wi h oc ahed al geome y. In his polyhed on, he
chlo ide ions occupy he axial si es, wha p oduces an elonga ion
o he oc ahed on.
Al hough he oc ahed al geome y is no e y common
o lan hanide ions, gi en ha hey end o achie e highe
coo dina ion numbe s, he s uc u e o [DyCl2(Ph3AsO)4]Cl
is e y simila o ha p e iously desc ibed o o he
[LnX2(R3PO)4]X (X =Cl o B ) compounds (Glazie e al., 2004;
Bowden e al., 2010; Pla , 2017), wi h dis ances and angles in he
usual ange, and ha do no me i u he conside a ion.
[Dy(OT )2(MePh2PO)4]OT ·THF (6·THF)
The uni cell o 6·THF con ains wo c ys allog aphically di e en
bu chemically equi alen [Dy(OT )2(MePh2PO)4]+ca ions,
which will be called 6A and 6B, espec i ely, and CF3SO−
3
coun e ions and THF as sol a e. Each [Dy(OT )2(MePh2PO)4]+
ca ion has an in e sion cen e , loca ed a he dysp osium a om.
Besides, one o hese [Dy(OT )2(MePh2PO)4]+ca ions, he
one con aining Dy2 (6B), has he coo dina ed i la e anions
diso de ed o e wo si es (80 and 20% occupancy).
An ellipsoid diag am o one o he [Dy(MePh2PO)4(OT )2]+
ions (6A) is shown in Figu e 7 and main bonds dis ances and
angles in Table S6.
In bo h [Dy(MePh2PO)4(OT )2]+ca ions, he DyIII ions
a e coo dina ed o ou phosphine oxides and o wo i la e
ions, which ac as monoden a e ligands. This gi es ise o an
O6oc ahed al en i onmen abou he dysp osium cen e . I
should be no ed ha h ee lan hanoid complexes o empi ical
o mula [Ln(OT )2(Ph3PO)4](OT )(Ln =Ce, Nd, Lu) ha e been
FIGURE 7 | Ellipsoids diag am (50% p obabili y) o [Dy(MePh2PO)4(OT )2]+
in 6A. Only he Dy, P, and dono O-a oms ha e been labeled, o cla i y. Colo
code: Dy, ligh blue; C, g ay; F, luo escen yellow; H, ligh g ay; O, ed; P,
o ange.
F on ie s in Chemis y | www. on ie sin.o g 7Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
p e iously epo ed, and in Ce and Nd complexes one o he
i la e ligands ac s as biden a e chela e and he o he one as
monoden a e (Fawce e al., 2002; Be he e al., 2003), leading
o coo dina ion numbe 7, while in 6bo h i la e dono s ac
as monoden a e. In ac , he s uc u e o 6is mo e simila
o ha o [Lu(OT )2(Ph3PO)4](OT ) (Fawce e al., 2002),
whe e he me al ion is also in an oc ahed al en i onmen . This
seems o be ela ed o he smalle size o DyIII compa ed wi h
CeIII and NdIII ha , in spi e o he less s e ically hinde ed
phosphine oxide in 6(MePh2PO) espec o he Ce and
Nd complexes (Ph3PO), canno alloca e se en a oms in i s
su oundings.
The sho es Dy-O dis ances in bo h 6A and 6B ca ions
co espond o Dy-Ophosphine oxide, he weakes bonds being he
Dy-O i la e ones, as expec ed in iew o he poo abili y o he
i la e anions o ac as ligands. Thus, he polyhed a a e axially
elonga ed. Ne e heless, i should be no ed ha while in 6A all
he Dy-Ophosphine oxide dis ances a e e y close, in 6B he s eng h
o hese bonds di e s signi ican ly (Dy2-O21 dis ance o ca. 2.19
Å and Dy2-O20 dis ance o ca. 2.24 Å) and he Dy-O i la e bond
is a bi s onge in 6B han in 6A, showing a less elonga ed
oc ahed on.
[Tb(h a)3(Ph3PO)2] (7) and [Dy(h a)3(Ph3PO)2] (8)
The c ys al s uc u es o bo h complexes a e e y simila , and
hey will be discussed oge he . The main di e ence be ween
hem is ha he luo ide a oms o one CF3moie y o one i la e
ligand a e diso de ed o e wo si es (56/44%) in 7, while his does
no occu in 8. Ellipsoids diag ams o 7and 8a e shown in
Figu e S2,Figu e 8, espec i ely, and main bond dis ances and
angles in Table S7.
FIGURE 8 | Ellipsoids diag am (50% p obabili y) o 8. Only he Dy, P, and
dono O-a oms ha e been labeled, o cla i y. Colo code: Dy, ligh blue; C,
g ay; F, luo escen yellow; H, ligh g ay; O, ed; P, o ange.
The s uc u e o bo h compounds esembles ha o
[Eu(h a)3(Ph3PO)2] (Hasegawa e al., 2013). Thus, in bo h
complexes he lan hanoid ion is coo dina ed o h ee biden a e
chela e hexa luo oace ylace ona e anions and o wo phosphine
oxides, gi ing ise o an O8en i onmen . Calcula ions o he
deg ee o dis o ion o he MO8en i onmen wi h espec o an
ideal eigh - e ex polyhed on wi h he SHAPE so wa e (Llunell
e al., 2005, 2010; Ruiz-Ma ínez e al., 2008), indica es ha he
geome y is close o squa e an ip ism in bo h cases. In hese
polyhed a, all he dis ances and angles a e in he expec ed ange
and do no me i u he conside a ion.
Magne ic P ope ies
The di ec cu en (dc) magne ic suscep ibili y s udies o all he
complexes (1-8) we e eco ded unde a magne ic ield o 1000
Oe in he empe a u e ange 2–300 K. The plo s o χMT s. T
o he complexes a e shown in Figu e S3. A oom empe a u e
(300 K), he χMTp oduc s o he e bium complexes 1and 7a e
in he ange 11.81–12.37 cm3kmol−1while o he dysp osium
complexes 2-6 and 8 a y om 14.05 o 14.93 cm3kmol−1. All
hese alues a e simila o he expec ed ones o one isola ed
TbIII (4 8,7F6) o DyIII (4 9,6H15/2) ion o 11.82 and 14.17
cm3kmol−1, espec i ely.
The expe imen al χMT alues emain basically cons an (1,
2), dec eases e y slowly (6-8) o dec ease a bi quickly (3-
5·1.25H2O) upon cooling up o 50 K, and mo e apidly below
his empe a u e in all cases. This beha io appea s o be mos ly
owing o he depopula ion o he MJsuble els o he lan hanoid
ion, and possibly o e y weak in e molecula in e ac ions, which
could be he cause o he mo e ma ked dec ease in χMTa low
empe a u e.
The a ia ion o he magne iza ion wi h he ield a 2 K o 1-8
(Figu e S3) a e akin in all cases, and shows a qui e as inc ease o
he M/NµB alue a low ield, wi hou a aining sa u a ion a 5 T.
The M/NµB alue a he la ges applied ield o 5 T (in he 5.07–
5.77 Ms/NµB ange) is ema kably lowe han he p edic able
one o isola ed TbIII o DyIII ions (Ms/NµB=gjJ), which can
be asc ibed o he c ys al- ield e ec s ha p omo e a signi ican
magne ic aniso opy.
Al e na ing cu en (ac) magne ic suscep ibili y
measu emen s we e also done o mic oc ys alline samples o
compounds 1-8, wi h he aim o knowing hei low- empe a u e
magne ic elaxa ion beha io . No one o he compounds show
ou -o -phase ac suscep ibili ies (χ′′) peaks a a equency o
1400Hz a ze o ield, indica ing he absence o SMM beha io .
Ne e heless, i should be no ed ha , some imes, applica ion o
an ex e nal s a ic ield (Ruiz e al., 2012), in o de o elimina e
he mixing o he g ound ±Ms le els, can pa ially o comple ely
supp ess he quan um unneling o he magne iza ion (QTM),
allowing o obse e he slow elaxa ion p ocess h ough he
eal he mally ac i a ed ene gy ba ie (U). The e o e, his
app oxima ion was es ed wi h he aim o checking he quan um
unneling e ec s.
Acco dingly, a iable empe a u e al e na ing cu en (ac)
suscep ibili ies we e measu ed o all he compounds a 1400Hz
and unde a small dc ield o 1000 Oe. In hese condi ions, he
χ” s. T cu e only shows a clea ne peak o 2,4, and 8
F on ie s in Chemis y | www. on ie sin.o g 8Sep embe 2018 | Volume 6 | A icle 420
Fondo e al. SMMs o Phosphine- and A sine-Oxides
(Figu e S4), bu in he case o 4 he peak appea s a 2.5 K a a
equency o 1400 Hz. Thus, in iew o hese esul s, gi en ha
he magne ic esponse o 4seems oo weak, a iable equency
and empe a u e ac suscep ibili ies we e measu ed only o 2
and 8wi h he applica ion o an expe imen al op imal dc ield
o 1000 Oe. As shown in Figu e 9, in hese condi ions 2and
8show equency and empe a u e-dependence o he ou -o -
phase suscep ibili y, wi h maxima o χ” below 5 and 5.8 K o
2and 8, espec i ely.
This indica es ha bo h complexes (2and 8) a e ield-
induced single ion magne s. In addi ion, i should be no iced
ha Figu e 9 e eals ha χ” goes o ze o below he maxima
FIGURE 9 | F equency dependence o ou -o -phase componen o he ac suscep ibili y (χ”) unde a dc applied ield o 1000 Oe a di e en empe a u es o 2 (A) and
8 (C); and A henius plo o 2 (B) and 8 (D). The solid lines ep esen he bes i o simple A henius o O bach-Raman equa ion.
FIGURE 10 | Cole–Cole plo in 1000 Oe dc ield o 2 (A) and 8 (B).
F on ie s in Chemis y | www. on ie sin.o g 9Sep embe 2018 | Volume 6 | A icle 420