Unusual fluorides of silverat high oxidation states

LECTURE VII

Frequency occurrence of different oxidation states

in the compounds of the Group 11 elements

–1 0 1 2 3 4 5 6 7 Cu Cu CuCl CuF2 KCuF4 Cs2CuIVF6 Ag Ag AgF AgF2 AgF3 Cs4AgIIIAgVF12 Au CsAu Au AuCl Au(SbF6)2 CsAuCl4 AuF5 (AuF6) (AuF7)

CuCl frequent AgF2 less frequent AuCl rare AuF5 very rare

Why fluorides of Ag(2+) i Ag(3+)?

Relativistic effects for Au(0)

5d nonrelat

6s relat

5d relat10

16s nonrelat

1. Yellow Au(0) and compounds of Au(1+);

2. Increased acidity of – supposedly soft – Au(1+);

3. Existence of Au–1 (CsAu);4. The M–H bond length is some 0.2 Å

shorter in AuH than in AgH;5. The highest oxidation state of Au is

most probably (7+);6. Au(2+) has significant tendency for

disproportionation.

Ag2+ ~ Cu2+ (d9)Ag3+ ~ Cu3+ (d8)

F1– ~ O2– (s2p6)

1. High–TC superonductivity in the hole–doped oxides of Cu2+;

2. Our previous theoretical predictions on large vibronic coupling in the systems built of hard acids and bases

3. Possibility of the „magic electronic state“ in the systems exhibiting avoided crossing between the „neutral“ and „ionic“ states

Why not Au(2+)?

1. Ag2+ is a very strong oxidizer. A) Ag2+ solvated in anhydrous HF oxidizes Xe0 to Xe2+, C6F6 to C6F6

+, and oxidizes CF3CF=CF2 quantitatively to CF3CF2CF3.

B) AgF2 oxidizes fullerene to C60F44 (AgF up to C60F18).C) Ag2+ is unknown in oxides and chlorides.

2. Ag3+ is an enormously strong oxidizer. A) Compounds of Ag3+ are best obtained by use of F radicals.

B) AgF2+ solvated in anhydrous HF oxidizes Kr0 to Kr2+, PtF6

– to PtF6, and together with Ni4+ is the best oxidizer available to chemistry.

C) Compounds of Ag3+ easily evolve F2 upon heating. AgF3 is thermodynamically unstable.

Enormous oxidizing properties of Ag(2+) i Ag(3+)

Cu3+ + O2– Cu2+ + O1– 2 O1– O2

2–

Ag3+ + F1– Ag2+ + F0 2 F0 F2

Ag3+ + F1– Ag2+ + F0

• energetic and spatial proximity of the Ag(4d) and F(2p) orbitals

• significant covalency of the Ag–F bonds !

• electronegativity of Ag3+ is close to that of F itself !

Cu Ag

O2

F

1+

2+

3+

1+

2+

3+

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 181 1

H2

He2 3

Li4

Be5B

6C

7N

8O

9F

10Ne

3 11Na

12Mg

13Al

14Si

15P

16S

17Cl

18Ar

4 19K

20Ca

21Sc

22Ti

23V

24Cr

25Mn

26Fe

27Co

28Ni

29Cu

30Zn

31Ga

32Ge

33As

34Se

35Br

36Kr

5 37Rb

38Sr

39Y

40Zr

41Nb

42Mo

43Tc

44Ru

45Rh

46Pd

47Ag

48Cd

49In

50Sn

51Sb

52Te

53I

54Xe

6 55Cs

56Ba

* 71Lu

72Hf

73Ta

74W

75Re

76Os

77Ir

78Pt

79Au

80Hg

81Tl

82Pb

83Bi

84Po

85At

86Rn

7 87Fr

88Ra

** 103Lr

104Rf

105Db

106Sg

107Bh

108Hs

109Mt

110Uun

111Uuu

112Uub

113Uut

114Uuq

115Uup

116Uuh

117Uus

118Uuo

* 57La

58Ce

59Pr

60Nd

61Pm

62Sm

63Eu

64Gd

65Tb

66Dy

67Ho

68Er

69Tm

70Yb

** 89Ac

90Th

91Pa

92U

93Np

94Pu

95Am

96Cm

97Bk

98Cf

99Es

100Fm

101Md

102Nb

The Cu/O vs Ag/F analogy

Binary fluorides of Ag

• AgF – photography colorless, NaCl

structure

• AgF3 – potent oxidizer brown red, unique helical AuF3

structure

• AgF2 – organic synthesis brown

ferromagnetic, monoclinic CuF2

• Ag2F – superconductivity green, inverse Cd(OH)2

structure

• Ag1+[Ag3+F4] – metastable • [Ag2+][AgF4

–]2 – red–brown unique ribbon

structure

• [AgF+][AgF4–] –

brown kinked 1D AgF+ chains

• AgF2–x – ???

• AgF1–x – defected structure yellow to

yellow–brown

Ternary & higher fluorides of Ag

• Ag(I):

• Ag(II):

• Ag(III)

:

2 Ag2C2 x AgF x 9 AgNO3 x H2O

Isolated Ag2+ centers(i) [Ag2+][AuF4

–]2; (ii) [Ag2+][MF6–]2, M=Bi, Sb, Ru,

Nb, Ta; (iii) [Ag2+][MF6–], M=Ge, Sn, Pb, Ti, Zr, Hf,

Rh, Pd, Pt, Mn, Cr; (iv) Ag3M2F14 & K3Ag2M4F23 M=Zr, Hf; (v) NaAgZr2F11.Infinite [AgF+] chains (straight or kinked)

(i) [AgF+][MF4–], M=Au, B; (ii) [AgF+][MF6

–], M=Bi, Sb, As, Au, Ir, Ru; (iii) [AgF+]2[AgF4

–][MF6–], M=Au,

Pt, Ru, Sb, As; (iv) [AgF+][M3M’3F19–], M=Cd, Ca, Hg,

M’=Zr, Hf; (v) MAgM’F6, M=Cs, Rb, K, M’= Al, Ga, In, Tl, Sc, Fe, Co.Infinite [AgF2] planes

(i) MAgF3, M=Cs, Rb, K; (ii) M2AgF4, M=Cs, Rb, K, Na.

Infinite [AgF3–] chains

NaAgF3

Isolated [AgF42–] squares

(i) MAgF4, M=Ba, Sr, Ca, Cd, Hg; (ii) Ba2AgF6.

Isolated LS [AgF4–] squares, or HS octahedron

(i) MAgF4, M=Cs, Rb, K, Na, Li, O2+, XeF5

+; (ii) Cs2KAgF6.

Crystal structures

Isolated Ag2+ centers: [Ag2+][SbF6

–]2

Puckered [AgF2] planes: AgF2

Infinite [AgF+] chains: [AgF+][BF4

–]Infinite [AgF2] planes:

[KF][AgF2]

Unique [AgF3] helix: AgF3

Isolated [AgF4–]

squares: KAgF4

Ag(3+): very short 1.89

Coordination environment of Ag(2+):

Ag(1+): very long 2.47

Electronic structure of several fluorides of Ag

–0.28 eV

EF

–0.72 eV

–2.02 eV

12%

Ag(4d)

34%

60%

contribution to the “ligand band”

Hypothesis

• DFT computations confirm that the Ag(4d) and F(2p) orbitals exhibit significant energetic and spatial proximity, and they strongly mix with each other in higher fluorides of Ag

• the Ag–F bonds are indeed significantly covalent in these compounds

• highly untypical situation takes place in the fluorides of Ag3+: more 4d states go to the “ligand band” than to the “metal band” (avoided crossing) !

Conclusions from Theory

• properly hole– or electron–doped fluorides of Ag2+ may be high– temperature superconductors, similar to their copper–oxide analogues, if quasi–2D structure is provided, and if defect localization is avoided

• self–doped fluorides of Ag2+ may exhibit metallic conductivity

Encouragement

“You may be surprised to learn that I have been looking for a superconductor in the Ag/F system for the past 8 years because of observations that we made in 1992. Briefly, we noted that whenever we prepared a [AgF]+[MF6] salt and washed it with anhydrous HF, the magnetic susceptibility exhibited a sharp drop at 63 K, suggestive of a superconducting transition caused by an impurity. Since this anomaly (it looks like a Meissner effect) was independent of M = Sb, As, Au, I assumed that the impurity was a mixed oxidation-state AgII/AgIII fluoride. The material that exhibits the 63 K anomally, does not produce identifying lines in the X-ray diffraction pattern (the parent materials give sharp strong patterns). My surmise has therefore been that the quantity present is small (< 5%). This surmise is obviously not valid if the material is non-crystalline. This set in train a set of investigations (...). My first and still favoured guess was that the 63 K diamagnetic phenomenon was caused by an electron-oxidized AgF2 sheet-structure [ i.e. [AgF2]n+, n<1] intercalated (perhaps non-stoichiometrically) by [AgF4] species. I also allowed that [MF6] could be an intercalating species. It is my belief that some disorder in the placement of the anionic charges is necessary, if hole localization is to be avoided. (...) It was this set of thoughts that caused me to look at the oxidation of AgF2 with [O2]+ salts, unfortunately we only obtained the linearly coordinated [AgF]2[MF6][AgF4] salts. The [AgF]2[MF6][AgF4] salts do not show the anomaly until they are washed with anhydrous HF (i.e. solvolysed). We never obtained an intercalated sheet structure, like that of Au[AuF4]2Au[SbF6]2. It could be that an off-stoichiometry silver relative of the latter is the desired material.”

When we showed him a draft of this paper, Prof. Bartlett described further his experimental search for superconductivity in Ag/F compound in a private communiation

to us (August 2000):

Experiments

Birmingham /UK/

Synthesissolid state & AHF/F2

Magnetic susceptibility measurements /SQUID/

Leicester /UK/Ljubljana /Slovenia/

XRDP 19F NMR Microwave cavity perturbationXPS

Feedback for synthesis

ESR

ICP MASS

Core XPS

Table 1. Position of the Ag(3d5/2) and F(1s) peaks (eV) (referenced to the C(1s) signal) in

the XPS spectra of several fluorides of Ag.

Compound AgF AgF2 c–AgF2 KAgF3 KAgF4

Ag(3d5/2) 368.20 367.85 368.95 368.70 368.35

F(1s) 683.35 683.35 683.49

685.79

683.47 686.25

0

1000

2000

3000

4000

5000

6000

1 2 3 4 5 6 7 8 9 10 11

E /eV

Inte

nsi

ty /c

ou

nts

KAgF3AgF2KAgF4AgF

Valence region XPS

Table 2. Parameters of the component bands in the valence region of the XPS

spectra for several fluorides of Ag: position of the band (referenced to the

C(1s) signal) (eV), its half width (eV), and intensity at maximum (counts).

These are results of the peak–fitting procedure using Gaussian peak profiles.

Band 2

(“ligand band”)

Band 1

(“metal band”)

Compound Position

/eV/

½ width

/eV/

Intensity

/counts/

Position

/eV/

½ width

/eV/

Intensity

/counts/

AgF 7.77 1.58 774 5.26 1.45 3808

c–AgF2 7.29 2.52 1087 5.57 1.98 3315

AgF2 6.97 3.46 1258 4.90 1.82 3287

KAgF3 6.85 3.12 1422 5.13 1.86 2916

KAgF4 7.27 3.42 1665 5.21 1.62 2389

Table 3. Ratio of the contributions of the Ag(4d) states to the „metal

band” and to the „ligand band” for several fluorides of Ag.

Comparison of experimental and theoretical results.

Ratio of contributions of Ag(4d) to Metal band : ligand band

Compound Theory /DFT/ Experiment /XPS/

AgF 88:12 82:18

c–AgF2 66:34 70:30

AgF2 66:34 58:42

KAgF3 53:47 51:49

KAgF4 40:60 40:60

Theory vs experiment

Microwave cavity perturbation

-6.E+06

-4.E+06

-2.E+06

0.E+00

2.E+06

4.E+06

6.E+06

20 30 40 50 60 70 80 90

-2.86E+06

-2.84E+06

-2.82E+06

-2.80E+06

-2.78E+06

-2.76E+06

0 50 100 150 200 250

AgF2

FM insulating

PM insulating

SQUID

0.E+00

2.E-07

4.E-07

6.E-07

8.E-07

1.E-06

050100150200

T /K

ZFC1.34E-06

1.35E-06

1.36E-06

1.37E-06

1030507090

T /K

ZFCFC

2.12E-06

2.16E-06

2.20E-06

2.24E-06

050100

T /K

2.20E-06

2.21E-06

2.22E-06

2.23E-06

2.24E-06

FCZFC

0.0E+00

1.0E-09

2.0E-09

3.0E-09

4.0E-09

5.0E-09

010203040

T /K

ZFC

a b

c d

Magnetic susceptibility / arbitrary units

T /K T /K

T /K T /K

0

FM

SC

ESR- zero field signal

- g2 signal

“BeAgF4”

XRDP

20 40 60 80

5000

10000

15000

20000

*** **

*

Cou

nts

2

†

#

† AgF2 ‡ AgF # - i -BeF2

* ???

†

†

† † †

† †

† † ‡

‡ ‡ ‡ ‡

*

#

#

#

#

#

# #

# * * *

# #

*

Conclusions

What next?

• identification of superconducting phase and synthesis of a pure compound + repeated XRDP and magnetic susceptibility measurements

• electrical resistivity contact or non–contact measurements ?

• high–pressure attempts to metallize fluorides of Ag2+, and mixed–valence fluorides of Ag2+/Ag3+

• Epitaxial growth of AgF2, molecular spacers, etc. …???

• observations of sudden drops in the magnetic susceptibility of a large number of samples in the BeAgF system

• possible superconductivity – and the attendant Meissner–Ochsenfeld effect – at temperatures ranging from 8.5 K to 64 K

• composition and structure of the phase(s) responsible for magnetic anomalies is unknown

• Li[AgF3–] … [BeF2][AgF2] … [AgF+][BF4

–] … ? Surface of AgF2 has been modified?

• KAgF3 is metallic above 70 K

T /K

0

20

40

60

80

100

120

140

160

180

200

220

Low Temp. SC

Medium Temp.SC

High Temp. SC

Ambient Temp.SC

Absolute Zero

Nb3Geclassical SC alloy

23 K

Liquid nitrogen77K

Lowest recordedtemp. on Earth

185 K

Acetone/CO2

cooling mixture195 K

LixHfNCl25.5 K

HgBa2Ca2Cu3O8+x

(under pressure)166 K

HgBa2Ca2Cu3O8+x

134 K

YBa2Cu3O7 x

93 K

Ba1 xKxBiO3

35 K

Rb2CsC60

35 K

h+ doped C60

52 K

YPd2B2C23 K

[Be/Ag/F]64 K

MgB2

39 K

h+ doped C60: CHBr3

117 K

Typical room temperature (25 oC)

Room Temp. SC

Literature

Review & theory:

Grochala W, Hoffmann R, Real and Hypothetical Intermediate-Valence Fluoride AgII/AgIII and AgII/AgI Systems as Potential Superconductors, ANGEW CHEM INT ED ENGL 40 (15): 2743-2781 2001

Experiment:

Grochala W, Edwards PP, Meissner–Ochsenfeld Superconducting Anomalies in the Be–Ag–F System, submitted to ANGEW CHEM INT ED ENGL

Acknowledgements

• The Cornell Theory Center (USA) /computational grant/

• The Daresbury Lab (UK) /experimental time at SCIENTA/

•The Cornell Center for Materials Research (USA) /DMR-

9632275/• The NSF (USA) /CHE 99-70089/

• The Royal Society (UK) /Postdoctoral and Research Fellowships/• The Crescendum Est–Polonia Foundation (Poland) /Research Stipend/

• Prof. Roald Hoffmann /Cornell, USA/• Prof. Peter P. Edwards /Birmingham, UK/• Prof. Neil Bartlett /Berkeley, USA/• Prof. Evgenii Antipov /Moscow, Russia/• Prof. Eric G. Hope & Prof. John Holloway /Leicester, UK/• Prof. Boris Žemva & Dr. Zoran Mazej /Ljubljana, Slovenia/• Prof. Russ G. Edgell /Oxford, UK/• Dr. Simon Kitchin /Birmingham, UK/

• Dr. Adrian Porch /Cardiff, UK/• Dr. Peter Kroll /Cornell, USA/• Prof. Kevin Smith /Boston, USA/• Prof. Andrew Harrison & Dr. Konstantin Kamenev /Edinburgh, UK/• Prof. David Jefferson /Cambridge, UK/• Prof. Miguel Moreno /Santander, Spain/• Prof. Berndt G. Mueller /Giessen , Germany/

• My wife