Supplementary Figure 1. Ellipsoid plot of [Cu2(m-xpt)2(NO3)2](PF6)2 (1). Cu···Cu =

6.843(2) Å.

Supplementary Figure 2. Ellipsoid plot of [Cu2(m-xpt)2Cl2](PF6)2 (2). Cu1···Cu2 =

7.615(2) Å.

Supplementary Figure 3. Electrochemical Studies. Cyclic voltammograms of [Cu2(m-

xpt)2(NO3)2](PF6)2 (1, black); and [Cu2(m-xpt)2Cl2](PF6)2, (2, blue) in DMF containing 0.1

M Bu4NPF6 recorded on a static glassy carbon disc working electrode with a Pt wire

auxiliary electrode and Ag/AgCl reference electrode at 25 °C at a scan rate of 50 mV s–

1.

Supplementary Figure 4. Reduction of Cu(II) dimer to Cu(I), showing the

disappearance of Cu(II). A portion of the electronic spectrum of a 3.76 mM solution of

[Cu2(m-xpt)2(NO3)2(PF6)2, 1, in DMF (blue), with increasing amounts of added sodium

ascorbate: 0.25 eq (black) to 1.25 eq (red).

Supplementary Figure 5. Reduction of Cu(II) dimer to Cu(I), showing the

appearance of Cu(I). A portion of the electronic spectrum of a 0.20 mM solution of

[Cu2(m-xpt)2(NO3)2(PF6)2, 1, in DMF (blue), with increasing amounts of added sodium

ascorbate: 0.25 eq (black) to 1.25 eq (red). λmax = 384 nm.

Supplementary Figure 6. Reaction of Cu(I) dimer with CO2. A portion of the

electronic spectrum of a 0.12 mM solution of [Cu2(m-xpt)2](PF6)2, 3, in DMF showing the

disappearance of Cu(I) by its reaction with CO2. Compare Figure 3, which shows the

appearance of Cu(II) during the same reaction.

(a) (b) Supplementary Figure 7. IR spectra: (a) [Cu2(m-xpt)2(μ-C2O4)](PF6)2 (blue) and

[Cu2(m-xpt)2(μ-13C2O4)](PF6)2 (red), and (b) difference spectrum (13C labeled minus

unlabeled) showing ν13CO = 1651 cm−1.

Supplementary Figure 8. Crystals of compound 4. Crystals of the oxalate-bridged

compound [Cu2(m-xpt)2(μ-C2O4)](PF6)2, 4, synthesized starting from: (a) [Cu2(m-

xpt)2(NO3)2](PF6)2, 1 (crystal size ca. 0.3 mm); and (b) [Cu2(m-xpt)2Cl2](PF6)2, 2. (The

blue-green crystals in sample (b) were found to be of the starting complex, 2.)

Supplementary Table 1. Electrochemical parameters. The potential of the ferrocene

(Fc/Fc+) reference redox couple under these conditions was 0.56 V vs. Ag/AgCl. The

peak separation ΔEp for the Fc/Fc+ couple ranged from 60-65 mV (10, 25 mV s−1) to 75

mV (100 mV s−1) under the same conditions. Thus, the CuII/CuI waves for 1 and 2 (see

also Supplementary Fig. 3) are quasireversible, approaching reversible behavior at low

scan rates; and the redox process is more readily reversible for 2 than for 1.

Complex Scan Rate (mV s−1)

Epa Epc ΔEp E1/2 (vs. Ag/AgCl)

E1/2 (vs. Fc/Fc+)

1 10 0.33 0.24 0.09 0.29 −0.27

1 25 0.34 0.23 0.11 0.29 −0.27

1 50 0.38 0.20 0.18 0.29 −0.27

1 100 0.40 0.18 0.22 0.29 −0.27

2 10 0.32 0.24 0.08 0.28 −0.28

2 25 0.33 0.24 0.09 0.28 −0.28

2 50 0.33 0.23 0.10 0.28 −0.28

2 100 0.35 0.22 0.13 0.29 −0.27

Supplementary Table 2. Crystal Data and Structure Refinement Parameters.

Compound 1 2

deposition no. CCDC 1000457 CCDC 1000458

formula [Cu2(m-xpt)2(NO3)2](PF6)2 ·3.5CH3CN

[Cu2(m-xpt)2Cl2](PF6)2

·4.44C3H7NO

M 1329.93 1601.17

crystal system Monoclinic Orthorhombic

space group P21/c Pna21

a/Å 11.4090(16) 13.810(2)

b/Å 12.8455(18) 19.380(3)

c/Å 23.411(3) 24.711(3)

/deg 108.575(5) 90

V/Å3 3252.3(8) 6613.6(16)

Z 2 4

T/K 100.0(5) 90.0(5)

Dcalc/g cm−3 1.358 1.608

crystal dimensions/mm

0.35 x 0.17 x 0.10 0.04 x 0.14 x 0.31

Radiation CuK MoK

θ limits/deg 3.98 – 59.20 1.65 – 25.71

reflns, measd /unique/obsd

22792/4639/3982 52940/12074/7203

F(000) 1340 3278

μ/mm−1 2.066 0.871

Rint 0.0468 0.0525

R[I>2σ(I)] 0.1013 0.0830

Rw (all data) 0.3540 0.2604

GOF 1.624 1.053

Supplementary Table 2 (continued).

Compound 4 4a 5

deposition no. CCDC 984468 CCDC 984469 CCDC 984470

formula [C46H36Cu2N16O4](PF6)2 ·2C3H7NO

[C46H36Cu2N16O4](PF6)2

·4CH3CN [C44H40Cu2N16O2](NO3)4

·2C3H7NO·3.34H2O

M 1440.12 1458.14 1406.50

crystal system

Orthorhombic Orthorhombic Monoclinic

space group Cmca Cmca P21/n

a/Å 24.1471(17) 24.461(7) 11.4361(13)

b/Å 11.7107(7) 11.959(3) 22.691(3)

c/Å 20.5949(12) 20.008(5) 11.9848(13)

/deg 90 90 106.847(5)

V/Å3 5823.8(6) 5853(3) 2976.6(6)

Z 4 4 2

T/K 100.0(5) 100.0(5) 100.0(5)

Dcalc/g cm−3 1.642 1.655 1.569

crystal dimen-sions/mm

0.45x0.17x0.04 0.30x0.22x0.13 0.12x0.19x0.21

radiation MoK MoK CuK

θ limits/deg 1.69-30.05 2.04-30.23 4.32-69.67

reflns, measd /unique/obsd

44218/4357/3592 31270/4418/3607 52513/14803/12358

F(000) 2928 2960 1455

μ/mm−1 0.891 0.886 1.691

Rint 0.0310 0.0447 0.0741

R[I>2σ(I)] 0.0527 0.0346 0.0637

Rw (all data) 0.1531 0.0900 0.1928

GOF 1.033 1.031 1.024

Supplementary Methods

X-ray Crystallography. Intensity data were collected at low temperature on a Bruker

Kappa Apex-II DUO CCD diffractometer fitted with an Oxford Cryostream chiller. MoK

( = 0.71073 Å) radiation with a TRIUMPH curved graphite monochromator was used

for 2, 4, and 4a, and CuK ( = 1.54184 Å) radiation from an I S microfocus source

with QUAZAR multilayer optics was used for 1 and 5. Data reduction included

absorption corrections by the multiscan method, with SADABS,1 or TWINABS1 for 5.

The structures were determined by direct methods and difference Fourier techniques

and refined by full-matrix least squares, using SHELXL-97.2 All non-hydrogen atoms

were refined anisotropically except for the minor component in a disordered nitrate in 1

and 5, C atoms in 2, and the minor component of the disordered PF6− anion in 1 (vide

infra). For 2, low data quality did not allow anisotropic refinement of the C atoms.

Disordered solvents were removed using the SQUEEZE3 procedure for 2, resulting in

non-integral solvent stoichiometry (see the CIF file for details). Normal refinement

procedures for 2 led to several unreasonable distances within the m-xpt ligands.

Therefore, for the final refinement, the two sets of ligand atoms were restrained to yield

similar bond distances and angles. For 4 and 4a, the PF6− anion is disordered into two

orientations, and a single solvent molecule (DMF for the crystals prepared by reaction of

Cu(I) (3) with CO2; CH3CN for those prepared by reaction of Cu(II) (1 or 2) with oxalate)

is disordered across a mirror plane. The crystal of 5 was a nonmerohedral twin by 180º

rotation about real axis [−1 0 1]. Refinement was vs. HKLF5 data, and twin components

were present approximately 52/48%. This structure has a disorder involving a nitrate ion

in two positions and a partially occupied water molecule associated with one of them.

All H atoms were placed in idealized positions, except for water hydrogen atoms. Water

H atoms were refined with restrained O-H distances, but the H atoms of the water

molecule in the nitrate/water disorder of 5 could not be located. Crystal data and

structure refinement parameters are presented in Supplementary Table 2.

Supplementary References:

1 SADABS (University of Göttingen, Germany). 2 Sheldrick, G. M. A short history of SHELX. Acta Crystallogr A 64, 112-122

(2008). 3 Spek, A. L. Structure validation in chemical crystallography. Acta Crystallogr.,

Sect. D 65, 148-155 (2009).