Methods
General methods and materials
All manipulations were performed under a N 2 atmosphere in a Vacuum
Atmospheres glovebox or under a N 2 or Ar atmosphere using stand-
ard Schlenk techniques. The solvent 1,2-difluorobenzene (DFB) was
deoxygenated by purging with argon for 1 h and dried using a com-
mercial solvent purification system designed by JC Meyer Solvent
Systems. The solvents 1,2-dimethoxyethane (DME) and bis(2-meth-
oxyethyl) ether (diglyme) were purchased from Sigma-Aldrich, dried
over Na/benzophenone (DME) or 4 Å molecular sieves (diglyme), and
degassed via three successive freeze-pump-thaw cycles. The compound
2,2′-bipyridine-5,5′-dicarboxylic acid (H 2 bpydc) was synthesized using
a previously published procedure^31. The compounds ZrCl 4 , Ni(DME)Br 2 ,
Ni(DME)Cl 2 , NiBr 2 , CoCl 2 and FeCl 2 were purchased from commercial
vendors (Sigma-Aldrich for ZrCl 4 , Ni(DME)Br 2 and Ni(DME)Cl 2 ; Strem
for NiBr 2 , CoCl 2 and FeCl 2 ) and used as received. All other chemicals
were purchased from commercial vendors and used as received unless
otherwise noted. Inductively coupled plasma optical emission spec-
trometry (ICP-OES) analysis was performed on a Perkin Elmer Optima
7000 DV instrument at the University of California, Berkeley, micro-
analytical facility. UV-Vis diffuse reflectance spectra were collected
using a CARY 5000 spectrophotometer interfaced with Varian Win
UV software. The samples were contained in a Praying Mantis air-free
diffuse reflectance cell and dispersed in non-absorbing BaSO 4 matrix.
The Kubelka–Munk conversion (F(R) versus wavenumber) of the raw
diffuse reflectance spectrum (R versus wavenumber) was obtained by
applying the formula F(R) = (1 − R)^2 /2R.
Synthesis of Zr 6 O 4 (OH) 4 (bpydc) 6 (1)
This material was synthesized as a microcrystalline powder using a
previously published procedure^17. Typically, a 2 l round bottom flask
equipped with a Schlenk adaptor, glass stoppers and a magnetic
stir bar was charged with H 2 bpydc (6.11 g, 25.0 mmol), benzoic acid
(224 g, 2.00 mol), and N,N-dimethylformamide (DMF; 1.00 l) from a
newly opened bottle. The resulting mixture was purged with dry Ar
for 30 min. Solid ZrCl 4 (5.83 g, 25.0 mmol) was then added, after which
the mixture was purged with dry Ar for an additional 30 min. Deionized
water (820 μl, 45.5 mmol) was added and the mixture was heated with
magnetic stirring for 5 days at 120 °C under a N 2 atmosphere. After
allowing the mixture to cool to room temperature, the solvent was
decanted and the resulting white microcrystalline powder was washed
by soaking three times in 1 l aliquots of fresh DMF for 24 h at 120 °C,
followed by solvent exchange with tetrahydrofuran (THF) via Soxhlet
extraction for 3 days. The THF-solvated powder was filtered under dry
Ar, followed by heating at 120 °C under dynamic vacuum for 24 h to give
fully desolvated 1. The powder X-ray diffraction pattern and Langmuir
surface area (2,700 m^2 g−1; N 2 , 77 K) of the material were found to be
consistent with those reported in the literature^17.
Single crystals of 1 were synthesized following a previously reported
procedure^16 and characterized by single-crystal X-ray diffraction.
Note that refinement of the linker occupancies in the single-crystal
and powder X-ray diffraction structures resulted in occupancies that
range from 76.8% to 100%, consistent with previous reports of missing
linker defects in Zr 6 O 4 (OH) 4 (bpydc) 616 and other zirconium metal–
organic frameworks^32.
General procedure for loading 1 with NiX 2 in diglyme
X− = Cl−, Br−. Single crystals of 1 (<0.1 mg) suspended in diglyme were
transferred into a 4 ml PTFE-capped vial. Most of the solvent was
decanted, followed by addition of excess metal source (Ni(DME)Cl 2 ,
Ni(DME)Br 2 , or NiBr 2 ; 5–10 mg; >50 equiv) and diglyme (3 ml). The mix-
ture was allowed to react for 1 month at 120 °C, resulting in a colour
change of the crystals to pale yellow. The crystals were then character-
ized by single-crystal X-ray diffraction.
Stoichiometric reactions were performed on microcrystalline
powder samples of 1 in the presence of single crystals that were later
characterized by crystallography. Single crystals of 1 (<0.1 mg) sus-
pended in diglyme were transferred into a 4 ml PTFE-capped vial. Most
of the solvent was decanted, followed by addition of microcrystalline
1 (60 mg), metal source (1.0–3.25 equiv Ni(DME)Cl 2 , Ni(DME)Br 2 , or
NiBr 2 per bpydc2– in microcrystalline 1 ) and diglyme. The mixture was
allowed to react for 1 month at 120 °C, resulting in a colour change of
both the crystals and the powder to pale yellow. Most of the solution
was removed by pipette and the crystals were subsequently soaked
three times in 3 ml of fresh DME at room temperature (~32 °C) for 24 h.
In cases where unreacted metal halide solids were observed, these were
removed by carefully transferring a slurry of the framework into a new
vial before each wash. A slurry containing most of the microcrystalline
powder was separated from the crystals and pipetted into a new vial,
after which the solvent was removed under reduced pressure at 80 °C to
give a microcrystalline powder sample of the NiX 2 -loaded framework.
The remaining single crystals were then used for single-crystal X-ray
diffraction experiments.General procedure for loading 1 with MX 2 in 10% (v/v) DME in
DFB
MX 2 = FeCl 2 , CoCl 2 and NiBr 2. Single crystals of 1 (<0.1 mg) suspended in
diglyme were transferred into a thick-walled borosilicate tube. Most of
the solvent was decanted, followed by addition of excess metal source
(FeCl 2 , CoCl 2 or Ni(DME)Br 2 ; 5–10 mg; >50 equiv), DME (0.30 ml) and
DFB (2.70 ml). The reaction mixture was degassed by three freeze–
pump–thaw cycles, after which the tube was flame-sealed and then
placed in an oven preheated to 120 °C. The mixture was allowed to
react for 1 month at this temperature, resulting in a colour change
of the crystals (purple for FeCl 2 , blue for CoCl 2 , and pale yellow for
Ni(DME)Br 2 ). The crystals were then characterized by single-crystal
X-ray diffraction.
Stoichiometric reactions were performed on microcrystalline pow-
der samples of 1 in the presence of single crystals that were later char-
acterized by crystallography. Single crystals of 1 (<0.1 mg) suspended
in diglyme were transferred into a thick-walled borosilicate tube. Most
of the solvent was decanted, followed by addition of microcrystalline
1 (60 mg), metal source (3.25 equiv FeCl 2 , CoCl 2 , or Ni(DME)Br 2 ), DME
(0.30 ml), and DFB (2.70 ml). The reaction mixture was degassed by
three freeze–pump–thaw cycles, after which the tube was flame-sealed
and then placed in an oven preheated to 120 °C. The mixture was allowed
to react for 1 month at this temperature, resulting in a colour change
of both the crystals and the powder (purple for FeCl 2 , blue for CoCl 2 ,
and pale yellow for Ni(DME)Br 2 ). Most of the solution was removed by
pipette and the crystals were subsequently soaked three times in 3 ml
of fresh DME at room temperature (~32 °C) for 24 h. In cases where
unreacted metal halide solids were observed, these were removed by
carefully transferring a slurry of the framework into a new vial before
each wash. A slurry containing most of the microcrystalline powder was
separated from the crystals and pipetted into a new vial, after which
the solvent was removed under reduced pressure at 80 °C to give a
microcrystalline powder sample of the MX 2 -loaded framework. The
remaining single crystals were then used for single-crystal X-ray dif-
fraction experiments.Single-crystal X-ray diffraction
X-ray diffraction analysis was performed on single crystals coated with
Paratone-N oil and mounted on a MiTeGen loops. The crystals were
frozen at 100 K by an Oxford Cryosystems Cryostream 700 Plus. Data
were collected at beamline 11.3.1 at the Advanced Light Source at Law-
rence Berkeley National Laboratory using synchrotron radiation
(λ = 0.8856 and 0.9537 Å) on a Bruker D8 diffractometer equipped with
either a Bruker PHOTON100 CMOS detector or a Bruker PHOTON II
CMOS detector. Raw data were integrated and corrected for Lorentz