SUPPORTING INFORMATION

Generic Post-Functionalization Route from

Amino-Derived Metal-Organic Frameworks

Marie Savonnet, Delphine Bazer-Bachi, Nicolas Bats, Erwann Jeanneau, Vincent Lecocq,

Catherine Pinel, David Farrusseng*

Universit Lyon 1 , IRCELYON, Institut de recherches sur la catalyse et

lenvironnement de Lyon, UMR 5256 CNRS;2 avenue Albert Einstein; F-69626

Villeurbanne Cedex, France, IFP-Lyon, BP n3, 69360, Solaize,France, Universit

Lyon1, Centre de Diffractomtrie, 69629 Villeurbanne Cedex, France

Chemicals

General

S1- Experimental Details for the Synthesis of DMOF-NH2

S2 Effect of solvent guest inclusion on powder X-ray Diffraction of DMOF-NH2

S3 -Experimental Details for the Post-Functionalization of DMOF-NH2

S4- Kinetic of DMOF-N3 formation

S5 - Experimental Details for the Synthesis of MIL-68(In)-NH2

S6 - Experimental Details for the Post-Functionalization of MIL-68(In)-NH2

Chemicals

All chemicals were used as received without any further purification: N,N-

dimethylformamide, DMF (Aldrich, 99.8%), Zinc nitrate, Zn(NO3)24H2O (Merck,

98.5%), 2-aminoterephthalic acid, NH2-bdc (Alfa Aesar, 99%), dimethyl-2-

aminoterephthalate (Aldrich, 97%), dichloromethane, CH2Cl2 (Acros Organics, 99.99%),

Indium nitrate, In(NO3).4H2O (Alfa Aesar, 99.99%), 1,4-diazabicyclo[2.2.2] octane,

DABCO (Aldrich, 98%), THF (Aldrich, 99%), tBuONO (Aldrich), TMSN3 (Aldrich,

99.5%), tetrakis(acetonitrile)copper(I)hexafluoro-phosphate (CuI(CH3CN)4PF6) (Aldrich),

phenylacetylene (Aldrich, 98%), DCl (Aldrich, 99%D), DMSO-d6 (Eurisotop, 99.8%D).

General

NMR spectra were recorded on a Bruker Avance 250 spectrometer operating at 250 MHz

for 1H. The following abbreviations are used: s, singlet; d, doublet; t, triplet; m, multiplet.

IR spectra were recorded on a Fourier Transform Vector 22 Bruker spectrometer in KBr

pellets in the 400-4000 cm-1 region. The following abbreviations are used: s, strong; m,

medium; w, weak; br, broad.

Powder X-ray diffraction patterns were recorded using a Bruker D5005 diffractometer

(BraggBrentano geometry, graphite monochromator, Cu K radiation).

N2 isotherms at 77K were performed using a BELSORP-max (BEL Japan) and samples

were desorbed one night at 100C under vacuum.

The thermogravimetric analyses have been carried out on a SETARAM type Setsys Evolution

12 apparatus (heating rate of 2 C min-1) under a pure air flow coupled with a mass

spectrometer PFEIFFER, type Omnistar.

S1 - Experimental Details for the Synthesis of DMOF-NH2

Zn(NO3)24H2O (0.781 g, 3.00 mmol) and 2-amino-1,4-benzenedicarboxylic acid (NH2-bdc,

0.554 g, 3.03 mmol) were dissolved in 75 mL of dimethylformamide (DMF). DABCO (0.542

g, 4.815 mmol) was then added to the solution, which immediately generated a large amount

of white precipitate. The mixture was filtered using a Pyrex glass funnel of fine porosity.

The powder was washed with 3 8 mL of DMF followed by 3 8 mL of CH2Cl2. The

powder was then soaked in 10 mL of CH2Cl2 for three days with fresh CH2Cl2 added every 24

h. After three days of soaking, the powder was dried under vacuum at room temperature for

one night. The mass of dried DMOF-NH2 obtained was 300 mg.

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 7.13 (d, 1H, J = 8.3Hz); 7.47 (s, 1H); 7.79 (d,

1H, J = 8.3Hz), 3.52 (s, 6.68H, DABCO)

Relative concentration of NH2-bdc and DABCO in DMOF-NH2: 1:0.56. (Expected value: 1:0.5)

IR data (KBr pellet), (cm-1): 3456 (br), 3345 (br), 2952 (w), 2878 (w), 1666(s), 1632 (s),

1577 (s), 1496 (s), 1435 (s), 1376 (s), 1256 (m), 1102 (m), 1056 (w), 1015 (w), 834 (w), 810

(w), 772 (m), 704 (w), 661 (w), 593 (w).

Figure 1: SEM photograph of the DMOF-NH2 powder.

S2 - Effect of solvent guest insertion on the powder X-Ray diffraction of DMOF-NH2

DMOF-NH2 was obtained as indicated in S1. After soaking in THF or THF/CH2Cl2, DMOF-

NH2 is less XRD crystalline. Drop of surface areas (1320 to 110 m/g) and microporous

volumes from 0.54 to 0.02 cm3/g is observed.

10 20 302 /

Figure 2: Powder X-ray diffraction patterns of (black) DMOF-NH2, (red) DMOF-NH2 soaked in THF and (blue) DMOF-NH2 soaked in THF and then in CH2Cl2.

S3 - Experimental Details for the Post-Functionalization of DMOF-NH2

In a typical synthesis, the freshly dried DMOF-NH2 (80 mg, 0.27 mmol equiv of -NH2) was

placed into a vial (10 mL capacity) with 3.0 mL of THF, 0.22 mL (1.84 mmol, 7 eq) of

tBuONO and 0.2 mL (1.51 mmol, 6 eq) of TMSN3. The sample was left to react for one night

at room temperature to produce the azide intermediate corresponding compound .The reaction

was quenched by decanting the solvent. Excess reactants were removed by washing three

times in THF followed by three times in CH2Cl2. Drying at room temperature yielded of

yellow powder of DMOF-N3. The mass of dried DMOF-N3 obtained was 65 mg (for

characterization purposes).

COOH

COOH

N3

Crystal data for DMOF-N3 [Zn2(C8H4O4N3)2(C6H12N2)]: quadratic system, P4/m with a = b =

10.837 , c = 9.614 and alpha=beta=gamma=90.

Crystal data for DMOF-1 [Zn2(C8H4O4)2(C6H12N2)]: tetragonal, P4/m, a= b = 10.929

c=9.608 (Angew. Chem. Int. Ed. 2004, 43, 5033 5036).

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 7.73-7.83 (m, 3H, ArH), 3.56 (s, 6.5H,

DABCO).

IR data (KBr pellet), (cm-1): 3430 (br), 3122 (w), 2123(s), 1626 (s), 1484 (w), 1383 (s),

1279 (m), 1167 (w), 1053 (s), 987 (w), 891 (w), 842 (m), 801 (s), 770 (s), 716 (w), 687 (w),

660 (w), 615 (w), 537 (w).

One pot post-functionalization of DMOF-NH2

In a typical synthesis, the freshly dried DMOF-NH2 (80 mg, 0.27 mmol equiv of -NH2) was

placed into a vial (10 mL capacity) with 3.0 mL of THF, 0.22 mL (1.84 mmol, 7 eq) of

tBuONO and 0.2 mL (1.51 mmol, 6 eq) of TMSN3. The sample was left to react for one night

at room temperature to produce the azide intermediate corresponding compound.

Phenylacetylene (0.96 mL, 8.8 mmol, 36 eq) and an amount of CuI(CH3CN)4PF6) (48 mg,

0.26 mmol, 1 eq) in 1 mL of THF were added and the mixture was stirred continuously for 24

h. After decantation, the supernate was removed. The solid was washed three times by THF (x

8 ml) and three times by CH2Cl2 (x 8 ml) in order to remove unreactive substrates. The solid

was then dried under vacuum at room temperature to yield the final yellow compound,

DMOF-fun. The mass of dried DMOF-fun obtained was 60 mg.

COOH

COOH

N

NN

1

2 34

5 6

7

89

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 9.14 (s, 1H, H4), 8.18 (m, 2H, H3, H2), 8.03 (d,

1H, J= 8Hz, H1), 7.92 (d, 2H, J= 7Hz, H5, H9), 7.45 (m, 3H, H6, H7, H8), 3.52 (s, 6H,

DABCO)

13C NMR, 62.8 MHz, (DCl/D2O/DMSO-d6) : 122.8, 125.2, 125.6, 126.4, 128.2, 128.9, 129,

130.1, 130.4, 130.9, 132.1, 134.1, 135.1, 146.6, 165.4, 166

IR data (KBr pellet), (cm-1): 3416 (br), 3140 (w), 3053 (w), 1605 (s), 1463 (w), 1383 (s),

1298 (w), 1157 (m), 1053 (w), 915 (w), 839 (s), 767 (s), 695 (s), 558 (w).

Elemental analysis for as-modified DMOF-fun (90%)

[Zn2(C16H9O4N3)1,8(C8H3O4N3)0,2(C6H12N2)] 0.8CH2Cl2

Calculated: Zn 14.33, C 49.21, H 3.35, N 12.35;

Found: Zn 14.29, C 46.46, H 3.24, N 12.58, Cu: 0.64

TGA data: calculated weight loss for 0.8 CH2Cl2: 8.1%; found: 8.1%.

Elementary analysis shows that 0.64% of Cu remains in the DMOF-fun with 90% of grafting

rate. It corresponds to 3 Cu for 100 triazolate functions.

10 20 30 40

2 /

Figure 3: Powder X-ray diffraction patterns of (red) DMOF-NH2, (black) DMOF-N3 and (blue) DMOF-fun.

4000 3500 3000 2500 2000 1500 1000 500wavenumber (cm-1)

Figure 4: IR spectra of (red) DMOF-NH2 and (black) DMOF-N3, (blue) DMOF-fun

Molecular Modelling

The DMOF functionalized structure has been obtained starting from the crystallographic data

of DMOF-1. The NH2 groups of DMOF-1 were replaced with (4-phenyl-1,2,3-triazol-1-yl)

groups. In order to relax the obtained structure, a geometric optimization followed by a NPT

Molecular Dynamics simulation has been performed. The Universal Force Field (UFF) (A. K.

Rappe, C. J. Casewit, K. S. Colwell, W. A. Goddard, W. M. Skiff; J. Am. Chem. Soc. 1992,

114, 10024-10035) combined with the self consistent charge equilibration method (Qeq)

(A.K. Rappe, W.A. Goddard; J. Phys. Chem. 1991, 95, 3358) as implemented in the Accelrys

MS Forcite package was used to mimic the energetic interactions of the system.

Figure 5: Modelling of DMOF-fun (100% grafting)

In order to estimate the sterical demand caused by the introduction of the 4-phenyl-1,2,3-

triazol-1-yl groups, the different energetic contributions where compared to those obtained for

the DMOF-1 at the end of a geometrical optimization employing the same methodology as in

the case of the DMOF functionalized.

DMOF-1 DMOF-fun Intramolecular

Energy (kcal/mol) 2555 3446 891 (+35 %) *

Van der Waals (kcal/mol)

221 133 -88 (-40 %)

Electrostatic Energy (kcal/mol)

-3033 -2666 367 (+12 %)

Table 1: Energetic contributions for DMOF-1 and DMOF-fun after geometric optimization allowing cell deformation. * Without subtracting the intramolecular interactions corresponding to the 1,2,3-triazol-1-yl groups.

The cell parameters obtained for both materials at the end of the respective geometric

optimizations are as well compared in Table 2:

Lx () Ly () Lz ()

DMOF-1 20.889 20.785 19.109

DMOF-fun 20.813 20.810 19.001

Table 2: cell parameters for DMOF-1 and DMOF-functionalized after geometric optimization allowing cell deformation.

As it can be expected the functionalization causes an increment of the intramolecular energy

(35%) at the same time that it generates a slightly higher electrostatic repulsion (10%). In the

first case, it should be stressed that the totality of these increments cannot be attributed to

sterical effects since an important part comes already from the addition of the new

intramolecular interactions of the new groups. In order to account for this effect the

intramolecular energy of the 1,2,3-triazol-1-yl groups has been computed separately following

the previously described methodology. Once this correction has been done the increment

obtained is only 126 kcal/mol (5%). These values seem quite reasonable taking into account

the size of the 4-phenyl-1,2,3-triazol-1-yl groups. In addition, no significant changes are

observed for the cell parameters confirming the feasibility of a complete functionalization.

Figure 6: Positive mode MS performed after digestion of DMOF-fun

Figure 7: N2 isotherm at 77K of DMOF-fun at different degrees of modification

Degree of modification (%)

BET surface area (m/g)

Microporous volume (m3/g)

0 1320 0.54

15 485 0.18

40 400 0.15

75 310 0.10

100 244 0.08

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20

Time (h)

Degr

ee o

f mod

ifica

tion

(%)

Figure 8: Kinetic of DMOF-fun (90%) formation.

Synthesis of the corresponding linkers in solution

COOMe

COOMe

NH2

COOMe

COOMe

N3

COOMe

COOMe

N

NN

tBuONOTMSN3

THF

PhenylacetyleneCu(ACN)4PF6

THF

The same protocol was applied to dimethyl-2-aminoterephthalate in solution in order to obtain

reference 1H NMR spectra.

Dimethyl-2-aminoterephthalate (0.190 g, 0.91 mmol) was dissolved in THF (4 mL) in a 25

mL round-bottomed flask and cooled to 0C in an ice bath. To this stirred mixture was added

tBuONO (141 mg, 0.16 mL, 1.37 mmol), followed by a dropwise addition of TMSN3 (126

mg, 0.14 mL, 1.10 mmol). The resulting solution was stirred at room temperature for one

night. Phenylacetylene (140 mg, 0.150 mL, 1.37 mmol) and CuI(CH3CN)4PF6 (48 mg, 0.26

mmol) were then added, and this mixture was stirred overnight at room temperature. The

mixture was concentrated under vacuum and the organics were extracted by CH2Cl2 and

washed with water, saturated NaHCO3 (aq) and brine. After drying over Na2SO4, the solvent

was taken off under reduced pressure. The yield obtained was 80%.

COOMe

COOMe

N

NN

1

3

4

5 6

7

892

10

11

1H NMR 250 MHz, (DMSO-d6) : 9.22 (s, 1H, H4), 8.24 (d, J= 7Hz, 2H, H3, H2), 8.07 (d, 1H,

J= 8.5Hz, H1), 7.95 (d, 2H, J= 8 Hz, H5, H9), 7.45 (m, 3H, H6, H7, H8), 3.93 (s, 3H, H10), 3.69

(s, 3H11)

IR data (KBr pellet), (cm-1): 3140 (w), 2960 (w), 2841 (w), 1737 (s), 1639 (w), 1575 (w),

1454 (m), 1400 (m), 1291 (m), 1250 (m), 1084 (s), 798 (m), 746 (m), 682 (m), 488 (w)

9,5 9,0 8,5 8,0 7,5 7,0

ppm

Figure 9: 1H NMR spectra of (black) corresponding linker in solution and (red) digested DMOF-fun. S4- Kinetic of DMOF-N3 formation

DMOF-NH2 (80 mg, 0.27 mmol equiv of -NH2) was placed into a vial (10 mL capacity) with

3.0 mL of THF, 0.22 mL (1.84 mmol, 7 eq) of tBuONO and 0.2 mL (1.51 mmol, 6 eq) of

TMSN3. The sample was left to react for 24h at room temperature with different samplings (at

2.5, 5, 8.5, 12 and 24h).

0 5 10 15 20 25

0

20

40

60

80

100

Con

vers

ion

(%)

Time (h)

Figure 10: Kinetic of DMOF-N3 formation.

S5 - Experimental Details for the Synthesis of MIL-68(In)-NH2

MIL-68(In)-NH2 was obtained by precipitation reaction in a Pyrex beaker (capacity 100 mL)

of a mixture of 4.82 mL (4.14 mmol) of 0.86 M of indium nitrate in DMF and 10.06 mL (3.32

mmol) of 0.33 M 2-aminoterephthalic acid (NH2-bdc) in DMF. The reaction mixture was

stirred for 5 minutes then 4.83 mL (6.67 mmol) of 1.38 M 4-diazabicyclo[2.2.2] octane

(DABCO) in DMF were added. The reaction mixture was stirred for 120 minutes at room

temperature. The obtained precipitate was washed with DMF at 160 C and soaked in

dichloromethane during 24h. The mass of dried MIL-68(In)-NH2 obtained was 880 mg.

Crystal data of MIL-68(In)-NH2: orthorhombic system, Cmcm with a = 21.804 ; b = 37.517

; c = 7.207 and alpha=beta=gamma=90.

Crystal data of MIL-68(In): orthorhombic system, Cmcm with a = 21.7739 ; b = 37.677 ; c

= 7.233 and alpha=beta=gamma=90 (Inorg. Chem. 2008, 47, 11892-11901).

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 7.80 (d, 1H, J = 8.3Hz), 7.44 (s, 1H), 7.15 (d,

1H, J = 8.3Hz)

2 /

IR data (KBr pellet), (cm-1): 3487 (br), 3379(br), 1556(s), 1434 (s), 1381 (s), 1366 (w),

1256 (m), 1144(w), 1042 (w), 966 (w), 829 (w), 770 (m), 550 (w)

10 20 30

Figure 11: Powder X-ray diffraction patterns simulated (red) from MIL-68(In).cif file (Inorg. Chem. 2008, 47, 11892-11901) and (black) experimental MIL-68(In)-NH2.

S6 - Experimental Details for the Post-Functionalization of MIL-68(In)-NH2

In a typical synthesis, the freshly dried MIL-68(In)-NH2 (80 mg, 0.26 mmol equiv of -NH2)

was placed into a vial (10 mL capacity) with 3.0 mL of solvent (THF) and 1.48 mL (12.48

mmol, 48 eq) of tBuONO and 1.3 mL (9.88 mmol, 38 eq) of TMSN3. The sample was left to

react 6 hours at room temperature to produce the azide MOF. The reaction was quenched by

decanting the solvent. Excess reactants were removed by washing three times in THF

followed by three times in CH2Cl2. Drying at room temperature yielded a yellow powder of

MIL-68(In)-N3. The mass of dried MIL-68(In)-N3 obtained was 70 mg (for characterization

purposes).

Crystal data: orthorhombic system, Cmcm with a = 21.855 ; b = 37.433 ; c = 7.209 and

alpha=beta=gamma=90.

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 7.73-7.83 (m, 3H, ArH)

IR data (KBr pellet), (cm-1): 3414 (br), 2123 (s), 1569 (s), 1495 (w), 1384 (s), 1279 (m),

1153 (w), 1085 (w), 798 (w), 769 (s), 657 (w), 552 (m).

One pot post-functionalization of MIL-68(In)-NH2

In a typical synthesis, the freshly dried MIL-68(In)-NH2 (80 mg, 0.26 mmol equiv of -NH2)

was placed into a vial (10 mL capacity) with 3.0 mL of solvent (THF) and 1.48 mL (12.48

mmol, 48 eq) of tBuONO and 1.3 mL (9.88 mmol, 38 eq) of TMSN3. The sample was left to

react 6 hours at room temperature to produce the azide MOF. Phenylacetylene (1.92 mL, 17.6

mmol, 74 eq) and an amount of CuI(CH3CN)4PF6) (96 mg, 0.52 mmol, 2.2 eq) in 1.5 mL of

THF were added and the mixture was stirred continuously for 24 h. After decantation, the

supernatant was removed. The solid was washed three times by THF (x 8 ml) and three times

by CH2Cl2 (x 8 ml) in order to remove unreactive substrates. Drying under vacuum at room

temperature yielded a yellow powder of MIL-68(In)-fun. The mass of dried MIL-68(In)-fun

obtained was 60 mg.

1H NMR 250 MHz, (DCl/D2O/DMSO-d6) : 9.13 (s, 1H, H4), 8.20 (d, 1H, J= 6.5Hz, H2), 8.14

(s, 1H, H3), 8.04 (d, 1H, J= 8Hz, H1), 7.92 (d, 2H, J= 6Hz, H5, H9), 7.42 (m, 3H, H6, H7, H8 +

15H excess of phenylacetylene blocked into the pores).

IR data (KBr pellet), (cm-1): 3422(br), 3049 (w), 2980 (w), 2120 (w), 1570 (s), 1482 (m),

1405 (s), 1298 (m), 1143 (m), 1042 (m), 898 (w), 836 (w), 745 (m), 683 (m), 566 (w), 512

(w).

4000 3500 3000 2500 2000 1500 1000 500

Wavenumber (cm-1)

Figure 12: IR spectra of (red) MIL-68(In)-NH2 , (black) MIL-68(In)-N3 and (blue) Mil-68(In)-fun

9,0 8,5 8,0 7,5 7,0 6,5 6,0

ppm

Figure 13: 1H NMR spectra of (red) MIL-68(In)-NH2 , (black) MIL-68(In)-N3 and (blue) Mil-68(In)-fun

9,5 9,0 8,5 8,0 7,5 7,0 6,5 6,0ppm

Figure 14: 1H NMR spectra of (red) MIL-68(In)-NH2 , (black) MIL-68(In)-N3 and (blue) Mil-68(In)-fun (40%)

In the DMOF system, all substrate in excess can easily be removed by washing treatments

(3xTHF + 3xCH2Cl2). In contrast, for high grafting rate (>90%) in MIL-68(In), excess of

substrate are trapped in the channels (S6, Fig13). However, for lower grafting rate (40%) as it

is shown in S6, Fig.14, all substrate in excess can be removed by washing treatments. This

difference between DMOF and MIL-68(In) can be explained by the different porous structure.

The DMOF has a 3 dimensional pore structure whereas MIL-68(In) has a 1D channel

structure. Therefore, it is much easier to evacuate guest molecules in DMOF although it has a

smaller pore size. For MIL-68(In), when the grafting rate is high (90%), the guest molecules

are trapped into the channels which become tight due to the high sterical demands. We

believe that the aromatic ring do not rotate anymore. For much lower grafting rate (S6,

Fig.14), we think that substrates can still pass from one channel to an other to allow a

complete evacuation of the excess. This hypothesis is supported by N2 physisorption

measurements at 77K on MIL-68(In) systems (S6, Fig 16). It clearly shows a marked drop of

the surface area and micropore volume when the degree of modification goes from 50% to

80%.

10 20 30 40 2 /

Figure 15: Powder X-ray diffraction patterns of (red) MIL-68(In)-NH2, (black) MIL-68(In)-N3 and (blue) MIL-68(In)-fun.

Figure 16: N2 isotherm at 77K of MIL-68(In)-fun at different degrees of modification

0 200 400 600-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

Wei

ght l

oss

(%)

Temperature C

Figure 17: Thermogravimetric analysis (TGA) of as-synthesized DMOF-NH2 The TGA data indicate that DMOF-NH2 loses H2O and DMF (23.8%) in the temperature

range of 100200C, and the resulting porous framework (57.7%) starts to decompose after

300C.

Degree of

modification (%)

BET surface

area (m/g)

Microporous

volume (m3/g)

0 1260 0.48

50 571 0.21

80 120 0.03

0 200 400 600-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

Wei

ght l

oss

(%)

Temperature (C)

Figure 18: Thermogravimetric analysis (TGA) of as-modified DMOF-fun (50%)

The TGA data indicate that DMOF-fun loses its guest molecules CH3CN and CH2Cl2 (8.1%)

in the temperature range of 15120C, then the unstable azide groups are decomposed

between 120-220C (4%). The resulting porous framework (63%) starts to decompose after

220C.

0 200 400 600-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

Wei

ght l

oss

(%)

Temperature (C)

Figure 19: Thermogravimetric analysis (TGA) of as-synthesized MIL-68(In)-NH2

The TGA data indicate that MIL-68(In)-NH2 loses its guest molecules CH2Cl2 (9.1%) in the

temperature range of 15120C, and the resulting porous framework (49%) starts to

decompose after 350C.

0 200 400 600-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

Wei

ght l

oss

(%

)

temperature (C)

Figure 20: Thermogravimetric analysis (TGA) of as-modified MIL-68(In)-fun (30%)

The TGA data indicate that MIL-68(In)-fun loses its guest molecules H2O (3.5%) in the

temperature range of 15120C, then the unstable azide groups are decomposed between 120-

220C (8%). The resulting porous framework (52%) starts to decompose after 270C.