Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
1. 1,5-Diazidonaphthalene (1)
There are three isomers for 1. The computational results of three isomers obtained at the
B3LYP/6-31G* level were shown below. The calculated and observed FT-IR spectra indicated
that the most stable isomer(1) would be a dominant species in the matrix.
Optimized geometries and energies of three isomers of 1 calculated at the B3LYP/6-31G* level.
Bond lengths in angstroms. Energies in hartrees.
N
BOND LENGTH
BOND ANGLE
N
1.141
1.236 N
172.9
N
N
N
1.420
1.377
1.432
1.382
115.4 123.6
1.419
1.410
1.432
120.2 122.1 121. 0
120.9 119. 9 117.9 120.1
1.410
1.431
1.382
120.1
1.419
117.9 119. 9 120.9
121. 0 122.1 120.2
123.6 115.4
1.377
1.420
N
118.5 N
1.236
N
118.5
N 1.141
N
N
172.9
Isomer (1)
-713.070543 (0.0 kcal·mol -1)
N
1.142
N
N
170. 6
N
1.236
N
121. 2 N
1.425
1.434
1.377
123. 6 115. 4
120. 3 119. 1 121. 0
120. 4 119. 1 118. 2 119. 8
1.381
1.421
1.409
1.436
1.409
120. 6 117. 4 120. 1 121. 0
120. 8
120. 5
121. 5
115. 4 123. 5
1.431
1.382
1.418
1.376
1.419
N
118. 4
N
1.237
N
N
N 1.141
N
Isomer (2)
172. 8
-713.064323 (+3.8 kcal·mol -1)
1
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
N 1.142
N
N 1.236
170.6
N
N
1.426
1.434
1.376
121.2 N
1.380
123.5 115.3
1.420
1.408
1.441
120.6 122.7 121.0
120.6 119.4 117.7 120.3
1.408
1.434
1.380
1.420
120.3 117.7 119.4 120.6
120.6
121.0
122.7
115.3 123.5
1.376
1.426
N
N 121.2
1.236 N
N
1.142 N
170.6
N
Isomer (3)
-713.057615 (+7.9 kcal·mol -1)
Table. Energiesa of three isomers of 1 estimated at the B3LYP/6-31G* level.
Isomer(1)
Isomer(2)
Isomer(3)
a
HF
ZPEa
-713.070543
-713.064323
-713.057615
0.150960
0.150853
0.150662
E / kcal·mol-1
0.0
+3.8
+7.9
Zero-point energies estimated at the UB3LYP/6-31G* level (scaled by 0.9806).
2
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Figure. Observed FT-IR spectra of 1 in an argon matirx at 11 K and the theoretical IR spectra of
three isomers of 1 calculated at the B3LYP/6-31G* level.
3
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Table. Observed and calculated IR data of 1.
Isomer(1)
Isomer(2)
 / cm I / km·mol
-1
518.4
521.7
627.7
690.0
19.3
5.9
5.6
72.4
765.0
774.5
859.7
20.0
83.4
19.2
1059.2
1148.4
1198.9
1218.7
20.5
5.1
8.7
5.7
1311.2
1325.4
326.9
33.1
1393.6
270.4
1499.4
-1
76.9
1563.7 5.3
1580.6
34.3
2176.3 1928.9
2180.7 678.5
3069.5
3078.3
3083.7
3109.2
16.0
10.6
18.0
7.9
Isomer(3)
/ cm I / km·mol
-1
502.3
519.4
7.5
13.4
611.9
673.7
727.6
8.7
26.0
20.6
767.8
771.9
861.5
881.3
1052.6
1146.7
1182.3
13.4
78.2
20.9
13.3
6.9
5.5
5.4
1314.7
6.7
1316.3
1333.8
1350.0
1396.4
1404.5
1501.7
-1
Obs.
/ cm I / km·mol
-1
-1
/ cm-1
520.4
23.3
535
679.2
32.2
707
767.0
769.2
859.9
49.5
34.0
27.1
781
876
1054.2
1152.2
1179.4
1211.1
1309.9
5.4
6.1
11.4
6.1
33.9
1067, 1079
213.7
48.5
11.6
235.7
9.5
57.5
1330.7
157.4
1294
1313
1398.2
244.3
1409
1503.1
48.4
1516, 1512
1583.1
2173.1
50.9
881.0
1584.7
2174.1
71.9
1273.5
1593
2116, 2177
3071.4
3080.5
3085.6
3097.7
3119.3
7.1
15.2
8.2
7.6
5.2
3100.5
3118.8
12.0
9.8
The IR bands with intensity > 5 km·mol-1 are listed.
4
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
2. 1-Azido-5-nitrenonaphthalene (2)
There are two isomers for 2. The computational results of two isomers obtained at the
UB3LYP/6-31G* level were shown below. The calculated and observed FT-IR spectra indicated
that the most stable isomer(1) would be a dominant species in the matrix.
Optimized geometries and energies of three isomers of 1 calculated at the B3LYP/6-31G* level.
Bond lengths in angstroms. Energies in hartrees.
BOND LENGTH
BOND ANGLE
N
1.311
N
115.6 123. 5
1.428
1.390
1.388
121. 9
120. 7
120. 7
121. 1 120. 0 118. 0 120. 0
1.424
1.392
1.404
1.425
120.7
120. 7 118. 6 120. 3
120. 1
118. 6
120. 9
120. 1 121. 2
1.471
1.432
1.385
1.407
1.418
118.5 N
N
1.237
N
N
N 1.141
N
172.7
Isomer (1)
-603.545141(0.0 kcal·mol -1)
<S2 >=2.065
N
N
1.311
1.384
1.471
121. 3 119. 9
120. 3
120. 3
118. 7
120. 2 120. 6 119. 0 120. 4
1.431
1.406
1.402
1.430
1.390
120. 3 117. 4 119. 1 121. 3
120. 6
121. 2
123. 3
115. 0 124. 2
1.431
1.387
1.426
1.391
1.423
N 121. 5
N
1.236
N
1.142
170. 6
N
N
N
Isomer (2)
-603.538512 (+4.0 kcal·mol -1)
<S2 >=2.066
5
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Figure. Observed FT-IR spectra of 1 in an argon matirx at 11 K and the theoretical IR spectra of
6
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
three isomers of 1 calculated at the B3LYP/6-31G* level.
7
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Table. Observed and calculated IR data of 1.
Isomer(1)
 / cm
-1
517.6
588.2
725.8
762.5
763.4
I / km·mol
Isomer(2)
-1
10.6
10.1
15.0
78.6
9.9
892.8
14.3
1044.4
1127.5
1187.3
9.0
8.3
11.4
1248.4
14.6
1317.7
1338.8
179.8
60.1
1393.4
1428.0
12.0
53.7
1543.1
1580.9
2180.6
3071.0
69.1
11.6
923.0
6.6
3074.2
3084.5
3099.7
7.9
7.0
8.1
Obs.
/ cm
I / km·mol
517.2
11.8
677.5
698.0
760.8
761.6
8.8
10.2
26.9
43.1
800.2
893.2
898.0
7.4
8.5
9.3
1132.5
1171.7
1204.1
1247.1
1283.5
9.8
12.9
7.2
16.0
20.2
1158
1135, 1140
1324.7
1348.0
1354.1
34.3
92.2
8.7
1292
1300, 1310
1354
1424.9
1472.7
1540.2
1580.7
2173.4
48.8
12.5
46.3
35.0
642.2
1438, 1446
1467
1558
3079.4
3101.6
7.2
8.0
-1
The IR bands with intensity > 6 km·mol-1
8
-1
/ cm-1
769
783
911
1259
1271
2122
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
3. 1,5-Dinitrenonaphthalene (3)
As described in the main text, the state order in the dinitrene is difficult to reproduce. The
results obtained at several computational levels were listed below. The computation at the
CAS(6,6)/6-31G* // UB3LYP/6-31G* level gave the wrong state order. The geometry optimized
using CAS method gave the enhanced bond alternation.
Table. Relative energiesa of dinitrene 3 with different spin multiplicities estimated by several
computational methods.
UB3LYP/6-31G*
CCSD(T)/6-31G*
// UB3LYP/6-31G*
(Singlet)b
Quintet
Triplet
Singlet
Quintet
Triplet
Singlet
a
Energya
<S2>
Energya
<S2>
+41.4 kcal·mol-1
+8.0 kcal·mol-1
+6.3 kcal·mol-1
-494.025716c
0.00
6.06
2.03
1.78
+22.0 kcal·mol-1
+16.3 kcal·mol-1
-492.506586
6.66
3.07
0.00
CAS(6,6)/6-31G*
CAS(6,6)/6-31G*
// UB3LYP/6-31G*
// CAS(6,6)/6-31G*
+10.4 kcal·mol-1
-7.6 kcal·mol-1
-490.985133
+12.2 kcal·mol-1
+0.3 kcal·mol-1
-491.002923
CASMP2(6,6)/6-31G*
// CAS(6,6)/6-31G*
+4.6 kcal·mol-1
+0.4 kcal·mol-1
-490.985133
Energies were corrected by zero-point energies estimated at the UB3LYP/6-31G* level (scaled
by 0.9806).
c
Estimated by the spin-restricted method.
c
Estimated by the symmetry-broken spin-unrestricted method without sum correction.
9
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Optimized geometries and energies of three isomers of 1 calculated at the B3LYP/6-31G* level.
And at the CAS(6,6)/6-31G* level. Bond lengths in angstroms. Energies in hartrees.
BOND LENGTH
N
Quintet
N
1.277
Triplet
1.323
1.470
1.423
1.494
1.462
1.398
1.440
1.425
1.385
1.455
1.354
1.407
1.367
N
N
Singlet (RB3LYP/6-31G*)
N
1.240
1.477
1.576
1.431
1.351
1.440
1.361
N
Singlet (UB3LYP/6-31G* Guess =Mix)
BOND ANGLE
N
BOND LENGTH
N
1.298
120.0 121.9
118.2 121.1
1.448
1.478
1.373
1.434
1.414
120.6
120.4
121.3
121.0
118.5
1.394
N
N
Singlet (CAS(6,6)/6-31G*)
N
N
1.251
119.7 122.5
117.8 121.5
1.478
1.496
120.5
120.8
121.4
121.8
117.7
1.346
1.473
1.465
1.335
N
N
10
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Figure. Observed IR spectra of 3 in an argon matrix at 11K and theoretical IR spectra of
11
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
1,5-dinitrenonaphthalene
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
with
different
12
spin
multiplicities.
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
Table. Observed and calculated IR data of 3.
CASSCF(6,6)/6-31G*a
UB3LYP/6-31G*
 / cm-1
I / km·mol-1
/ cm-1
Obs.
/ cm-1
Au
451.3
5.0
441.3
Bu
Bu
Au
Bu
483.4
503.1
742.0
751.8
54.7
14.8
82.9
19.3
540.9
551.9
781.0
827.0
507
Bu
Bu
Bu
Bu
Bu
Bu
Bu
Bu
Bu
908.3
1036.6
1148.2
1246.9
1376.4
1436.6
1479.7
1529.4
3076.7
20.5
19.3
26.1
33.0
12.4
32.9
24.4
5.2
15.6
983.7
1092.2
1275.5
1435.5
1548.6
1737.8
1810.8
1889.6
3376.5
1033
1161
1212
1279
1393
Bu
3106.9
12.3
3404.4
The IR bands with intensity > 5 km·mol-1 are listed.
CASSCF calculation using Gaussian98.
13
a
750
1466
1506
IR intensities could not obtained by
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
4. ESR measutment
ESR spectra were measured on a JEOL JES-RE3X spectrometer equipped with an
optical transmission cavity equipped with Air Products LTD-3 liquid helium transfer system. All
measurements were carried out at 78 K.
Diazido precursor was dissolved in purified 1-methyltetrahydrofuran. The solution (4.6
mM) was placed within a quartz 4mm o.d. ESR sample tube and degassed by freeze-pump-thaw
cycles. The sample set to the cavity at 78K was irradiated with an USHIO UI-501C xenon lamp
through Toshiba UV-D36B band-pass filter (360 <  < 380 nm).
Observed spectrum is shown below. Signals at 622 and 611 mT were ascribed to triplet
mononitrenes. Signals ascribable to a dinitrene with higher spin multiplicity were not observed.
The reason for absence can be explained as follows: since the ground state of
1,5-dinitrenonaphthalene was a singlet diradical and it has a large ES–T, the population of the
thermally-excited triplet state is too low to detect. To consider this interpretation, we theoretically
estimated the relative energies of 2,7-dinitrenophenanthrene 5 whose nitreno groups are
connected to rigid aromatic hydrocarbons. This dinitrene showed the highest ES–T value (0.93
kcal·mol-1)3 among the investigated quinonoidal dinitrenes1: the ES–T value obtained
experimentally by the temperature-dependent ESR study. The predicted ES–T value of 5 (5.7
kcal·mol-1) is smaller than that of 3 (6.3 kcal·mol-1) at the same computational level (see ESI).
The observed ESR signals due to the thermally excited triplet state were quite weak even for 5.3
Figure. The ESR spectra of 3 in a glassy MTHF matrix observed upon irradiation with a xenon
lamp through a band-pass filter (360 <  < 380 nm) for 15 min.
14
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
5. Computational study of the dinitrenes
In the quinonoidal dinitrenes, the ground state is the singlet diradical state which lies
slightly below the triplet state: their energy gaps between singlet and triplet states (ES-T) are
rather small (<1 kcal·mol-1). The state order of the dinitrene with different spin multiplicities was
often difficult to reproduce in the computational study because of the multireference character of
singlet diradicals.
However, calculations using the symmetry-broken spin-unrestricted DFT
(UDFT) method could give the correct state order as reported by Lahti et al.(J. Phys. Chem. A.,
2001, 105, 251.)
These results suggest that the UDFT method can describe successfully the
electronic state of singlet diradicals.
A major drawback of the UDFT method for treating
diradicals is that the singlet state can be highly spin-contaminated. Indeed, the UDFT solution for
3 showed the <S2> value of 1.78, indicating the admixure of the triplet and quintet states, and
showed relatively large deviations in the wavenumbers in IR bands. However, the spectral
features of observed IR spectra were qualitatively reproduced by the UDFT computation.
If
almost the same amount of spin contamination is shown in two different systems, the estimated
ES-T values could be compared qualitatively. In the literature by Lahti et al., dinitrenobiphenyl
(bpn) and dinitrenostylbene (stn) showed <S2> values of 1.61 and 1.51, respectively.
The
experimental ES-T values of bpn and stn were 0.75 and 0.47, respectively (J. Am. Chem. Soc.,
1997, 119, 2187.), whereas the predicted ES-T values were 3.2 and 2.7, respectively. Although
only one example is available, it seems that the UDFT method may qualitatively predict the
relative ES-T values for dinitrenes despite of the overestimation.
To consider the interpretation of ES-T of the dinitrene 3, we compared the
theoretically-estimated energies of 3 in different spin multiplicities with those of two other
quinonoidal dinitrenes, 1,4-dinitrenobenzene 4 and 2,7-dinitrenophenanthrene 5, whose nitreno
groups are connected to rigid aromatic hydrocarbons. The obtained results are summarized in the
following table.
Table. Relative energiesa of 3, 4, and 5 with different spin multiplicities estimated at the
UB3LYP/6-31G* level.
Quintet
Triplet
3
4
+8.0 kcal·mol-1 <S2>=6.06
+6.3 kcal·mol-1 <S2>=2.03
+33.1 kcal·mol-1 <S2>=6.02
+1.5 kcal·mol-1 <S2>=2.02
15
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
Singletb
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
-493.768975
<S2>=1.78
-340.317779
<S2>=1.11
5
Quintet
Triplet
Singlet b
+7.5 kcal·mol-1 <S2>=6.07
+5.7 kcal·mol-1 <S2>=2.03
-647.665910
<S2>=1.79
a
All energies were corrected by zero-point energies estimated at the UB3LYP/6-31G* level
(scaled by 0.9806).
b
Estimated by the symmetry-broken spin-unrestricted method without sum correction.
Among the three, 3 and 5 showed the similar spin contamination. Therefore, the experimental
results are discussed from these computational results. The dinitrene 5 showed the highest ES–T
value (0.93 kcal·mol-1)3 among the investigated quinonoidal dinitrenes.1 The predicted ES–T
value of 5 (5.7 kcal·mol-1) is smaller than that of 3 (6.3 kcal·mol-1). This computational results
suggested that 3 would show the high ES–T value as well as 5.
The difference in the ES–T values of 1,5-dinitrenonaphthalene 3, 1,4-dinitrenobenzene 4, and
2,7-dinitrenophenanthrene 5 could be explained on the basis of the spin distributions that were
estimated by the unrestricted method (UB3LYP/6-31G* level). Atomic spin density values of 3,
4, and 5 are depicted in Figure. In all compounds, the spin densities on nitrogen atoms are more
than 1.00, which reveals the spin alignment due to the one-center interaction between and 
electrons of a nitrogen atom. The spin densities of nitrogen atoms in 3, 4, and 5 are in order of 5
> 3 > 4, which is different from the order in the ES–T values (3 > 5 > 4). Whereas spin densities
on the carbon atom adjacent to nitrogen atoms (underlined) in 3, 4, and 5 are in the same order as
that in the ES–T values. This value would reflect two polarization effects: (1) polarization effect
from the radical center on the adjacent nitrogen atom. (2) polarization effect from the other
radical center through aromatic system. The latter effect would be lager in 3 than in 5, which
resulted in the change of the order. Indeed, the larger spin polarization (0.3248) was observed for
3 on the central carbon atoms in the naphthalene framework in comparison with that for 4
(0.1846) and 5 (0.2604) (Figure). These results suggested that the larger spin polarization on the
aromatic system, in particular, on the atoms on the pathway connecting two radical centers,
enhanced the stability of the singlet state of 3.
16
Supplementary Material (ESI) for Chemical Communications
This journal is © The Royal Society of Chemistry 2001
Supplementary Information
T. Sato, et al. “Preparation of 1,5-dinitrenonaphthalene in cryogenic matrices”
1.3954
-0.3612
-0.2937
0.2199
0.3879
0.3248
-0.3248
-0.3879
-0.2199
0.2937
0.3612
3
-1.3954
0.1846
-0.1846
-0.2181
-1.1532
0.2181
1.1532
0.1846
-0.1846
4
0.1377
-0.1377
-0.1878
0.4066
1.4524
0.1878
0.2604
-0.4066
-1.4524
-0.2604
0.2818
-0.2818
0.2530
-0.1594
0.1594
-0.2530
5
Figure. Atomic spin density values of 3, 4, and 5 estimated
at the UB3LYP/6-31G* level.
17