Carbon Dioxide Copolymerization Study with a Sterically Encumbering Naphthalene-Derived Oxide

Article abstract of DOI:10.1021/acscatal.5b01375

Poly(1,4-dihydronaphthalene carbonate) has been prepared via the catalytic coupling of carbon dioxide and 1,4-dihydronaphthalene oxide using chromium(III) catalysts. The copolymer formation is found to be greatly dependent on the steric environment around the metal center. Traditional (salen)Cr^IIIX/cocatalyst systems bearing bulky t-butyl groups hinder the approach of the large monomer, significantly diminishing polymer chain growth and providing the entropically favored cyclic byproduct in excess. In contrast, employing the sterically unencumbered azaannulene-derived catalyst, (tmtaa)Cr^IIIX/cocatalyst system (tmtaa = tetramethyltetraazaannulene) shows polymer selectivity close to 90% with three times the activity (TOF = 20-30 h^-1). With the use of a bifunctional (salen)Cr^III catalyst, even higher polymer selectivity (>90%) can be observed. The complete synthesis of a new bifunctional tetraazaannulene ligand for a more effective catalyst is also described herein.

Full text of DOI:10.1021/acscatal.5b01375

Research Article
pubs.acs.org/acscatalysis
Carbon Dioxide Copolymerization Study with a Sterically
Encumbering Naphthalene-Derived Oxide
Donald J. Darensbourg* and Samuel J. Kyran
Department of Chemistry, Texas A&M University, 3255 TAMU, College Station, Texas 77843, United States
S
* Supporting Information
ABSTRACT: Poly(1,4-dihydronaphthalene carbonate) has
been prepared via the catalytic coupling of carbon dioxide
and 1,4-dihydronaphthalene oxide using chromium(III)
catalysts. The copolymer formation is found to be greatly
dependent on the steric environment around the metal center.
Traditional (salen)CrIIIX/cocatalyst systems bearing bulky t-
butyl groups hinder the approach of the large monomer,
significantly diminishing polymer chain growth and providing
the entropically favored cyclic byproduct in excess. In contrast,
employing the sterically unencumbered azaannulene-derived
catalyst, (tmtaa)CrIIIX/cocatalyst system (tmtaa = tetramethyl-
tetraazaannulene) shows polymer selectivity close to 90% with three times the activity (TOF = 20−30 h−1). With the use of a
bifunctional (salen)CrIII catalyst, even higher polymer selectivity (>90%) can be observed. The complete synthesis of a new
bifunctional tetraazaannulene ligand for a more effective catalyst is also described herein.
KEYWORDS: carbon dioxide, 1,4-dihydronaphthalene oxide, polycarbonate, copolymerization, steric effect
produce the cyclic byproduct is enhanced for epoxides with a π
system near the methine carbon as a result of a resonance-
stabilized transition state that provides a lower barrier for
polymer backbiting, as seen through computations.6 The use of
a bifunctional (salen)CoIII catalyst, 2, then proved fruitful in
producing PIC with high selectivity (>99%) at 25 °C.4 The
ammonium side arm of 2 prevents disassociation of the growing
anionic polymer chain, keeping it closer to the metal center via
electrostatic interactions. This encourages long-chain prop-
agation, averting the detrimental polymer backbiting process
leading to cyclic carbonates.1k Furthermore, the built-in
cocatalyst helps increase the activity at lower loadings along
with the thermal stability of the complex. This gave us PIC in
higher MWs with Mn reaching 10 000 g/mol. The catalytic
activity still remained low, with TOFs of no more than 12 h−1.
We felt the need to improve upon the results described
above and decided to look for other aromatic epoxides that
could fare well in the catalysis. Our focus then shifted to 2,3-
epoxy-1,2,3,4-tetrahydronaphthalene or 1,4-dihydronaphtha-
lene oxide (DNO). There were a few reasons that made
DNO an appropriate monomer for the next study: (1) The
aromatic ring is one carbon away from both of the methine
carbons, which will diminish the electronic effects that lead to
polymer backbiting. (2) DNO can be viewed as a benzene
fused with a cyclohexene oxide; cyclohexene oxide selectively
produces the copolymer because it has a large activation barrier
INTRODUCTION
■
Epoxides employed in the catalytic copolymerization reactions
with CO2 are largely aliphatic and alicyclic in nature for
producing polycarbonates.1 The most common and well-
studied in the literature are propylene oxide (PO) and
cyclohexene oxide (CHO) along with their functionalized
derivatives. Poly(propylene carbonate) (PPC) and poly-
(cyclohexene carbonate) (PCHC) have been commercialized
in recent years and are used as sacrificial binding agents,
additives in coatings, and as low-molecular-weight PC-polyol
precursors for polyurethane synthesis.2 Of late, there has been
an interest in preparing CO2-derived polycarbonates that
display high thermal resistance comparable to that of the
widely applicable industrial standard BPA polycarbonate (Tg =
154 °C). We began work in this area when we first presented
poly(indene carbonate) with a Tg of 138 °C as a result of its
fused cyclopentene and benzene ring in the polymer backbone
(Figure 1).3,4 Lu and co-workers then described the preparation
of CO2 copolymers with 3,5-dioxaepoxides in which their use of
4,4-dimethyl-3,5,8-trioxabicyclo[5.1.0]octane (CXO), in partic-
ular, gave the respective polycarbonate, PCXC with a Tg of 140
°C.5
The copolymerization of indene oxide with CO2 proved to
be a challenge. Our initial attempts with the (salen)CoIII-2,4-
dinitrophenoxide catalyst, 1 (Figure 2) along with an onium
salt cocatalyst at its normal working temperature range (25−40
°C) was met with the production of only the entropically
favored product, cis-indene carbonate.3 After lowering the
temperature to 0 °C, a polymer selectivity of ∼60% was
observed with very low TOFs (2 h−1). The tendency to
Received: July 2, 2015
Revised: August 4, 2015
© XXXX American Chemical Society
5421
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Figure 1. (left to right) CO2-derived PCs: poly(indene carbonate); PCXC; poly(1,4-dihydronaphthalene carbonate); and the industrial PC derived
from phosgene, BPA polycarbonate.
Figure 2. Cobalt(III) catalysts: (a) (salen)CoIII-2,4-dinitrophenoxide, 1; (b) bifunctional (salen)CoIII, 2; and (c) biphenol-linked dinuclear CoIII
complex, 3.
for making cyclic carbonate;6,7 hence, DNO could have an
analogous property. (3) The resulting polymer could display a
higher glass transition temperature than PIC as a result of its
bulkier backbone.
to 1,4-dihydronaphthalene oxide (DNO) suitable for CO2
coupling (Scheme 2).11,12 A pale yellow solid that melts
Scheme 2. Preparation of DNO from Naphthalene
We had previously worked with DNO as early as 2004, at
which time we communicated the crystal structure of cis-1,4-
dihydronaphthalene carbonate (DNC) and showed this cyclic
carbonate to be the only isolable product from the catalysis of
CO2 with DNO using a (salen)CrIII catalyst.8 We have now
reexamined this reaction with various chromium(III) catalysts
and report here the successful preparation of poly(1,4-
dihydronaphthalene carbonate), PDNC (Scheme 1). Parallel
above 40 °C, DNO would require the addition of a solvent such
as CH2Cl2 or toluene for the copolymerization reaction with
CO2 at ambient temperature, as was previously done with
indene oxide (MP = 32 °C). Surprisingly, catalysts 1 and 2
showed no sign of a coupling reaction to produce either cyclic
or polycarbonates. Long reaction times (up to 5 days), varying
reaction temperature (0−25 °C with solvents; neat reactions
with 1 and 2 at 45 and 70 °C, respectively), and varying
cocatalyst initiators (2,4-dinitrophenoxide, azide, chloride,
trifluoroacetate) under pressures of 25−35 CO2 bar proved
ineffective at our typically employed catalyst loading of 0.2%.
However, Lu and co-workers show the catalysis to occur at a
higher loading of 1 (0.5%), albeit with a TOF of only 15 h−1.9
Hence, it was decided to examine the catalysis using the CrIII
complexes.
Scheme 1. CO2 Coupling Reaction with DNO
to our efforts, Lu and co-workers have very recently
communicated their preparation of the same copolymer by
employing a dinuclear cobalt(III) catalyst, 3 (Figure 2c).9 They
show high-molecular-weight isotactic PDNC to exhibit a Tg of
150 °C, very close to that of BPA polycarbonate (Figure 1).
Their work also substantiates our further findings regarding a
strong steric influence from the catalyst when copolymerizing a
large monomer such as DNO which is described in detail (vide
infra).
Gratifyingly, a reaction at 70 °C with (salen)CrIIICl, 4
(Figure 3), and an equivalent of PPNCl (PPN = bis-
(triphenylphosphine)iminium) as cocatalyst at a loading of
0.2% provided poly(1,4-dihydronaphthalene carbonate),
PDNC, with a polymer selectivity of 78% and a TOF of 8
h−1 (Table 1, entry 4). The working temperatures for
chromium catalysts are higher than cobalt (≥70 °C), so the
use of solvents can be abated because DNO is a liquid at that
temperature. Solvents are known to lower both the conversion
and polymer selectivity due to epoxide dilution, and it is also
seen here (Table 1, entries 1−3). When the more active
bifunctional (salen)CrIII catalyst, 8, is used at a loading of
RESULTS AND DISCUSSION
■
Naphthalene, the readily available polycyclic aromatic hydro-
carbon occurring as the largest component in coal tar (10 wt
%),10 can be reduced to 1,4-dihydronaphthalene and epoxidized
5422
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Figure 3. (Salen)CrIII catalysts used for CO2/DNO coupling.
Figure 4. Cocatalyst effect on product selectivity for CO2/DNO
coupling with (salen)CrCl, 4.
0.13%, both higher polymer selectivity (93%) and TOFs (29
h−1) are observed at 100 °C (Table 1, entry 12). A molecular
weight of 6700 g/mol displaying a Tg of 136 °C was noted.
Lowering the loading of 8 to 0.1% failed to produce higher
MWs (Table 1, entry 13).
Here, when a DNO molecule gets ring-opened and
carboxylated, the nucleophilic cocatalysts become increasingly
competitive to coordinate and displace the one-mer unit rather
than allow chain propagation via ring-opening of another
epoxide due to unfavorable steric effects (Figure 5). The
displaced one-mer unit can then backbite to give cyclic
carbonate. This would explain the observed higher cyclic
carbonate selectivity for the more nucleophilic azide cocatalyst
versus chloride, a better leaving group, if backbiting were the
exclusive reason. Prior results showing only cyclic carbonate
formation may well be due to the use of 2.25 equiv of N-methyl
imidazole (NMI), another good nucleophile (Table 1, entry
8).8 On the other hand, the lack of catalysis with the cobalt
complexes in our earlier attempts can presumably be due to its
smaller metal center, more effectively shrouded by the salen
ligand toward an approaching DNO molecule. The effective
ionic radii of six-coordinate, low-spin CoIII and CrIII are 54.5
and 61.5 pm, respectively.15 Considering spherical volume,
which is of a cubic function, CoIII is nearly a third smaller than
the size of CrIII. Although we note that increased catalyst
loading plays a role in enhancing polymer production, as seen
with both the CoIII catalyst, 1 (our results vs Lu’s, vide supra),
and the CrIII catalysis (Table 1, entries 3 vs 7), the trend in
Cocatalyst Effects. We noticed an odd occurrence for the
catalysis carried out with the binary salenCrIII catalyst/
cocatalyst systems. Specifically, there seemed to be a large
cocatalyst effect on the resulting product ratio based on the
identity of the nucleophile and the amount utilized in the
reaction (Table 1, entries 4−7). More PDNC was obtained
when employing PPNCl vs PPNN3 but changing to 2 equiv of
the cocatalysts drastically increased the cyclic product with less
than 5% polymer selectivity seen for two azides. Figure 4
depicts this observation more clearly. The role of a cocatalyst is
to help initiate the ring-opening of an epoxide, and the addition
of 2 equiv can also help accelerate the copolymerization, as was
shown previously in a CHO/CO2 study.13 On the basis of the
data at hand, we believe this trend originates as a result of a
steric hindrance from the catalyst. It is now much harder for a
bulkier DNO, vs a PO or CHO molecule, to access the metal
center surrounded by four t-butyl groups in catalyst 4. In fact,
we have shown that a highly sterically encumbered catalyst
bearing six t-butyl groups, including on the R1 and R2 positions
of the salen backbone, shows little activity for CHO/CO2
coupling.14
Table 1. Copolymerization of 1,4-Dihydronaphthalene Oxide and CO2 with (salen)CrIII Catalysts
a
b
c
c
d
e
f
cat./cocat. [loading]
CO2 (bar) T (°C) [solvent] time (h) % conv. % PDNC TOF (h−1
)
Mn (g/mol) [theor Mn]
PDI
1.06
Tg (°C)
g
1
4/TBAN3 [1:2:250]
4/TBAN3 [1:2:250]
4/TBAN3 [1:2:250]
4/PPNCl [1:1:500]
4/PPNN3 [1:1:500]
4/PPNCl [1:2:500]
4/PPNN3 [1:2:500]
5/NMI [4:9:1000]
6/PPNCl [1:1:500]
6/PPNN3 [1:2:500]
7/PPNCl [1:1:500]
8 [1:750]
45
30
25
30
30
30
30
55
30
30
30
25
25
70 [toluene]
6
6
37
36
47
24
32
41
54
20
25
36
17
58
17
47
45
67
78
61
28
4
16
15
19
8
2600 [7900]
4000 [14 800]
3500 [17 500]
119
2
3
4
5
6
70 [CH2Cl2]
70
6
1.06
1.07
128
70
15
14
14
14
12
14
18
14
15
7
70
11
15
20
3
70
7
70
h
8
80 [toluene]
< 1
76
5
9
70
70
70
9
3000 [18 200]
1.09
1.11
10
11
12
13
10
6
60
93
83
i
100
100
29
23
5300; 6700 [76 200]
130; 136
8 [1:1000]
3000 [25 300]
a
b
Loading ratio depicts catalyst/cocatalyst/DNO, respectively and catalyst/DNO for last two entries. 1 mL solvent per 1 g of DNO, except entry 8,
c
d
1
which used 1.6 mL Tol/g DNO. Determined using H NMR spectroscopy. Turnover frequency of DNO to products (PDNC and DNC) as
determined via H NMR spectroscopy. Determined via gel permeation chromatography in THF. Theoretical Mn = (M/I) × (% conversion) ×
(MW of DNC). TBAN3 used for its solubility in toluene Data based on isolated yields from ref 8. NMI is N-methyl imidazole. A low-MW
e
f
1
g
h
i
polymer sample was recovered from the CH2Cl2/MeOH workup solution after initial precipitation of the high-MW polymer.
5423
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Figure 5. Cartoon representation of the increased nucleophilic competition of a cocatalyst due to the steric hindrance felt by an approaching DNO.
The gray spheres depict the t-butyl groups of catalyst 4.
product selectivity because of steric hindrance cannot be ruled
out.
catalyst 4 (Table 2, entries 2−5; Figure 7). Lower catalyst
loadings (0.13%) show reduced activity but maintain moderate
to good polymer selectivity (Table 2, entries 6, 7). These
results suggest that as we expand our epoxide toolset toward
bulkier monomers for producing polycarbonates with high
thermal resistance, the catalysts we employ need to evolve, as
well, particularly taking steric positioning into consideration.
With the use of their biphenol-linked dinuclear CoIII catalyst,
3, Lu and co-workers were able to prepare PDNC with high
activity and selectivity. For example, at a loading of 0.2% and 2
equiv of PPNDNP as cocatalyst, they observed a TOF of 125
h−1, yielding only the copolymer after 4 h at 25 °C.9 A TOF of
900 h−1 was also noted for a 30 min reaction at an elevated
temperature of 50 °C. In a prior work, they have shown the
copolymerization with catalyst 3 proceeds via an intramolecular
bimetallic mechanism.17 This involves an alternating chain
growth through the carbonate attack of the copolymer from
one metal center onto an activated epoxide on the next metal.
This cooperative nature gives 3 its superior catalytic rate and
polymer selectivity as compared with a monometallic
mechanism, as seen with 1. Important to note here is the
ability of 3 to accommodate both the growing polymer chain
and an epoxide displaying its open environment with very little
steric impact. In fact, on the basis of experimental and
theoretical results, it was observed that the substituent size on
the dinuclear CoIII complex and the width of catalytic cavity
(influenced by the linker utilized) clearly dictated the activity of
CHO/CO2 copolymerization (Figure 8).17 The t-butyl groups
in the 5-positions are found to be remote from the catalytic
centers, whereas when the substituents on the 3-positions are
changed from an H to t-butyl to adamantyl groups, the TOFs
drop from 1409 to 1269 to 173 h−1. The activities mentioned
go along with the biphenol linker, which displays a fairly wide
cavity arising from the large endo phenol−phenol dihedral
angle of about 140° (Co−Co distance ≈ 8.2 Å). When a
binaphthol bridge is used instead, the opening of the catalytic
cavity is reduced as a result of an endo naphthol−naphthol
dihedral angle of no more than 90° (Co−Co distance ≈ 6.5 Å).
Consequently, the resulting activity is poorer with TOFs
around 50 h−1 for PCHC production.17
Steric Effects. If the observed cocatalyst effect is indeed a
reflection of the sterics around the metal center, then a better
polymer selectivity and activity should be achieved by using
(salen)CrIII complexes bearing smaller substituents. For this
purpose, we used catalyst 6 with smaller methoxy groups in the
5-positions with an equivalent of PPNCl, and the resulting
polymer selectivity was now 15% higher when compared with
catalyst 4 (Table 1, entry 9 vs 5; note: catalyst 6 has a
coordinated azide), but if the cocatalyst was switched to 2 equiv
of PPNN3, it quickly shut down polymer production, with
PDNC selectivity no greater than 5% (Table 1, entry 10). This
indicated that the change in the steric environment is slight
when reducing the substituent size in the 5-positions alone.
Catalyst 7 with hydrogens vs methoxy groups and an ethylene
vs cyclohexene backbone did not show any improvements,
either (Table 1, entry 11). Replacing all four t-butyls with
smaller substituents is not a good choice because it results in a
poorer solubility of the chromium catalyst in the epoxide
medium, thus leading to poorer activity.14 Facing this limitation
with (salen)CrIII complexes, we chose to look at a different
ligand altogether for optimizing the polymerization condition.
A tetramethyltetraazaannulene chromium complex, (tmtaa)-
CrIIICl, 9 was shown to be highly active (TOF = 1500 h−1) for
the copolymerization of CHO with CO2 by our group in
2007.16 The less sterically congested and planar motif of this
azaannulene ligand seemed to be a suitable candidate in
producing PDNC with high selectivity (Figure 6). Upon
Figure 6. Tetraazaannulene-derived catalyst, (tmtaa)CrIIICl, 9 (left),
vs the sterically crowded catalyst, (salen)CrIIICl, 4 (right).
Computational Insight. Additional understanding on the
formation of the 1,4-dihydronaphthalene carbonates was
obtained with the help of computations. Thermodynamically,
CHO and DNO behave similarly, having nearly the same
calculated enthalpies and free energies of reactions with CO2 to
give poly- and cyclic carbonates (Scheme S1). When it comes
to kinetics, PCHC has the highest free energy barrier among all
polycarbonates toward carbonate backbiting to produce cis-
CHC.6 Indeed, very little cyclic byproduct is seen exper-
imentally, which has made CHO an initial benchmark for
performing a catalytic run at 0.2% loading of 9 with an
equivalent of PPNCl, we found nearly 90% polymer selectivity
with three times the activity of the binary (salen)CrIII complex,
3 (Table 2, entry 1 vs Table 1, entry 4). The resulting PDNC
also had a MW of 5800 g/mol with a Tg of 136 °C, comparable
to that obtained with the bifunctional (salen)CrIII catalyst, 8. A
series of experiments varying the cocatalyst and their loading
with 9 further showed that although there was more cyclic
product with 2 equiv, it was not a drastic change, as seen with
5424
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Table 2. Copolymerization of DNO and CO2 with (tmtaa)CrIIICl, 9
a
b
b
c
d
e
cat./cocat. [loading]
CO2 (bar) T (°C) time (h) % conv.
% PDNC
TOF (h−1
)
Mn (g/mol) [theor Mn]
PDI
Tg (°C)
1
2
3
4
5
6
7
9/PPNCl [1:1:500]
9/PPNCl [1:1:500]
9/PPNN3 [1:1:500]
9/PPNCl [1:2:500]
9/PPNN3 [1:2:500]
9/PPNCl [1:1:750]
9/PPNN3 [1:1:750]
30
30
30
30
30
30
30
70
70
70
70
70
70
70
14
7
63
40
42
44
55
27
23
89
84
85
58
71
80
47
23
28
30
32
39
15
13
5800 [53 400]
4700 [31 000]
4600 [34 100]
1.11
1.10
1.10
136
7
7
7
4500 [36 500]
3800 [31 226]
1.11
1.09
14
14
a
b
c
Loading ratio depicts catalyst/cocatalyst/DNO, respectively. Determined using 1H NMR spectroscopy. Turnover frequency of DNO to products
d
e
(PDNC and DNC) as determined via 1H NMR spectroscopy. Determined via gel permeation chromatography in THF. Theoretical Mn = (M/I) ×
(% conversion) × (MW of DNC).
that the only energy minimum conformation observed is boat-
shaped. Hence, the free energy barrier for carbonate backbiting
is lower for PDNC at 22.7 kcal/mol. This lower barrier may
also explain why some cis-DNC (the only isomer seen
experimentally) is produced even under the best catalytic
conditions thus far with CrIII complexes that work at higher
temperatures.
A Bifunctional Azaannulene Ligand. The highest
molecular weight of PDNC isolated in the current study is
6700 g/mol, displaying a Tg of 136 °C. Comparing the Tg
values to those of low-MW PIC shows PDNC to be greater by
∼5 °C because of its increased bulk (Figure 10). The true
measure of a polymer’s thermal transition is past 10 000−
15 000 g/mol, whereby its Tg becomes independent of the
MW. Using catalyst 3, this was shown to be 150 °C for PDNC.9
Figure 7. Cocatalyst effect on product selectivity for CO2/DNO
With a similar interest in making higher MWs using CrIII
coupling with (tmtaa)CrCl, 9.
complexes, we began to look into newer and effective ligand
designs. We decided to prepare a tetraazaannulene ligand with
ammonium side arms, much like the salen bifunctional ligands.
With less sterics and the added benefit of the pendant arms that
prevent polymer backbiting, a CrIII or even a CoIII complex
bearing this ligand could prove to be an effective catalyst in
producing high-MW PDNC.
Another advantage of a bifunctional tetraazannulene
chromium(III) catalyst would be the range of epoxides that
could now be utilized. Previously, we had shown catalyst 9,
testing new catalysts for CO2 copolymerization reactions. This
is the result of an endergonic conformational change of its
cyclohexene backbone from a chair to boat conformation (Δ =
4.7 kcal/mol) prior to achieving its backbiting transition state
(Δ = 21.1 kcal/mol). This gives PCHC an overall barrier of
25.8 kcal/mol; other aliphatic polycarbonates exhibit 18−20
kcal/mol.6 A sketched representation of the conformations are
indicated in Figure 9. In the case of PDNC, the benzene ring
fused to the cyclohexene ring reduces the latter’s flexibility such
Figure 8. Illustration of dinuclear CoIII complexes and its catalytic cavity (top). Higher activity observed when the cavity is more open and bearing
smaller substituents.17
5425
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Figure 9. Sketched representations: (a) Total energetic barrier of 25.8 kcal/mol for carbonate backbiting of PCHC from its lowest energy chair
conformation. (b) With the boat conformation as the energy minimum, PDNC has a lower backbiting barrier of 22.7 kcal/mol. Free energies
calculated using the CBS-QB3 composite method.6 The anionic polymer chain, X, is modeled as methyl carbonate.
dihalo compounds. Although we did not have success with the
dichloro derivative, reaction with the dibromo compound
proceeds smoothly, giving us a diammoniumtetraazaannulene
(dataa) ligand in a high yield.
The overall synthesis is relatively quick (5 steps) without any
tedious purification as otherwise seen with the preparation of
the asymmetrical bifunctional salen ligands.20 The presence of
more than one pendant ammonium arm on the ligand for a
bifunctional catalyst has been previously shown to be highly
active, as well.21 This diammonium ligand, dataa, has been fully
characterized; a preliminary crystal structure is presented in
Figure S1. There is also room for modification of the length of
the pendant arms (furan vs pyran derived aldehyde), the
electronics (substituted OPDs), and the bulkiness of the onium
salt (various tertiary amines), which will make it desirable for
Figure 10. Tg vs MW comparison of PDNC and PIC.
optimization studies. Our preliminary work shows the
(tmtaa)CrIIICl, to be successful in producing polycarbonates
with high selectivity and activity for CHO and 4-vinyl-CHO,
but other epoxides such as PO and styrene oxide (SO)
coordination of dataa onto a cobalt center by refluxing with
Co(OAc)2·4H2O in DMF to give the (dataa)CoII complex.
Thus far, we have been unable to oxidize the (dataa)CoII
produced mostly the cyclic carbonate product.16 This is a result
complex by various means; however, further work on the
of a more electron-rich metal center bearing four nitrogen
preparation of CoIII and CrIII complexes and their use in
donors. The growing anionic polymer chain is thus not as
catalysis for copolymerizing various epoxides with CO2 is being
tightly bound, leading to a higher propensity to backbite.
pursued.
Electrostatic interactions with the ammonium side arms of a
bifunctional catalyst can then mitigate this process. CHO is
unaffected because it is reticent toward backbiting, as described
in the previous section.
CONCLUDING REMARKS
■
Our renewed interest in exploring the CO2 coupling reaction
with 1,4-dihydronaphthalene oxide was in part due to the
demanding but nevertheless a successful catalytic copolymer-
ization observed with a related bicyclic monomer, indene
oxide.3,4 Results from the latter also indicated advantages of
using 1,4-dihydronaphthalene oxide over indene oxide. An
earlier attempt at making poly(1,4-dihydronaphthalene carbo-
nate) involved the use of a binary (salen)CrIII catalyst/
cocatalyst system optimized at that time for making poly-
(propylene carbonate) and poly(cyclohexene carbonate).1f,8
This mostly afforded the cyclic byproduct, cis-1,4-dihydronaph-
thalene carbonate, which we assumed to be the result of an
electronic difference when compared with its simpler analog,
cyclohexene oxide, that primarily provides the copolymer. With
a more detailed study, we have shown here that poly(1,4-
dihydronaphthalene carbonate) can indeed be produced with
high selectivity using chromium(III) catalysts.
The first step toward preparing this new ligand involves the
synthesis of a dihydroxytetraazaannulene, dhtaa, compound via
the ring-opening and condensation reaction of 3,4-dihydro-2H-
pyran-5-carbaldehyde and o-phenylenediamine (OPD) as
shown previously by Breitmaer and co-worker (Scheme 3).18
This results in a tetraazaannulene bearing two propyl arms with
an OH group at the end, which can be further transformed for
our purposes. Mesylation of this dihydroxy compound provides
the dimesylatetetraazaannulene (dmtaa) in near quantitative
yields. We found the use of ionic liquids, 1-butyl-3-
methylimidazolium chloride and bromide salts, ([BMIM]X; X
= Cl, Br) to successfully convert the dimesylate compound into
the dichloro- and dibromotetraazaannulene, dctaa and dbtaa, as
per the protocol developed by Chae and co-workers for
transformation of sulfonate esters.19 Single crystals of dctaa
suitable for X-ray diffraction were grown from the slow cooling
of a hot MeCN solution; its structure is shown in Figure 11.
The final step involves the alkylation of NEt3 in DMF with the
The formation of the copolymer is highly dependent on the
steric environment provided by the ligands in a catalyst. A
5426
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
Scheme 3. Five-Step Synthesis of a Diammonium Tetraazaannulene Ligand (dataa)
(salen)CrIIICl, with polymer selectivity reaching 90%. The
competitive binding of the cocatalyst is also subdued, reflecting
the more spacious catalyst. When utilizing a bifunctional
salenCrIII complex, copolymer selectivity of >90% is achieved,
but the molecular weights remain low, around 6500 g/mol.
Computations indicate that poly(1,4-dihydronaphthalene car-
bonate) is somewhat more susceptible to carbonate backbiting,
with a lower energy barrier (22.7 kcal/mol) than poly-
(cyclohexene carbonate) (25.8 kcal/mol). This is a result of
the fused benzene ring of the former, which prevents the
cyclohexyl ring from adopting the lower-energy chair
conformation.
We have also developed a new bifunctional tetraazaannulene
ligand that can be easily prepared and further optimized.
Supporting two pendant ammonium arms and having less steric
hindrance with respect to salen, this ligand can yield a CrIII or
CoIII catalyst that could display enhanced activity and polymer
selectivity. This may then prove fruitful for producing high-
molecular-weight polycarbonates from sterically demanding
monomers, such as 1,4-dihydronaphthalene oxide, and realizing
high thermally resistant properties in CO2 derived polycar-
bonates.
Figure 11. X-ray crystal structure of dctaa. Thermal ellipsoids shown
at 50% probability level.
bulkier monomer, 1,4-dihydronaphthalene oxide, now faces a
large steric hindrance at a CrIII center surrounded by a salen
ligand bearing four t-butyl groups. As a result, its chain growth
is inhibited because of an increased preference for displacement
by the coordination of the smaller cocatalysts. The displaced
anionic polymer chain then proceeds to backbite, yielding cyclic
carbonate. Increasing the cocatalyst ratio to 2 equiv or
employing more nucleophilic anions produces cyclic carbonates
almost exclusively.
EXPERIMENTAL SECTION
■
Materials and Methods. Unless otherwise specified, all
syntheses and manipulations were carried out on a double-
manifold Schlenk vacuum line under an atmosphere of argon or
in an argon-filled glovebox. Tetrahydrofuran was purified by an
MBraun Manual Solvent Purification system packed with Alcoa
F200 activated alumina desiccant. Triethylamine (EMD) was
freshly distilled from CaH2 prior to use. Naphthalene (Aldrich);
t-BuOH (Alfa Aesar); m-chloroperoxybenzoic acid, mCPBA
The use of a less sterically demanding tetramethyltetraa-
zaannulene catalyst, (tmtaa)CrIIICl, increases the catalytic
activity to 3-fold (23 vs 8 h−1) in comparison with
5427
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
(Aldrich); 1-butyl-3-methylimidazolium chloride and bromide
salts, [BMIM]X, X = Cl, Br (Alfa Aesar); methanesulfonyl
chloride (Acros); and dry N,N-dimethylformamide (Aldrich)
were used as received. (R,R)-N,N′-bis(3,5-di-tert-butyl salicyli-
dene)-1,2-cyclohexane diaminochromium(III) chloride,
(salen)CrCl, was purchased from Strem. Other (salen)CrCl
derivatives;14 (salen)CoDNP;22 bifunctional salen CrIII and
CoIII catalysts,20 [PPN]DNP23 and [PPN]N3,24 were prepared
as per the literature. The tetramethyltetraazaannulene CrIII
complex, (tmtaa)CrCl, was synthesized per a modification of
Cotton’s procedure.16 6,13-Bis(3-hydroxypropyl)-dibenzo[b,i]-
[1,4,8,11]tetraazacyclotetradecine or dihydroxytetraazaannu-
lene (dhtaa) was prepared starting from 3,4-dihydro-2H-
pyran.18,25 Research grade 99.999% carbon dioxide supplied
in a high-pressure cylinder and equipped with a liquid dip tube
was purchased from Airgas. The CO2 was further purified by
passing through two steel columns packed with 4 Å molecular
sieves that had been dried under vacuum at ≥200 °C.
Measurements. NMR spectra were recorded on either a
Varian INOVA 300 or 500 MHz spectrometer. Infrared spectra
were obtained on a Bruker Tensor 27 FTIR spectrometer.
Elemental analyses were determined by Atlantic Microlab
(Norcross, GA). Molecular weight determinations (Mn and
Mw) were carried out with a Malvern Modular GPC apparatus
equipped with ViscoGEL I-series columns (H + L) and a model
270 dual detector consisting of RI and light scattering detectors.
Tg’s were measured using a Mettler Toledo polymer DSC
equipped with a liquid nitrogen cooling system and 50 mL/min
purge of dry nitrogen gas. Samples (∼8 mg) were weighed into
40 μL aluminum pans and subjected to two heating cycles. The
first cycle covered the range from 25 to 200 °C at 10 °C/min
and was then cooled to 0 °C at −10 °C/min. Midpoint Tg data
were obtained from the second heating cycle, which ranged
from 0 to 300 °C at a heating rate of 5 °C/min.
(80 mL), and the biphasic medium was allowed to stir at room
temperature for 4 h. The solution was chilled again and treated
with additional CHCl3 (100 mL) and mCPBA (14.0 g, 0.081
mol). Continued stirring for 17 h resulted in the complete
conversion of starting material, as seen via TLC. The reaction
mixture was transferred to a 1000 mL Erlenmeyer flask, and the
acid generated was neutralized by addition of more NaHCO3
solution (70 mL), stirring until bubbling ceased. Excess
mCPBA was then destroyed with a saturated sodium thiosulfate
solution (150 mL, added in 50 mL increments and tested with
starch paper). Finally, distilled water (50 mL) was added, the
mixture was stirred, and the organic layer was allowed to
separate. The organic layer was further washed with water (3 ×
70 mL) in a separation funnel, dried with anhydrous MgSO4,
and filtered to give a clear yellow solution. Following removal
of solvent, the crude epoxide was purified via flash
chromatography with gradient elution of ethyl acetate in
hexane on a silica column using Teledyne ISCO’s CombiFlash
Rf 150. DNO was obtained as a pale yellow solid (14 g) with a
total yield of about 50%. Characterizations match that
8
1
previously reported. mp = 40−42 °C. NMR in CDCl3: H δ
3.15−3.40 (m, 4H, CH2), 3.50 (s, 2H, CH), 7.05−7.27 (m, 4H,
aromatic H); 13C{1H}: δ 29.7, 51.7, 126.5, 129.2, 131.4. Anal.
Calcd for C10H10O: C, 82.16; H, 6.89. Found: C, 82.25; H,
6.90.
Poly(1,4-dihydronaphthalene carbonate) (PDNC):
Representative Coupling Reaction of CO2 and 1,4-
Dihydronaphthaleneoxide. A stir bar, (tmtaa)CrCl (3.0
mg, 7.0 μmol), PPNCl (3.8 mg, 6.6 μmol), and 1,4-
dihydronaphthalene oxide (0.500 g, 3.42 mmol, 500 equiv)
were added to a 12 mL stainless steel autoclave reactor that had
been previously dried at 150 °C for 6 h. The reactor was
pressurized to 27 bar and heated to 70 °C in an oil bath with
magnetic stirring. After 7 h, the reactor was cooled in an ice
bath and depressurized slowly. Sublimated epoxide was
observed on the inside of the reactor cap, but it was very
2,3-Epoxy-1,2,3,4-tetrahydronaphthalene or 1,4-Di-
hydronaphthalene Oxide (DNO). The first step involving
reduction of naphthalene was prepared per the literature:11
Sodium (11.6 g, 0.504 mol) as small cut and flattened pieces
was added over a period of 15 min to a solution of naphthalene
(25.0 g, 0.195 mol) in THF (250 mL). After stirring for 30 min,
the blue solution was chilled in an ice bath and treated
dropwise with t-BuOH (47 mL, 0.491 mol) dissolved in THF
(50 mL) over the next 20 min. The solution was allowed to stir
for 2.5 h, and the THF was removed in vacuo after removing
the excess sodium pieces. The resulting pinkish red oil was
taken into 150 mL of Et2O and washed with distilled H2O (5 ×
100 mL) until the organic layer turned clear yellow. Caution:
The first addition of water needs to be done very slowly as the
flask heats up, causing Et2O to boil when the salt dissolves. The
organic layer was further washed with brine (3 × 50 mL) and
dried over anhydrous CaCl2 by stirring overnight. Note: CaCl2
also removes excess t-BuOH. Filtration and removal of Et2O
1
minimal (10−30 mg). A H NMR spectrum of the reaction
mixture was taken immediately. The crude mixture was
dissolved in CH2Cl2 and added dropwise to an acidified
methanol (10% HCl) solution to precipitate the polymer. The
polymer was collected via filtration and purified twice more by
precipitation from CH2Cl2/MeOH. The pale yellow to off-
white PDNC was dried in vacuo before further analysis by GPC
and DSC. Note: A 50/50 mix of C6D6 and CDCl3 was used to
obtain a well-resolved NMR spectrum to determine the
polymer/cyclic carbonate ratio in the crude mixture. The
peaks of both species overlap in CDCl3, making it harder to do
so, and using only C6D6 underestimates the amount of polymer
because it is not completely soluble in benzene. See Figure S2.
1
NMR in CDCl3: H δ 3.05 (br s, 2H, CH2), 3.40 (br s, 2H,
CH2), 5.23 (br s, 2H, CH), 7.05−7.31 (m, 4H, aromatic H);
13C{1H}: δ 31.3, 73.7, 126.5, 128.6, 131.7, and 153.5; 1H NMR
(C6D6/CDCl3): δ 2.75 (br s, 2H, CH2), 3.13 (br s, 2H, CH2),
5.06 (br s, 2H, CH), 6.70−7.00 (m, 4H, aromatic H). IR
(CH2Cl2): 1753 cm−1. Anal. Calcd for (C11H10O3)n: C, 69.46;
H, 5.30. Found: C, 69.36; H, 5.48.
1
give a viscous yellow oil. H NMR on the crude product
showed ∼80% 1,4-dihydronaphthalene, and the rest is made up
of the 1,2-isomer and tetralin. The product composition varies
depending on the amount of excess sodium and reaction time
involved, but the crude product can be used for the next step
without additional purification.
Epoxidation of 1,4-dihydronaphthalene was carried out in a
manner similar to a literature preparation:12 mCPBA (17.5 g,
0.101 mol) was added slowly to an ice-cooled solution of 1,4-
dihydronaphthalene (16 g, 0.123 mol) in CHCl3 (100 mL). It
was then treated with a saturated aqueous NaHCO3 solution
cis-1,4-Dihydronaphthalene Carbonate (cis-DNC). Pure
cis-DNC was obtained after removal of the acidified methanol
filtrate from a polymer workup under reduced pressure and
recrystallization of the residue from methanol/hexane.
Characterization of the white product matches that previously
8
1
described. mp = 155 °C. H NMR in CDCl3: δ 2.90 (s, 2H,
CH2), 3.15 (s, 2H, CH2), 5.19 (s, 2H, CH), 7.20−7.30 (m, 4H,
5428
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
aromatic H); in C6D6/CDCl3: δ 2.22 (s, 2H, CH2), 2.62 (s, 2H,
CH2), 4.36 (s, 2H, CH), 6.75−7.05 (m, 4H, aromatic H). IR in
CH2Cl2: 1807 (s), 1794 (sh); in THF: 1811 (s) cm−1. Anal.
Calcd for C11H10O3: C, 69.46; H, 5.30. Found: C, 69.47; H,
5.49.
solution was allowed stir for 3 days, after which the solvents are
removed under reduced pressure. The resulting red residue was
washed down the sides of the flask with CH2Cl2 (<5 mL) and
treated with THF (50 mL). It was stirred for a few minutes,
filtered, and dried in vacuo to give the product as a red solid
(0.462 g, 92%). NMR in DMSO-d6: 1H δ 1.17 (t, 18H,
NCH2CH3), 1.81 (m, 4H, −CH2−), 2.27 (t, 4H, −CH2−), 3.13
(m, 4H, −CH2−), 3.25 (q, 12H, NCH2CH3), 6.93 (m, 4H,
aromatic H), 7.31 (m, 4H, aromatic H), 7.91 (m, 4H, CH),
13.64 (t, 2H, NH); 13C{1H}: δ 7.31, 23.7, 28.8, 52.2, 55.6,
106.7, 114.1, 124.4, 136.9, 148.0. ESI-MS (m/z): 286.22 [(M −
2Br−)2+]. Anal. Calcd for C36H56Br2N6·3H2O: C, 54.96; H,
7.94; N, 10.68. Found: C, 54.12; H, 7.29; N, 10.54. X-ray
quality crystals were grown via vapor diffusion of methyl t-butyl
ether into a MeOH solution.
Synthesis of 6,13-Bis(3-methanesulfonylpropyl)-
dibenzo[b,i][1,4,8,11]tetraazacyclotetradecine or Dime-
sylatetetraazaannulene (dmtaa). A 500 mL round-bottom
flask was charged with dhtaa (0.955 g, 2.36 mmol) and dry
THF (250 mL). To this red mixture, NEt3 (2.0 mL, 14 mmol)
followed by methanesulfonyl chloride (1.0 mL, 13 mmol) were
added using a syringe. The reaction was stirred at room
temperature until complete conversion of the dihydroxy
compound was observed via TLC (24−32 h). The solvent
was removed under reduced pressure, and the crude mixture
washed thoroughly with distilled H2O (150 mL) to remove
triethylammonium chloride. Further washings using MeOH (50
mL) and Et2O (50 mL) and drying in vacuo gives the orange-
ASSOCIATED CONTENT
■
S
* Supporting Information
1
red product (1.28 g, 97%). H NMR (DMSO-d6): δ 1.89 (p,
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b01375.
4H, −CH2−), 2.32 (t, 4H, −CH2−), 3.19 (s, 6H, CH3), 4.24 (t,
4H, −CH2−), 6.90 (m, 4H, aromatic H), 7.28 (m, 4H, aromatic
H), 7.85 (m, 4H, CH), 13.61 (t, 2H, NH). Poor solubility
precluded the collection of 13C NMR. ESI-MS (m/z): 561.20
(MH+). mp = 198 °C (dec). Anal. Calcd for C26H32N4O6S2: C,
55.70; H, 5.75; N, 9.99. Found: C, 54.79; H, 5.86; N, 9.55.
Synthesis of 6,13-Bis(3-chloropropyl)-dibenzo[b,i]-
[1,4,8,11]tetraazacyclotetradecine or Dichlorotetraa-
zaannulene (dctaa). A 30 mL tall vial with a stir bar was
packed with dmtaa (0.600 g, 1.07 mmol), followed by [BMIM]
Cl (1.92 g, 11.0 mmol) on top. The vial was placed in a 90 °C
oil bath to let the ionic liquid melt (∼15 min). Stirring was then
begun, and after 24 h, TLC showed complete conversion of the
dimesylate to the dichloro compound. Note: If needed, some
acetonitrile can be used to rinse down the sides of the vial
during the course of the reaction to achieve complete mixing
after lowering the temperature to 80 °C. The resulting brown
sludge was taken into CH2Cl2 (125 mL), washed with 20%
aqueous NaCl solution (3 × 100 mL), and dried with
magnesium sulfate. The solvent was then removed in vacuo,
and the residue was washed with Et2O (75 mL) to give a brown
product after drying in vacuo (0.325 g, 69%). NMR in CDCl3:
1H δ 1.95 (p, 4H, −CH2−), 2.41 (t, 4H, −CH2−), 3.63 (t, 4H,
−CH2−), 6.92 (m, 4H, aromatic H), 7.05 (m, 4H, aromatic H),
7.63 (m, 4H, CH), 13.71 (t, 2H, NH); 13C{1H}: δ 29.8, 34.5,
43.9, 106.0, 113.5, 124.3, 137.5, 147.0. ESI-MS (m/z): 441.18
(MH+). mp = 224 °C (dec). X-ray quality crystals were grown
via slow cooling of a hot MeCN solution.
Synthesis of 6,13-Bis(3-bromopropyl)-dibenzo[b,i]-
[1,4,8,11]tetraazacyclotetradecine or Dibromotetraa-
zaannulene (dbtaa). Prepared in a manner similar to that
for dctaa described above by using [BMIM]Br. A brownish red
product was isolated in >60% yield. Characterizations matched
those previously reported.26 1H NMR (DMSO-d6): δ 2.02 (p,
4H, −CH2−), 2.34 (t, 4H, −CH2−), 3.55 (t, 4H, −CH2−),
6.91 (m, 4H, aromatic H), 7.28 (m, 4H, aromatic H), 7.85 (m,
4H, CH), 13.61 (t, 2H, NH). Anal. Calcd for C24H26Br2N4: C,
54.36; H, 4.94; N, 10.57. Found: C, 55.22; H, 5.23; N, 10.42.
Synthesis of 6,13-Bis[3-(N,N,N-triethylammonium)-
propyl)]-dibenzo[b,i][1,4,8,11]tetraazacyclotetradecine
Dibromide or Diammoniumtetraazaannulene (dataa). A
100 mL Schlenk flask charged with dbtaa (0.365 g, 0.688
mmol) was treated with 50 mL of DMF and NEt3 (10 mL, 72
mmol) and placed in an oil bath set at 50 °C. The dark red
ORTEP of the preliminary structure of dataa, calculated
enthalpies of reaction, and NMR spectra of PDNC and
DNC (PDF)
X-ray structural data for dctaa (CIF)
AUTHOR INFORMATION
Corresponding Author
*Phone: 979.845.2983. Fax: 979.845.0158. E-mail: djdarens@
chem.tamu.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge the financial support of the National
Science Foundation (CHE-1057743) and the Robert A. Welch
Foundation (A-0923). Special thanks to all past and current
group members who have synthesized the many catalysts used
herein. We would also like to thank Dr. Andrew Yeung for
calculating the energy barriers.
REFERENCES
■
(1) Several comprehensive reviews cover the copolymerization of
epoxides and CO2. For example, see: (a) Darensbourg, D. J.;
Holtcamp, M. W. Coord. Chem. Rev. 1996, 153, 155−174. (b) Coates,
G. W.; Moore, D. R. Angew. Chem., Int. Ed. 2004, 43, 6618−6639.
(c) Sugimoto, H.; Inoue, S. J. Polym. Sci., Part A: Polym. Chem. 2004,
42, 5561−5573. (d) Darensbourg, D. J.; Mackiewicz, R. M.; Phelps, A.
L.; Billodeaux, D. R. Acc. Chem. Res. 2004, 37, 836−844. (e) Chisholm,
M. H.; Zhou, Z. J. Mater. Chem. 2004, 14, 3081−3092. (f) Dare-
nsbourg, D. J. Chem. Rev. 2007, 107, 2388−2410. (g) Klaus, S.;
Lehenmeier, M. W.; Anderson, C. E.; Rieger, B. Coord. Chem. Rev.
2011, 255, 1460−1479. (h) Kember, M. R.; Buchard, A.; Williams, C.
K. Chem. Commun. 2011, 47, 141−163. (i) Lu, X. B.; Darensbourg, D.
J. Chem. Soc. Rev. 2012, 41, 1462−1484. (j) Lu, X.-B.; Ren, W.-M.;
Wu, G.-P. Acc. Chem. Res. 2012, 45, 1721−1735. (k) Darensbourg, D.
J.; Wilson, S. J. Green Chem. 2012, 14, 2665−2671. (l) Paul, S.; Zhu,
Y.; Romain, C.; Brooks, R.; Saini, P. K.; Williams, C. K. Chem.
Commun. 2015, 51, 6459−6479. (m) Ang, R.-R.; Sin, L. T.; Bee, S.-T.;
Tee, T.-T.; Kadhum, A. A. H.; Rahmat, A. R.; Wasmi, B. A. J. Cleaner
Prod. 2015, 102, 1−17.
(2) http://www.novomer.com/; http://empowermaterials.com/;
http://www.econic-technologies.com/; http://www.materialscience.
bayer.de/Projects-and-Cooperations/CO2-Project; Accessed June 25,
2015.
5429
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430

ACS Catalysis
Research Article
(3) Darensbourg, D. J.; Wilson, S. J. J. Am. Chem. Soc. 2011, 133,
18610−18613.
(4) Darensbourg, D. J.; Wilson, S. J. Macromolecules 2013, 46, 5929−
5934.
(5) Liu, Y.; Wang, M.; Ren, W.-M.; He, K.-K.; Xu, Y.-C.; Liu, J.; Lu,
X.-B. Macromolecules 2014, 47, 1269−1276.
(6) Darensbourg, D. J.; Yeung, A. D. Macromolecules 2013, 46, 83−
95.
(7) Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C. J. Am.
Chem. Soc. 2003, 125, 7586−7591.
(8) Darensbourg, D. J.; Fang, C. C.; Rodgers, J. L. Organometallics
2004, 23, 924−927.
(9) Liu, Y.; Wang, M.; Ren, W.-M.; Xu, Y.-C.; Lu, X.-B. Angew. Chem.,
Int. Ed. 2015, 54, 7042−7046.
(10) Naphthalene and Hydronaphthalenes. Ullmann’s Encyclopedia of
Industrial Chemistry [Online]; Wiley-VCH: Weinheim, Published
March 15, 2003. http://onlinelibrary.wiley.com/doi/10.1002/
14356007.a17_001.pub2/abstract. Accessed June 25, 2015.
(11) Menzek, A.; Altundas,
752−753.
(12) Li, R.; Jansen, D. J.; Datta, A. Org. Biomol. Chem. 2009, 7, 1921−
1930.
̧ A.; Gultekin, D. J. Chem. Res. 2003, 2003,
̈
(13) Darensbourg, D. J.; Mackiewicz, R. M. J. Am. Chem. Soc. 2005,
127, 14026−14038.
(14) Darensbourg, D. J.; Mackiewicz, R. M.; Rodgers, J. L.; Fang, C.
C.; Billodeaux, D. R.; Reibenspies, J. H. Inorg. Chem. 2004, 43, 6024−
6034.
(15) Shannon, R. D. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,
Theor. Gen. Crystallogr. 1976, A32, 751−767.
(16) Darensbourg, D. J.; Fitch, S. B. Inorg. Chem. 2008, 47, 11868−
11878.
(17) Liu, Y.; Ren, W.-M.; Liu, C.; Fu, S.; Wang, M.; He, K.-K.; Li, R.-
R.; Zhang, R.; Lu, X.-B. Macromolecules 2014, 47, 7775−7788.
(18) Hanke, R.; Breitmaier, E. Chem. Ber. 1982, 115, 1657−1661.
(19) Liu, Y.; Xu, Y.; Jung, S. H.; Chae, J. Synlett 2012, 23, 2692−
2698.
(20) (a) Darensbourg, D. J.; Tsai, F.-T. Macromolecules 2014, 47,
3806−2813. (b) Wu, G.-P.; Xu, P.-X.; Lu, X.-B.; Zu, Y.-P.; Wei, S.-H.;
Ren, W.-M.; Darensbourg, D. J. Macromolecules 2013, 46, 2128−2133.
(21) (a) Noh, E. K.; Na, S. J.; S, S.; Kim, S.-W.; Lee, B. Y. J. Am.
Chem. Soc. 2007, 129, 8082−8083. (b) S, S.; Min, J. K.; Seong, J. E.;
Na, S. J.; Lee, B. Y. Angew. Chem., Int. Ed. 2008, 47, 7306−7309.
(22) Lu, X.-B.; Shi, L.; Wang, Y.-M.; Zhang, R.; Zhang, Y.-J.; Peng,
X.-J.; Zhang, Z.-C.; Li, B. J. Am. Chem. Soc. 2006, 128, 1664−1674.
(23) Wu, G.-P.; Wei, S.-H.; Lu, X.-B.; Ren, W.-M.; Darensbourg, D. J.
Macromolecules 2010, 43, 9202−9204.
(24) Demadis, K. D.; Meyer, T. J.; White, P. S. Inorg. Chem. 1998, 37,
3610−3619.
(25) Breitmaier, E.; Ullrich, F.-W.; Potthoff, B.; Bohme, R.; Bastian,
̈
H. Synthesis 1987, 1987, 1−9.
́ ́ ́
(26) Pawlica, D.; Radic Stojkovic, M.; Sieron, L.; Piantanida, I.;
Eilmes, J. Tetrahedron 2006, 62, 9156−9165.
5430
DOI: 10.1021/acscatal.5b01375
ACS Catal. 2015, 5, 5421−5430