Synthesis and characterization of cyclopentadienylgallium amide compounds as potential single source precursors to GaN

Article abstract of DOI:10.1016/j.poly.2014.09.008

The synthesis, spectroscopic characterization, and single crystal X-ray structures of [(η⁵-C₅Me₄H)₂Ga(μ₂-NH₂)] (1), [(η⁵-C₅Me₅)₂Ga(μ₂-

Full text of DOI:10.1016/j.poly.2014.09.008

Polyhedron 84 (2014) 216–222
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Synthesis and characterization of cyclopentadienylgallium amide
compounds as potential single source precursors to GaN
a
b
a
b
a,
⇑
P. Perrotin , Bretni S. Kennon , Brendan Twamley , Joel S. Miller , Pamela J. Shapiro
a
Department of Chemistry, University of Idaho, Moscow, ID 83844-2343, USA
Department of Chemistry, University of Utah, Salt Lake City, UT 84112-0850, USA
b
a r t i c l e i n f o
a b s t r a c t
5
Article history:
The synthesis, spectroscopic characterization, and single crystal X-ray structures of [(
g
-C Me H) Ga
5
4
2
Received 13 June 2014
Accepted 9 September 2014
Available online 12 October 2014
5
5
5
(
(
l
l
2
-NH
2
)] (1), [(
g
-C
5 5
Me )
2
Ga(
2
l -NH
2
)]
2
(2), [(
g
5
-C Me
5
)Ga(Cl)(
l
2 2
-NH )]
3
(4), [(
g
-C
5
5
Me )
2
Ga
2
-N(H)Et)]
2
(5) are described. Preliminary studies of the thermal and photochemical decomposition
of compound 2 were performed to evaluate its suitability as single source precursors to GaN.
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Cyclopentadienylgallium
Gallium amide
GaN
Thermolysis
Crystal structure
⁄
⁄
1
. Introduction
[Cp Ga(
l
-NXylyl)]
2
and [Cp Ga(
l
-N
3
)N(SiMe
3
)
2
]
2
,
which were
⁄
prepared by reaction of Cp Ga with XylylN
tively [12].
3
and TMSN respec-
3
The cyclopentadienyl ligand and its derivatives have been very
important to fundamental organometallic chemistry and homoge-
neous catalysis. Most commonly these ligands are associated with
transition metals; however, cyclopentadienyl compounds of main
group metal and metalloid elements have also received consider-
able attention for their diverse and unusual structures and their
fluxional behavior [1–5]. These compounds are also of interest as
potential precursors in metal–organic chemical vapor deposition
2
. Materials and methods
2.1. General procedures
0
0
⁄
⁄
Cp
3
Ga [13], Cp
3
Ga (Cp = C
5
Me
4
H) [14], Cp Ga (Cp = C Me )
3 5 5
⁄
[
15] and [Cp
2
Ga(l
-Cl)]
2
[16] were prepared as described in the liter-
(
MOCVD) [3], as ligands for transition metals [4], as synthetic
ature. Anhydrous ammonia was purchased from Matheson TriꢀGas
and degassed by applying three freeze–pump–thaw cycles immedi-
ately prior use. Ethylamine (97%) was purchased from Aldrich, dried
precursors to organometallic complexes [5], in catalysis [6], and
as molecular precursors to materials [7].
In the work described herein, cyclopentadienyl gallium com-
3
over Na(s) and degassed multiple times to remove NH contamina-
pounds containing -NH
2
ligands were prepared as potential single
t
tion. BuLi (1.7 M solution in pentane) was purchased from Acros
Organics. Toluene, benzene, hexanes, and pentane were dried and
distilled from sodium/benzophenone immediately prior to use.
source precursors (SSPs) to GaN. High temperatures are generally
needed to break down inorganic or organometallic precursors to
GaN films and nanomaterials, limiting control over the size and
crystallinity of the GaN nanomaterials and thin films [8,9]. Since
cyclopentadienyl gallium bonds have been shown to be signifi-
cantly more labile than other Ga–C bonds as well as Ga–H bonds
3 6 6
CDCl (D, 99.8%) and C D (D, 99.5%) were purchased from Cam-
bridge Isotope laboratories and stored over 4A molecular sieves.
All compounds were handled and stored in a nitrogen-filled glove-
box and were manipulated using argon and vacuum Schlenk tech-
niques. A calibrated gas bulb was used to control the amount of
ammonia that was introduced to the reaction vessels, as described
elsewhere [17]. IR spectra of the compounds were recorded on a
Thermo Nicolet Avatar 370 FT-IR instrument. NMR spectra were
[
2,9–11], the use of cyclopentadienyl ligands in lieu of alkyl ligands
in the gallium precursors presented a possible means of lowering
the decomposition temperature of the SSP.
Prior to this work, only two cyclopentadienylgallium
compounds containing
a
Ga–N bond had been reported:
1
13
recorded on a Bruker Avance 300 MB (300 MHz H; 75.4 MHz, C)
1
13
or a Bruker Avance 500 (500 MHz H; 125 MHz C) spectrometer.
1
13
⇑
Corresponding author. Tel.: +1 781 418 1840; fax: +1 781 272 5225.
E-mail address: shapiro@netzsch.com (P.J. Shapiro).
H and C chemical shifts were referenced to solvents. Differential
scanning calorimetry (DSC) measurements were performed on a
http://dx.doi.org/10.1016/j.poly.2014.09.008
277-5387/Ó 2014 Elsevier Ltd. All rights reserved.
0

P. Perrotin et al. / Polyhedron 84 (2014) 216–222
217
TA Model Q2000 DSC instrument. Thermogravimetric analyses were
performed on a TA Model Q500 thermal gravimetric analyzer (TGA)
filtered and the precipitate was washed three times with THF
(30 mL). The combined volatiles from the filtrate were removed
under reduced pressure and the residue was washed with pentane
(
ambient to 1000 °C, ±0.1 lg) located in a nitrogen-filled glovebox.
The TGA was equipped with a Pfeiffer Thermostar GSD301T3 quad-
rupole mass spectrometer to identify gaseous products with masses
less than 300 amu. Single crystal X-ray analyses were performed on
a Bruker/Siemens SMART APEX instrument. Crystals of each com-
pound were removed from the flask and covered with a layer of
hydrocarbon oil. A suitable crystal was selected, attached to a glass
fiber and placed in the low-temperature nitrogen stream [18]. Cell
parameters were retrieved using SMART [19] software and refined
using SAINTPLUS [20] on all observed reflections. Data reduction and
correction for Lp and decay were performed using the SAINTPLUS
software. Absorption corrections were applied using SADABS [21].
Structures was solved by direct methods and refined by least squares
(3 ꢂ 30 mL), affording 2 as a fine white powder (Yield: 221 mg, 47%).
1
3 5 3 5
H NMR (CDCl , 303 K, 300 MHz): d 1.65, 1.49 (30H, s, C (CH ) ),
1
1.17 (2H, s, NH
1.51 (30H, s, C
303 K, 75 MHz): d 119.9 (C
NMR (CDCl
2
) ppm. H NMR (THF-d
(CH ), 1.37 (2H, s, NH
(CH ) 12.2 (C
8
, 303 K, 300 MHz): d 1.66,
1
3
5
3
)
5
2
) ppm. C NMR (CDCl
3
,
5
3
)
5
5
(CH N
3
)
5
) ppm. 14
3
, 303 K, 36.1 MHz): ꢁ80.2 ppm. IR (KBr disk): 3378
ꢁ1
(m), 3309 (w) cm (N–H).
Single crystals of compound 2 for X-ray analysis were grown
over 5 days from a saturated CHCl
3
solution cooled at ꢁ20 °C.
⁄
2.4. Preparation and characterization of [Cp Ga(Cl)(
l
-NH )] (4)
2
3
2
⁄
method on F using the SHELXTL program package [22].
[Cp Ga(
l-Cl)]2 (1.00 g, 1.33 mmol) was dissolved in 30 mL
2
toluene. The flask was cooled to ꢁ196 °C and NH
3
(g) (1.1 eq,
0
2
.2. Preparation of [Cp
2
Ga(
l
2
-NH
2
)]
2
(1)
2.93 mmol) was added via a calibrated gas bulb. The mixture was
allowed to warm to room temperature and stirred for 4 h. The mix-
ture was then heated at 80 °C for 16 h, during which time the solu-
tion turned yellow and a white precipitate formed. The reaction was
cooled to room temperature and filtered. The precipitate was
washed with pentane (2 ꢂ 20 mL). The combined filtrate and wash-
Using a calibrated glass bulb, one equivalent of NH
to a frozen solution of Cp
3
was added
0
3
Ga (400 mg; 0.92 mmol) in 25 mL of tol-
uene cooled at ꢁ196 °C. The reaction mixture was allowed to warm
to room temperature while stirring and was stirred overnight at
room temperature. The solution was filtered to remove any
insoluble material, and the volatiles were removed under reduced
pressure. The residue was dissolve in ca. 7 mL toluene and the
solution was stored at ꢁ20 °C for 1 week, affording crystals of com-
ings were dried under reduced pressure, affording compound 4 as a
1
white powder (Yield: 301 mg, 44%).
H NMR (CDCl3, 303 K,
13
300 MHz): d 1.76 (45H, s, C (CH ) ), 1.01(6H, bs, NH ) ppm.
5
3 5
2
C
NMR (C D , 303 K, 75 MHz): d 120.7 (C (CH ) ) 12.5 (C (CH ) )
6
6
5
3 5
5
3 5
ꢁ1
pound 1 (Yield: 36 mg, 11%) which were isolated by decantation of
ppm. IR (KBr disk): 3247 (m) cm (N–H). Single crystals of 4 for
X-ray analysis were obtained by slow evaporation of a solution of
the compound in benzene solution at room temperature in a N2
atmosphere.
1
the solvent and further dried under vacuum. H NMR (C
6
D
6
, 303 K,
H), 1.86 (s,
, 303 K,
5 MHz): d 132.3 (s), 131.3 (s), 58.2 (s), 14.7 (s), 12.0 (s) ppm. IR
3
1
7
00 MHz): d 2.90 (s, 2 H, C
5
Me
4
H), 1.92 (s, 12H, C
) ppm. 13C NMR (C
D
6 6
5 4
Me
2H, C Me H), 0.84 (bs, 1.7H, NH
5
4
2
⁄
ꢁ1
2.5. Preparation and characterization of [Cp Ga(
l-NHEt)]2 (5)
(
KBr disk): 3442 (m), 3309 (m) cm
(N–H). Single crystals of
2
compound 1 for X-ray analysis were grown over a week from a
saturated toluene solution cooled at ꢁ20 °C.
Using a calibrated glass bulb, 1 equivalent of EtNH
2
was added
⁄
at ꢁ196 °C to a frozen solution of Cp
3
Ga (400 mg, 0.84 mmol) in
⁄
toluene. The solution was allowed to warm to room temperature
and stirred for 15 h. The colorless solution was then filtered and
the volatiles were removed under vacuum. The residue was washed
2.3. Preparation of [Cp
2
2
Ga(l-NH )]
2
(2)
with pentane (3 ꢂ 10 mL) and dried under vacuum to afford 177 mg
2
.3.1. Method 1
1
(
55%) of 15 as a white powder. H NMR (C
6
D
6
, 303 K, 300 MHz): 2.89
In a glovebox, a 50 mL round bottom flask was charged with
⁄
2 3 5 3 5
(4H, t, J = 7.2 Hz, NCH CH ), 1.80 (30H, s, C (CH ) , 1.12 (6H, q,
0
.590 g of Cp
3
Ga (1.24 mmol). The flask was then connected to a
1
3
J = 7.2 Hz, NCH
(CH ), 40.4 (NCH
IR (KBr disk): 3277 (m) cm (N–H). Anal. Calc. for C44
8.77; H, 9.44; N, 3.65. Found: C, 69.16; H: 9.60; N: 3.52%. Single
2
CH
3
). C NMR (CDCl
3
, 303 K, 75 MHz): d 120.9
CH ), 13.7 (C (CH ) ppm.
72Ga : C,
calibrated gas volume. On a Schlenk line 25 mL of benzene was con-
densed onto the solid at ꢁ78 °C. The solid was allowed to dissolve at
room temperature and the flask was cooled to ꢁ196 °C and the gas
above the frozen solution was evacuated. One equivalent of ammo-
nia gas was measured into the gas bulb and then condensed onto the
reaction flask. The reaction mixture was then allowed to warm up
gradually to room temperature and stirred for 3 h. A white precipi-
tate formed and the solution became slightly yellow. The solution
was filtered, and the precipitate was dried and washed twice with
(
C
5
3
)
5
2
CH ), 19.6 (NCH
3
2
3
5
3 5
)
ꢁ
1
H
2 2
N
6
crystals of 15 were grown over 5 days from a saturated toluene solu-
tion at ꢁ20 °C.
3. Results and discussion
20 mL pentane. Drying under vacuum offered 0.332 g of 2 (75%) as
3.1. Synthesis of cyclopentadienyl-gallium amide compounds
a fine white powder.
Our investigations started by following the reaction of Cp
with an equivalent of ammonia in CDCl , C or toluene-d
8
3
Ga
by
2
.3.2. Method 2
3
6
D
6
1
In a similar manner as described in Section 2.3.1, ammonia
H NMR. Regardless of the solvents used, the elimination of
approximately 1.5 equivalents of cyclopentadiene was observed
along with the formation of intractable, presumably polymeric
materials. Since increasing the steric bulk of the cyclopentadienyl
ligands should improve the solubility and prevent oligomerization
of the ammonolysis product, gallium substrates with methyl-
(
1.33 mmol) was measured into a gas bulb and then condensed
⁄
into
a
flask containing
a
frozen solution of [Cp
2
Ga(
l
-Cl)]
2
(
500 mg, 0.665 mmol) in 50 mL THF that was cooled at ꢁ196 °C.
The reaction mixture was allowed to warm to room temperature
and stirred for 2.5 h. An atmosphere of argon was admitted to the
flask, and the solution was cooled to ꢁ78 °C. A 1.7 M pentane solu-
substituted cyclopentadienyl rings were examined.
t
0
tion of BuLi (780
l
L, 1.33 mmol) was added dropwise to the flask
Reaction of Cp
3
Ga with NH
3
(1:1) in toluene, followed by work
with a microsyringe over 5 min. The mixture was allowed to warm
gradually to room temperature and was stirred overnight, affording
a colorless solution with a white precipitate. The solution was
up and slow recrystallisation from toluene at ꢁ20 °C yielded
0
[Cp
2
Ga(l
-NH )]
2 2
(1) (Eq. (1)). Although the reaction to form
compound 1 was reproducible and essentially quantitative when

2
18
P. Perrotin et al. / Polyhedron 84 (2014) 216–222
⁄
⁄
2
GaCl]
Scheme 1. Cp
3
Ga and [Cp
2 3
reactions with NH .
monitored in an NMR tube by 1H NMR spectroscopy, isolation of
the compound proved challenging and yields were invariably low.
0
0
0
Cp Ga þ NH
3
ꢁ! 0:5½Cp GaNH
2
ꢃ þCp H
ð1Þ
3
2
2
ð1Þ
The use of pentamethylcyclopentadienyl gallium species pro-
vides a more reliable route to gallium amide compounds as shown
⁄
in Scheme 1. Reaction of Cp
3
Ga with ammonia in toluene afforded
⁄
[
Cp
2
Ga(
l
-NH
2
)]
2
(2) in 75% isolated yield. One drawback to this
synthetic pathway is the low yields of the Cp
presented an appealing alternative since
it is prepared in high yield (P70%) and the all of the precious Cp
⁄
3
Ga starting material
⁄
(
ca. 40%). [Cp
2
Ga(
l
-Cl)]
2
⁄
ligand would be conserved in forming target compound 2. Attempts
⁄
to prepare compound 2 by reacting [Cp
2
Ga(
l
-Cl)]
2
with sodium
amide in THF or Et2O were unsuccessful, yielding only a complex
⁄
mixture of products. Instead, [Cp
2
Ga(
l
-Cl)]
2
was reacted with an
⁄
equivalent of ammonia to afforded the adduct [Cp
2
3
GaCl(NH )] (3),
which was not isolated but was characterized by 1H NMR spectros-
copy (d 1.76 (s, 30H, C Me ), 1.08 (bs, 2.8H, NH ) ppm). An in situ
5
5
3
dehydrohalogenation reaction of 3 with t-butyllithium afforded
compound 2 with an isolated yield of 47%. In the absence of t-butyl-
⁄
lithium, compound 3 eliminated Cp H slowly (days) at room temper-
⁄
ature or in a few hours at 80 °C to form [Cp (Cl)Ga(
2
l-NH )]
3
(4). If
not isolated, compound
4 formed an unidentified precipitate,
possibly a higher aggregate species, over time.
3.2. Spectroscopic characterization of cyclopentadienyl-gallium amide
compounds
2
Only the presence of the NH signal at 0.84 ppm for compound
1
1
Cp
differentiated its H NMR spectrum from that of its precursor,
0
Ga. The room temperature 1H NMR spectrum of 2 in CDCl
exhibited two broad signals for the Cp methyl substituents at
3
3
⁄
1
.65 ppm and 1.49 ppm, with a broad peak at 1.17 ppm for the
two protons of the amido ligand, indicative of a fluxional process
in solution. A kinetic analysis or freezing out of the fluxional pro-
cess by variable temperature 1H NMR was precluded by the poor
2 2
solubility of compound 1 in CD Cl and THF-d8, not to mention
the coordinating nature of THF and the interference of its residual
⁄
proton signal (1.78 ppm) with that of the Cp methyl substituents.
2
The presence of the –NH ligand in compounds 1 and 2 was
Fig. 1. Thermal ellipsoid drawings of compound 1 (top) and compound 2 (bottom)
with atom numbering schemes. Displacement ellipsoids are drawn at the 30%
also confirmed by the presence of N–H stretching bands
ꢁ
1
ꢁ1
(
3
1:
m
N–H(sym) = 3442 cm
,
m
N–H(asym) = 3382(m) cm , 2:
m
N–H(sym)
=
probability level. Hydrogen atoms, except for those in the NH
omitted for clarity.
2
moieties, have been
378 cmꢁ1,
m
N–H(asym) = 3309 cm ) in their infrared spectra.
ꢁ1

P. Perrotin et al. / Polyhedron 84 (2014) 216–222
219
in response to the increasing substitution of the cyclopentadienyl
ring and the amide ligand. Conversely, there is decreasing trend
for the Cipso–Ga–Cipso angles to accommodate the increasing N–
Ga–N bond angles.
A thermal ellipsoid drawing of the molecular structure of com-
pound 4 is shown in Fig. 3. Selected bond lengths and angles are
presented in Table 3. The compound is a trimer with
2
l amide
bridges. The geometry of the cyclic core is shown more clearly in
Fig. 4. The six membered ring with alternating Ga and N atoms
adopts a boat shape with all three chlorine ligands in axial
positions, two on the same side of the ring and one on the opposite
⁄
side. The more sterically demanding Cp ligands occupy equatorial
positions. The Ga–Cl bonds ranging from 2.198 Å to 2.217 Å are sig-
nificantly longer than the terminal Ga–Cl bonds in [Cl
2
Ga(
l
-Cl)]
2
⁄
(
2.06 Å) [23], [Cp Ga(
l
-Cl)Cl]
2
(2.12 Å) [16], but comparable to
-N )]
those in [HClGa(
2.177 Å) [25], [Cl
lGa( -N(H)SiMe )]
l-N
3
)] (avg. 2.171 Å) [24], [MeClGa(
4
l
3 4
(
2
Ga(
l
2
-N(H)SiMe
3
)]
2
(2.150 Å) [26], and [MeC-
⁄
l
2
3
2
(2.212 Å) [27]. The Cp –Ga bonds are similar
in length to those in compound 2. A side view of the cyclic core
of 4 is shown in Fig. 4. The Ga–N–Ga angles (avg. 121.7°) are signif-
icantly wider than the N–Ga–N angles (av. 104.3°). The Ga–N bond
lengths range from 1.944 Å to 1.984 Å (avg. 1.97) and fall within
Fig. 2. Thermal ellipsoid drawing of compound 5. Displacement ellipsoids are
drawn at the 30% probability level. Hydrogen atoms have been omitted for clarity.
2 2 3
the values reported for cyclotrigallazane, [Ga(l -NH )] [28]. In
3
.3. X-ray crystallographic characterization of cyclopentadienyl-
the boat-shaped six-membered ring, Ga(1), N(3), N(2), and Ga(2)
are coplanar (mean deviation from a plane: 0.0263 Å) and atoms
N(1) and Ga(3) are bent out of the plane by 0.6155 Å and
0.8982 Å, respectively.
gallium amide compounds
Thermal ellipsoid drawings of the molecular structures of com-
pounds 1 and 2 are shown together in Fig. 1 for visual comparison.
In all previously reported compounds with a [Ga(
the six-membered rings adopt a different conformation: twisted
sofa in [Me Ga( -NH )] [29], chair in [H Ga( -NH )] [30], or
planar in [ Bu Ga( -NH [31]. The conformation adopted in the
solid state by 4 appears to be influenced by both inter and intramo-
lecular hydrogen bonding. As depicted in Fig. 4, a total of 3 inter-
molecular hydrogen bonds ranging from 2.47 to 2.80 Å were
found between two molecules of 14. In addition, there is an
intramolecular hydrogen bond between Cl(2) and N(1)H (2.81 Å)
2 3
l-NH )] core,
The molecular structure of compound 5 is shown in Fig. 2. All three
1
compounds are
2
l -amide bridged dimers with
g
coordination of
2
l
2
3
2
l
2 3
t
all four cyclopentadienyl ligands. Selected bond lengths and angles
for the three structures are compared in Table 2. Crystallographic
data are listed in Table 1. Although the geometries of the three
structures are similar, there is a trend of increasing Ga–N, Ga–Ga,
and Ga–Cipso bond lengths and increasing Ga–N–Ga and N–Ga–
N bond angles from compound 1 to compound 2 to compound 5
2
l
2 3
)]
Table 1
Crystal data and structure refinement for compounds 1, 2, 4, and 5.
1
2
4
5
Empirical formula
Formula weight
T (K)
C
36
H
56Ga
N
2 2
C
40
H
64Ga
N
2 2
C
30
H
51
C
l3Ga
3
N
3
C
44 2 2
H72Ga N
656.27
86 (2)
712.37
86 (2)
769.25
87 (2)
768.48
89 (2)
Wavelength (Å)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
0.71073
triclinic
P1ꢀ
0.71073
triclinic
P1ꢀ
0.71073
monoclinic
P2(1)/c
0.71073
triclinic
P1ꢀ
8.3685 (4)
10.3895 (5)
10.7569 (5)
78.554 (1)
69.865 (1)
74.856 (1)
841.50 (7)
1
8.4233 (6)
10.6279 (8)
11.7286 (9)
78.033 (1)
76.655 (1)
68.604 (1)
942.38 (12)
1
11.7032 (5)
10.0215 (8)
11.0807 (9)
11.3758 (9)
61.166
74.516 (1)
68.620 (1)
1024.01 (14)
1
1.246
21.1615 (10)
14.3708 (7)
90
103.318 (1)
90
3463.3 (3)
4
1.475
a
(°)
b (°)
(°)
c
3
V (Å )
Z
q
l
3
calc (Mg/m )
1.295
1.627
1.255
1.458
ꢁ
1
(mm
)
2.572
1.347
F(000)
348
380
1584
412
Crystal size (mm)
Crystal color and habit
h range (°)
0.37 ꢂ 0.20 ꢂ 0.10
0.29 ꢂ 0.28 ꢂ 0.08
colorless plate
80–25.25
ꢁ10 6 h 6 10, ꢁ12 6 k 6 12,
ꢁ14 6 l 6 14
12039
0.12 ꢂ 0.10 ꢂ 0.08
colorless parallelepiped
1.75–25.25
ꢁ14 6 h 6 14, ꢁ25 6 k 6 25,
ꢁ17 6 l 6 17
39429
0.23 ꢂ 0.22 ꢂ 0.08
colorless plate
2.06–25.25
ꢁ12 6 h 6 12, ꢁ13 6 k 6 13,
ꢁ13 6 l 6 13
15260
colorless parallelepiped
2.03–27.50
Index ranges
ꢁ10 6 h 6 10, ꢁ13 6 k 6 13,
ꢁ
13 6 l 6 13
Reflections collected
12830
Independent reflections (Rint
)
3866 (0.0327)
3866/0/189
1.107
3401 (0.0247)
3401/0/209
1.105
6271 (0.0807)
6271/0/349
1.092
3706 (0.0233)
3706/0/232
1.079
Data/restraints/parameters
2
Goodness-of-fit (GOF) on F
R [I > 2
wR [I > 2
Largest difference in peak and
r
(I)]
0.0360
0.0830
0.663 and ꢁ0.786
0.0314
0.0830
1.227 and ꢁ0.515
0.0485
0.1082
1.144 and ꢁ0.487
0.0300
0.0752
triclinic
r(I)]
ꢁ
3
hole (e Å
)

2
20
P. Perrotin et al. / Polyhedron 84 (2014) 216–222
that probably has an important influence on the boat
conformation.
of 46% of the initial sample mass, was around 58 °C. The second
mass loss step, corresponding to around 20% of the initial sample
mass, commenced around 292 °C. The total mass loss of 66% is less
than the 74% mass loss expected for complete elimination of the
3.4. Preliminary thermal and photochemical decomposition studies of
⁄
Cp ligands.
compounds 2 and 5
⁄
⁄
The mass spectrum showed Cp (m/z = 135) and HCp
⁄
⁄
(m/z = 136) evolution peaking at 280–290 °C. Cp and HCp loss
Jutzi and coworkers demonstrated photochemical cleavage of
also occurred at ꢄ375 °C (peak value). Varying, small levels of
⁄
the Cp –Ga bond in the compound
(
R = 2,6-diisopropylphenyl or 2,6-diethylphenyl)
8
GaN(R)CH = CHN(R)] [32]. Similarly, irradiation of a THF-d solu-
[
2
tion of compound 2 over 12 h with the full spectrum of a medium
⁄
⁄
pressure mercury arc lamp led to the formation of Cp H, Cp
the 2,3,4,5-tetramethylfulvene (CH = C Me ). Over the course of
the photolysis, concentrations of the Cp dimer and tetramethylful-
2
and
2
5
⁄
4
⁄
⁄
vene increased at the expense of Cp H. It has been shown that Cp H
⁄
undergoes decomposition under irradiation to form H
CH = C Me ).
Due to the poor solubility of 2 in THF and the instability of the
2
, Cp
2
and
(
2
5
4
more solubilizing chloroform solvent to irradiation, it was not pos-
⁄
sible to accurately evaluate the extent of Cp H elimination. For this
⁄
reason, the more soluble compound [Cp
2
Ga(l
-NHEt)]
2
(5) was pre-
pared to model the photochemistry of compound 2. Irradiation of 5
in either toluene-d
Cp Ga specie(s) and Cp H, Cp
8
, C
6
D
6
, or pyridine-d
5
led to the formation of new
⁄
⁄
⁄
1
2
and (CH
2
= C
5
Me
4
). H NMR measure-
ments of the reaction products indicated that roughly one equivalent
⁄
of Cp ligand was eliminated from 2 under the irradiation. However,
well-defined gallium-containing reaction products could not be iso-
lated from the reaction mixture. Efforts to crystallize the products
afforded only glassy materials that would not redissolve in any
solvent.
The thermal decomposition behavior of 2 was examined by DSC
and TGA. The DSC and TGA thermograms are shown in Fig. 5. The
DSC curve exhibits a relatively sharp endothermic event peaking
at 296 °C and a broader, more complex endotherm with a peak at
Fig. 3. Thermal ellipsoid plot of 4. Displacement ellipsoids are drawn at the 30%
probability level. Hydrogen atoms have been omitted for clarity.
4
10 °C.
The identities of the fragments liberated under thermal decom-
position of 2 were determined by mass spectrometry measure-
ments of the gases evolved during the TGA analyses. The onset
temperature for the first mass loss step, corresponding to a loss
Table 2
Selected bond lengths (Å) and angles (°) for compounds 1, 2, and 5.
1
2
5
Ga(1)–N(1)
Ga(1)–N(1A)
Ga(1)–Ga(1A)
Ga(1)–Cispo
Ga(1)–Cispo
2.0000 (19)
2.0044 (18)
2.9431 (5)
2.021 (2)
2.022 (2)
94.60 (8)
2.0209 (17)
2.0220 (18)
3.023
2.041 (2)
2.036 (2)
96.78 (7)
83.22 (7)
119.65 (9)
2.0389 (18)
2.0431 (18)
3.056
2.0679 (19)
2.075 (2)
96.96 (7)
83.04 (7)
117.83 (8)
Ga(1)–N(1)–Ga(1A)
N(1)–Ga(1)–N(1A)
85.40 (8)
120.13 (9)
C
ispo–Ga(1)–Cispo
C
ipso: C(1) and C(10) for 1; C(1) and C(11) for 2 and 5.
Table 3
Selected bonds length (Å) and angles (°) for 4.
Ga(1)–N(1)
Ga(1)–N(3)
Ga(2)–N(1)
Ga(2)–N(2)
Ga(3)–N(2)
Ga(3)–N(3)
N(1)–Ga(1)–N(3)
Ga(1)–N(1)–Ga(2)
N(1)–Ga(2)–N(2)
1.982 (4)
1.944 (4)
1.983 (4)
1.948 (4)
1.977 (4)
1.984 (4)
105.88 (17)
123.1 (2)
105.09 (17)
Ga(1)–Cl(1)
Ga(2)–Cl(2)
Ga(3)–Cl(3)
Ga(1)–C(1)
Ga(2)–C(11)
Ga(3)–C(21)
Ga(2)–N(2)–Ga(3)
N(2)–Ga(3)–N(3)
Ga(1)–N(3)–Ga(3)
2.2061 (13)
2.2172 (12)
2.1984 (13)
2.020 (5)
2.0126 (5)
2.009 (5)
120.9 (2)
102.02 (16)
121.0 (2)
3 3
Fig. 4. (Top) Side view of Ga N core of 4 with thermal ellipsoids at 30%. (Bottom)
View of two neighboring cyclic cores showing hydrogen bonding.

P. Perrotin et al. / Polyhedron 84 (2014) 216–222
221
Fig. 5. (Top) DSC and (bottom) TGA thermograms for compound 2.
low molecular weight species (m/z = 16, 18, 42) were also observed
as were peaks for m/z = 121, 105, 93, 79, 65, and 53. The latter
peaks are consistent with the fragmentation of Cp H [33]. The total
endotherm peaking at 410 °C, multiple overlapping events appear
to be occurring at this temperature, one of which could be the for-
mation of GaN.
⁄
mass loss observed during the TGA experiments (76%) is consistent
with the decomposition of 2 into GaN, although further experi-
ments are necessary to substantiate this suggestion.
4. Summary and conclusions
⁄
The DSC and TG-MS measurements combined show that Cp
Ammonolysis of tricyclopentadienylgallium compounds occur
readily at or below room temperature to form the corresponding
dicyclopentadienyl-gallium amide compounds. By contrast, reflux-
loss from compound 2 occurs in a stepwise manner but is still
⁄
incomplete at temperatures as high as 375 °C. While further Cp
elimination probably contributes to
a
portion of the DSC
2 2
ing in toluene is required for the formation of [Me Ga(l-NH )]3

2
22
P. Perrotin et al. / Polyhedron 84 (2014) 216–222
(c) G. Linti, H.L. Schnöckel, Coord. Chem. Rev. 206 (2000) 285;
⁄
from Me
3
GaꢀNH
3
. Cp
2
3
GaClꢀNH . This is a clear illustration of the
(
(
d) C. Gemel, T. Steinke, M. Cokoja, A. Kempter, R.A. Fischer, Eur. J. Inorg. Chem.
2004) 4161.
greater lability of cyclopentadienyl ligands compared to alkyl ligands
in organometallic compounds of the main group metals. The more
forcing conditions required for the ammonolysis of Cp
compare to Cp
enyl-main group metal linkage is affected strongly by other ligands
on the metal.
[
5] (a) S. Harder, Coord. Chem. Rev. 176 (1998) 17;
(b) T.P. Hanusa, Organometallics 21 (2002) 2559.
⁄
2
GaCl as
⁄
[6] (a) M. Bochmann, D.M. Dawason, Angew. Chem., Int. Ed. 35 (1996) 2226;
b) S.-J. Lee, P.J. Shapiro, B. Twamley, Organometallics 25 (2006) 5582.
3
Ga shows that the reactivity of the cyclopentadi-
(
[
7] (a) E.K. Byrne, L. Parkanyl, K.H. Theopold, Science 241 (1988) 332;
(b) T. Douglas, K.H. Theopold, Inorg. Chem. 30 (1991) 594;
(c) K. Soulantica, A. Maisonnat, M.-C. Frome, B. Chaudret, Angew. Chem., Int.
Ed. 40 (2001) 448;
0
Steric stabilization of the ammonolysis products with bulky Cp
⁄
and Cp ligands enabled the isolation of discrete dicyclopentadie-
nylgallium amide dimers. Higher aggregation of the ammonolysis
product to a trimer, and ultimately higher aggregates, occurred
(
d) K. Soulantica, L. Erades, M. Sauvan, F. Senocq, A. Maisonnat, B. Chaudret,
Adv. Funct. Mater. 13 (2003) 553;
e) Y. Guari, K. Soulantica, K. Philippot, C. Thieuleux, A. Mehdi, C. Reye, B.
Chaudret, R.J.P. Corriu, New J. Chem. 27 (2003) 1029;
f) K. Soulantica, A. Maisonnat, F. Senocq, M.-C. Fromen, M.-J. Casanove, B.
Chaudret, Angew. Chem., Int. Ed. 40 (2001) 2984;
g) H.J. Haugan, W. Yu, S.T. Lee, A. Petrou, B.D. McCombe, K.S. Brewer, J.F. Lees,
O.T. Beachley, J. Cryst. Growth 244 (2002) 157;
h) J.W. Elam, A.B.F. Martinson, M.J. Pellin, J.T. Hupp, Chem. Mater. 18 (2006)
3571.
(
⁄
when a chloride ligand was used in place of one of the Cp ligands.
(
⁄
Investigation of the thermal decomposition of [Cp
2
Ga(l
2 2
-NH )]
⁄
(
showed that Cp ligand loss occurs in a stepwise manner, with the
majority of the ligand elimination occurring below 300 °C and the
remainder occurring around 375 °C.
(
⁄
Photolytic cleavage of one of the Cp
rings from
[8] (a) M. Cokoja, H. Parala, M.-K. Schroter, A. Birkner, M.W.E. van den Berg, W.
⁄
Grunert, R.A. Fischer, Chem. Mater. 18 (2006) 1634;
[
Cp
2
Ga(
l
2
-NH
2
)] was also demonstrated, although the structure of
(
b) O.I. Mi c´ i c´ , S.P. Ahrenkiel, D. Bertram, A.J. Nozik, Appl. Phys. Lett. 75 (1990)
78;
(c) A. Manz, A. Birkner, M. Kolbe, R.A. Fischer, Adv. Mater. 12 (2000) 569;
d) K. Sardar, M. Dan, B. Schwenzer, C.N.R. Rao, J. Mater. Chem. 15 (2005)
175;
e) G. Pan, M.E. Kordesch, P.G. Van Patten, Chem. Mater. 18 (2006) 3915;
(f) H.S. Park, S.D. Waezsada, A.H. Cowley, H.W. Roesky, Chem. Mater. 10 (1998)
251;
g) J.W. Hwang, S.A. Hanson, D. Britton, J.F. Evans, K.F. Jensen, W.L. Gladfelter,
the resulting gallium product was not identified. Examination of
the thermal degradation of this unidentified photolysis product is
worthwhile and could offer a different, more effective route to
GaN involving photolysis of the cyclopentadienylgallium amide
single source precursor, followed by thermal degradation of the
photolysis product(s).
4
(
2
(
2
(
Chem. Mater. 2 (1990) 342.
[9] D.A. Neumayer, A.H. Cowley, A. Decken, R.A. Jones, V. Lakhotia, J.G. Ekerdt, J.
Am. Chem. Soc. 117 (1995) 5893.
Acknowledgments
[
[
10] G.E. Coates, J. Chem. Soc. (1951) 200.
11] O.T. Beachley Jr., T.L. Royster Jr., J.R. Arhar, A.L. Rheingold, Organometallics 12
(1993) 1976.
P.P. thanks the University of Idaho College of Graduate Studies
for a graduate research fellowship. The Bruker (Siemens) SMART
APEX diffraction facility was established at the University of Idaho
with the assistance of the NSF-EPSCoRprogram and the M.J.
Murdock Charitable Trust, Vancouver, WA.
[
[
[
12] P. Jutzi, B. Neumann, G. Reumann, H.G. Stammler, Organometallics 18 (1999)
2037.
13] O.T. Beachley Jr., T.D. Getman, R.U. Kirss, R.B. Hallock, W.E. Hunter, J.L. Atwood,
Organometallics 4 (1985) 751.
14] (a) H. Schumann, R. Nickel, R. Weimann, J. Organomet. Chem. 468 (1994) 43;
(
(
b) O.T. Beachley Jr., M.T. Mosscrop, M.R. Churchill, J. Organomet. Chem. 626
2001) 59.
Appendix A. Supplementary data
[
[
15] O.T. Beachley Jr., R.B. Hallock, Organometallics 6 (1987) 170.
16] O.T. Beachley Jr., R.B. Hallock, H.M. Zhang, J.L. Atwood, Organometallics 4
CCDC 1011762, 1011761, 1011763, and 1011764 contain the
supplementary crystallographic data for compounds 1, 2, 4, and
, respectively. These data can be obtained free of charge via
http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
deposit@ccdc.cam.ac.uk.
(
1985) 1675.
[
17] B.J. Burger, J.E. Bercaw, in: A.L. Wayda, M.Y. Darensbourg (Eds.), Experimental
Organometallic Chemistry, American Chemical Society, Washington, DC, 1987.
18] H. Hope, Prog. Inorg. Chem. 41 (1995) 1.
19] SMART: v. 5.626, Bruker Molecular Analysis Research Tool, Bruker AXS, Madison,
WI, 2002.
: v. 6.45, Data Reduction and Correction Program, Bruker AXS,
Madison, WI, 2001.
21] SADABS: v. 2.01, An Empirical Absorption Correction Program, Bruker AXS Inc.,
Madison, WI, 2001.
22] G.M. Sheldrick, SHELXTL: v. 6.10, Structure Determination Software Suite, Bruker
AXS Inc., Madison, WI, 2001.
5
[
[
[20] SAINTPLUS
[
[
References
[
1] (a) P. Jutzi, Chem. Rev. 86 (1986) 983;
[23] S.C. Wallwork, I.J. Worrall, J. Chem. Soc. (1965) 1816.
[24] J. McMurran, J. Kouvetakis, D.C. Nesting, D.J. Smith, J.L. Hubbard, J. Am. Chem.
Soc. 120 (1998) 5233.
[25] J. Kouvetakis, J. McMurran, C. Steffek, T.L. Groy, J.L. Hubbard, Inorg. Chem. 39
(2000) 3805.
[26] W.R. Nutt, J.A. Anderson, J.D. Odom, M.M. Williamson, B.H. Rubin, Inorg. Chem.
24 (1985) 159.
[27] W.R. Nutt, R.B. Stimson, M.F. Leopold, B.H. Rubin, Inorg. Chem. 21 (1982) 1909.
[28] C.J. Carmalt, Coord. Chem. Rev. 223 (2001) 217 (and references therein).
[29] M.J. Almond, M.G.B. Drew, C.E. Jenkins, D.A. Rice, J. Chem. Soc., Dalton Trans.
(1992) 5.
(
(
(
(
(
(
(
b) P. Jutzi, Comments Inorg. Chem. 6 (1987) 123;
c) P. Jutzi, J. Organomet. Chem. 400 (1990) 1;
d) H. Schumann, Pure Appl. Chem. 63 (1991) 813;
e) P. Jutzi, J. Dahlhaus, Coord. Chem. Rev. 137 (1994) 179;
f) D.J. Burkey, T.P. Hanusa, Comments Inorg. Chem. 17 (1995) 41;
g) P.J. Shapiro, Coord. Chem. Rev. 189 (1999) 1;
h) P. Jutzi, N. Burford, in: A. Togni, R.L. Halterman (Eds.), Metallocenes:
Synthesis, Reactivity, and Applications, Wiley-VCH, Weinheim, 1998;
(
(
i) P. Jutzi, N. Burford, Chem. Rev. 99 (1999) 969;
j) P. Jutzi, G. Reumann, J. Chem. Soc., Dalton Trans. (2000) 2237.
[
2] P.J. Shapiro, in: P.J. Shapiro, D.A. Atwood (Eds.), Group 13 Chemistry: From
Fundamentals to Application, American Chemical Society, Washington, DC,
[30] J.P. Campbell, J.W. Hwang, V.G. Young Jr., R.B. Von Dreele, C.J. Cramer, W.L.
Gladfelter, J. Am. Chem. Soc. 120 (1998) 521.
2
002.
3] W.A. Wojtczak, P.F. Fleig, M.J. Hampden-Smith, Adv. Organomet. Chem. 40
1996) 215.
4] (a) R.A. Fischer, J. Weiss, Angew. Chem., Int. Ed. 38 (1999) 2830;
b) L.O. Schebaum, P.J. Shapiro, in: P.J. Shapiro, D.A. Atwood (Eds.), Group 13
[31] D.A. Atwood, A.H. Cowley, P.R. Harris, R.A. Jones, S.U. Koschmieder, C.M. Nunn,
Organometallics 12 (1993) 24.
[32] T. Pott, P. Jutzi, W.W. Schoeller, A. Stammler, H.G. Stammler, Organometallics
20 (2001) 5492.
[33] NIST Standard Reference Database 69: NIST Chemistry WebBook. <http://
webbook.nist.gov/cgi/cbook.cgi?ID=C4045447&Mask=200>.
[
[
(
(
Chemistry: From Fundamentals to Application, American Chemical Society,
Washington, DC, 2002;