Platinum ω-Alkenyl Compounds as Chemical Vapor Deposition Precursors: Synthesis and Characterization of Pt[CH2CMe2CH2CH═CH2]2and the Impact of Ligand Design on the Deposition Process

Article abstract of DOI:10.1021/acs.chemmater.0c03226

We describe the synthesis and characterization of three platinum(II) ω-alkenyl complexes of stoichiometry Pt[CH2CMe2(CH2)xCH═CH2]2 where x is 0, 1, or 2, as well as some related platinum(II) compounds formed as byproducts during their synthesis. The ω-alkenyl ligands in all three complexes, cis-bis(η1,η2-2,2-dimethylbut-3-en-1-yl)platinum (2), cis-bis(η1,η2-2,2-dimethylpent-4-en-1-yl)platinum (3), and cis-bis(η1,η2-2,2-dimethylhex-5-en-1-yl)platinum (4), bind to Pt by means of a Pt-alkyl sigma bond at one end of the ligand chain and a Pt-olefin pi interaction at the other; the olefins reversibly decomplex from the Pt centers in solution. The good volatility of 3 (10 mTorr at 20 °C), its ability to be stored for long periods without decomposition, and its stability toward air and moisture make it an attractive platinum chemical vapor deposition (CVD) precursor. CVD of thin films from 3 shows no nucleation delay on several different substrates (SiO2/Si, Al2O3, and VN) and gives films that are unusually smooth. At 330 °C in the absence of a reactive gas, the precursor deposits platinum containing 50% carbon, but in the presence of a remote oxygen plasma, the amount of carbon is reduced to below the Rutherford backscattering spectroscopy (RBS) detection limit without affecting the film smoothness. Under hot wall CVD conditions at 250 °C in the absence of a co-reactant, 72% of the carbon atoms in 3 are released as hydrogenated products (largely 4,4-dimethylpentenes), 22% are released as dehydrogenated products (all of which are the result of skeletal rearrangements), and 6% remain in the film. Some conclusions about the CVD mechanism are drawn from this product distribution.

Full text of DOI:10.1021/acs.chemmater.0c03226

pubs.acs.org/cm
Article
Platinum ω‑Alkenyl Compounds as Chemical Vapor Deposition
Precursors: Synthesis and Characterization of
Pt[CH CMe CH CHCH ] and the Impact of Ligand Design on the
2
2
2
2 2
Deposition Process
Sumeng Liu, Zhejun Zhang, Danielle Gray, Lingyang Zhu, John R. Abelson, and Gregory S. Girolami*
Cite This: https://dx.doi.org/10.1021/acs.chemmater.0c03226
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı Supporting Information
ABSTRACT: We describe the synthesis and characterization of three
platinum(II) ω-alkenyl complexes of stoichiometry Pt-
[
CH CMe (CH ) CHCH ] where x is 0, 1, or 2, as well as some related
2
2
2 x
2 2
platinum(II) compounds formed as byproducts during their synthesis. The ω-
alkenyl ligands in all three complexes, cis-bis(η ,η -2,2-dimethylbut-3-en-1-
yl)platinum (2), cis-bis(η ,η -2,2-dimethylpent-4-en-1-yl)platinum (3), and cis-
bis(η ,η -2,2-dimethylhex-5-en-1-yl)platinum (4), bind to Pt by means of a Pt−
alkyl sigma bond at one end of the ligand chain and a Pt−olefin pi interaction at
the other; the olefins reversibly decomplex from the Pt centers in solution. The
good volatility of 3 (10 mTorr at 20 °C), its ability to be stored for long periods
1
2
1
2
1
2
without decomposition, and its stability toward air and moisture make it an attractive platinum chemical vapor deposition (CVD)
precursor. CVD of thin films from 3 shows no nucleation delay on several different substrates (SiO /Si, Al O , and VN) and gives
2
2
3
films that are unusually smooth. At 330 °C in the absence of a reactive gas, the precursor deposits platinum containing 50% carbon,
but in the presence of a remote oxygen plasma, the amount of carbon is reduced to below the Rutherford backscattering
spectroscopy (RBS) detection limit without affecting the film smoothness. Under hot wall CVD conditions at 250 °C in the absence
of a co-reactant, 72% of the carbon atoms in 3 are released as hydrogenated products (largely 4,4-dimethylpentenes), 22% are
released as dehydrogenated products (all of which are the result of skeletal rearrangements), and 6% remain in the film. Some
conclusions about the CVD mechanism are drawn from this product distribution.
2
4
INTRODUCTION
nent of data storage media, and as electrical contacts in
■
25−27
medical devices.
Though widely studied, one major drawback of (C H Me)-
PtMe3 as a CVD precursor is that it suffers from long
Some 30 years ago, our research group was the first to show
that the organoplatinum complex (C H )PtMe was a useful
5
4
5
5
3
precursor for the chemical vapor deposition (CVD) of
7,8,11,28,29
nucleation delays.
This problem can be traced to the
1
platinum thin films. Over the years, many other Pt CVD
high metal−ligand bond strengths and consequent high kinetic
2
−4
precursors have been described,
but a close variant of
barriers associated with eliminating ligands from adsorbed
(
(
C H )PtMe , the methylcyclopentadienyl complex
29,30
5
5
3
(C H Me)PtMe molecules.
Such elimination reactions
5
4
3
5
C H Me)PtMe , has become the most widely employed.
5
4
3
typically occur more quickly on platinum surfaces than they do
on the underlying substrate, owing to the highly catalytic
Like (C H )PtMe , it is stable toward air and moisture, it is
5
5
3
5
,8,31
thermally stable at room temperature, and its vapor pressure
nature of platinum.
As a result, deposition initiates to form
6
,7
(
53 mTorr at 23 °C) is relatively high. Pure platinum films
can be grown by CVD or atomic layer deposition (ALD) from
C H Me)PtMe and several other precursors, provided that a
islands on the surface, which subsequently grow rapidly.
Similar island growth behavior has been observed for several
10,32−36
(
other Pt CVD precursors.
Eventually, the islands
5
4
3
reactive gas is present, such as hydrogen, oxygen, ozone, or an
coalesce, but the long nucleation delays, and resulting sparse
6
−11
oxygen plasma;
these reactive gases assist in the removal of
excess carbon from the films.
Received: August 6, 2020
Revised: October 2, 2020
2
−4,12,13
Today, the CVD
and atomic layer deposition
of platinum thin films from organoplatinum
precursors are of interest for a wide variety of applications,
1
4−17
(
ALD)
1
8
including the fabrication of gate electrodes and other
8
,19
20−23
microelectronic structures,
as catalysts,
as a compo-
©
XXXX American Chemical Society
https://dx.doi.org/10.1021/acs.chemmater.0c03226
A
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
3
5
nucleation, mean that it is difficult to make continuous films
that are also very thin; instead, at the point of full coalescence,
the films are typically relatively thick and rough. Although the
deposition of conformal Pt films has been achieved by means
of an alternating ALD/oxygen plasma process from
Grignard reagent. Similar reactions with the analogous β-
stabilized Grignard reagents (or in one case with an
alkyllithium) afford the corresponding compounds cis-bis-
1
2
1
2
(
η ,η -2,2-dimethylbut-3-en-1-yl)platinum (2), cis-bis(η ,η -
,2-dimethylpent-4-en-1-yl)platinum (3), and cis-bis(η ,η -
,2-dimethylhex-5-en-1-yl)platinum (4). Compounds 2 and 3
are light yellow liquids, whereas 4 is a low-melting solid; all
three compounds can be purified by column chromatography
on silica gel with pentane as the eluent.
1
2
2
2
(
C H Me)PtMe , nucleation delays of ∼50 cycles are still
5
4
3
31
observed.
The nucleation delay problem can potentially be resolved by
employing a precursor whose ligands can more easily eliminate
16
than those in (C H Me)PtMe . For example, the “open-Cp”
5
4
3
All three bis(ω-alkenyl)platinum compounds 2, 3, and 4 can
be stored indefinitely under an argon atmosphere at −20 °C.
Of the three, the dimethylbutenyl derivative 2 is the least
stable: it reacts with air and water and decomposes within days
at room temperature under Ar. Of intermediate stability is the
dimethylhexenyl compound 4, which is air and water stable but
decomposes thermally in several days at room temperature. In
contrast, the most stable is the dimethylpentenyl compound 3,
which can be stored at room temperature for 1 month in air
with only slight darkening and with no observable changes in
the NMR spectrum. Thus, 3 is far more thermally stable than
its unmethylated analog 1, which decomposes by β-hydrogen
precursor (2,4-dimethylpentadienyl)(ethylcyclopentadienyl)-
ruthenium(II) shows enhanced nucleation compared to
3
7
bis(ethylcyclopentadienyl)ruthenium(II), and the amino-
alkene precursor dimethyl(N,N-dimethyl-3-buten-1-amine)
platinum shows enhanced nucleation compared to (C H Me)-
5
4
3
8
PtMe3.
In this context, we have previously described the square-
planar platinum(II) precursor in which two coordination sites
are occupied by weakly coordinating olefin groups: Pt-
1
2
[
(
(CH ) CHCH ] or cis-bis(η ,η -pent-4-en-1-yl)platinum
2
3
2 2
3
5
1). Although this pentenyl compound is highly volatile and
affords relatively clean platinum films, it decomposes
thermally within 2 weeks at room temperature
3
5
35,39
and thus
39
elimination over 2 weeks at room temperature.
is not ideal as a CVD precursor. The thermal sensitivity stems
from the availability of a low-barrier mechanism to eliminate a
ligand: β-hydrogen elimination initiated by decomplexation of
the olefin groups:
In the present study, we sought to prepare new platinum
CVD precursors that are stable at room temperature but still
retain the lability of the alkenyl ligands, with the idea that a
low-barrier pathway for dissociation of a ligand would lead to
4
0−42
fast nucleation.
Our strategy was to replace the β-
hydrogen atoms in 1 with methyl groups: this “β-stabilization”
strategy has long been employed to increase the thermal
43−47
stability of organometallic complexes.
We now report the synthesis and characterization of three
new “β-stabilized” platinum complexes of stoichiometry
Pt[CH CMe (CH ) CHCH ] where x is 0, 1, or 2: cis-
2
2
2 x
2 2
The melting points are ca. −15 °C for 2, ca. 20 °C for 3, and
1
2
1
2
bis(η ,η -2,2-dimethylbut-3-en-1-yl)platinum (2), cis-bis(η ,η -
4
6−47 °C for 4; all three compounds 2−4 are volatile, with 2
1
2
2
,2-dimethylpent-4-en-1-yl)platinum (3), or cis-bis(η ,η -2,2-
having the highest volatility. Such low melting points are useful
for CVD precursors because the delivery rate of liquid
dimethylhex-5-en-1-yl)platinum (4). We also describe the
growth of the platinum films from 3, which is a particularly
interesting CVD precursor owing to its good volatility, its low
melting point, its stability toward air and water, and its much
longer shelf life compared to the “β-unstable” parent
compound 1. Notably, CVD of thin films from 3 shows no
3
,7
precursors can be easily controlled and reproduced. The
vapor pressure of 3, which is 12 mTorr at 23 °C and 250
mTorr at 80 °C, follows the equation log p = 6.19−2400/T
(
pressure is in Torr and temperature is in K). The vapor
nucleation delay on several different substrates (SiO /Si,
pressure of 3 is only slightly lower than that of the commercial
precursor (C H Me)PtMe , which has P = 53 mTorr at 23
2
Al O , and VN) and gives films that are unusually smooth. In a
2
3
5
4
3
vap
7
separate paper, we will describe the mechanism by which 3
decomposes thermally in solution. Those results provide
additional insights into the relationship between the structure
of the precursor and the CVD growth kinetics and morphology
of the resulting platinum thin films.
°C. The enthalpy of evaporation of 3, ΔH = 11 ± 2 kcal
mol (see the Supporting Information, SI), is similar to the
0.5 ± 0.2 kcal mol value reported for (C H Me)PtMe .
vap
−
1
−
1
7
1
5 4 3
Cyclooctadiene-Containing Byproducts. Treatment of
COD)PtCl2 with the butenyl Grignard reagent, (2,2-
(
dimethylbut-3-en-1-yl)magnesium bromide, gives poor yields
RESULTS AND DISCUSSION
of 2; instead, the major product is the monoalkylated
■
1
compound bromo(η -2,2-dimethylbut-3-en-1-yl)(1,5-cyclooc-
Preparation of β-Stabilized Bis(ω-alkenyl)platinum(II)
Complexes. In our earlier work, cis-bis(η ,η -pent-4-en-1-yl)
platinum (1) was generated by treatment of (COD)PtCl2
1
2
tadiene)platinum (5). Good yields of 2 require the use of
the corresponding alkyllithium reagent (see the Experimental
Section).
(
COD = 1.5-cyclooctadiene) with two equivalents of a
B
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Table 1. Crystallographic Data for the New Compounds
2
3
4
5
6
formula
C H Pt
C H Pt
14 26
C H Pt
16 30
C H BrPt
14 23
C H Pt
20 34
1
2
22
FW (g mol−1)
λ (Å)
361.38
0.71073
100
389.44
0.71073
100
417.49
0.71073
100
466.32
0.71073
100
469.56
0.71073
100
T (K)
crystal system
space group
a (Å)
b (Å)
c (Å)
β (deg)
V (Å3)
Z
monoclinic
P21/n
monoclinic
P21/c
monoclinic
C2/c
monoclinic
P21
monoclinic
C2/c
6.2341(2)
10.3493(3)
18.9360(6)
92.5422(8)
1220.52(7)
4
5.8662(3)
21.591(1)
22.360(1)
91.629(2)
2830.9(2)
8
6.0482(4)
13.5208(8)
19.287(1)
95.364(2)
1570.3(2)
4
9.7170(4)
11.9545(5)
13.2835(6)
110.537(2)
1445.0(1)
4
18.3933(6)
9.0744(3)
11.3108(4)
104.951(1)
1823.9(1)
4
ρcalcd (g cm−3)
μ (mm−1)
R (int)
1.967
11.455
0.037
1.827
9.884
0.030
1.766
8.916
0.036
2.144
12.450
0.041
1.710
7.687
0.053
abs corr method
max/min transm factors
data/restraints/params
GOF on F2
multi-scan
0.334/0.141
3042/0/140
1.231
0.0278
0.0703
numerical
0.457/0.038
7109/0/316
1.170
0.0142
0.0293
numerical
0.535/0.066
1952/0/90
1.210
0.0118
0.0304
multi-scan
0.747/0.516
11517/1/294
1.070
0.0189
0.335
multi-scan
0.512/0.419
2274/0/101
1.125
0.0172
0.0374
a
R1 [I > 2σ(I)]
b
wR (all data)
2
max/min Δρelectron (eÅ−3)
1.68/−2.27
0.62/−0.49
2 2
0.89/−0.78
0.93/−1.01
0.52/−1.13
a
R = ∑ ||F | − |F ||/|∑|F |. wR = [∑w(F − F ) /∑(F ) ]1/2.
b
2
2
c
2
1
o
c
o
2
o
o
a
Table 2. Selected Bond Distances and Angles for the New Compounds
2
3
4
5
6
CC/Å
1.370(8)
2.068(5)
2.259(6)
2.296(6)
94.9(2)
1.366(3)
2.058(2)
2.301(2)
2.250(2)
88.81(8)
106.6(2)
1.373(3)
2.077(2)
2.272(2)
2.257(2)
88.2(1)
1.326(6)
2.072(3)
N/A
1.324(4)
2.083(2)
N/A
N/A
88.6(1)
111.3(2)
Pt−C /Å
α
b
Pt−Colefin /Å
Pt−Colefin /Å
c
N/A
89.8(2)
d
C −Pt−C ′/deg
α
α
C −C −C /deg
100.0(4)
112.3(2)
110.7(3)
α
β
γ
a
b
c
d
Some of the numbers given are averages over chemically equivalent bonds. Methine carbon. Methylene carbon. C −Pt−Br angle.
α
The reaction of (COD)PtCl with the dimethylhexenyl
2
Grignard reagent generates 4 in only 10% isolated yield
because a second compound is also formed: (1,5-
1
cyclooctadiene)bis[η -(2,2-dimethylcyclopentyl)methyl] plati-
num (7). This latter compound (which was isolated a mixture
of three diastereomers) is the result of insertion of the olefin
group into the Mg-C bond before alkylation of the Pt center.
We show in the SI (Section 2) that 4 does not convert to 7 in
We also found that the adduct (1,5-cyclooctadiene)bis(η1-
,2-dimethylbut-3-en-1-yl)platinum (6) is obtained when the
butenyl compound 2 is sublimed at room temperature in the
presence of ∼1 equiv of COD onto a cold (−80 °C) surface
under vacuum. In both 5 and 6, the olefin groups of the
butenyl ligands are not coordinated to Pt. The COD adduct 6
is in equilibrium with 2 and free COD in solution: when pure 6
2
benzene-d in the presence of 1,5-cyclooctadiene, whereas the
6
hexenyl Grignard reagent readily cyclizes (i.e., undergoes an 5-
48
exo-trig insertion reaction).
Solid-State Structures of the β-Stabilized Bis(ω-
alkenyl)platinum(II) Complexes. Crystallographic data for
compounds 2−6 are listed in Table 1, and selected bond
distances and angles are given in Table 2. The structures of 2−
is dissolved at millimolar concentrations in toluene-d at room
8
temperature, it is almost completely dissociated to 2 and free
COD (Figure S5.4). The temperature dependence of the
4
(Figure 1) all contain two ω-alkenyl ligands in which the α-
equilibrium constant K = [6]/([2][COD]) shows that ΔH =
eq
1
carbon atom is sigma-bound to platinum (η ) and the CC
35
−
1
−1 −1
5
± 1 kcal mol and ΔS = 29 ± 5 cal mol ·K (Figure S5.6).
The large entropy change is consistent with the dissociation of
one particle to two:
2
bond at the other end of the chain is pi-bound (η ), as found
previously for Pt[(CH ) CHCH ] .
2
3
2 2
Compounds 2−4 all adopt solid-state structures with
(
idealized) C symmetries, in which the two α-carbon atoms
2
are mutually cis. The C −Pt−C ′ angle of 94.9(2)° in the
α
α
butenyl compound 2 is significantly larger than the ∼90°
angles seen in the pentenyl and hexenyl compounds 3 and 4
(
Table 2). This difference reflects the short chain length (and
C
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
1
2
1
2
Figure 1. Left: Crystal structures of (a) cis-bis(η ,η -2,2-dimethylbut-3-en-1-yl)platinum (2), (b) cis-bis(η ,η -2,2-dimethylpent-4-en-1-yl)platinum
(
1
2
3), and (c) cis-bis(η ,η -2,2-dimethylhex-5-en-1-yl)platinum (4). Right: Views showing different tilt angles of the CC bond vectors with respect
to the square plane in the butenyl compound 2, the pentenyl compound 3, and the hexenyl compound 4. The ellipsoids are drawn at the 35%
probability level, except the hydrogen atoms are depicted as arbitrarily sized spheres.
consequent small bite angle) of the chelating butenyl ligands in
the angle between the CC vector and the normal the tilt
8
2.
angle). For olefin complexes of d square-planar metals, there
The C −C −C bond angles of 100.0(4) and 106.6(2)° in 2
is little electronic preference for any particular CC
α
β
γ
and 3, respectively, are compressed so as to enable the CC
bond to coordinate to platinum despite the short chain lengths.
The C −C −C bond angle of 112.3(2)° in 4 (for which the
orientation, but a tilt angle of 0° minimizes interligand steric
62−65
repulsions.
In contrast, the CC vectors in 2−4 are all
tilted significantly away from this preferred orientation and
instead form tilt angles of 52 and 35° (average = 44°) in the
butenyl compound 2 (Figure 1), 29° in the pentenyl
compound 3, and 12° in the hexenyl compound 4. For
comparison, the tilt angles in the unmethylated pentenyl
compound 1 were 34° and 23° (average = 29°). The
presence of two different tilt angles in 2 and also in 1 is
undoubtedly a consequence of packing effects; both
compounds are completely symmetrical in solution (see
below).
α
β
γ
longer chain length reduces the ring strain) is similar to those
in the COD adducts 5 and 6 (Figure 2), in which the ω-alkenyl
ligands are unidentate and thus unaffected by ring strain.
The rings formed by the pentenyl groups in 3 adopt chair
conformations. In contrast, twist-boat conformations are seen
for most other complexes, including the unmethylated
compound 1, in which a bidentate ligand with a 5-atom
backbone is bound to a metal by a sigma bond at one end of
35
6
0
2
49−59
the chain and an η -CC bond at the other.
We propose
that the 2,2-dimethyl substituents in 3 cause the change in ring
In the butenyl compound 2, the short chain length clearly
prevents the CC bond vectors from being oriented in the
normal direction. Similar conformational constraints are
known to cause such tilting in other square-planar platinum(II)
complexes, which in the extreme can cause coordinated CC
bonds to be oriented a full 90° away from the normal, i.e., the
conformation, as has been seen in some related com-
6
0,61
pounds.
Possibly for the same reason, the ω-alkenyl rings
in the hexenyl compound 4 adopt boat-chair conformations,
whereas in the butenyl compound 2, they adopt idealized
envelope conformations.
6
5,66
In four-coordinate platinum(II) olefin complexes, the CC
CC bond lies in the square plane.
As the number of
bond vector typically is normal to the square plane (we will call
carbon atoms in the ligand backbone increases, there are more
D
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
compound 2 remains planar despite having a Pt−C contact
distance that is normal for Pt−olefin complexes. The planarity
suggests that the Pt−C interaction with this carbon atom is
largely nonbonding, a hypothesis that is consistent with the
negligible Pt−C coupling constant to this carbon as seen in the
solution NMR spectrum (see below).
The small elongation of the CC bonds and the significant
influence of steric and packing effects on the conformations of
2
, 3, and 4 suggest that the olefin−Pt interactions are relatively
weak in these complexes; we will return to this issue in a
subsequent section.
Isomerism of the β-Stabilized Platinum(II) ω-Alkenyl
Complexes in Solution. Interestingly, we find that the
butenyl and hexenyl complexes 2 and 4 exist in solution as a
68
mixture of two species, even though only one species is
observed in the solid state. For all three compounds 2−4, three
diastereomers are possible: two diastereomers of C symmetry
2
in which the two olefins present the same face to Pt (re−re or
si−si; these two alternatives are enantiomers and thus give the
same NMR spectrum) and one diastereomer of C symmetry in
s
which the two olefins present different faces to Pt (si-e). For
the rest of this discussion, we will refer to the re−re and si−si
diastereomers as “the C isomer” and the si−re diastereomer as
2
“
the C isomer.”
s
In the solid state, compounds 2−4 all adopt the C structure.
2
In solution, only a single isomer of the pentenyl compound 3 is
evident from the NMR spectra, which certainly is also the C2
1
Figure 2. Top: Crystal structure of bromo(η -2,2-dimethylbut-3-en-1-
yl)(1,5-cyclooctadiene)platinum (5). Bottom: Crystal structure of
structure (if the C isomer is present, its concentration is less
s
1
bis(η -2,2-dimethylbut-3-en-1-yl)(1,5-cyclooctadiene)platinum (6).
The ellipsoids are drawn at the 35% probability level, except the
hydrogen atoms are depicted as arbitrarily sized spheres.
than 1%). For solutions of the butenyl and hexenyl compounds
2
and 4, however, significant amounts of both the C and C
2 s
isomers are present. Thus, in toluene-d at room temperature,
the Pt NMR spectrum of 3 shows only one singlet at δ
8
1
95
1
95
−
3777, whereas the Pt NMR spectrum of 2 shows two
conformational degrees of freedom and the CC bond
vectors form smaller tilt angles.
Unlike our pentenyl and hexenyl compounds 3 and 4,
however, many square-planar compounds are known in which
an olefin at the end of a five- or six-atom chelating chain binds
peaks at δ −4019 and δ −4110, in a ratio of approximately 2:1,
195
and the Pt NMR spectrum of 4 shows two peaks at δ −3778
and δ −3797, in a ratio of approximately 10:1 (Figure 3). For
the rest of this discussion, we will assume that the predominant
49−59
species in solution for 2 and 4 is the C isomer seen in the
2
to a metal with a tilt angle of essentially zero.
We believe
that this perpendicular structure is not adopted by the alkenyl
ligands in 3 and 4 because it would lead to unfavorable inter-
and intraligand repulsions involving the 2-methyl groups and
the cis-double bonds.
The CC bond lengths in compounds 2−4 of ∼1.37 Å
(
Table 2) are similar to the CC bond lengths seen in other
67
platinum(II) olefin complexes such as Zeise’s salt. The bonds
are about ∼0.045 Å longer than the ∼1.325 Å CC bond
lengths seen for noncoordinated alkenes such as those in the
COD adducts 5 and 6. As expected, the CC bond distances
1
2
in 2−4 are similar to those in other platinum(II) η ,η -pent-4-
en-1-yl complexes in which the CC bond is trans to non-π
52,60
donors such as NHCs or phosphines,
but slightly shorter
than those in which the CC bond is trans to π donors such
4
9−51
as chlorine or oxygen (∼1.40 Å).
It is well known that coordination of olefins to metal centers
Figure 3. 195Pt NMR spectra of the butenyl compound 2 (top), the
causes the carbon atoms to rehybridize so that the four
pentenyl compound 3 (middle), and the hexenyl compound 4
(bottom) at 20 °C in toluene-d . Two isomers (C and C ) exist in
67
substituents are no longer co-planar with the CC bond.
8
2
s
Interestingly, the olefinic methine group in the butenyl
solution for 2 and 4, but not for the pentenyl compound 3.
E
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
solid state; as we will see, this assumption can explain the
relative amounts of the two isomers as a function of chain
length, whereas the opposite assumption cannot.
NMR spectroscopy. Nevertheless, as we will show below, this
isomer must be present transiently in solution, as an
intermediate in an intramolecular dynamic process.
The presence in solution of only the C isomer for the
Structure and Bonding of the β-Stabilized Platinum-
2
1
13
pentenyl compound 3 but both the C and C isomers for 2
(II) ω-Alkenyl Complexes in Solution. The H and
C
2
s
1
13
1
and 4 is also evident in the H and C{ H} NMR spectra.
Thus, the H NMR spectrum of the butenyl compound 2 at
NMR resonances in 2−4 have been assigned based on the peak
patterns, coupling constants, and one-dimensional (1D)
nuclear Overhauser effect spectroscopy (NOESY) NMR
spectra (see Figure 5). For all three compounds, the following
1
−
60 °C in toluene-d shows twice the number of expected
peaks (discussed in detail below), which can be assigned to the
presence of two isomers in a 2:1 concentration ratio (Figure
8
4
). A variable temperature NMR study of 2 shows that the
1
Figure 5. H NMR spectrum of the pentenyl compound 3 in toluene-
1
Figure 4. H NMR spectra in the olefinic region of the butenyl
compound 2 (top left) and the hexenyl compound 4 (top right) at
d8 at −20 °C.
−
60 °C in toluene-d . The C /C isomer ratios for 2, 3, and 4 are
8 2 s
.9:1, >100:1, and 11:1, respectively.
1
discussion will initially focus on the spectra of the C isomer,
2
which is the only isomer present for the pentenyl compound 3,
and is the major isomer for the butenyl and hexenyl
compounds 2 and 4. All of the NMR data (reported below
relative amount of the minor (C ) isomer increases slightly
with temperature. Fitting the temperature dependence of the
s
equilibrium constant Keq = [C -2]/[C -2] to the van’t Hoff
for solutions in toluene-d ) are consistent with structures in
2
s
8
1
2
equation gives the following parameters: ΔH = −0.3 ± 0.5 kcal
which the alkenyl ligands coordinate to Pt in an η ,η fashion.
Thus, the H NMR spectra of the C isomers of 2−4 each
display three signals between δ 3 and 5, which are assigned to
−
1
−1 −1
1
mol , ΔS = 0 ± 2 cal mol ·K (Figure S6.2). The small
values for the enthalpy and entropy changes show that the two
isomers are rather similar thermodynamically. Likewise, two
2
the olefinic protons (Table 3). The shielded chemical shifts
1
13
1
3
isomers are also seen in the low-temperature H and C{ H}
and the diminished J
coupling constants (compared to
HH
NMR spectra of the hexenyl compound 4, in approximately a
those seen for free olefins) clearly show that the CC bonds
35,55−59,69,70
1
0:1 ratio. Fitting the temperature dependence of the
are coordinated to platinum in solution.
In the
1
3
1
equilibrium constant Keq = [C -4]/[C -4] to the van’t Hoff
C{ H} NMR spectrum of the butenyl compound 2 at −60
2
s
−
1
equation gives ΔH = −0.8 ± 0.5 kcal mol , ΔS = 1 ± 2 cal
°C, small Pt−C coupling constants are observed for the
−
1
−1
1
mol ·K (Figure S6.14). The C isomer is favored over the
olefinic carbons: J is 26 Hz for the terminal methylene
2
PtC
C isomer at room temperature by a free-energy difference ΔG
s
−
1
−1
1
13
of 0.3 ± 0.8 kcal mol for 2 and by 1.1 ± 0.8 kcal mol for 4.
Table 3. Selected H and C NMR Data for the C Isomers
2
The C isomer is not observed for the pentenyl compound 3.
s
of the Butenyl Compound 2, the Pentenyl Compound 3,
We propose that, for all three compounds, the energy of the
minor isomer is dictated by the energy penalty that arises from
steric repulsions between the two olefin groups. The small bite
angle of the butenyl ligand in 2 means that the two olefins are
far from one another, so that the nonbonded repulsion
between the two olefin groups in each of the isomers is small.
The result is that both isomers have about the same energy.
The longer chain length of the hexenyl ligand in 4 means that
the olefinic C=C vectors in 4 are nearly perpendicular to the
square plane of the Pt center; the nonbonded repulsions
and the Hexenyl Compound 4 in Toluene-d at 20 °C
8
Unless Otherwise Specified
a
2
3
4
1H δ (α-CH
1H 2JPtH (α-CH
)
1.32, 0.91
86, 105
4.39
2.25, 1.64
37, 103
4.51
2.30, ∼1.80
b
2
)/Hz
2
62, 101
1
b
H δ (−CH)
∼4.14
1
1
H δ (CH )
3.65, 3.51
15.2, 8.8
5.46
3.57, 3.40
14.7, 8.0
43.3
4.28, 3.39
15, 9.1
46.7
2
3
H J (olefinic)/Hz
HH
13C δ (α-CH2)
between the two olefin groups in the C and C isomers are
13C 1JPtC (α-CH )/Hz
2
s
610
844
861
2
probably not small, but owing to the perpendicular orientation
they are likely to be rather similar, so that again the two
isomers have about the same energy.
13
C δ (−CH)
97.37
74.76
0
104.67
88.76
16
104.51
82.86
13
C δ (CH )
2
13
1
1
c
C J (−CH)/Hz
20
PtC
For the pentenyl compound 3, the bite angle is sufficiently
c
3C 1JPtC (CH )/Hz
26
29
28
2
large to place the olefins in close proximity in the C isomer,
s
but the conformational constraints of the β-methyl groups and
the five-carbon backbone make steric clashing of the olefin
a
At −60 °C. At room temperature, the two isomers are exchanging
b
c
195
rapidly. Overlapping with other resonances. At −70 °C. The Pt
1
3
1
groups unavoidable, particularly for the C isomer of 3. The
s
satellites for the olefinic carbons of 2 in the low-temperature C{ H}
73
energy of this isomer is therefore higher, which results in it
being present in solution in amounts too small to detect by
NMR spectrum are broad owing to chemical shift anisotropy, so
these Pt−C coupling constants are somewhat less precise.
F
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
carbon and 0 Hz (!) for the methine carbon. This zero
coupling constant to the methine carbon is consistent with the
observation in the solid state that this carbon atom is planar
and apparently unhybridized. One possibility is that, owing to
the constraints of the short chain length, the methine p-orbital
is located close to the nodal surfaces of the linear combinations
of the valence orbitals on Pt; as a result, the overlap of the
7
1,72
methine carbon with Pt is relatively weak.
1
Small J
coupling constants are also observed in the
PtC
1
3
1
C{ H} NMR spectra of the pentenyl compound 3 (29.4 Hz
for the methylene carbon and 17 Hz for the methine carbon at
0 °C) and the hexenyl compound 4 (20 Hz for the methylene
2
carbon and 28 Hz for the methine carbon at −70 °C). The
small Pt−C coupling constants in 2−4 show that the CC
bonds remain bound to platinum in solution, but suggest that
the interactions are relatively weak.
We describe all of the JPtC coupling constants to the olefin
groups in 2−4 as “small” because in other platinum(II) cis-
dialkyl diolefin complexes they are usually much larger: 55 Hz
Figure 6. (a) View of the pentenyl compound 3 along the C −C
β
α
bond, (b) Newman projection of 3 along the C −C bond, (c) view
β
α
1
of 3 along the C −C bond, and (d) Newman projection of 3 along
γ
δ
the C
−C bond. In (a) and (c), the platinum atoms are in red and
δ
γ
the carbon atoms are in black. Hydrogen atoms are depicted as
arbitrarily sized spheres.
in (COD)PtMe , 46 Hz in the norbornadiene compound
2
7
4
(
(
NBD)PtMe2, and 44.6 Hz in the neopentyl compound
7
5
1
Coordination Dynamics of the Pt−Olefin Bond. In
compounds 2−4, the binding of the platinum atom to one face
COD)Pt(CH CMe ) . Interestingly, the JPtC coupling
2
3 2
constants to the olefinic carbons in 3 are also smaller than
those of 50 and 40 Hz in the unmethylated analog
of the olefin makes the two hydrogen atoms of each -CH - unit
2
7
6
in the alkenyl ring (and the two methyl groups of the CMe
Pt[(CH ) CHCH ] , 1. This latter comparison suggests
2
2
3
2 2
1
unit) diastereotopic. At low temperatures, the H NMR spectra
that the 2,2-dimethyl substitution in complexes 2−4 weakens
the Pt−olefin bonding, perhaps by a combination of electronic
factors (the trans influence of the neo-alkyl groups is larger
than that of n-alkyl groups) and steric factors (the 2-methyl
groups force the rings to adopt conformations that are less
favorable for Pt−olefin overlap).
Long-range Pt−H couplings to ligand protons in platinum
complexes are known to follow a Karplus-type dependence on
show that these pairs of H atoms (or pairs of Me groups) are
indeed chemically inequivalent, as mentioned above. For all
three compounds, however, an exchange process occurs that
7
7
1
causes the H NMR peaks arising from these groups initially to
broaden and then to coalesce as the temperature is raised.
Because the terminal vinyl protons of 3 show no broadening,
the only mechanism that can explain these exchange processes
is decomplexation of the olefin from the Pt center and
78,79
the torsional angle between the Pt−C and C−H bonds,
and we can use this dependence to assign the H NMR
8
1,82
1
recoordination through the other face of the olefin;
such
83
1
exchange processes have been called “alkene flipping”. This
same exchange process also results in the exchange of the C2
and C isomers of 2 and 4 and requires that the C isomer of 3
resonances in 2−4. For example, in the H NMR spectrum of
the pentenyl compound 3 at −20 °C, one of the α-CH
2
s
s
protons, one of the β-methyl groups, and one of the γ-CH
2
also be formed transiently in solution, although in amounts too
small to detect. We note that similar exchange processes have
protons, exhibits a large JPt−H coupling constant, whereas the
partner group in each diastereotopic pair shows little or no
1
2
84
been seen in other η ,η -ω-olefin complexes of platinum.
4
observable JPt−H coupling (in fact, 3 shows a four-bond J
Pt−H
For the pentenyl compound 3, which exists in solution
coupling of 10 Hz to the equatorial 2-Me group). If we assume
that the pentenyl ligands adopt chair conformations as
observed in the solid state (Figure 6), then the larger couplings
can be assigned to the equatorial groups (which form torsion
angles relative to the Pt−C bond that are close to 180°) and
the smaller couplings to their axial counterparts (which
describe torsion angles that are close to 90°). These
assignments have been confirmed by a series of 1D NOE
experiments (Figure S4.7).
1
exclusively as the C isomer, the H NMR line shapes of the
2
diastereotopic methyl groups (Figure 7) were simulated in
terms of a two-site exchange process to determine the olefin
85
face exchange rate as a function of temperature. A fit of 24
rates between 11 and 69 °C to the Eyring equation gives
‡
−1
‡
activation parameters of ΔH = 19 ± 1 kcal mol and ΔS = 9
−1 −1
±
3 cal mol ·K for the exchange process between the
methyl groups in 3 (Figure S6.11 and S6.12). Because the rate
of exchange is half of the rate of olefin decomplexation (see the
Experimental Section), the activation parameters for olefin
1
The J coupling constant to the α-CH group of 610 Hz
PtC
2
seen for the butenyl compound 2 is significantly smaller than
those of 844 and 861 Hz observed for the pentenyl and butenyl
compounds 3 and 4, respectively. The latter are similar to the
‡
−1
‡
decomplexation are ΔH = 19 ± 1 kcal mol and ΔS = 10 ±
−1 −1
3
cal mol ·K .
Similarly small but positive activation entropies of 12−14 cal
1
−
1
−1
JPtC coupling constant of 844 Hz seen in the neopentyl
mol ·K seen for olefin face exchange processes in some 4-
75
13
86,87
complex (cod)Pt(CH CMe ) . The C NMR chemical shift
pentenyl zirconocene compounds
were proposed to
2
3 2
for the α-CH group in 2 of δ 5.5 is also quite different from
indicate a mechanism in which the metal−olefin bond is
broken without the concomitant association of solvent
molecules (solvent-assisted mechanisms should have negative
activation entropies). We propose, therefore, that olefin face
2
those of δ 43.3 and 46.7 seen for 3 and 4. We attribute these
differences to the short chain length of the butenyl ligand in 2:
the greatly perturbed interligand angles evidently affect the
Pt−C σ interactions, as well as the values of the chemical shift
exchange in 3 also occurs by breaking the platinum−olefin
8
0
41,88
tensor.
bond
without the involvement of solvent. The resulting
G
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
conformational flexibilities) are ∼3−8 cal mol K−1 when the
−1
exchange process is purely intramolecular, but are 8−12 eu
8
1,82
more negative if solvent plays a role.
−
1
−1
The 4 to 10 cal mol ·K entropies of activation for 2, 3,
and 4 (Table 4) indicate that olefin face exchange in all three
ω-alkenyl compounds takes place by a mechanism involving a
transition state in which the metal−olefin bond is broken but
(
most likely) replaced with an agostic interaction between Pt
8
1,83,93,95−98
and one of the ω-hydrogen atoms.
At 20 °C, the
rate of the olefin decomplexation increases in the following
−
1
−1
order: 6 s for the pentenyl compound 3, ∼40 s for the
−
1
hexenyl compound 4, and ∼3000 s
for the butenyl
compound 2. Notably, the rate of thermal decomposition of
1
2
the platinum η ,η -2,2-dimethyl-ω-alkenyl complexes at room
temperature increases in the same order (i.e., 3 < 4 < 2). This
correlation suggests that the intermediate formed during the
olefin face exchange is also an intermediate in the thermolysis
pathway.
1
Figure 7. Variable-temperature (VT)- H NMR spectrum of the
pentenyl compound 3 in toluene-d8.
Owing to its higher thermal stability compared to the other
new alkenyl compounds, the pentenyl compound 3 shows the
best potential as a CVD precursor, a topic to which we now
turn.
intermediate is either a 3-coordinate species with an empty site
89−91
(
if the olefin simply dissociates)
or a 4-coordinate C−H σ
complex (if the metal−olefin bond is broken but is replaced
with an agostic interaction).
have been proposed as intermediates in olefin face exchange
83,92,93
Cold Wall Chemical Vapor Deposition of Thin Films
from 3. Cold wall CVD experiments were performed in a
In fact, C−H σ complexes
99,100
6
93
0
83,92,94
high-vacuum chamber described elsewhere,
and the onset
processes in certain d rhenium and d zirconium
2
of nucleation and film growth were monitored by real-time
complexes. Compared to the ground-state η -olefin complex,
the transition state that leads to the C−H σ-complex
intermediate should have increased degrees of freedom due
to the expansion of the ring size, so that a small and positive
entropy of activation is expected.
101
spectroscopic ellipsometry (SE).
Film thicknesses were
measured by scanning electron microscopy (SEM), and growth
rates (in atoms per area) were measured by Rutherford
backscattering spectroscopy (RBS). The films were grown
from the pentenyl compound 3, which was kept in a reservoir
maintained at 75 °C and transported to the growth chamber
with an argon carrier gas.
For the butenyl and hexenyl compounds 2 and 4, analysis of
1
the H NMR line shapes is more complicated because of the
effects of coupling to platinum and also because these
We find that, at substrate temperatures between 275 and 350
compounds exist in solution as a mixture of the C and Cs
2
°
C, smooth and reflective films can be grown from 3 on
isomers (in ratios of ∼2:1 and ∼10:1, respectively) some of
1
various substrates (SiO /Si, Al O , and VN). For example, on
whose H NMR resonances overlap. For both compounds, we
2
2
3
a SiO /Si substrate at 330 °C, a 17 nm film with a rms
found that the rate is best determined by fitting the line shapes
2
1
3
1
roughness of 1.7 nm was deposited in less than 10 min (Figure
of selected C{ H} NMR resonances. For 2, fitting of the
8
). Interestingly, unlike growth from (C H Me)PtMe , which
olefin face exchange rate at 11 different temperatures between
5
4
3
7
,8,11,28,29
‡
shows significant nucleation delays,
growth from 3
−
46.4 and 22.2 °C to the Eyring equation gives ΔH = 13.8 ±
−
1
‡
−1 −1
occurs without delay on all substrates tested (Figure 8).
While maintaining the same substrate reservoir temperature
and carrier gas flow rate, the rate of deposition (Å/min)
increases exponentially with an increasing temperature
between 275 and 330 °C (Figure S8.2); this result shows
that growth under these conditions is in the reaction-limited
regime. The deduced activation energy of 18 kcal/mol (0.78
0
.6 kcal mol and ΔS = 4 ± 2 cal mol ·K for the C
−1
2
‡
‡
isomer and ΔH = 13.5 ± 0.6 kcal mol and ΔS = 4 ± 2 cal
−
1
−1
mol ·K for the C isomer (Figures S6.5−S6.9). For 4, fitting
s
of the olefin face exchange rate at 8 different temperatures
‡
between −2.8 and 33.6 °C to the Eyring equation gives ΔH =
−
1
‡
−1 −1
1
7 ± 1 kcal mol , ΔS = 5 ± 4 cal mol ·K for the C
−1 −1
2
‡
‡
isomer and ΔH = 16 ± 1 kcal mol , ΔS = 6 ± 4 cal mol ·
−
1
K
for the C isomer (Figures S6.16−S6.18).
eV) is considerably smaller than the >1.5 eV activation energy
s
8
The solvent is unlikely to participate in these exchange
seen for growth from (C
H
5
Me)PtMe
4
3
. In addition, the onset
processes: activation entropies for olefin face exchange in metal
complexes of the 2-methylhex-5-en-2-olate ligand (which is
structurally analogous to the 2,2-dimethylhex-5-en-1-yl ligand
in 4 in terms of having the same chain length and similar
temperature for CVD growth from 3 is ∼25 °C lower than that
5
of ∼300 °C for (C H Me)PtMe . All of these observations are
5
4
3
consistent with the conclusion that 3 is a more reactive
molecule under CVD conditions than the cyclopentadienyl
Table 4. Activation Parameters for CC Bond Decomplexation in Toluene-d , and Correlation With Thermal Stability, for
8
the Butenyl Compound 2, the Pentenyl Compound 3, and the Hexenyl Compound 4
2
3
4
‡
−1
ΔH , kcal mol
13.8(6), C 13.5(6), Cs
19(1)
10(3)
16(1)
17(1), C 16(1), C
2
2
s
ΔS , cal mol ·K−1
‡
−1
4(2), C 4(2), Cs
5(4), C 6(4), C
s
2
2
‡
−1
ΔG , kcal mol at 20 °C
thermal stability at 20 °C
12.6(8), C 12.3(8), Cs
16(2), C 14(2), C
2
2
s
unstable
stable for weeks
unstable
H
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Figure 8. (a) SEM image of a Pt film grown from the pentenyl compound 3 on silicon at 330 °C (thickness: 19 nm); (b) SEM image of a Pt film
grown from the pentenyl compound 3 on silicon at 330 °C in the presence of an oxygen plasma (thickness: 11 nm); (c) real-time ellipsometry
curve (at 2.65 eV) of Pt growth at 330 °C on SiO /Si, VN, and Al O . The platinum precursor is introduced at the time of 0. The ellipsometry
2
2
3
signal responds immediately to the introduction of the precursor, which suggests that there is no nucleation delay on these three substrates; (d)
atomic force microscopy (AFM) image of a Pt film grown without and with oxygen plasma from the pentenyl compound 3 at 330 °C on an Al O
2
3
substrate.
compound, as expected owing to the presence of the labile ω-
alkenyl groups.
The purity, crystallinity, and resistivity of the film grown in
the presence of the oxygen plasma also show significant
improvements (Figure S8.3 and S8.4): the oxygen and carbon
contents are below the RBS detection limit (variously
As is typically seen for organoplatinum CVD precursors, the
films obtained from 3 contain both platinum and carbon when
the depositions are conducted in the absence of oxidative or
reductive co-reactants. RBS analysis shows that the films grown
under such conditions consist of ∼50 atom % platinum with
the balance being carbon. Because of the presence of the
9
estimated to be between 5 and 30%) vs the ∼50% carbon
content seen for the films grown from 3 in the absence of a
plasma. The films grown in the presence of the remote oxygen
plasma contain nanocrystalline platinum with crystalline
domains of ∼7 nm in size (Figure S8.4) and are reasonably
smooth: the rms roughness of 2.2 nm is comparable with the
1.7 nm value seen in the absence of the plasma (Figure 8). The
use of the oxygen plasma also lowers the resistivity from 830 to
450 μΩ·cm for a 12 nm film deposited on Al O . It may prove
1
02−105
latter,
a 12 nm film deposited at 330 °C on Al O has a
2 3
relatively high resistivity of 830 μΩ·cm vs the 10.6 μΩ·cm
10
value seen for bulk platinum.
Remote Oxygen Plasma-Assisted Chemical Vapor
Deposition of 3. The quality of films grown by CVD from
organoplatinum precursors is usually improved, and the
deposition rate is often increased, by addition of oxidants or
reductants such as O or H to the gas stream, which react with
2
3
possible to obtain films with even faster growth rates, lower
carbon contents, and lower resistivities by employing a
stronger plasma source or a different oxidant such as ozone.
Hot Wall Chemical Vapor Deposition of Thin Films
from 3. To obtain information about the chemistry that
occurs during the CVD growth from 3, CVD experiments were
performed in a static vacuum (∼10 mTorr) vertical hot wall
deposition system equipped with a liquid nitrogen-cooled
2
2
5
,8,106
7,10
excess carbon to remove it as COx
or C H .
We find,
x
y
however, that addition of reactant gases such as O or H to a
2
2
flux of 3 has almost no effect on the deposition rate. This
finding suggests that the Pt surface generated from 3 is
poisoned in such a way that it is unable to activate these
molecules. We therefore investigated lowering the carbon
content of the films by employing a reactant that does not rely
on the catalytic ability of the platinum surface: atomic oxygen
107
108
NMR tube to trap the organic byproducts. The precursor
reservoir was kept at 75−80 °C to achieve reasonable growth
rates, and the hot zone was heated either to 250 ± 10 or to 300
± 10 °C. At both temperatures, smooth and reflective films can
be grown on several different substrates (including silicon,
gold, and glass).
9
generated by a low-power remote plasma. Using the same
cold wall equipment, we found that deposition from 3 at 330
°C in the presence of a remote oxygen plasma (see the
experimental section) also occurs with no nucleation delay,
and that the rate of deposition is increased to ∼220% of its
value in the absence of the plasma.
The morphology and composition of the films grown in the
static hot wall CVD apparatus are analogous to those obtained
in the cold wall experiments in the absence of an oxygen
I
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
plasma (Figure S9.1−S9.6). Glancing angle XRD analysis
shows that the ∼300 nm thick films grown at 250 °C contain
nanocrystalline platinum, with crystalline domains ∼3.5 nm in
size. The X-ray photoelectron spectra (XPS) show that the
films grown on glass consist of 47% Pt, 40% C, and 13% O and
the films grown on gold consist of 56% Pt, 38% C, and 5% O
Scheme 1. Fate of the Carbon and Hydrogen Atoms during
the CVD of the Films from the Pentenyl Precursor 3
a
(
the oxygen probably coming from postgrowth air exposure).
In addition, for a film grown at 250 °C on glass, the electrical
resistivity was 950 μΩ-cm, which is similar to the conductivity
of the films grown at 330 °C in the cold wall apparatus.
1
Quantitative H NMR spectroscopy shows that the principal
a
The products that are the result of skeletal rearrangements are
indicated with the wavy lines.
organic products generated at 250 °C under hot wall CVD
conditions are 4,4-dimethyl-1-pentene (0.40 equiv per mole of
3
consumed), 4,4-dimethyl-2-pentene (0.80 equiv), and 2,2-
genated) dimethylpentenyl ligands to produce 4,4-dimethyl-
pentenes and 2,2-dimethylpentane, which dissociate from the
surface and are carried away.
dimethylpentane (0.24 equiv) (Figure S10.1−S10.7). Togeth-
er, these three products account for 72% of the carbon atoms
in 3. All of these products are generated by adding hydrogen
atoms to the 2,2-dimethylpentenyl ligands in 3. When this
CVD experiment is repeated in an apparatus whose surfaces
have been deuterated (by treatment first with 2 M NaOD in
D O and then with 2 M DCl in D O), the organic byproducts
contain no deuterium above natural abundance levels, as
judged by H NMR spectroscopy. This experiment shows that
In contrast, the dehydrogenated ligands have no easy
pathway to desorb from the surface because the 2,2-dimethyl
substituents block low-energy pathways such as β-hydrogen
elimination and formation of conjugated dienes. Other possible
dehydrogenation products, such as acetylenes or allenes, are
either not formed or are too strongly bound to the surface to
desorb. Instead, the only way in which the dehydrogenated
ligands from 2 can desorb from the surface is by undergoing
processes that involve making or breaking C−C bonds,
2
2
2
the hydrogen that appears in the hydrogenated ligands does
not come from adventitious sources such as the walls of the
CVD chamber and must instead come from the precursor.
Together, these experiments show that, when 3 is thermolyzed
at 250 °C under CVD conditions, 72% of the pentenyl ligands
113,116
reactions that are known to occur on Pt surfaces.
In
the present case, these skeletal rearrangements lead to the
desorption of methylenecyclobutanes, cyclobutenes, penta-
dienes, and toluene, as well as isobutylene.
are converted to C products (4,4-dimethylpentene isomers
7
and 2,2-dimethylpentane) by acquiring hydrogen atoms from
the remaining 28% of the pentenyl ligands. In other words,
some of the pentenyl ligands must be dehydrogenated by C−H
bond activation, a process that readily occurs on Pt surfaces at
We conclude this section with a comparison of the CVD
behavior of the unmethylated compound 1 and its β-stabilized
analog 3. In contrast to the byproduct distribution seen for 3
(which consists mostly of hydrogenated ligands), for the
unmethylated analog 1, both hydrogenated ligands (pentene
isomers) and dehydrogenated ligands (pentadiene isomers) are
5,109−115
these temperatures.
Most of the pentenyl ligands that are dehydrogenated under
CVD conditions at 250 °C are also released as volatile organic
products. Among these dehydrogenated byproducts are the C7
hydrocarbons 1,1-dimethyl-3-methylenecyclobutane (0.04
equiv per mole of 3 consumed), 1,3,3-trimethylcyclo-1-butene
35
generated in roughly equal amounts. A key attribute is that
both types of products (hydrogenated ligands and dehydro-
genated ligands) have low-energy pathways to form volatile
desorbates. The ease of removing both hydrogenated and
dehydrogenated ligands from the surface is consistent with the
relatively low carbon content (Pt/C ratio of ∼4.6:1) in the
films grown from 1 in the absence of co-reactants. Compared
to the CVD of 3, the rapid desorption of most of the ligands
during the CVD of 1 should generate a more active Pt surface,
which will catalyze additional deposition from 1. Such
autocatalytic behavior should lead to the formation of rough
films because growth is faster than nucleation. In fact, rough
films indicative of autocatalytic CVD are exactly what we see
(
0.06 equiv), 2,4-dimethyl-1,3-pentadiene (0.08 equiv), 1,1-
dimethyl-2-methylenecyclobutane (0.06 equiv), and toluene
0.02 equiv) (See Supporting Information, Section S10). For
(
all of these compounds, the dehydrogenation is accompanied
by a rearrangement of the pentenyl skeleton by means of
breaking and/or making C−C bonds. These dehydrogenated
C products, together with the small amount of isobutylene
7
and higher-molecular-weight (most likely C ) products
1
4
formed, account for 22% of the carbon atoms in 3. Combined
with the 72% of the carbon atoms in 3 that are released as
hydrogenated hydrocarbons, we find that 94% of the carbon
atoms are released during CVD as volatile byproducts; the
remaining 6% of the carbon atoms, or about one carbon atom
per Pt, remain in the Pt film. The finding that just under one
carbon atom per platinum is retained in the film is consistent
with the XPS results.
In summary, the NMR and XPS data enable us to determine
what happens to essentially all of the atoms in the precursor
under CVD conditions; the stoichiometry of the reaction is
shown in Scheme 1.
35
for 1.
Comparison of CVD Behavior of 3 and (C H Me)-
5
4
PtMe . There are three important differences in the CVD of Pt
3
from 3 and from (C H Me)PtMe : (1) CVD from 3 initiates
5
4
3
very rapidly, whereas growth from (C H Me)PtMe suffers
5
4
3
from lengthy nucleation delays, (2) co-flow of O or H has
2
2
almost no effect on the rate of the film deposition from 3, but
8
greatly increases the rate of deposition from (C H Me)PtMe ,
5
4
3
and (3) films obtained from 3 are very smooth, but those
obtained from (C H Me)PtMe are quite rough.
5
4
3
We attribute the difference in nucleation delay to the lability
of the alkenyl ligands in 3: specifically, we propose that
decomplexation of the olefin groups from Pt, which we know
This product distribution is consistent with a mechanism in
which about a quarter of the dimethylpentenyl ligands
5
,109−114
90
dehydrogenate on the surface.
The liberated hydrogen
has a low barrier, forms highly reactive Pt species that serve
atoms combine on the surface with the other (nondehydro-
as nucleation initiators. In contrast, (C H Me)PtMe is
5
4
3
J
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
kinetically very inert and has no low-energy pathways to break
when the depositions are conducted in the absence of a co-
reactant. Interestingly, by employing a remote oxygen plasma,
the amount of carbon impurities is significantly reduced and
the resistivity of the films is lowered, without significantly
degrading the smooth film morphology. The formation of
carbon impurities is related to the reactions that 3 undergoes
under CVD conditions: activation of the C−H bonds of some
of the pentenyl ligands leads to the formation of surface
hydrogen atoms, which react with other ligands to generate
2,2-dimethylpentenes and 4,4-dimethylpentane. These hydro-
genated species desorb from the surface, but the dehydro-
genated 2,2-dimethylpent-4-en-1-yl ligands can desorb only by
undergoing higher-energy skeletal rearrangements and, as a
a Pt−C bond; the result is a long nucleation delay in ALD and
8
a higher activation energy for growth in CVD.
To explain the other two differences, we note that CVD
from organoplatinum precursors inevitably leads to the
formation of carbonaceous adsorbates, and such species in
116,117
turn are known to influence the reactivity of the surface.
In this context, we know that 3 and (C H Me)PtMe leave a
5
4
3
similarly large amount of carbon impurities in the film when
10,118
used as single-source precursors.
Thus, the difference
between these two precursors must not lie in the amount of
carbon generated, but instead must reflect a difference in its
chemical nature.
It is known that step sites on Pt surfaces are able to break
result, they deactivate surface sites that normally react with H
or O and catalyze further deposition from 3. This behavior is
2
116,119
H−H and C−H bonds
and promote the activation of
O , and that kink sites are able to break C−C as well as H−
2
1
20
unlike that seen for other organoplatinum precursors, which
show no such poisoning effect. Further evidence in support of
a poisoning effect is that removal of the carbonaceous species
through the use of an oxygen plasma leads to faster growth
rates. The combination of fast nucleation and slow (i.e., non-
autocatalytic) growth is what is responsible for the highly
2
116
H and C−H bonds. We propose that the inability of the Pt
films grown from 3 to activate O and H arises because the
2
2
step and kink sites become poisoned by the dehydrogenated
pentenyl ligands, which cannot desorb unless they undergo a
skeletal rearrangement. Although the kink sites can promote
C−C bond activation and can be reactivated when the
rearranged ligand desorbs, this reactivation is slow because the
skeletal rearrangement has a high barrier. When the reactivated
kink sites bind to another dehydrogenated ligand, they become
temporarily deactivated again. The net effect is that both the
step and kink sites are poisoned, so that the surface is largely
unable to chemisorb O or H .
122
smooth films that are deposited from 3.
All in all, the results provide an interesting example of how
ligand engineering can be used to develop CVD precursors
that afford films with desired properties. In addition to tuning
the reactivity of the precursor molecule itself, the design can
also dictate what organic groups are delivered onto the surface
and those groups can affect the film morphology by changing
the reactivity of the surface.
2
2
The poisoning can also explain why the films obtained from
this precursor are so smooth: the poisoning inhibits
autocatalytic deposition of 3, which is known to occur with
most Pt precursors. The resulting slower deposition rate,
combined with the fast nucleation rate, creates circumstances
in which growth occurs by constant renucleation, a mechanism
EXPERIMENTAL SECTION
■
All experiments were carried out in a vacuum or under an argon
atmosphere using standard Schlenk techniques unless otherwise
specified. All glassware was oven-dried before use. Solvents (pentane,
diethyl ether) were distilled under a nitrogen atmosphere from
sodium/benzophenone immediately before use. Magnesium turnings
1
21
that is known to lead to enhanced film smoothness.
Concluding Remarks. We have made three new platinum-
1
23
(
II) compounds containing chelating ω-alkenyl ligands cis-
(Fisher) were used as received. The compounds (COD)PtCl
,
2
4-
1
24
1
2
1
2
bromo-3,3-dimethyl-1-butene,
(2,2-dimethylbut-3-en-1-yl)-
bis(η ,η -2,2-dimethylbut-3-en-1-yl)platinum (2), cis-bis(η ,η -
1
24
124
1
2
lithium,
5-bromo-4,4-dimethyl-1-pentene,
and 6-bromo-5,5-
2
,2-dimethylpent-4-en-1-yl)platinum (3), and cis-bis(η ,η -2,2-
1
24
dimethyl-1-hexene,
were prepared as described elsewhere.
dimethylhex-5-en-1-yl)platinum (4). All three compounds are
dynamic in solution: the CC bonds reversibly decomplex at
rates that are fast on the NMR time scale, with the pentenyl
compound 3 showing the slowest decomplexation rates and
the greatest thermal stability.
Benzene-d and toluene-d were purchased from Cambridge Isotope
6
8
Laboratories in 1 mL ampoules and used without purification.
@
SilicaFlash P60 silica gel was purchased from Silicycle.
Elemental analyses were performed by the University of Illinois
Microanalytical Laboratory. The FTIR spectra were acquired on a
Thermo Nicolet IR200 spectrometer as mineral oil mulls between
KBr plates and processed using the OMNIC software package with
automatic baseline corrections. Melting points were acquired on a
Thomas−Hoover Uni-Melt apparatus in sealed capillaries under an
In addition to being thermally stable under ambient
conditions, compound 3 is volatile and stable toward air and
water and as such is a potentially useful new CVD precursor
for platinum. At temperatures of 275−350 °C in vacuum, it
nucleates very well on SiO /Si, Al O , and VN to deposit
1
2
13
argon atmosphere. The 1D H, H, and C NMR data were recorded
on a Varian Inova 400 spectrometer at 9.39 T, a Varian Inova 500
spectrometer at 11.74 T, a Varian Inova 600 spectrometer at 14.09 T,
or a Bruker Avance III HD spectrometer equipped with a 5 mm
2
2
3
highly smooth thin films; this behavior is unlike most other
platinum CVD precursors, which generally form large particles
due to nucleation delays and autocatalytic growth kinetics.
Compared to (C H Me)PtMe , which exhibits long nucleation
1
1
BBFO CryoProbe at 11.74 T. The 1D H− H NOE data and the
1
95
5
4
3
Pt NMR data were recorded on a Varian Inova 600 spectrometer at
14.09 T. Chemical shifts are reported in δ units (positive shifts to
7
28
delays in both CVD and ALD, deposition from 3 is
characterized by a much shorter nucleation delay. This effect is
probably a consequence of the availability of a low-barrier
process in 3 (initiated by olefin decomplexation) to initiate the
CVD mechanism; no such low-barrier process is available for
the (C H Me)PtMe precursor. The fast rate of nucleation for
1
13
higher frequency) relative to tetramethylsilane (TMS) ( H, C), set
by assigning appropriate shifts to residual solvent signals or to an
1
95
external standard of aqueous 1.0 M K PtCl
( Pt) by sample
2
6
replacement. X-ray crystallographic data were collected by the staff of
the G. L. Clark X-ray Laboratory at the University of Illinois. The gas
chromatography−mass spectrometry (GC−MS) spectra were col-
lected on a GC−MS system (Agilent Inc, Palo Alto, CA) consisting of
an Agilent 7890 gas chromatograph, an Agilent 5975 mass selective
detector, and a HP 7683B autosampler by the staff of the Roy J.
Carver Biotechnology Center at the University of Illinois. The peaks
5
4
3
3
also makes it a promising CVD precursor to deposit Pt
nucleation layers.
As is common for all organoplatinum CVD precursors, the
films obtained from 3 contain both platinum and some carbon
K
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
were evaluated using the AMDIS 2.71 (NIST, Gaithersburg, MD)
program and identified with the aid of the libraries NIST08 (NIST,
MD) and W8N08 (Palisade Corporation, NY).
was charged with magnesium turnings (4.0 g, 170 mmol), diethyl
ether (20 mL), and 1,2-dibromoethane (9 drops). The mixture was
heated to reflux for 12 h and then was treated with 5-bromo-4,4-
dimethylpent-1-ene (4.25 g, 24.0 mmol) in one portion while still hot.
The mixture was heated to reflux for a further 12 h and then was
filtered. The solid residue was extracted with diethyl ether (2 × 10
mL), filtering each time. The filtrates were combined to give a light
yellow solution, which was used in the next step. Yield: 24 mL of a
0.41 M solution (41%).
Isothermal thermogravimetric analysis (TGA) data were collected
on a TA Instruments Q50 TGA system by the staff of the Materials
Research Laboratory at the University of Illinois. Dynamic TGA data
were collected on a Cahn TherMax 500 TGA instrument. SEM data
were collected on a Hitachi S-4800 High Resolution SEM instrument.
GI-XRD data were collected on a PANalytical MRD instrument and
processed with the JADE software package. XPS data were collected
on a PHI 5400 instrument and processed with the CasaXPS software
1
2
cis-Bis(η ,η -2,2-dimethylpent-4-en-1-yl)platinum (3). To a
suspension of (COD)PtCl (1.50 g, 3.99 mmol) in diethyl ether (20
2
package. Before being analyzed, the films were treated with ozone for
mL) at 0 °C was added (2,2-dimethylpent-4-en-1-yl)magnesium
bromide (23 mL of a 0.41 M solution in diethyl ether; 9.4 mmol)
dropwise with vigorous stirring. The mixture was stirred at 0 °C for 2
h and then was warmed to room temperature. The mixture was
filtered, and the solid was extracted with diethyl ether (20 mL). The
filtrate and the filtered extract were combined and then the solvent
was removed under vacuum. The resulting slurry was extracted with
pentane (2 × 20 mL). The extracts were filtered, combined, quenched
with water (5 drops), and dried over magnesium sulfate. Pentane was
removed on a rotary evaporator, and the product was purified by silica
+
1
min and then sputtered with Ar for 4 min to remove surface carbon
contamination arising from atmospheric exposure. Rutherford back-
scattering spectra were collected on a 3SDH Pelletron instrument.
Atomic force microscopy data were collected on a Cypher S AFM
microscope. Electrical conductivities were measured by the four-point
probe method.
1
2
cis-Bis(η ,η -2,2-dimethylbut-3-en-1-yl)platinum (2). To a
solution of (COD)PtCl (0.28 g, 0.75 mmol) in diethyl ether (20
2
124
mL) at −78 °C was added (2,2-dimethylbut-3-en-1-yl)lithium (20
mL of a 0.09 M solution in pentane, 1.8 mmol). The mixture was
stirred at −78 °C for 5 h and was allowed to warm to −20 °C over 5
h. The solvents were slowly removed under vacuum, keeping the
temperature at −20 °C to prevent vaporization of the product. The
residue was extracted with pentane (2 × 20 mL), and the extracts
gel column chromatography using pentane as the eluent (R = 0.57).
f
The removal of the solvent from the eluate gave a light yellow oil. The
oil was recrystallized from ethanol (∼0.2 mL) at −20 °C to give the
product as a light yellow solid. Yield: 0.99 g (64%). Anal. calcd for
C H Pt: C, 43.2; H, 6.73. Found: C, 42.9; H, 6.56. Mp ∼20 °C.
1
4
26
1
95
1
1
were filtered, combined, cooled to 0 °C, and quenched with H O (5
2
Pt{ H} NMR (129 MHz, C D , 20 °C): δ −3777 (s). H NMR
7
8
drops). The resulting solution was dried over Na SO at 0 °C, cooled
2
4
(500 MHz, C D , 20 °C): δ 4.51 (m, 2 H, −CH), 3.57 (d, 2 H,
7
8
to −78 °C, and filtered. The filtrate was evaporated under vacuum at
3
2
2
3
J
= 14.7 Hz, J = 39.6 Hz, CH ), 3.40 (d, 2 H, J = 8.0 Hz,
2 HH PtH
HH
PtH
2
HH
−
20 °C to give a brown oil. The product was purified by column
2
2
JPtH = 29.5 Hz, CH ), 2.25 (d, J = 11.8 Hz, J = 36.6 Hz, 2
chromatography on silica gel with pentane as the eluent (R = 0.30).
2
3
f
H, Pt-CH equatorial), 1.91 (dd, J = 11.4 Hz, J = 4.2 Hz, 2 H,
2
HH
HH
Slow evaporation of the eluate at −20 °C gave the product as a white
solid, which melts to form a light yellow liquid at room temperature.
Yield: 0.16 g (58%). Anal. calcd for C H Pt: C, 39.9; H, 6.14.
2
2
3
-CH equatorial), 1.64 (d, J = 11.7 Hz, J = 102.9 Hz, 2H, Pt-
CH axial), 1.47 (dd, J = 11.4 Hz, J = 9.6 Hz, 2 H, 3-CH
axial), 1.24 (s, 6 H, 2-Me axial), 1.13 (s, 6 H, JPtH = 9.9 Hz, 2-Me
equatorial). C{ H} NMR (126 MHz, C D , 20 °C): δ 104.67 (s,
2 HH PtH
2
3
2
HH
HH
2
1
2
26
4
1
95
1
Found: C, 40.3; H, 6.06. Mp ca. −15 °C. Pt { H} NMR (129 MHz,
13
1
7
8
C D , 20 °C): δ −4019 (br, C isomer, relative intensity = 1), −4110
7
8
s
1
1
JPtC = 16 Hz, −CH), 88.76 (s, J = 29.4 Hz, CH ), 51.14 (s,
PtC
2
(
br, C isomer, relative intensity = 2). At room temperature, the C
2
2
1
1
2-C), 48.69 (s, 3-CH
2
), 43.30 (s, JPtC = 843.6 Hz, 1-CH
2
), 33.08 (s,
and C isomers are exchanging rapidly: H NMR (500 MHz, C D , 20
s
7
8
3
3
3
2
JPtC = 99.6 Hz, 2-Me equatorial), 31.99 (s, J = 32.3 Hz, 2-Me
PtC
°
C): δ 4.39 (dd, 2 H, J = 15.3, 8.8 Hz, J = 40.8 Hz, −CH),
HH
PtH
3
2
axial).
3
.67 (d, 2 H, J = 15.3 Hz, J = 38.6 Hz, CH ), 3.46 (d, 2 H,
HH
PtH
2
1
2
3
cis-Bis(η ,η -2,2-dimethylhex-5-en-1-yl)platinum (4). This
compound was prepared from 6-bromo-5,5-dimethyl-1-hexene via
the Grignard reagent according to the procedure for synthesizing cis-
JHH = 8.8 Hz, CH ), 1.08 (br, 12 H, 2-Me), 0.87 (br, 4 H, Pt-
2
13
1
CH2). C{ H} NMR (126 MHz, C D , 20 °C): δ 99.72 (br,
−
7
8
1
2
CH), 74.69 (s, J = 25.3 Hz, CH ), 37.76 (s, J = 113.7
PtC
2
PtC
1
2
3
1
bis(η ,η -2,2-dimethylpent-4-en-1-yl)platinum. After being purified by
Hz, 2-C), 32.59 (s, J = 40.6 Hz, 2-Me), 5.51 (s, J = 614 Hz, Pt-
PtC PtC
column chromatography using pentane as the eluent (R = 0.50), it
f
CH2).
was further purified by slow evaporation of pentane from an ethanol/
pentane solution at −78 °C. The resulting thermally sensitive yellow
crystals were collected and dried. Yield: 51 mg (10%). Anal. calcd for
Separate NMR peaks can be seen for the two isomers at low
1
temperatures: H NMR (500 MHz, C D , −60 °C): Major isomer: δ
7
8
3
2
4
.39 (dd, 2 H, J = 15.2, 8.8 Hz, J = 39 Hz, −CH), 3.65 (d, 2
HH
PtH
3
2
3
C H Pt: C, 46.0; H, 7.24. Found: C, 46.2; H, 6.91. Mp 46−47 °C.
H, J = 15.2 Hz, J = 41 Hz, CH ), 3.51 (d, 2 H, J = 8.8
16 30
HH PtH 2 HH
1
95
1
2
2
2
Pt{ H} NMR (129 MHz, C D , 20 °C): δ −3778 (s, C isomer,
Hz, J = 19 Hz, CH ), 1.32 (d, 2 H, J = 10 Hz, J = 86 Hz,
7
8
s
PtH
2
HH
PtH
1
2
relative intensity =1), -3797 (s, C isomer, relative intensity = 5). H
NMR (500 MHz, C D , 20 °C): Major isomer: δ 4.28 (d, 2 H, J
Pt-CH ), 1.22 (s, 6 H, 2-Me), 1.07 (s, 6 H, 2-Me), 0.91 (d, 2 H, J
2
2
HH
3
2
13
1
=
=
10.0 Hz, J = 105 Hz, Pt-CH2). C{ H} NMR (126 MHz, C D ,
7
8
HH
PtH
7
8
3
1
1
J
5 Hz, CH ), 4.14 (m, 2 H, −CH), 3.39 (d, 2 H, J = 9.1 Hz,
2
(
0 °C): δ 97.37 (s, −CH), 74.76 (s, J = 26 Hz, CH ), 37.82
2
HH
PtC
2
2
2
2
2
3
= 34.2 Hz, CH ), 2.30 (d, J = 9.3 Hz, J = 64 Hz, 2 H,
s, J = 112 Hz, 2-C), 32.45 (s, 2-Me, J cannot be obtained due
PtH
2
HH
PtH
2
PtC
PtC
3
3
2
to peak overlapping), 32.45 (s, J = 48 Hz, 2-Me, JPtC), 5.46 (s,
JPtC = 610.0 Hz, 1-CH2).
Pt-CH
2
equatorial), 2.14 (m, 2 H, 4-CH
2
), 1.80 (d, JHH = 7 Hz, JPtH
PtC
1
= 105 Hz, 2H, Pt-CH
axial), 1.55−1.75 (m, 4 H, 3-CH
& 4-CH ),
2
2 2
3
1
2
.20 (br, 2H, 3-CH ), 1.15 (br, 6 H, 2-Me equatorial), 1.00 (br, 6 H,
-Me axial). C{ H} NMR (151 MHz, C D , 20 °C): δ 104.51 (br,
Minor isomer: δ ∼4.37 (m, −CH), 3.78 (d, 2 H, J = 15.3 Hz,
2
HH
1
3
1
2J = 36 Hz, CH ), 3.25 (d, 2 H, J = 8.7 Hz, J = 15 Hz, 
2
2
7
8
PtH
2
HH
PtH
1
2
2
−CH), 82.86 (br, CH ), 46.66 (s, J
PtC
= 860.9 Hz, Pt-CH2),
CH ), 1.29 (s, 6 H, 2-Me), ∼1.18 (d, 2 H, J = 10 Hz, J = ∼70
2
2
HH
PtH
3
2
2
40.07 (br, 2-C), 37.72 (s, J = 97 Hz, 2-Me equatorial), 37.33 (br,
Hz, Pt-CH ), 1.04 (s, 6 H, 2-Me), 0.78 (d, 2 H, J = 10 Hz, J =
PtC
2
HH
PtH
3
13
1
JPtC = 20 Hz, 3-CH ), 28.86 (br, 2-Me axial), 28.05 (br, 4-CH ).
1
(
20 Hz, Pt-CH2). C{ H} NMR (126 MHz, C D , 20 °C): δ 100.80
s, −CH), 74.93 (s, J = 25 Hz, CH ), 37.54 (s, J = 115
Hz, 2-C), 32.79 (s, J = 56 Hz, 2-Me, J ), 32.50 (s, 2-Me, J
PtC
2
2
7
8
1
2
Minor isomer is not observable at room temperature, but a second
PtC
2
PtC
3
3
3
13
set of peaks are clearly shown in the C NMR spectrum in toluene-d
PtC
PtC
8
1
13
1
cannot be obtained due to peak overlapping), 5.00 (s, J = 608.0
at −30 °C (Figure S4.16). Major isomer: C{ H} NMR (151 MHz,
PtC
Hz, 1-CH2).
C
7
D
8
, −30 °C): δ 104.06 (br, −CH), 82.48 (br, CH
), 46.64 (s,
2
1
2
Note: 2,2-dimethylbut-3-en-1-yllithium reacts with silicone grease,
so Krytox grease must be used in order to obtain appreciable yield for
JPtC = 853.85 Hz, Pt-CH
2
), 40.12 (br, JPtC = 38.82 Hz, 2-C), 37.87
3
3
(s, J = 98.16 Hz, 2-Me equatorial), 36.78 (br, J = 20 Hz, 3-
PtC
PtC
124
current scale of reaction.
2,2-Dimethylpent-4-en-1-yl)magnesium Bromide. A 250
mL three-necked flask equipped with a water-cooled reflux condenser
CH ), 28.73 (br, 2-Me axial), 27.91 (br, 4-CH , J = 14.15 Hz).
2
2
P
t
C
1
3
1
(
Minor isomer: C{ H} NMR (151 MHz, C D , −30 °C): δ 106.27
7
8
(br, −CH), 81.36 (br, CH ), 47.28 (s, Pt-CH ). Other
2
2
L
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
resonances cannot be unambiguously identified due to peak
overlapping.
Activation parameters were determined by fitting the rates as a
function of temperature to the Eyring equation. The errors in the rate
constants of ca. 20% were estimated on the basis of subjective
judgments of the sensitivities of the fits to changes in the rate
constants. The temperature of the NMR probe was calibrated using a
General Crystallographic Procedure. The following details
were common to all of the crystal structure determinations; for details
about individual compounds, see Supporting Information, Section S3.
Crystals mounted on glass fibers with either Krytox or Paratone oil
were transferred onto the diffractometer and kept at −173 °C in a
cold nitrogen gas stream. Intensity data were collected on a Bruker
Apex II diffractometer equipped with a CCD detector or a Bruker D8
Venture kappa diffractometer equipped with a Photon 100 CMOS
detector. Standard peak search and indexing procedures gave rough
cell dimensions. Data were collected with an area detector using the
measurement parameters listed in Table 1. The measured intensities
were reduced to structure factor amplitudes and their estimated
standard deviations by correction for background, scan speed, and
Lorentz and polarization effects. No corrections for crystal decay were
necessary. Systematically absent reflections were deleted and
symmetry equivalent reflections were averaged to yield the set of
unique data. The structure was solved by direct methods
127
methanol temperature standard,
and the estimated error in the
temperature measurements was 1 K. Uncertainties in the activation
1
28
parameters were estimated from propagation of error formulae.
1
For the pentenyl compound 3, the H NMR spectra were collected
in toluene-d at 24 temperatures between 11 and 69 °C. The H
1
8
NMR line shapes of the resonances due to the diastereotopic β-Me
groups were simulated to extract the olefin face exchange rate. The
natural linewidth (FWHM = 2.6 Hz) was measured from the
resonance for the axial methyl group at −22 °C, a temperature at
which the exchange is in the slow-exchange limit. Because the
1
95
1
equatorial methyl resonance shows significant Pt− H coupling, the
exchange rates in the slow-exchange regime (11−36 °C) were
determined by fitting the line shape of only the axial methyl
resonance. Above 36 °C, as coalescence is approached, the equatorial
1
95
methyl resonance is sufficiently broad that the Pt satellites are no
longer evident, but we found that we could successfully simulate the
line shapes in this temperature range by assigning the equatorial
methyl resonance an effective natural linewidth (FWHM) of 4.0 Hz.
For the butenyl and hexenyl compounds 2 and 4, analyses of the
(
SHELXTL); subsequent least-squares refinement and difference
Fourier calculations revealed the positions of all of the non-hydrogen
atoms and the olefinic hydrogen atoms (for 2, 3, and 4) on the ω-
alkenyl ligands. The analytical approximations to the scattering factors
were used, and all structure factors were corrected for both real and
imaginary components of anomalous dispersion. The methyl groups
were allowed to rotate about the C−C axis to find the best least-
squares positions. The displacement parameters for methylene and
methine hydrogens were set equal to 1.2 times Ueq for the attached
1
variable temperature H NMR spectra are made more difficult by two
factors: both compounds exist in solution as a mixture of two isomers
1
95
and the peaks suffered from overlapping and coupling to Pt. We
found that it was far easier to simulate the variable temperature
1
3
1
C{ H} NMR line shapes for exchange between the C and C
carbon; those for methyl hydrogens were set to 1.5 times U . In the
2
s
eq
1
3
1
isomers. For 2 and 4, most of the C{ H} NMR resonances show
final cycle of least-squares, independent anisotropic displacement
factors were refined for the non-hydrogen atoms. Final refinement
parameters are given in Table 1. A final analysis of variance between
the observed and calculated structure factors showed no apparent
errors.
1
95
13
Pt− C coupling, but because couplings to the α-carbon resonances
1
for the C and C isomers are so large ( J ∼600 Hz for 2 and ∼800
Hz for 4), the line shapes of the central features of these resonances
are well separated from (and unaffected by overlap with) their Pt
satellites. In addition, for 2, we also studied the olefinic methine
2
s
PtC
1
95
1
1
1
2
H{ H} NOE Spectra of cis-Bis(η ,η -2,2-dimethylpent-4-en-
1
1
1
-yl)platinum (3). The 1D H{ H} NOE NMR spectra of 3 were
carbon resonances of the C
Pt− C coupling.
For 2, the natural line widths of the C{ H} NMR resonances
2
and C
s
isomers, which show no
1
95
13
measured at −20 °C by means of a NOESY 1D experiment with a 500
ms mixing time. At this temperature, the olefin face exchange rate for
the Pt−olefin interaction is slow. The NOE results confirm the
assignments made on the basis of the magnitudes of the Pt−H
1
3
1
(FWHM = 1.6 Hz for the C isomer and 2.2 Hz for the C isomer)
s
2
1
3
1
were taken from the C{ H} NMR spectrum at −62 °C, which is at
the slow-exchange limit. For 4, the slow-exchange natural linewidth of
coupling constants. Thus, when the axial Pt-CH resonance at δ 1.59
2
1
3
1
was irradiated, sizable NOE signals are seen for the resonances at δ
the C{ H} NMR resonances (FWHM = 2.3 Hz) for the C
s
isomer
was taken from the C{ H} NMR spectrum at −21 °C. Because the
resonance for the C isomer at −21 °C is broad (probably due to
slowing of ring conformational motion), the slow-exchange natural
line widths (FWHM = 2.4 Hz) for the C isomer were extrapolated
1
3
1
2
.21 and at δ 1.19, which are assigned to the equatorial Pt-CH2
resonance and the equatorial 2-Me resonances, respectively (Figure
2
S4.7). When the equatorial Pt-CH resonance at δ 2.21 was irradiated,
2
both methyl resonances show sizable NOEs. When the equatorial γ-
2
1
3
1
from the linewidth seen in the C{ H} NMR spectrum at −8 °C by
assuming that this resonance narrows at a lower temperature in the
same way as does the C isomer. The equilibrium constant K = [C ]/
CH resonance at δ 1.89 was irradiated, the olefinic methine peak and
2
both methyl peaks show sizable NOEs.
Variable Temperature 1H NMR Spectra and NMR Simu-
lations. Simulations of variable temperature NMR line shapes to
determine the olefin face exchange rates were performed with the
s
2
1
[C ] was measured by H NMR spectroscopy over temperatures (−83
s
to −41 °C for 2, −83 to 21 °C for 4) at which the two isomers are
slowly exchanging; the activation parameters obtained from a van’t
Hoff plot were used to extrapolate this equilibrium constant to higher
temperatures (Figure S6.2 and S6.14), and the resulting mole
fractions were used in the simulations.
For 3, the rate measured based on the line broadening of the
methyl resonances is actually the rate of converting one enantiomer of
the C2 isomer to the other. The rate of interest (olefin
85
program WINDNMR. Before the simulations were performed, the
experimental spectra were phased and baseline-corrected using the
125
NUTS software package (Acorn NMR Inc.). The chemical shift
difference between the slow-exchange resonances is linearly depend-
ent on temperature, and we assumed that the low-temperature trend
in the difference could be extrapolated to temperatures above the
coalescence point. The rates of exchange as a function of temperature
were determined from visual comparisons of the experimental spectra
with computed trial line shapes. At each temperature, the exchange
rate was the value that gave the best fit of the calculated to the
observed line shape. For a two-site exchange, the WINDNMR
package parameterizes the “rate” of exchange as the sum of the
decomplexation) corresponds to the rate of conversion of the C2
isomer to the C isomer, which based on the scheme below should be
s
twice of the rate of the interconversion between the C isomers,
2
because each time a C isomer is converted to a C isomer, it converts
2
s
to its enantiomer with a probability of 50%. Because the rate of
decomplexation is twice the measured rate, the entropy of activation
of decomplexation should be ∼1 eu larger than the value obtained by
fitting the measured rate to the Eyring equation.
126
forward and backward rate constants (kab + kba). For a first-order
exchange between two sites of equal population, the actual rate
constant (i.e., either kab or k ) is half of this parameter. For a first-
ba
order exchange between two sites of unequal population, the actual
rate constants are given by k (i.e., leaving the a site) = (mole fraction
ab
of the b site)(kab + k ) and similarly for k .
ba
ba
M
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
1
2
Volatility of cis-Bis(η ,η -2,2-dimethylpent-4-en-1-yl)plati-
num (3). The vapor pressure and molar enthalpy of sublimation of
3
growth time. The carbon content in the films was measured by RBS.
Surface roughness was measured by atomic force microscopy (Asylum
Research Cypher) using a silicon nitride cantilever in contact mode.
Thermolysis of 3 in the Condensed State. The thermolysis of
3 in the condensed state begins at ∼145 °C and is largely complete by
were measured from the isothermal evaporation rates by means of
the following equation, which is a modified version of the Langmuir
equation for free evaporation in vacuum:
129
1
75 °C, as judged by thermogravimetric analysis under an atmosphere
p = k(dm/dt) T/M
of N . The final mass after pyrolysis is 45% of the original mass; for
2
comparison, 50% of the mass of 3 is Pt.
where p is the vapor pressure, dm/dt is the mass loss rate, T is the
absolute temperature, and M is the molecular weight. The parameter k
is an instrument-specific factor that depends in a complex way on the
experimental conditions used to carry out the isothermal measure-
ments; its value can be determined by measuring the isothermal
evaporation rate of a calibration standard at different temperatures
and comparing the evaporation rate with the reported vapor pressure
at each temperature. Because 3 liquefies with gentle heating, 1-
methylnaphthalene (which is also a liquid at room temperature) was
chosen as the standard; from its reported temperature-dependent
The ∼5% mass loss seen during the ∼10 min ramp to 120 °C
(
Figure S7.5) is not due to thermolysis of 3 but can be ascribed to a
small amount of vaporization; thermolysis of 3 in C D follows a
6
6
unimolecular decomposition mechanism with a first-order rate
−4
−1
133
constant of 9 × 10 min at 120 °C, and at this rate it would
take much longer (∼60 min) for ∼5% of 3 to decompose at this
temperature.
We estimate that it will take ∼400 h at 75 °C and ∼200 h at 80 °C
for 5% of 3 to decompose.
Thermolysis of 3 under Static Vacuum CVD Conditions. To
collect and analyze quantitatively the gaseous products generated
130
vapor pressure
and evaporation experiments conducted on our
−3
instrument, we found that the parameter k has a value of 5.4 × 10
under CVD conditions, 3 was thermolyzed under static vacuum (10
Torr K1
/2
g
3/2
mol min .
1/2
1
107,134
mTorr) in a previously described glass apparatus.
This
The isothermal evaporation rate of 3 was measured on the same
instrument with the same purge gas flow rate, and from the calibration
we determined the vapor pressure as a function of temperature. A fit
apparatus consisted of a solvent reservoir, a precursor reservoir, a
vertical deposition zone, and an NMR tube. The apparatus was
modified from the previous design by placing four indentations in the
vertical hot zone so that a substrate could be coated during
thermolysis and subsequently analyzed to assess the film properties.
The hot zone was heated with heating tape and the temperature was
regulated with a variable transformer. Temperatures were measured
with a K-type thermocouple held next to the hot zone under the
heating tape. The residual protons of commercial C D (99.5%) were
1
31
of the data to the August equation
(
gives log p = 6.19−2400/T
pressure is in Torr and temperature is in K). For example, the vapor
pressure of 3 is 10 mTorr at 20 °C and 250 mTorr at 80 °C. The
vapor pressure of 3 at 80 °C is higher than the 45 mTorr vapor
pressure reported for the related platinum(II) dialkyl diolefin complex
20
(
COD)PtMe2.
6
6
The molar enthalpy of sublimation ΔH was determined to be 11
vap
used as an integration standard. The concentration of protons in this
1
±
2 kcal mol− from a fit of ln p to 1/T according to the Clausius−
−1
solvent (0.284 mmol g ) was assessed by dissolving 1,2,4,5-
Clapeyron equation
tetrachlorobenzene (32.7 mg, 0.151 mmol) in the C D solvent
6
6
1
(
0.959 g) and measuring the relative intensity of the H NMR signals,
ΔHvap
ln p = B −
where B is a constant. The enthalpy of evaporation of 3 is similar to
employing a 150 s post-acquisition delay to eliminate integration
errors due to long relaxation times for the aromatic protons.
The solvent reservoir was charged with C D (0.850 g), which was
RT
6
6
−
1
degassed with 3 freeze−pump−thaw cycles; the solvent reservoir was
isolated from the rest of the apparatus until the deposition step was
complete. Because 3 is low melting, it is difficult to load into the
apparatus as the pure material. As a result, 3 (69.0 mg, 0.177 mmol)
was dissolved in pentane (5 mL) and the solution was transferred into
the precursor reservoir with a syringe. The pentane solvent was
carefully removed at 0 °C, a temperature at which the evaporation of
precursor is negligible. CVD growth was conducted at 250 ± 10 °C,
with the precursor reservoir maintained at 80 °C.
that of 10.5 ± 0.3 kcal mol for the commercial precursor
7
(
C H Me)PtMe .
5 4 3
Cold Wall CVD Experiments. CVD experiments were performed
99,100
in a cold wall high-vacuum chamber described elsewhere.
The
precursor was loaded into a glass reservoir, which was kept at 75 °C
during deposition experiments. A flow rate of 20−30 sccm argon
(
99.999%) was used as a carrier; the precursor was delivered into the
chamber through a 0.4 cm i.d. stainless steel tube pointing toward the
substrate. The total chamber pressure (Ar and precursor, measured by
the manometer attached to the chamber) was kept at 2 to 3 mTorr. A
remote microwave plasma source consisting of a 9.5 mm i.d. Pyrex
tube with a 2.45 GHz Evenson cavity located outside of the
Byproducts from the thermolysis were trapped in the NMR tube,
which was cooled with liquid nitrogen. After the deposition was
complete, the known amount of C D solvent was transferred into the
6
6
107,134
132
NMR tube.
As judged from quantitation of the contents of the
chamber was used to generate atomic oxygen from research grade
NMR tube, all (105 ± 10%) of the protons in the precursor were
accounted for in the byproducts collected in the NMR tube,
indicating that all of the volatile products had been trapped. The
dioxygen gas (99.999%). For these experiments, the flow of O was
2
regulated with a mass flow controller to establish a partial pressure in
the chamber of 1.5 mTorr. The net plasma power (forward minus
reflected) was 70 W.
1
identities of the products were determined from their H, COSY, and
HSQC NMR spectra and GC−MS data (Figure S10.1−S10.9).
Several different substrates were tested. 10 nm films of Al O were
2
3
grown on Si (100) by ALD from trimethylaluminum and water at 80
C. Vanadium nitride (VN) was grown by in situ CVD from
tetrakis(dimethylamido)vanadium and ammonia at 330 °C on Si
100). Silicon (Si(100) with native oxide) and Al O substrates were
cleaned by UV/ozone to remove ambient contaminations for 10 min
before loading into the chamber. All substrates were radiatively
heated. The temperatures were measured with a K-type thermocouple
attached to the sample holder.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.chemmater.0c03226.
°
■
*
sı
(
2
3
Crystallographic data (CIF)
Crystallographic data (CIF)
Crystallographic data (CIF)
Crystallographic data (CIF)
Crystallographic data (CIF)
Syntheses and characterization of bromo(2,2-dimethyl-
but-3-en-1-yl)(1,5-cyclooctadiene)platinum (5), bis(η -
2,2-dimethylbut-3-en-1-yl)(1,5-cyclooctadiene)platinum
The onset of nucleation was monitored by real-time spectroscopic
ellipsometry (SE), and the changes in the angle Ψ were reported at a
single energy, 2.65 eV, which provides the greatest sensitivity to the
101
onset of nucleation, as discussed previously.
Film thickness was
measured by scanning electron microscopy (Hitachi 4800) and
Rutherford backscattering spectroscopy, RBS (3SDH Pelletron). The
film growth rates were calculated by dividing the RBS thickness by the
1
N
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
1
REFERENCES
(
6), and bis[η -(2,2-dimethylcyclopentyl)methyl](1,5-
■
cyclooctadiene)platinum (7); details of crystallography
characterizations; H, C, Pt, HSQC NMR, IR, and
UV/vis spectra of bis(η ,η -2,2-dimethyl-ω-alkenyl)-
platinum compounds 2, 3, and 4; H and C VT-
NMR spectra, NMR line shape fits, and Eyring plot of
bis(η ,η -2,2-dimethyl-ω-alkenyl)platinum compounds,
and van’t Hoff plot of the C and C isomers of 2 and
; measured vapor pressure of 3 at different temper-
(
1) Gozum, J. E.; Pollina, D. M.; Jensen, J. A.; Girolami, G. S.
1
13
195
″Tailored″ Organometallics as Precursors for the Chemical Vapor
Deposition of High-Purity Palladium and Platinum Thin Films. J. Am.
Chem. Soc. 1988, 110, 2688−2689.
1
2
1
13
(
2) Doppelt, P. Why is Coordination Chemistry Stretching the
1
2
Limits of Micro-Electronics Technology? Coord. Chem. Rev. 1998,
78−180, 1785−1809.
3) Hampden-Smith, M. J.; Kodas, T. T. Chemical Vapor
Deposition of Metals: Part 1. An Overview of CVD Processes.
Chem. Vap. Depos. 1995, 1, 8−23.
4) Jones, A. C.; Hitchman, M. L. Overview of Chemical Vapour
1
(
2
s
4
atures and related plots; film characterization data,
including the film growth rate, Arrhenius plot, SEM
images, RBS spectra, and glancing angle (1°) XRD
spectra of the films grown in the cold wall apparatus;
film characterization data, SEM image, XPS spectra, and
glancing angle (1°) XRD spectra of the films grown in
(
Deposition. In Chemical Vapour Deposition: Precursors, Processes and
Applications; Jones, A. C.; Hitchman, M. L., Eds.; The Royal Society
of Chemistry: Cambridge, 2009; pp 1−36.
(5) Mackus, A. J. M.; Leick, N.; Baker, L.; Kessels, W. M. M.
Catalytic Combustion and Dehydrogenation Reactions during Atomic
Layer Deposition of Platinum. Chem. Mater. 2012, 24, 1752−1761.
1
the hot wall apparatus; H, COSY, HSQC NMR, and
GC−MS spectra of the thermolysis products generated
in the static vacuum CVD apparatus, with the assign-
ment of each species; spectroscopic evidence showing
(
6) Xue, Z.; Strouse, M. J.; Shuh, D. K.; Knobler, C. B.; Kaesz, H. D.;
Hicks, R. F.; Williams, R. S. Characterization of
Methylcyclopentadienyl)trimethylplatinum and Low-Temperature
Organometallic Chemical Vapor Deposition of Platinum Metal. J.
Am. Chem. Soc. 1989, 111, 8779−8784.
(7) Xue, Z.; Thridandam, H.; Kaesz, H. D.; Hicks, R. F.
Organometallic Chemical Vapor Deposition of Platinum. Reaction
Kinetics and Vapor Pressures of Precursors. Chem. Mater. 1992, 4,
(
that 3 does not react with H (PDF)
2
AUTHOR INFORMATION
■
Corresponding Author
Gregory S. Girolami − School of Chemical Sciences, University
of Illinois at Urbana-Champaign, Urbana, Illinois 61801,
United States; orcid.org/0000-0002-7295-1775;
Email: girolami@scs.illinois.edu
1
(
62−166.
8) Hiratani, M.; Nabatame, T.; Matsui, Y.; Imagawa, K.; Kimura, S.
Platinum Film Growth by Chemical Vapor Deposition Based on
Autocatalytic Oxidative Decomposition. J. Electrochem. Soc. 2001, 148,
C524−C527.
Authors
(9) Knoops, H. C. M.; Mackus, A. J. M.; Donders, M. E.; van de
Sanden, M. C. M.; Notten, P. H. L.; Kessels, W. M. M. Remote
Plasma ALD of Platinum and Platinum Oxide Films. Electrochem.
Solid-State Lett 2009, 12, G34−G36.
Sumeng Liu − School of Chemical Sciences, University of Illinois
at Urbana-Champaign, Urbana, Illinois 61801, United States;
orcid.org/0000-0002-2133-2122
Zhejun Zhang − Department of Materials Science and
Engineering, University of Illinois at Urbana-Champaign,
Urbana, Illinois 61801, United States
Danielle Gray − School of Chemical Sciences, University of
Illinois at Urbana-Champaign, Urbana, Illinois 61801, United
States; orcid.org/0000-0003-0059-2096
Lingyang Zhu − School of Chemical Sciences, University of
Illinois at Urbana-Champaign, Urbana, Illinois 61801, United
States; orcid.org/0000-0002-6657-271X
(
10) Dryden, N. H.; Kumar, R.; Ou, E.; Rashidi, M.; Roy, S.;
Norton, P. R.; Puddephatt, R. J.; Scott, J. D. Chemical Vapor
Deposition of Platinum: New Precursors and Their Properties. Chem.
Mater. 1991, 3, 677−685.
(
11) Dendooven, J.; Ramachandran, R. K.; Devloo-Casier, K.;
Rampelberg, G.; Filez, M.; Poelman, H.; Marin, G. B.; Fonda, E.;
Detavernier, C. Low-Temperature Atomic Layer Deposition of
Platinum Using (Methylcyclopentadienyl)trimethylplatinum and
Ozone. J. Phys. Chem. C 2013, 117, 20557−20561.
(12) Pierson, H. O. Introduction and General Considerations. In
Handbook of Chemical Vapor Deposition; Pierson, H. O., Ed.; William
Andrew Publishing: Oxford, 1992; pp 1−16.
John R. Abelson − Department of Materials Science and
Engineering, University of Illinois at Urbana-Champaign,
Urbana, Illinois 61801, United States
(
13) Green, M. L.; Levy, R. A. Chemical Vapor Deposition of Metals
for Integrated Circuit Applications. JOM 1985, 37, 63−71.
14) Hamalainen, J.; Ritala, M.; Leskela, M. Atomic Layer
Deposition of Noble Metals and Their Oxides. Chem. Mater. 2014,
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.chemmater.0c03226
(
̈
̈
̈
̈
2
6, 786−801.
Author Contributions
All authors have given approval to the final version of the
manuscript.
(15) George, S. M. Atomic Layer Deposition: An Overview. Chem.
Rev. 2010, 110, 111−131.
(
and Reaction Mechanisms for the Atomic Layer Deposition of Metal
Films. Coord. Chem. Rev. 2013, 257, 3271−3281.
(
16) Bernal Ramos, K.; Saly, M. J.; Chabal, Y. J. Precursor Design
Notes
The authors declare the following competing financial
interest(s): One patent related to this work has been filed:
Metal Complexes for Depositing Films and Method of Making
and Using the Same, U.S. Patent 20190077819 (Published
Mar. 14th, 2019).
17) Aaltonen, T.; Ritala, M.; Sajavaara, T.; Keinonen, J.; Leskela, M.
̈
Atomic Layer Deposition of Platinum Thin Films. Chem. Mater. 2003,
15, 1924−1928.
(18) Fang, Q.; Hodson, C.; Xu, C.; Gunn, R. Nucleation and
Growth of Platinum Films on High-k/Metal Gate Materials by
Remote Plasma and Thermal ALD. Phys. Procedia 2012, 32, 551−560.
(
19) Baum, T. H.; Comita, P. B. Laser-Induced Chemical Vapor
ACKNOWLEDGMENTS
■
1
Deposition of Metals for Microelectronics Technology. Thin Solid
Films 1992, 218, 80−94.
(20) Hierso, J.-C.; Serp, P.; Feurer, R.; Kalck, P. MOCVD of
Rhodium, Palladium and Platinum Complexes on Fluidized Divided
We thank the National Science Foundation under grants CHE
665191 and CHE 1954745 (to G.S.G.) and CMMI 1825938
(
to J.R.A.) for support of this research.
O
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Substrates: Novel Process for One-Step Preparation of Noble-Metal
Films Using Dimethyl (N,N-Dimethyl-3-Butene-1-Amine-N) Plati-
Catalysts. Appl. Organomet. Chem. 1998, 12, 161−172.
num and O Reactant. Chem. Mater. 2019, 31, 5056−5064.
2
(
21) Dossi, C.; Psaro, R.; Bartsch, A.; Fusi, A.; Sordelli, L.; Ugo, R.;
(39) Jiang, M.; Zhang, M.; Li, C.; Williams, C. T.; Liang, C. CVD of
Bellatreccia, M.; Zanoni, R.; Vlaic, G. Chemical Vapor Deposition of
Platinum Hexafluoroacetylacetonate Inside KL Zeolite: A New Route
to Nonacidic Platinum-in-Zeolite Catalysts. J. Catal. 1994, 145, 377−
Pt(C H ) to Synthesize Highly Dispersed Pt/SBA-15 Catalysts for
Hydrogenation of Chloronitrobenzene. Chem. Vap. Depos. 2014, 20,
146−151.
(40) Foley, P.; DiCosimo, R.; Whitesides, G. M. Mechanism of
Thermal Decomposition of Dineopentylbis(triethylphosphine)-
platinum(II): Formation of Bis(triethylphosphine)-3,3-dimethylplati-
nacyclobutane. J. Am. Chem. Soc. 1980, 102, 6713−6725.
(41) Brainard, R. L.; Miller, T. M.; Whitesides, G. M. Mechanisms of
Thermal Decomposition of trans-Chloroneopentylbis-
5 9 2
3
83.
22) Serp, P.; Kalck, P.; Feurer, R. Chemical Vapor Deposition
(
Methods for the Controlled Preparation of Supported Catalytic
Materials. Chem. Rev. 2002, 102, 3085−3128.
(
23) Choi, D. S.; Robertson, A. W.; Warner, J. H.; Kim, S. O.; Kim,
H. Low-Temperature Chemical Vapor Deposition Synthesis of Pt−
Co Alloyed Nanoparticles with Enhanced Oxygen Reduction
Reaction Catalysis. Adv. Mater. 2016, 28, 7115−7122.
(
1
tricyclopentylphosphine)platinum(II). Organometallics 1986, 5,
481−1490.
42) DiCosimo, R.; Moore, S. S.; Sowinski, A. F.; Whitesides, G. M.
(
(
24) Ryan, M. Platinum in Next-Generation Materials for Data
Cyclometalation of Dialkylbis(triethylphosphine)platinum(II) Com-
plexes: Formation of Pt,Pt-Bis(triethylphosphine)platinacycloalkanes.
J. Am. Chem. Soc. 1982, 104, 124−133.
Storage. Platinum Metals Rev. 2010, 54, 244−249.
(
25) Dorovskikh, S. I.; Zharkova, G. I.; Turgambaeva, A. E.; Krisyuk,
V. V.; Morozova, N. B. Chemical Vapour Deposition of Platinum
Films on Electrodes for Pacemakers: Novel Precursors and Their
Thermal Properties. Appl. Organomet. Chem. 2016, 31, No. e3654.
(43) Wilkinson, G. The Transition Metal to Carbon Sigma Bond.
Pure Appl. Chem. 1972, 30, 627−636.
(44) Collier, M. R.; Lappert, M. F.; Truelock, M. M. μ-Methylene
(
26) Woodward, B. K. Platinum Group Metals (PGMs) for
Transition Metal Binuclear Compounds: Complexes with Me SiCH -
Permanent Implantable Electronic Devices. In Metals for Biomedical
Applications; Baltzer, N.; Copponnex, T., Eds.; Woodhead Publishing:
Sawston, Cambridge, 2014; pp 130−147.
3
2
and Related Ligands. J. Organomet. Chem. 1970, 25, C36−C38.
(45) Wilkinson, G. The Long Search for Stable Transition Metal
Alkyls. Science 1974, 185, 109.
46) Yagupsky, G.; Mowat, W.; Shortland, A.; Wilkinson, G.
(
27) Morcos, B. M.; O’Callaghan, J. M.; Amira, M. F.; Van Hoof, C.;
(
Op de Beeck, M. Electrodeposition of Platinum Thin Films as
Interconnects Material for Implantable Medical Applications. J.
Electrochem. Soc. 2013, 160, D300−D306.
Trimethylsilylmethyl Compounds of Transition Metals. J. Chem. Soc.
D 1970, 1369−1370.
(
47) Lehmkuhl, H.; Naydowski, C.; Benn, R.; Rufinska, A.; Schroth,
́
(
28) Baker, L.; Cavanagh, A. S.; Seghete, D.; George, S. M.; Mackus,
1
2
5
G.; Mynott, R.; Kru
nickel-Komplexe. Chem. Ber. 1983, 116, 2447−2465.
48) Richey, H. G.; Rees, T. C. Facile Intramolecular Cyclization of
an Olefinic Grignard Reagent. Tetrahedron Lett 1966, 7, 4297−4301.
49) Fantasia, S.; Jacobsen, H.; Cavallo, L.; Nolan, S. P. Insertion of
̈
ger, C. η ,η -4-Alkenyl(η -cyclopentadienyl)-
A. J. M.; Kessels, W. M. M.; Liu, Z. Y.; Wagner, F. T. Nucleation and
Growth of Pt Atomic Layer Deposition on Al O Substrates Using
2
3
(
(
Methylcyclopentadienyl)-trimethyl Platinum and O Plasma. J. Appl.
2
Phys. 2011, 109, No. 084333.
29) Mackus, A. J. M.; Verheijen, M. A.; Leick, N.; Bol, A. A.;
(
(
a N-Heterocyclic Carbene (NHC) into a Platinum-Olefin Bond.
Kessels, W. M. M. Influence of Oxygen Exposure on the Nucleation of
Platinum Atomic Layer Deposition: Consequences for Film Growth,
Nanopatterning, and Nanoparticle Synthesis. Chem. Mater. 2013, 25,
Organometallics 2007, 26, 3286−3288.
(50) Da, T. T.; Hong Hai, L. T.; Meervelt, L. V.; Dinh, N. H.
Synthesis, Structure, and in vitro Cytotoxicity of Organoplatinum(II)
Complexes Containing Aryl Olefins and Quinolines. J. Coord. Chem.
1
(
905−1911.
30) Lien, C.; Konh, M.; Chen, B.; Teplyakov, A. V.; Zaera, F. Gas-
2
(
015, 68, 3525−3536.
Phase Electron-Impact Activation of Atomic Layer Deposition (ALD)
51) Vicente, J.; Chicote, M. T.; MacBeath, C.; Jones, P. G.
Precursors: MeCpPtMe . J. Phys. Chem. Lett. 2018, 9, 4602−4606.
3
Nucleophilic Attack of Carbonyl-Stabilized Phosphorus Ylides on
Neutral and Cationic Alkene Platinum(II) Complexes. Organo-
metallics 2003, 22, 1843−1848.
(
31) Erkens, I. J. M.; Verheijen, M. A.; Knoops, H. C. M.; Keuning,
W.; Roozeboom, F.; Kessels, W. M. M. Plasma-Assisted Atomic Layer
Deposition of Conformal Pt Films in High Aspect Ratio Trenches. J.
Chem. Phys. 2016, 146, No. 052818.
(52) Huynh, H. V.; Pham, V. T.; Chi, N. T. T. Cyclometallated
Platinum(II) Complexes with a Phenylpropene-Derived π/σ-Chelator
and N-Heterocyclic Carbenes. Eur. J. Inorg. Chem. 2017, 2017, 5650−
(
32) Elam, J. W.; Zinovev, A. V. V.; Pellin, M. J.; Comstock, D. J.;
Hersam, M. C. Nucleation and Growth of Noble Metals on Oxide
Surfaces Using Atomic Layer Deposition. ECS Trans. 2007, 3, 271−
5
655.
(
53) Cooper, M. K.; Guerney, P. J.; Elder, M.; McPartlin, M. The X-
r a y S t r u c t u r e o f t h e C o m p l e x
Dichloroorthovinyldiphenylarsineplatinum(II). J. Organomet. Chem.
977, 137, C22−C24.
54) Cooper, M. K.; Guerney, P. J.; McPartlin, M. The trans
2
78.
(
33) Elliott, S. D. Mechanism, Products, and Growth Rate of Atomic
Layer Deposition of Noble Metals. Langmuir 2010, 26, 9179−9182.
34) Chen, Y. J.; Kaesz, H. D.; Thridandam, H.; Hicks, R. F. Low-
1
(
(
Temperature Organometallic Chemical Vapor Deposition of Plati-
Influence and trans Effect of the Arsine Group. The X-ray Structures
num. Appl. Phys. Lett. 1988, 53, 1591−1592.
of Dichloro[diphenyl(o-vinylphenyl)arsine]platinum(II) and [1-(o-
1
(
35) Tagge, C. D.; Simpson, R. D.; Bergman, R. G.; Hostetler, M. J.;
D i p h e n y l a r s i n o p h e n y l ) - 2 - m e t h o x y e t h y l - A s , C ] -
Girolami, G. S.; Nuzzo, R. G. Synthesis of a Novel Volatile Platinum
Complex for Use in CVD and a Study of the Mechanism of Its
Thermal Decomposition in Solution. J. Am. Chem. Soc. 1996, 118,
(hexafluoroacetylacetonato)platinum(II). J. Chem. Soc., Dalton Trans.
1
980, 349−354.
(55) Abel, E. W.; Evans, D. G.; Koe, J. R.; Sik, V.; Bates, P. A.;
Hursthouse, M. B. Platinum Metal Complexes of Potentially
Chelating Alkene-Thioether and Alkene−Selenoethe rLigands: Syn-
thesis and Dynamic Nuclear Magnetic Resonance Study of
[MX {MeE(CH ) CHCH }](M = Pt or Pd; X = Cl, Br, or I; E
2
(
634−2643.
36) Malandrino, G.; Nigro, R. L.; Fragala,
Platinum (100) Films on Random Hastelloy C276. Chem. Vap. Depos.
999, 5, 59−61.
37) Shibutami, T.; Kawano, K.; Oshima, N.; Yokoyama, S.;
̀
I. L. MOCVD of
1
(
2
2
n
2
= S or Se; n= 2 or 3) and the X-ray Structure of cis-Dibromo(2-thia-6-
heptene)platinum(II), [PtBr {{MeS(CH ) CHCH }]. J. Chem.
Funakubo, H. Ruthenium Film with High Nuclear Density Deposited
by MOCVD Using a Novel Liquid Precursor. Electrochem. Solid-State
Lett. 2003, 6, C117−C119.
2
2
3
2
Soc., Dalton Trans. 1989, 985−989.
(56) McCrindle, R.; Ferguson, G.; Arsenault, G. J.; McAlees, A. J.;
Ruhl, B. L.; Sneddon, D. W. Reaction of Diazomethane with
Platinum(II) Chloride Complexes of Chelating Olefins. Crystal
(
38) Lee, W.-J.; Wan, Z.; Kim, C.-M.; Oh, I.-K.; Harada, R.; Suzuki,
K.; Choi, E.-A.; Kwon, S.-H. Atomic Layer Deposition of Pt Thin
P
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Structure Analyses of a-Chloro-b-(chloromethyl)-cd-(2,2,N,N-tetra-
methyl-3-buten-1-amine)platinum(II), a-Chloro-b-(chloromethyl)-dc-
(69) Kelly, R. D.; Brent Young, G. Synthesis and Spectroscopic
Characteristics of Bis(ethenyldimethylsilylmethyl)platinium(II) Com-
plexes Containing Nitrogen Donor Ligands. J. Organomet. Chem.
1989, 361, 123−138.
(70) Kelly, R. D.; Young, G. B. Synthesis and Spectroscopic
Characteristics of Bis(ethenyldimethylsilylmethyl)platinum(II) Com-
plexes Containing Tertiary Phosphine Ligands. Polyhedron 1989, 8,
433−445.
(71) Eisenstein, O.; Hoffmann, R. Activation of a Coordinated
Olefin toward Nucleophilic Attack. J. Am. Chem. Soc. 1980, 102,
(
(
2,2,N,N-tetramethyl-3-buten-1-amine)platinum(II), a-Chloro-b-
chloromethyl)-dc-(2,2-dimethyl-3-buten-1-yl methyl sulfide)-
platinum(II), and Dichloro[(2,2-dimethyl-4-penten-1-yl)-
methylsulfonium methylide-C,C,C]plantinum(II). Organometallics
1
(
986, 5, 1171−1178.
57) Abel, E. W.; Evans, D. G.; Koe, J. R.; Sik, V.; Hursthouse, M. B.;
Bates, P. A. Platinum Metal Complexes of Potentially Chelating
Alkene Thioether and Selenoether Ligands: the Synthesis and
6
(
148−6149.
Dynamic Nuclear Magnetic Resonance Study of [MX {E-
2
72) Eisenstein, O.; Hoffmann, R. Transition-Metal Complexed
[
(CH ) CR=CR ] }](M = Pt or Pd; X = Cl, Br, or I; E = S or Se;
2 n 2 2
Olefins: How Their Reactivity toward a Nucleophile Relates to Their
Electronic Structure. J. Am. Chem. Soc. 1981, 103, 4308−4320.
n= 2 or 3; R = H or Me) and the X-ray crystal structure of cis-Di-iodo-
5-thianona-1,8-diene)platinum(II), [PtI {S[(CH ) CHCH ] }]. J.
(
2
2
2
2 2
(
73) Still, B. M.; Kumar, P. G. A.; Aldrich-Wright, J. R.; Price, W. S.
Chem. Soc., Dalton Trans. 1989, 2315−2321.
1
95
Pt NMRTheory and Application. Chem. Soc. Rev. 2007, 36, 665−
86.
74) Clark, H. C.; Manzer, L. E.; Ward, J. E. H. 13C Nuclear
(
58) Abel, E. W.; Evans, D. G.; Koe, J. R.; Hursthouse, M. B.; Mazid,
6
M.; Mahon, M. F.; Molloy, K. C. Platinum Metal Complexes of
Potentially Chelating Alkene−Sulphur and Alkene−Selenium Li-
gands. The Synthesis by Chalcogen Dealkylation and X-ray Structures
of the Dimeric Complexes [{PtI(SCH CH CMeCH )} ] and
(
Magnetic Resonance Studies of Organometallic Compounds. IV. cis-
Dimethylplatinum(II) Derivatives. Can. J. Chem. 1974, 52, 1165−
2
2
2
2
1
(
170.
[
{PtI(PPh )(SCH CH CMCH )} ], and a DynamicNuclear Mag-
3
2
2
2
2
75) Brendel, M.; Engelke, R.; Desai, V. G.; Rominger, F.; Hofmann,
netic Resonance Study of [{PtI(L)(SCH CH CMeCH )} ][L =
2 2 2 2
P. Synthesis and Reactivity of Platinum(II) cis-Dialkyl, cis-Alkyl
Chloro, and cis-Alkyl Hydrido Bis-N-heterocyclic Carbene Chelate
Complexes. Organometallics 2015, 34, 2870−2878.
PPh , PPh Me, or As(CH SiMe ) ]. J. Chem. Soc., Dalton Trans. 1990,
3
2
2
3 3
1
(
697−1704.
59) Abel, E. W.; Evans, D. G.; Koe, J. R.; Hursthouse, M. B.; Mazid,
M. Dimethylplatinum Complexes of Polydentate Alkene−Sulfur and
Selenium Ligands. J. Chem. Soc., Dalton Trans. 1992, 663−667.
60) Ermer, S. P.; Struck, G. E.; Bitler, S. P.; Richards, R.; Bau, R.;
(76) Maudez, W.; Roy, C.; Tran, P. D.; Thurier, C.; Karmous, F.;
Doppelt, P. New Dimethyl(norbornadienyl)platinum(II) Precursors
−
(
for Platinum MOCVD. Chem. Vap. Depos. 2014, 20, 59−68.
1
(
77) J
for the olefinic carbons in (COD)PtR complexes fall in
Pt‑C 2
Flood, T. C. Kinetics and Conformation in the Reversible Insertion of
an Alkene into a Platinum-Carbon Bond in a Chelated (Pentenyl)
platinum Complex. Organometallics 1993, 12, 2634−2643.
the order neopentyl (38 Hz) <methyl (40 Hz) <trimethylsilylmethyl
42 Hz) <ethyl (47 Hz) <trifluoromethyl (56 Hz) <benzyl (65 Hz).
(
See Kelly, R. D.; Young, G. B.; M, H.; Clark, H. C.; Manzer, L. E.;
Stothers, J. B.; Ward, J. E. H. Carbon-13 Nuclear Magnetic Resonance
Studies of Organometallic Compounds. VII. 1,5-
Cyclooctadieneplatinum(II) Derivatives. J. Am. Chem. Soc. 1975, 97,
(
61) Zhao, P.; Incarvito, C. D.; Hartwig, J. F. Carbon-Oxygen Bond
Formation between a Terminal Alkoxo Ligand and a Coordinated
Olefin. Evidence for Olefin Insertion into a Rhodium Alkoxide. J. Am.
Chem. Soc. 2006, 128, 9642−9643.
7
21−727. Kelly, R. D.; Young, G. B. Synthesis and Spectroscopic
(
62) Albright, T. A.; Hoffmann, R.; Thibeault, J. C.; Thorn, D. L.
Characteristics of Bis(ethenyldimethylsilylmethyl)platinum(II) Com-
Ethylene complexes. Bonding, rotational barriers, and conformational
plexes Containing Tertiary Phosphine Ligands. Polyhedron 1989, 8,
preferences. J. Am. Chem. Soc. 1979, 101, 3801−3812.
433−445. Chisholm, M. H.; Clark, H. C.; Manzer, L. E.; Stothers, J.
(
63) Kurosawa, H.; Miki, K.; Kasai, N.; Ikeda, I. In-Plane Olefin
B.; Ward, J. E. H. Carbon-13 Nuclear Magnetic Resonance Studies of
Organometallic Compounds. VII. 1,5-Cyclooctadieneplatinum(II)
Derivatives. J. Am. Chem. Soc. 1975, 97, 721−727.
Coordination and Unusually Small Substituent Dependency of
3
Stability in (η -Allyl) (pentafluorophenyl) (olefin)platinum(II)
Complexes. Organometallics 1991, 10, 1607−1613.
(78) Hall, J. R.; Appleton, T. G. Complexes with Six-Membered
(
64) Ziegler, T.; Rauk, A. A Theoretical Study of the Ethylene-Metal
Chelate Rings. III. Factors Influencing the Values of the Platinum-
+
+
+
0
2+
Bond in Complexes between Cu , Ag , Au , Pt or Pt and Ethylene,
Based on the Hartree-Fock-Slater Transition-State Method. Inorg.
Chem. 1979, 18, 1558−1565.
3
4
Proton Coupling Constants JPt‑N‑C‑H and JPt‑N‑C‑CH3 in Diamine
Complexes of Platinum(II) and -(IV). Inorg. Chem. 1971, 10, 1717−
1
725.
(
65) Wright, L. L.; Wing, R. M.; Rettig, M. F.; Wiger, G. R.
(
79) Karplus, M. Contact Electron-Spin Coupling of Nuclear
Magnetic Moments. J. Chem. Phys. 1959, 30, 11−15.
80) Gay, I. D.; Young, G. B. Chemical Shift Anisotropy of
Organometallic Carbon. Organometallics 1996, 15, 2264−2269.
81) Carpentier, J.-F.; Wu, Z.; Lee, C. W.; Stromberg, S.;
Molecular Structure of Dichloro(5-methylenecycloheptene)platinum-
II). ″In-Plane″ Coordination of the Exocyclic Olefin. J. Am. Chem.
Soc. 1980, 102, 5949−5950.
66) Rakowsky, M. H.; Woolcock, J. C.; Wright, L. L.; Green, D. B.;
(
(
(
(
̈
Rettig, M. F.; Wing, R. M. In-plane Coordinated Double Bonds.
Molecular Structures, Spectroscopy, and Stability of 5-Methylenecy-
clooctene and 5-Methylenecycloheptene Complexes of Platinum(II).
Organometallics 1987, 6, 1211−1218.
0
Christopher, J. N.; Jordan, R. F. d Metal Olefin Complexes.
Synthesis, Structures, and Dynamic Properties of (C R ) Zr-
5
5 2
+
(
(
OCMe CH CH CHCH ) Complexes: Models for the Elusive
2
2
2
2
+
C R ) Zr(R)(Olefin) Intermediates in Metallocene-Based Olefin
5 5 2
(
67) Love, R. A.; Koetzle, T. F.; Williams, G. J. B.; Andrews, L. C.;
Polymerization Catalysis. J. Am. Chem. Soc. 2000, 122, 7750−7767.
Bau, R. Neutron Diffraction Study of the Structure of Zeise’s salt,
(
82) Carpentier, J.-F.; Maryin, V. P.; Luci, J.; Jordan, R. F. Solution
•
0
KPtCl (C H ) H O. Inorg. Chem. 1975, 14, 2653−2657.
3
2
4
2
Structures and Dynamic Properties of Chelated d Metal Olefin
5
1
(
68) The second species seen in solution is unlikely to be the isomer
Complexes {η : η -C R SiMe NtBu}Ti(OCMe CH CH CH
5
4
2
2
2
2
+
5
1
in which the two alkyl groups are mutually trans. Trans effects should
make such isomers much too high in energy relative to the cis
isomers; in addition, without the steric hindrance between the cis
double bonds, there would be no reason for the ratio between the two
isomers to vary non-monotonically with the length of the carbon
backbone, as is seen experimentally. We can also rule out the
possibility that the second species is a dimer or higher oligomer: the
unusual Pt-C coupling constants seen (especially in 2) are consistent
only with the presence of a bidentate ligand with a small ring size.
CH ) (R = H, Me): Models for the {η : η -C R SiMe NtBu}Ti-
2
5
4
2
+
(
R‘)(olefin) Intermediates in “Constrained Geometry” Catalysts. J.
Am. Chem. Soc. 2001, 123, 898−909.
(83) Stoebenau, E. J.; Jordan, R. F. Nonchelated Alkene and Alkyne
0
Complexes of d Zirconocene Pentafluorophenyl Cations. J. Am.
Chem. Soc. 2006, 128, 8638−8650.
(84) Abel, E. W.; Koe, J. R.; Hursthouse, M. B.; Malik, K. M. A.;
Mazid, M. A. Platinum and Rhodium Complexes of Dialkenyl
Telluroether Ligands: Synthesis, Dynamic Nuclear Magnetic Reso-
Q
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
nance and Crystal Structure of cis-Dibromo(3,3,7,7-tetramethyl-5-
tellura-3,7-disilanona-1,8-diene)platinum(II). J. Chem. Soc., Dalton
Trans. 1994, 2645−2650.
VCH Verlagsgesekhaft mbH and VCH Publishers Inc: Weinheim and
New York, 1994; p 147.
(103) Sambles, J. R. The Resistivity of Thin Metal FilmsSome
Critical Remarks. Thin Solid Films 1983, 106, 321−331.
(104) Basic Concepts. In The Electrical Resistivity of Metals and
Alloys; Rossiter, P. L., Ed.; Cambridge University Press: Cambridge,
1987; pp 1−29.
(
85) Reich, H. J. WinDNMR: Dynamic NMR Spectra for Windows.
J. Chem. Educ. 1995, 72, 1086.
86) Casey, C. P.; Carpenetti, D. W. Measurement of Barriers for
(
0
Alkene Dissociation and for Inversion at Zirconium in a d
Zirconium-Alkyl-Alkene Complex. Organometallics 2000, 19, 3970−
(105) Mayadas, A. F.; Shatzkes, M. Electrical-Resistivity Model for
Polycrystalline Films: the Case of Arbitrary Reflection at External
Surfaces. Phys. Rev. B 1970, 1, 1382−1389.
3
(
977.
87) Casey, C. P.; Carpenetti, D. W.; Sakurai, H. Models for
Intermediates in Metallocene-Catalyzed Alkene Polymerization:
(106) Aaltonen, T.; Rahtu, A.; Ritala, M.; Leskela, M. Reaction
̈
1
2
Alkene Dissociation from Cp Zr[η ,η -CH Si(CH ) CH CHCH ]-
Mechanism Studies on Atomic Layer Deposition of Ruthenium and
Platinum. Electrochem. Solid-State Lett 2003, 6, C130−C133.
(107) Jeffries, P. M.; Dubois, L. H.; Girolami, G. S. Metal-Organic
Chemical Vapor Deposition of Copper and Copper(I) Oxide from
Copper(I) tert-Butoxide. Chem. Mater. 1992, 4, 1169−1175.
(108) We are not able to trap byproducts quantitatively in our
dynamic CVD system, and we are aware that product distributions
obtained under static CVD conditions are not necessarily the same as
those under dynamic CVD conditions. For 3, however, the similarity
of the structure and properties of the films grown under the two
conditions suggests that the product distributions are not likely to be
significantly different.
2
2
3
2
2
2
[
B(C F ) ]. Organometallics 2001, 20, 4262−4265.
88) Frey, U.; Helm, L.; Merbach, A. E.; Romeo, R. High-Pressure
6
5 4
(
NMR Kinetics. Part 41. Dissociative Substitution in Four-Coordinate
Planar Platinum(II) Complexes as Evidenced by Variable-Pressure
High-Resolution Proton NMR Magnetization Transfer Experiments.
J. Am. Chem. Soc. 1989, 111, 8161−8165.
(
89) Wick, D. D.; Goldberg, K. I. C-H Activation at Pt(II) To Form
Stable Pt(IV) Alkyl Hydrides. J. Am. Chem. Soc. 1997, 119, 10235−
0236.
90) Ortuno, M. A.; Conejero, S.; Lledos, A. True and Masked
̃
́
1
(
Three-Coordinate T-shaped Platinum(II) Intermediates. Beilstein J.
Org. Chem. 2013, 9, 1352−1382.
(109) Janssens, T. V. W.; Zaera, F. Neopentyl Iodide on Pt(111) I.
Adsorption and Thermal Decomposition. Surf. Sci. 2002, 501, 1−15.
(110) Lee, T. R.; Whitesides, G. M. Heterogeneous, Platinum-
Catalyzed Hydrogenations of (Diolefin)dialkylplatinum(II) Com-
plexes. Acc. Chem. Res. 1992, 25, 266−272.
(111) Miller, T. M.; Izumi, A. N.; Shih, Y. S.; Whitesides, G. M.
Heterogeneous, Platinum-Catalyzed Hydrogenation of (Diolefin)-
dialkylplatinum(II) Complexes: Kinetics. J. Am. Chem. Soc. 1988, 110,
3146−3156.
(112) Miller, T. M.; McCarthy, T. J.; Whitesides, G. M. Deuterium-
Labeling Experiments Relevant to the Mechanism of Platinum-
Catalyzed Hydrogenation of (Diolefin)dialkylplatinum(II) Com-
plexes: Evidence for Isotopic Exchange via Platinum Surface
Hydrogen. The Stereochemistry of Reduction. J. Am. Chem. Soc.
1988, 110, 3156−3163.
(113) Miller, T. M.; Whitesides, G. M. Isotopic Exchange Reactions
Occurring in the Hydrogenation of (1,5-Cyclooctadiene)-
dialkylplatinum(II) Complexes Over Platinum Black. J. Am. Chem.
Soc. 1988, 110, 3164−3170.
(
91) Labinger, J. A. Platinum-Catalyzed C−H Functionalization.
Chem. Rev. 2017, 117, 8483−8496.
(
92) Yoder, J. C.; Bercaw, J. E. Chain Epimerization during
Propylene Polymerization with Metallocene Catalysts: Mechanistic
Studies Using a Doubly Labeled Propylene. J. Am. Chem. Soc. 2002,
1
24, 2548−2555.
93) Peng, T. S.; Gladysz, J. A. Mechanism of Equilibration of
Diastereomeric Chiral Rhenium Alkene Complexes of the Formula
(
5
+
‑
[
(η -C H )Re(NO)(PPh )(H CCHR)] BF . The Metal Traverses
5
5
3
2
4
Between Alkene Enantiofaces Without Dissociation! J. Am. Chem. Soc.
992, 114, 4174−4181.
94) Kelly, W. M.; Wang, S.; Collins, S. Polymerization of
1
(
Cyclopentene Using Metallocene Catalysts: Competitive Cis- and
Trans-1,3 Insertion Mechanisms. Macromolecules 1997, 30, 3151−
3
158.
(
95) Jones, W. D. Isotope Effects in C-H Bond Activation Reactions
by Transition Metals. Acc. Chem. Res. 2003, 36, 140−146.
96) Churchill, D. G.; Janak, K. E.; Wittenberg, J. S.; Parkin, G.
(
Normal and Inverse Primary Kinetic Deuterium Isotope Effects for C-
H Bond Reductive Elimination and Oxidative Addition Reactions of
Molybdenocene and Tungstenocene Complexes: Evidence for
Benzene σ-Complex Intermediates. J. Am. Chem. Soc. 2003, 125,
(114) Cahill, J. J.; Panayotov, V. G.; Cowen, K. A.; Harris, E.;
Koplitz, L. V.; Birdwhistell, K.; Koplitz, B. Development of a Method
for Investigating Carbon Removal Processes during Photoassisted
Film Growth Using Organometallic Precursors: Application to
Platinum. J. Vac. Sci. Technol., A 2007, 25, 104−109.
1
(
403−1420.
97) Jensen, M. P.; Wick, D. D.; Reinartz, S.; White, P. S.;
(115) Geyer, S. M.; Methaapanon, R.; Shong, B.; Pianetta, P. A.;
Bent, S. F. In Vacuo Photoemission Studies of Platinum Atomic Layer
Deposition Using Synchrotron Radiation. J. Phys. Chem. Lett 2013, 4,
176−179.
Templeton, J. L.; Goldberg, K. I. Reductive Elimination/Oxidative
Addition of Carbon-Hydrogen Bonds at Pt(IV)/Pt(II) Centers:
Mechanistic Studies of the Solution Thermolyses of TpMe Pt-
2
(
CH ) H. J. Am. Chem. Soc. 2003, 125, 8614−8624.
(116) Somorjai, G. A.; Blakely, D. W. Mechanism of Catalysis of
Hydrocarbon Reactions by Platinum Surfaces. Nature 1975, 258,
580−583.
3
2
(
98) Lersch, M.; Tilset, M. Mechanistic Aspects of C-H Activation
by Pt Complexes. Chem. Rev. 2005, 105, 2471−2526.
99) Jayaraman, S.; Klein, E. J.; Yang, Y.; Kim, D. Y.; Girolami, G. S.;
(
(117) Simonovis, J.; Tillekaratne, A.; Zaera, F. The Role of
Carbonaceous Deposits in Hydrogenation Catalysis Revisited. J.
Phys. Chem. C 2017, 121, 2285−2293.
Abelson, J. R. Chromium Diboride Thin Films by Low Temperature
Chemical Vapor Deposition. J. Vac. Sci. Technol., A 2005, 23, 631−
6
(
33.
(118) Gozum, J. E. Transition Metal Allyls and Hydrides as Chemical
Vapor Deposition Precursors; University of Illinois at Urbana-
Champaign: Urbana, IL, 1991.
100) Jayaraman, S.; Yang, Y.; Kim, D. Y.; Girolami, G. S.; Abelson,
J. R. Hafnium Diboride Thin Films by Chemical Vapor Deposition
from a Single Source Precursor. J. Vac. Sci. Technol., A 2005, 23,
(119) van Lent, R.; Auras, S. V.; Cao, K.; Walsh, A. J.; Gleeson, M.
1
(
619−1625.
A.; Juurlink, L. B. F. Site-Specific Reactivity of Molecules with Surface
101) Babar, S.; Mohimi, E.; Trinh, B.; Girolami, G. S.; Abelson, J.
Defectsthe Case of H Dissociation on Pt. Science 2019, 363, 155.
2
R. Surface-Selective Chemical Vapor Deposition of Copper Films
through the Use of a Molecular Inhibitor. ECS J. Solid State Sci.
Technol. 2015, 4, N60−N63.
(120) Cao, K.; van Lent, R.; Kleyn, A. W.; Kurahashi, M.; Juurlink,
L. B. F. Steps on Pt Stereodynamically Filter Sticking of O . Proc. Natl.
2
Acad. Sci. U.S.A. 2019, 116, 13862.
(121) Zhang, P.; Mohimi, E.; Talukdar, T. K.; Abelson, J. R.;
(
102) Zinn, A. A. Chemical Vapor Deposition of Tungsten. In The
Chemistry of Metal CVD; Kodas, T. T.; Hampden-Smith, M. J., Ed.;
Girolami, G. S. Iron CVD from Iron Pentacarbonyl: Growth
R
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
pubs.acs.org/cm
Article
Inhibition by CO Dissociation and Use of Ammonia to Restore
Constant Growth. J. Vac. Sci. Technol., A 2016, 34, No. 051518.
(
122) Babar, S.; Kumar, N.; Zhang, P.; Abelson, J. R.; Dunbar, A. C.;
Daly, S. R.; Girolami, G. S. Growth Inhibitor To Homogenize
Nucleation and Obtain Smooth HfB Thin Films by Chemical Vapor
2
Deposition. Chem. Mater. 2013, 25, 662−667.
(
123) McDermott, J. X.; White, J. F.; Whitesides, G. M. Thermal
Decomposition of Bis(phosphine)platinum(II) Metallocycles. J. Am.
Chem. Soc. 1976, 98, 6521−6528.
(
124) Liu, S.; Gray, D.; Zhu, L.; Girolami, G. S. Lithium−Olefin π-
Complexes and the Mechanism of Carbolithiation: Synthesis,
Solution Behavior, and Crystal Structure of (2,2-Dimethylpent-4-en-
1
(
-yl)lithium. Organometallics 2019, 38, 2199−2210.
125) Nuts; Acorn NMR Inc.: Livermore, CA, 2012.
(
126) Reich, H. J. WinDNMR help. https://www.chem.wisc.edu/
areas/reich/windnmrhlp/index.htm (accessed June 10th, 2019).
127) Ammann, C.; Meier, P.; Merbach, A. A Simple Multinuclear
NMR Thermometer. J. Magn. Reson. 1982, 46, 319−321.
128) Morse, P. M.; Spencer, M. D.; Wilson, S. R.; Girolami, G. S. A
(
(
•
••
Static α-CH M Interaction Observable by NMR Spectroscopy:
Synthesis of the Chromium(II) Alkyl [Cr (CH SiMe ) ] and its
2‑
2
2
3 6
Conversion to the Unusual ’Windowpane’ Bis(metallacycle) Complex
2
2‑
[
1
Cr(κ CC’-CH SiMe CH ) ]. Organometallics 1994, 13, 1646−
2
2
2 2
655.
129) Price, D. M. Vapor Pressure Determination by Thermog-
ravimetry. Thermochim. Acta 2001, 367−368, 253−262.
130) Sasse, K.; Jose, J.; Merlin, J.-C. A Static Apparatus for
(
(
Measurement of Low Vapor Pressures. Experimental Results on High
Molecular-Weight Hydrocarbons. Fluid Phase Equilib. 1988, 42, 287−
3
04.
(
131) August, E. F. Ueber die Berechnung der Expansivkraft des
Wasserdunstes. Ann. Phys. 1828, 89, 122−137.
132) Mohimi, E.; Zhang, Z. V.; Mallek, J. L.; Liu, S.; Trinh, B. B.;
(
Shetty, P. P.; Girolami, G. S.; Abelson, J. R. Low Temperature
Chemical Vapor Deposition of Superconducting Vanadium Nitride
Thin Films. J. Vac. Sci. Technol., A 2019, 37, No. 031509.
(
133) Liu, S.; Zhang, Z.; Abelson, J. R.; Girolami, G. S. “Platinum
ω-Alkenyl Compounds as Chemical Vapor Deposition Precursors.
Mechanistic studies of the Thermolysis of Pt[CH CMe CH CH
2
2
2
CH ] in Solution and the Origin of Rapid Nucleation,” Submitted to
2
2
Organometallics, in press . DOI: 10.1021/acs.organomet.0c00542.
134) Cheon, J.; Dubois, L. H.; Girolami, G. S. Mechanistic Studies
(
of the Thermolysis of Tetraneopentyltitanium(IV). 2. Solid State and
Ultra-High-Vacuum Studies of the Chemical Vapor Deposition of
TiC Films. J. Am. Chem. Soc. 1997, 119, 6814−6820.
NOTE ADDED AFTER ASAP PUBLICATION
■
This paper was published on the Web on October 27, 2020,
with reference 77 missing parts. The corrected version was
reposted on October 28, 2020.
S
https://dx.doi.org/10.1021/acs.chemmater.0c03226
Chem. Mater. XXXX, XXX, XXX−XXX