Experimental and First-Principles NMR Analysis of Pt(II) Complexes with O,O′-Dialkyldithiophosphate Ligands

Article abstract of DOI:10.1021/acs.jpca.6b09586

Polycrystalline bis(dialkyldithiophosphato)Pt(II) complexes of the form [Pt{S₂P(OR)₂}₂] (R = ethyl, iso-propyl, iso-butyl, sec-butyl or cyclo-hexyl group) were studied using solid-state ³¹P and ¹⁹⁵Pt NMR spectroscopy, to determine the influence of R to the structure of the central chromophore. The measured anisotropic chemical shift (CS) parameters for ³¹P and ¹⁹⁵Pt afford more detailed chemical and structural information, as compared to isotropic CS and J couplings alone. Advanced theoretical modeling at the hybrid DFT level, including both crystal lattice and the important relativistic spin-orbit effects qualitatively reproduced the measured CS tensors, supported the experimental analysis, and provided extensive orientational information. A particular correction model for the non-negligible lattice effects was adopted, allowing one to avoid a severe deterioration of the ¹⁹⁵Pt anisotropic parameters due to the high requirements posed on the pseudopotential quality in such calculations. Though negligible differences were found between the ¹⁹⁵Pt CS tensors with different substituents R, the ³¹P CS parameters differed significantly between the complexes, implying the potential to distinguish between them. The presented approach enables good resolution and a detailed analysis of heavy-element compounds by solid-state NMR, thus widening the understanding of such systems.

Full text of DOI:10.1021/acs.jpca.6b09586

Subscriber access provided by RYERSON UNIV
Article
Experimental and First-Principles NMR Analysis of Pt(II)
Complexes With O,O#-Dialkyldithiophosphate Ligands
Juho Roukala, Simon Orr, John V Hanna, Juha Vaara,
Alexander V. Ivanov, Oleg N. Antzutkin, and Perttu Lantto
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.6b09586 • Publication Date (Web): 30 Sep 2016
Downloaded from http://pubs.acs.org on October 10, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry A is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Experimental and First-Principles NMR
Analysis of Pt(II) Complexes With
O,O′-Dialkyldithiophosphate Ligands
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Juho Roukala,∗,† Simon Orr,‡ John V. Hanna,‡ Juha Vaara,† Alexander V. Ivanov,¶
Oleg N. Antzutkin,§,‡ and Perttu Lantto∗,†
†NMR Research Unit, University of Oulu, P.O. Box 3000, FI-90400 Oulu, Finland
‡Department of Physics, Warwick University, Gibbet Hill Rd., Coventry CV4 7AL, United
Kingdom
¶
Institute of Geology and Nature Management, Far Eastern Branch of the Russian Academy
of Sciences, Ryolochny Lane 1, Blagoveshchensk, 675000, Amur Region, Russia
§
Chemistry of Interfaces, Luleå University of Technology, Universitetsområdet C140, Porsön,
Luleå, Sweden
E-mail: juho.roukala@oulu.fi; perttu.lantto@oulu.fi
Phone: +358 (0)294 48 1312
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 2 of 30
1
2
3
4
Abstract
5
6
7
8
Polycrystalline bis(dialkyldithiophosphato)Pt(II) complexes of the form Pt{S2P(OR)2}2
(R = ethyl, iso-propyl, iso-butyl, sec-butyl or cyclo-hexyl group) were studied using
solid-state 31P and 195Pt NMR spectroscopy, to determine the influence of R to the
structure of the central chromophore. The measured anisotropic chemical shift (CS)
parameters for 31P and 195Pt afford more detailed chemical and structural information,
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
as compared to isotropic CS and -couplings alone. Advanced theoretical modeling at
J
hybrid DFT level, including both crystal lattice and the important relativistic spin–orbit
effects qualitatively reproduced the measured CS tensors, supported the experimental
analysis, and provided extensive orientational information. A particular correction
model for the non-negligible lattice effects was adopted, allowing one to avoid a severe
deterioration of the 195Pt anisotropic parameters due to the high requirements posed
on the pseudopotential quality in such calculations. While negligible differences were
found between the 195Pt CS tensors with different substituents R, the 31P CS param-
eters differed significantly between the complexes, implying potential to distinguish
between them. The presented approach enables good resolution and a detailed analysis
of heavy-element compounds by solid-state NMR, thus widening the understanding of
such systems.
Introduction
Both experimental nuclear magnetic resonance (NMR) measurements and calculations of the
chemical shift (CS) tensor parameters of heavy nuclei, such as 195Pt 207Pb, etc., still constitute
,
a challenge. The large chemical shift anisotropy (CSA), which can amount to a few thousand
ppm, causes very wide spectral features in solid-state NMR. On the other hand, relativistic
effects on the electronic structure have to be taken into account in the quantum-mechanical
calculations of the NMR parameters. Studies of heavy-atom induced effects on light atoms
have been carried out with one- or two-component relativistic methods1–5 and, latterly, with
four-component calculations.6–8 Instead, the investigation of the heavy nuclei themselves1,9,10
has been less frequent, particularly in larger compounds or at the fully relativistic level.11,12
Square-planar platinum(II) complexes featuring an intra-orbital dsp2 hybrid state of the
central platinum atom are of interest mainly in medical applications. The high antitumor ac-
tivity of cis−[Pt(NH3)2Cl2] (cisplatin, first generation) was first discovered by Rosenberg13,14
in 1969. Since then, practical and important properties have also been established in other
2
ACS Paragon Plus Environment

Page 3 of 30
The Journal of Physical Chemistry
1
2
3
4
5
planar Pt(II) compounds, e.g., cis−[Pt(NH3)2{(OOC)2C(CH2)3}] (carboplatin, second gen-
eration),15 and [Pt{1, 2−(NH2)2(C6H10)}(OOC−COO)], (oxaliplatin, third generation).16,17
6
7
8
9
Most of the presently known bis(dialkyldithiophosphato)platinum(II) complexes are liquid
substances.18 Therefore, only few crystalline platinum(II)
O,O -dialkyldithiophosphates have
′
previously been studied.18–24 A schematic representation of the complexes studied here is
shown in Figure 1. The CSA of both 195Pt and 31P nuclei can be useful for studying the
structures of these compounds, particularly in the case of Pt(II)-dialkyldithiophosphate
surface complexes, when other structural methods such as X-ray diffraction are not applicable
because of the amorphous nature of samples.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R
R
O
O
O
O
S
S
S
S
P
Pt
P
R
R
′
Figure 1: Schematic representation of the molecular structure of the Pt(II)
O,O -dialkyl
dithiophosphates. The alkyl group R is one of C2H5 (compound I), iso-C3H7 (II), iso-C4H9
(III), sec-C4H9 (IV), or cyclo-C6H11 (V).
A full analysis of the 195Pt and 31P CS tensors of five compounds of the form Pt{S2P(OR)2}2
21,23
is reported in this study. For platinum(II) complexes with R=iso-C4H9
,
R=iso-C3H7
22
and R=cyclo-C6H11
,
some of the experimental 195Pt magic-angle-spinning (MAS) NMR
data has been reported, while for compounds with R=C2H5 and R=sec-C4H9, the data are
missing until now. Previous reports have revealed a clear difference between the 195Pt CSA
25,26
of cooperite (PtS
)
and platinum(II) dialkyldithioposphates.21–23 The former compound
has a structure based on square planar [PtS4] chromophores while the latter complexes form
distorted squares, yet they maintain an almost axially symmetric 195Pt CSA. Without a
complete analysis of the platinum-195 CSA it has been, so far, impossible to determine
whether the type of alkyl ligands has a significant effect on the degree of deviation from the
square planar geometry.
As the 195Pt CSA is too large for NMR data acquisition with a single frequency window,
a summed Fourier-transform field-sweep technique27 was employed in this work. This allowed
for a fully automated acquisition of the entire lineshape without the need for normalizing
intensities with respect to the frequency of acquisition, as would be required with a traditional
frequency-stepping approach.28 To complement experiments, a computational study was also
performed in this work, yielding the 195Pt and 31P nuclear shielding tensors, as well as the
Pt–P spin–spin coupling constants, J
. Recently, Fu and Vaara calculated the relativistic 195Pt
and 31P shielding tensors of the in vacuo bis(diethyldithiophosphato)platinum(II) complex.29
To our knowledge, the present paper constitutes the first successful attempt at calculating the
solid-state NMR properties in this family of Pt(II) complexes. Relativistic effects, arising due
to the heavy metal, are here found to significantly affect the NMR properties of both 31P and
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 30
1
2
3
4
5
6
7
8
9
195Pt nuclei. A recent benchmark study30 has also shown the importance of relativistic effects
to the 31P chemical shifts. The largest effect of the crystal lattice is to modify the isotropic
chemical shifts of both 195Pt and 31P. The contribution to the CSA and asymmetry parameters
of these two types of nuclei is smaller, albeit noticeable, and should not be neglected if at
least qualitative agreement with experiments is sought after. In this work a good agreement
is obtained between theoretical and with experimental results, especially for phosphorus.
Additionally, the sign of the spin–spin coupling constants and the directions of the principal
axes of the CS tensors are revealed from the calculations. These are of crucial importance
for understanding the correlation between the structure and NMR spectra of the named
platinum(II)-dialkyldithiophosphate complexes, as well as other heavy element-containing
molecular crystals.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
Five crystalline complexes were obtained by reacting K2[PtCl4] and salts of the form
◦
K{S2P(OR)2} in aqueous solutions at 60 C overnight. The reaction products were filtered
leaving yellow precipitates that were washed and left to dry. The following platinum(II)-
dialkyldithiophosphate complexes were prepared using this protocol and abbreviated as:
Pt-dtp-ethyl [bis(diethyldithiophosphato)platinum(II), compound I], Pt-dtp-iso-propyl [bis(di-
iso-propyldithiophosphato)platinum(II), compound II], Pt-dtp-iso-butyl [bis(di-iso-butyl-
dithiophosphato)platinum(II), compound III], Pt-dtp-sec-butyl [bis(di-sec-butyldithiophos-
phato)platinum(II), compound IV] and Pt-dtp-cyclo-hexyl [bis(di-cyclo-hexyldithiophos-
phato)platinum(II), compound V]. The X-ray diffraction structures [see Tables S1–S7 in the
Supporting Information (SI)] of compounds III and IV contained two alternative conformations
each, which are denoted with A and B, for both complexes.
All 195Pt static field-sweep NMR data were obtained at 7.05 T using Chemagnetics CMX-
300 spectrometer and a modified Magnex magnet equipped with a secondary superconducting
coil capable of sweeping the field 0.5 T. The current to the secondary coil was controlled
by the spectrometer through a Lakeshaw magnet power supply. All measurements were
undertaken using a Bruker Z33V HP static probe, with approximately 200 mg of each sample
packed into 9.5 mm PTFE sample holders, which were accommodated in the horizontal
solenoid arrangement of this probe. In order to isolate the very rapidly decaying free induction
decay (FID) signal from probe ringing, a π/2–
τ
–
π
–
τ
Hahn echo sequence was used. The
lengths of the π/2 and pulses were 5 s and 10
π
µ
µ
s, respectively, and delay was 15 s. A
τ
µ
recycle delay of 4 s was implemented for complex II, while complexes I, III, IV and V required
a longer recycle delay of 35 s. All 195Pt isotropic chemical shifts, δiso, were referenced to the
sharp single resonance line of 1M K2Pt(CN)6 (aq) at δiso
= −3866 ppm (external reference).
The field sweeps were initiated with zero current in the secondary coil and a resultant field of
7.05 T, with the probe being tuned to 64.08 MHz, which is below the known low-frequency
edge of each measured 195Pt lineshape. For complexes II–V the current was then increased to
1.2963 A in six equal increments of 216 mA, each of which reduced the resultant field applied
at the sample. For complex II, 11000 transients were acquired at each field value, while for
complexes III–V, 5100 transients were acquired. For complex I, the current was increased
to 1.080 A in five equal increments of 216 mA with 7400 transients acquired at each field
4
ACS Paragon Plus Environment

Page 5 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
value. The current in the secondary coil is linked to the frequencies
F of the spins at 7.05
T by Amax
respectively, and
the secondary coil.31
The data at each field value was Fourier-transformed and processed with 5 kHz of exponen-
tial line broadening prior to shifting the frequency axis by ∆ Fmax−Fmin Nslices 1) =
50 kHz with respect to the previous spectrum. This shifted each spectrum to the corre-
sponding frequency at B0 = 7 05 T. The spectra were then summed to reconstruct the total
=
B0C
(
Fmax − Fmin
)/Fmin. In this case Fmax and Fmin are 64.38 and 64.08 MHz,
C
= 0 0254615 is a constant that depends on the particular properties of
.
F
=
−
(
)
/
(
−
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
.
lineshape. As this technique relies on the recreation of a lineshape with a greater span than
the equipment is capable of exciting at a single field value, the gap between each spectrum
must be considered carefully. If the spectra are widely enough separated, the lineshape will
become distorted by the excitation profiles of the individual spectra. Modeling the response
of both the spins and the spectrometer show that this effect becomes visually insignificant
when ∆
F is less than or equal to 1.5δF, where 2δF is the full width at half maximum of an
individual spectrum.31 It is observed that the FWHM width of an excitation and response
was ca. 75 kHz for a π/2 pulse length of 5 µs, so the condition ∆F < 56 kHz is fulfilled.
Each summed 195Pt broadline spectrum was simulated using a static CSA lineshape function
using the Dmfit software package.32
All 31P CPMAS NMR data were recorded at 7.05 T using a Chemagnetics CMX-300
console operating at a 31P Larmor frequency of 121.482 MHz. The 1H 31P cross-polarization
–
experiment was used in conjunction with continuous wave proton decoupling at 300.098 MHz
to suppress heteronuclear dipolar interactions. These measurements were undertaken using
a Bruker 4 mm HX probe, which enabled MAS frequencies of 2–4 kHz. The 1H π/2 pulse
length was 3.5
were indirectly referenced to 85% H3PO4
reference at δiso = 0
9 ppm.33 The 31P chemical shift anisotropy parameter, δaniso, and the
asymmetry parameter, , of the chemical shift tensor were calculated using the Mathematica
µ
s, the contact time 2 ms, and the recycle delay 3 s. The 31P chemical shifts
(
δiso = 0.0 ppm) via a secondary solid NH4H2PO4
.
η
program of Levitt and co-workers,34 based on an analysis of χ2 statistics. Plotting of the
diagrams was based on a quantitative analysis of the integrated intensity ratios of the spinning
sidebands in the spectra, recorded at two different spinning frequencies. The calculations
were performed using the Mathematica program (version 4.1.2).35,36
Computational Details
First-principles density functional theory (DFT) calculations of the NMR shielding ten-
sors
CASTEP37–40 and Amsterdam Density Functional41–46 (ADF) codes, respectively. The latter
program was also used in the computation of spin–spin -couplings, which were not available
σ were performed using both solid-state and single-molecule methodologies with the
J
in the version of CASTEP in use. Nonrelativistic calculations were performed with ADF
(see SI for more details). The zeroth-order regular approximation47,48 (ZORA) method was
utilized both in CASTEP and ADF to treat relativistic effects, through scalar-relativistic
(SR) pseudopotentials49 in the former, and at the 1-component SR and 2-component spin–
orbit (SO) levels in the latter one. The fully relativistic ReSpect50–52 program based on
4-component theory was used to benchmark the ADF calculations (see Tables S30 and S31 in
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 6 of 30
1
2
3
4
5
SI). A unit cell with 6
for crystal calculations, while 3
×
5
×
4
k
-points and a cutoff energy of 750 eV was used in CASTEP
-points were used for periodic calculations at the
× 3 × 3 k
6
7
8
9
molecular limit (a single molecule in a cubic cell with a side length of 20 Å). The molecular
calculations employed all-electron jcpl/TZP basis sets53 for Pt and P/other atoms in ADF,
and Dyall’s cvtz-level basis sets54–57 for all atoms in ReSpect. The parameter-free PBE58,59
density functional was used in all codes, with additional PBE060,61 hybrid functional calcula-
tions performed with ADF. Calculations were performed for both A and B conformations of
complexes III and IV.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The reported computational results (vide infra) were obtained using geometry-optimized
structures, where both the ions and the cell parameters were relaxed using CASTEP at the
“fine” level with ultrasoft pseudopotentials,62 a cutoff energy of 340 eV, and
k
-point grids
3×3×2 (compound I), 4×3×2 (II), 3×3×3 (III), 4×3×2 (IV), and 2×3×2 (V). The unit cells
of compounds I and V contained two molecules each, whereas compounds II–IV contained
one molecule per unit cell. The effect of geometry optimization to the NMR parameters was
studied at the SR solid-state level of theory (vide infra).
Crystal lattice effects were found to be non-negligible for the shielding tensors, necessitating
solid-state calculations. A molecular model, on the other hand, allowed evaluating the
relativistic effects due to the heavy platinum atom more thoroughly, especially the SO effect,
as well as assessing the impact of using a hybrid DFT functional. Both features are known
to affect the shielding tensors of heavy-element compounds.63,64 Lacking software capable of
performing solid-state NMR calculations at the SO-coupled relativistic level of theory, with
a hybrid DFT functional, a combination of solid-state and molecular approaches was used
here to arrive at the final results. This is in contrast to the other theoretical studies of 195Pt
shielding in solid-state complexes,65,66 where the relativistic and crystal lattice effects have
not been considered simultaneously.
For each complex, the crystal lattice effect was calculated in CASTEP as the difference
between a solid-state crystal calculation with the PBE functional, and another one where a
single molecule was isolated in a cubic cell with a side length of 20 Å. This was taken as a
correction to the base result, the shielding tensor calculated at the SO ZORA relativistic
level using a hybrid PBE0 functional, in ADF. The corrected total tensors were obtained by
summing the corresponding individual shielding tensor components σij as
SR PBE, isolated molecule
ij
corr.
ij
σ
= σiSjO PBE0, molecule + (σSR PBE, solid − σ
)
(1)
ij
after which the resulting principal components of the shielding tensors were obtained by
diagonalizing the symmetric part of the tensor. It should be noted that, with an increasing
cell size, especially the 195Pt CS parameters obtained from periodic calculations with an
isolated molecule in a large unit cell converged to clearly different values as compared to
those obtained from a molecular calculation at roughly the same level (SR PBE). This may
indicate a problem with the pseudopotential used in CASTEP; see SI for more details.
The shielding tensors are analyzed in terms of the isotropic shielding constant σiso, chemical
6
ACS Paragon Plus Environment

Page 7 of 30
The Journal of Physical Chemistry
1
2
3
4
shift δiso, the CSA parameter δaniso, and the asymmetry parameter η,
5
6
7
8
ref
iso
σ
− σiso
ref
ref
iso
ref
iso
δiso
=
≈ σ − σiso, when σ is small
(2)
(3)
(4)
1 − σ
iso
δaniso = δzz − δiso
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
δyy − δxx
η =
,
δzz − δiso
where δzz
, δyy and δxx are the principal components of the chemical shift, labeled in such
a way that |δzz − δiso| ≥ |δxx − δiso| ≥ |δyy − δiso|. Also the spin–spin coupling constants
2
J
(
195Pt
–
31P) were studied. Our results support the findings of a computational study of
-couplings in dioxaphosphorinanes,67 where DFT was shown to qualitatively
31P spin–spin
J
reproduce the experimentally observed J-coupling constants (see SI for more details).
The calculations were performed at an optimized geometry that represents a minimum
energy structure of the PBE functional, including the DFT-D dispersion correction using
the method of Tkachenko and Scheffler.68 The choice of the geometry (vide infra) was based
on SR CASTEP shielding tensor calculations at three different levels of optimization. First,
only hydrogen positions were optimized (H-opt., see Tables S8–S14 in SI) while keeping the
unit cell parameters fixed to their experimental values; then all ions were optimized (I-opt.,
Tables S15–S21 in SI), and finally also the unit cell was relaxed (A-opt., Tables S22–S28 in
SI).
Results and Discussion
Solid-state NMR Data
The 31P NMR signals of all the complexes under study, shown in Figure 2 (a), are clearly
split into three resonance lines with relative intensities ca. 1:4:1. A 1:1 doublet (satellite
resonance lines) is a result of 195Pt
–
31P spin–spin
J
-coupling with a coupling constant of ca.
430–450 Hz. The natural abundance of 195Pt (33.83%) explains the relative intensities of
the satellites. A single wide feature is observed in the spectrum of Pt{S2P(O-sec-C4H9)2}2
where the shape and asymmetry of the spectrum suggest a -coupled split signal broadened
,
J
to a much greater extent than the spectra of the other complexes. It is proposed that this is
caused by the presence of six chiral isomers of the Pt-dtp complex, as explained in Ref. 24.
Inspection of the sideband intensities in the 31P MAS NMR spectra [see Figure 2 (a)]
of all the complexes suggests a nearly rhombic symmetry of the chemical shift tensor of
all the complexes. Statistical estimates of the 31P CSA parameters were made by plotting
the χ2 diagrams of δaniso and
= 0
to a perfectly rhombic arrangement (
η
from the spinning sideband analysis. The range of values
57–0.98 (see Table 1 below) obtained for the complexes suggests a configuration close
= 1). The values of δaniso range from ca. 30 to 45
η
.
η
−
−
ppm for different Pt-dtp complexes in this study.
In contrast to the 31P NMR data, the lineshapes of the static 195Pt NMR spectra [see
Figure 2 (b)] of all the complexes correspond to an almost axially symmetric CS tensor.
They are of a form similar to the spectrum of polycrystalline cooperite25,26 (natural PtS) and
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 8 of 30
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: NMR spectra of Pt{S2P(OR)2}2 complexes: (a) 1H 31P CP MAS NMR spectra,
–
with a spinning frequency of 2 kHz. (b) Static 195Pt field-sweep NMR spectra (blue lines)
and static CSA models from Dmfit (black lines),32 5 kHz of broadening has been applied to
both the spectra and the models.
8
ACS Paragon Plus Environment

Page 9 of 30
The Journal of Physical Chemistry
1
2
Table 1: Calculated and experimental 31P shielding tensor parameters in com-
pounds I–V.
3
4
5
6
7
b
c
Compounda
Theor. σiso (ppm) Exp. δiso (ppm)
8
9
I
II
236.8
239.4
237.9
237.6
244.6
237.1
244.1
(ppm) Exp. δ
2.6
0.0
1.6
1.9
−5.2
2.4
105.0
108.5
104.3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
III A
III B
IV A
IV B
V
Compound
I
104.1
99.0
(ppm)
−3.5
0.0
II d
iso
II d
iso
Theor. δ
II
III A
III B
IV A
IV B
V
−4.2
−4.4
−4.7
−9.5
e
Compound Theor. δaniso (ppm) Exp. δaniso (ppm)
I
II
III A
III B
IV A
IV B
V
Compound
I
−31.3
39.3
32.4
−35.1 0.2
−42.7 0.2
−38.1 0.1
−36.4
44.5
−43.3 0.2
41.2
27.7
Theor. η
0.56
−30.8 0.2
Exp. η
0.57 0.03
0.75 0.02
0.73 0.01
II
0.91
0.94
0.63
0.66
0.95
0.68
III A
III B
IV A
IV B
V
0.89 0.02
0.98 0.02
aPt-dtp-R, R: ethyl (I), iso-propyl (II), iso-butyl (III), sec-butyl (IV), and cyclo-hexyl (V).
A and B refer to the two different conformations of the same complex.
bADF SO-ZORA calculation with the PBE0 hybrid functional, added with crystal lattice
correction from CASTEP and using optimized geometry (ions and cell parameters,
CASTEP).
cReferenced to the peak of ammonium dihydrogen phosphate at 0.9 ppm.33 See Refs. 24 (I,
IV), 22 (II, V), and 21 (III).
dReferenced to compound II.
ePositive sign in some of the calculated 31P CSAs results from a nearly symmetric NMR
spectrum, where a small shift in δyy may swap the labels of δxx and δzz.
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 30
1
2
3
4
5
6
7
8
9
suggest that the platinum atom is held in a configuration close to the square chromophores
(
PtS4) found in cooperite. However, the value of the δzz component of the 195Pt CS tensor
for these complexes is much smaller than that observed in PtS, leading also to a substantial
difference in the isotropic shifts. Fitting of the CSA lineshapes in dmfit was possible with a
good agreement for all the complexes. The systems were observed to have very similar values
of δaniso, within 140 ppm of one another (ca. 2300 ppm, see Table 2). The complex I has
a smaller anisotropy and, correspondingly, a more negative isotropic shift, which perhaps
indicates a closer crystal packing in the direction perpendicular to the PtS4 chromophore,
which would follow intuitively from the smaller ligand size. The complex IV has the largest
experimental anisotropy parameter, which could correspond to a weaker shielding from
looser molecular packing. This may be caused by disorder due to the chiral polymorphism
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of ligands, which is referred to earlier. The values of
η
(ca. 0.25) for the 195Pt CS tensor
have also been estimated for the first time and show little variance among the systems, and
no correlation with δaniso. This suggests that there is little variation in the extent of the
tetrahedral distortion of the PtS4 chromophore between the complexes. For comparison, in
cooperite25 these parameters are δaniso = 5873 ppm and
η = 0.37, reflecting the sensitivity
to the local structure around the platinum site. A clear dependence of the anisotropic
CS parameters on the S−P−S bond angle has also been demonstrated in polycrystalline
′
O, O -dialkyldithiophosphate complexes of Zn(II) and Ni(II),69 and Pb(II).70
Computational
The calculated isotropic shielding constants, chemical shifts, CSA parameters, and asymme-
tries in compounds I–V are shown in Tables 1 and 2 for 31P and 195Pt, respectively. The
chemical shifts are reported relative to the complex with the largest absolute experimental
chemical shift; this is compound II for 31P and compound I in the case of 195Pt. Scalar
relativistic effects were estimated using the molecular model and X-ray structures, and were
found to be important in all complexes (see SI and Tables S30 and S31 therein for more
details). The following discussion is based on results where the SR effects are accounted for.
Phosphorus-31
Figure 3 reveals that the calculations are found to yield good CSA and asymmetry parameters
for 31P in complexes I–V, as compared to experimental data, both in the solid-state and
molecular models. The crystal lattice corrected, SO-coupled relativistic level of calculation
offers a slight improvement over the other approaches (where either SO-coupling and hybrid,
or crystal lattice are missing), but due to the smallness of the various contributions, all the
methods produce rather similar results. Isotropic 31P shielding constants are in the range
237–245 ppm and gain positive contributions from SO coupling (ca. 19%), use of hybrid
functional (ca. 4%) and the crystal lattice (less than 1.5%) in all compounds (see Table S32
in SI). Their impact to the experimental observable, the isotropic chemical shift, is much
smaller because of the high similarity of the values among this group of molecules, leading
to large cancellation. The experimental chemical shifts relative to compound II are small,
ca.
−4 ppm for compounds I, III, and IV, and almost −10 ppm for V. Their computational
counterparts are in the same very narrow magnitude range, although no quantitative match
10
ACS Paragon Plus Environment

Page 11 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
Table 2: Calculated and experimental 195Pt shielding tensor parameters in com-
pounds I–V.
7
8
9
b
c
Compounda Theor. σiso (ppm) Exp. δiso (ppm)
I
II
5855.6
5784.7
5849.4
5869.8
5817.4
5788.0
5793.9
−4030
−4002.0
−4029.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
III A
III B
IV A
IV B
V
−3890
−3963.6
d
d
Compound
I
II
III A
III B
IV A
IV B
V
Theor. δiIso (ppm) Exp. δiIso (ppm)
0.0
70.9
6.3
−14.2
38.2
67.6
61.7
0.0
28.0
0.3
140.0
66.4
Compound Theor. δaniso (ppm) Exp. δaniso (ppm)
I
II
2318.3
2322.7
2392.3
2276.6
2311.2
2258.2
2231.9
Theor. η
0.29
2250.0
2340.0
2340.0
III A
III B
IV A
IV B
V
Compound
I
2390.0
2330.0
Exp. η
0.23
II
0.33
0.23
III A
III B
IV A
IV B
V
0.33
0.30
0.36
0.36
0.26
0.25
0.25
0.35
aPt-dtp-R, R: ethyl (I), iso-propyl (II), iso-butyl (III), sec-butyl (IV), and cyclo-hexyl (V).
A and B refer to the two different conformations of the same complex.
bADF SO-ZORA calculation with the PBE0 hybrid functional, added with crystal lattice
correction from CASTEP and using optimized geometry (ions and cell parameters,
CASTEP).
cReferenced to the peak of K2Pt(CN)6 (aq) at −3866 ppm.
dReferenced to compound I.
11
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 30
1
2
3
4
is obtained within it.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
31
Figure 3: Comparison of the calculated and experimental values of P chemical shift
parameters, δiso (CS, top), δaniso (CSA parameter, middle), and (asymmetry, bottom) in
η
compounds I–V. The three leftmost groups of bars in each panel correspond to calculations
at the SR level in solid state (PBE functional), SO ZORA level in a molecular model
(PBE0 functional), and the molecular results corrected with the crystal lattice contribution,
correspondingly. Columns annotated with an asterisk * are actually positive, but negated
here for easier comparison; see text for details. Bars annotated with A or B refer to two
different conformations of the same complex in the crystallographic unit cell.
The 31P CSA parameter is found to be positive in most calculations, with compounds
I and III B being the only ones having a negative CSA parameter at the crystal lattice
corrected SO ZORA level of calculation (see Table 1). This is due to the highly symmetric
31P NMR spectrum, where δyy
≈
(
δxx
+
δzz). Thus, even a small change in δyy relative to
1
2
12
ACS Paragon Plus Environment

Page 13 of 30
The Journal of Physical Chemistry
1
2
3
4
5
the two other components may swap the labels xx and zz, resulting in a sign change in the
parameter.
6
7
8
9
The asymmetry parameter is a delicate property to calculate, as demonstrated by the
mutually partly cancelling contributions due to geometry, hybrid functional and SO effects,
which are positive for 31P η in some compounds and negative for some (vide infra). The
crystal lattice corrected computed asymmetry parameters are, however, larger than 0.5 in all
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
complexes, with II, III A and IV B having η ≈ 0.9 (see Table 1), meaning nearly rhombic CS
tensors.
The computational results agree qualitatively with experiments in all the compounds,
and the remaining differences are expected to be, at least partly, due to errors introduced by
the chosen computational model. The different basis sets employed in the different programs
(plane waves in CASTEP, Slater-type functions in ADF, and Gaussian functions in ReSpect),
for example, are a source of error until the basis set limit, which is difficult to reach for
systems of this size. The numerical accuracy is also dependent on the integration grids and
convergence criteria, which may ultimately have an effect on the final results. Even if these
errors are small within a single program, their cumulative effect may be more visible when
the outcomes from several programs are combined into a single result.
Geometry optimization. The changes to the CS, as well as the CSA and asymmetry
parameters due to the optimization of the hydrogen atoms, then all ions, and finally also the
unit cell parameters, is shown graphically in Figure 4, where the difference of the parameters
with respect to the experimental values is plotted. The corresponding numerical data are
presented in Tables S30, S31, S32, and S33 in SI. The most notable benefit, obtained at the
point where the geometry was optimized for all the ions, is to reduce the greatly overestimated
31P CSA in compound IV to a level that agrees well with experiments. Optimization of only
hydrogens introduces very small changes, which nevertheless demonstrate the sensitivity of
NMR to structural modifications. Relaxing the unit cell parameters in addition to all ions
only has a little effect to the phosphorus CSA, mainly canceling the difference between the
A and B conformations of compound III, which was introduced by the cell-constrained ion
optimization. The asymmetry parameter is relatively more affected, but in a less consistent
way.
Solid-state and molecular models. The 31P chemical shifts relative to compound II
gain a relatively large positive contribution (up to ca. 2 ppm) from the crystal lattice, as seen
in Figure 5 (the corresponding numerical data are presented in Tables S32 and S33 in SI),
although the absolute changes are small due to small absolute values of the thus referenced CS
values. The contributions to the CSA and asymmetry parameters are smaller in comparison
to the two other contributions (exact exchange and SO-coupling), up to ca. 6 ppm and 0.16,
respectively. As seen in Figure 5, the main effect of the crystal lattice correction is to induce
small relative changes between the CS parameters in different compounds.
Hybrid DFT functional. The partial inclusion of Hartree–Fock exchange into the DFT
functional was studied at the SR level of theory in the molecular model. The use of the hybrid
functional PBE0 (25% exact exchange) increases slightly the 31P chemical shifts with respect
to compound II in all complexes, as shown in Figure 5, except IV B where the contribution
is zero. The magnitude of the CSA parameter is increased in all compounds except for IV A,
where the inclusion of HF exchange yields a very small decrease. The asymmetry parameter
is less consistently affected by the hybrid functional, with positive contributions in III A,
13
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 30
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Effect of geometry optimization to the 31P (panels on the left) and 195Pt (right)
chemical shift parameters. The plotted values show the difference in the parameters between
the experimentally observed value (the zero line) and the fully, partially, or unoptimized
(experimental X-ray) geometries. In the case of 31P CSA parameters, the magnitudes |δaniso
|
are shown, to avoid comparison between positive and negative values (see text for more
details). The large deviations of the 195Pt CSA and asymmetry parameters from zero are
due to the fact that CASTEP overestimates (underestimates) the calculated δaniso (η); see SI
for details.
14
ACS Paragon Plus Environment

Page 15 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: Contributions due to the crystal lattice, use of hybrid functional, and spin–orbit
coupling to the 31P (panels on the left) and 195Pt (right) CS parameters, using the fully
optimized geometries. The values denote the following differences: (solid-state)
in-vacuo) calculation (crystal lattice) at the periodic SR PBE level of theory, PBE0
(Hybrid functional) at the molecular SR level of theory, and SO SR calculation (SO coupling)
at the molecular level, with PBE0. In the case of the 31P CSA parameters, the magnitudes
|δaniso are shown, to avoid comparison between positive and negative values (see text for
−
(molecule-
−
PBE
−
|
more details). The total values of each property are obtained by tensor component-wise
summation as shown in (1), and not by summing these property differences, as explained in
the text.
15
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 30
1
2
3
4
IV A and V, and negative ones in the others.
Spin–orbit correction. While a relatively light element itself, phosphorus is also
influenced by the so-called HALA (heavy-atom effects on the light atom71–73) relativistic
effects due to the vicinity of the heavier atoms (S, Pt). The 31P chemical shift, referenced to
compound II, gains positive contributions from the SO-coupling at the ZORA level of theory
in all complexes, as seen in Figure 5, although the change in compound V is negligibly small.
In compounds I and III, the SO-contribution to CS is larger than the sum of the corresponding
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
hybrid and crystal lattice modifications. While there may appear a vague decreasing trend,
II
iso
from complex I to V, both in the
δ
(
31P) and also in the SO contribution therein, the number
of systems displaying this tendency in the present work is not large enough to draw any
definite conclusions. The phosphorus CSA and asymmetry parameters gains both positive
and negative contributions of varying magnitude, depending on the complex. Except for the
asymmetry parameter in compounds II and III A, the SO and hybrid functional contributions
to the 31P δaniso and
η have opposite signs, with the magnitude of the SO-coupling being
slightly smaller in most cases.
Based on the results, all the contributions are numerically significant, although there
is some cancellation between them in the total parameters, as seen in Figure 5. Thus, the
present approach of combining several calculations to include all the effects is justified, even
if it does introduce a possible source of error as noted earlier.
Platinum-195
Unlike the CS tensor parameters of 31P, those of 195Pt require much more attention to obtain
an agreement with experimental reference values for the right reasons. While the solid-state
CASTEP calculation with X-ray ion positions and cell parameters appeared to produce
qualitatively correct relative chemical shifts for the studied compounds, as referenced to
experimentally measured parameters, the approach failed to agree with the experimental CSA
and asymmetry parameters. Notably, the periodic molecular-limit (CASTEP) calculations
yielded much larger 195Pt CS anisotropy and much smaller asymmetry parameters as compared
to the SR molecular calculation (ADF), hinting at a problem with the pseudopotential used
in CASTEP (see SI for more details). Thus, the solid-state calculation did not provide a
good starting point for the theoretical modeling in this case. Instead, with an optimized
geometry, the all-electron molecular calculation at the SO-coupled ZORA level of theory using
the hybrid functional (PBE0) provided a better approximation, yielding already qualitative
agreement with the experiments (see Figure 6), and adding the crystal lattice effect further
improved the lineshape corresponding to the 195Pt CS tensor, as compared to experiments in
Table 2. As with phosphorus, the isotropic 195Pt shielding constants (5785–5870 ppm) and
the contributions therein are nearly equivalent in all compounds, with spin–orbit coupling
being the dominant one (ca. 55%) and crystal lattice the smallest (up to 1.7%). Unlike with
phosphorus, the use of hybrid functional decreases the platinum shielding constants slightly
in all compounds (by up to 7%, see Table S33 in SI). The following discussion is based on
the experimentally observable isotropic CS, where cancellation again results in much smaller
absolute contributions.
Considering the different complexes, little variation is found between them in the 195Pt
CSA and asymmetry parameters, as shown in Table 2. Especially the magnitude of the
16
ACS Paragon Plus Environment

Page 17 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
195
Figure 6: Comparison of the calculated and experimental values of
parameters, δiso (CS, top), δaniso (CSA parameter, middle), and
Pt chemical shift
η (asymmetry, bottom) in
compounds I–V. The three leftmost groups of bars in each panel correspond to calculations
at the SR level in solid state (PBE functional), SO ZORA level in a molecular model
(PBE0 functional), and the molecular results corrected with the crystal lattice contribution,
correspondingly. Bars annotated with A or B refer to two different conformations of the same
complex in the crystallographic unit cell.
17
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 30
1
2
3
4
5
6
7
8
9
platinum CSA parameter is roughly equal (between 2200 and 2400 ppm) in all complexes.
However, its value in compound I is consistently larger than in compound V (see Figure 6),
irrespective of the geometry optimization and corrections due to SO-contribution and the use
of hybrid functional. This is in contrast to experiments, even though the difference between
the CSA parameters of compounds I and V is small both computationally and experimentally.
The 195Pt asymmetry parameters also show only small variations (up to 0.07 between the
calculated values in I and IV) between the complexes, and while the calculated values are
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
slightly larger (up to ca. 0.1) than their experimental counterparts, they agree that
η is small,
as compared to the same parameter for 31P. These findings favor the idea that the central
part of the complexes is more or less invariant to the different ligands, and therefore also
the platinum shielding tensors are nearly equivalent in shape for Pt complexes with ligands
belonging to the same class, i.e., dialkyl dithiophospates. A large anisotropy combined with a
relatively small asymmetry translates to an axial, disc-shaped shielding tensor in the present
case.
Geometry optimization. As with phosphorus, the relaxation of the hydrogen positions
only introduces small changes, most visible in the 195Pt CS of compound II (relative to
compound I) and the CSA parameters in compounds II and IV, as shown in Figure 4. The
largest change to the CS parameters is induced by the ion optimization, leading to a notable
decrease of the CSs, and increase of the CSA parameters in all complexes. The asymmetry
parameters are hardly affected by the optimization, remaining closer to zero than unity in all
compounds. Relaxation of the cell parameters in addition to ions does not yield great changes,
but does increase the magnitude of the 195Pt CS in compound III relative to compound I. It
should be noted that in Figure 4 the geometry optimization causes the CSA parameters to
move further away from zero, i.e., the experimental reference values, because the CASTEP
calculations overestimate the parameter, which was attributed to the pseudopotential (see
below).
Solid-state and molecular models. Relative to the other contributions, the effect
of the crystal lattice is most notable in the isotropic 195Pt CS, as shown in Figure 5. The
crystal lattice modification of the CS in compound III is negligible, whereas in II it is the
dominant contribution of the three. The CSA and asymmetry parameters are relatively much
less affected in every complex, although the absolute contributions to δaniso of compounds I
and III A are still sizable (ca. 100 ppm), justifying the crystal lattice correction. Careful
consideration is required when estimating this contribution, because the CASTEP calculations
for a molecule in vacuum using the standard pseudopotentials yields radically different CSA
(ca. 3000 ppm in CASTEP vs. 2000 ppm in ADF, see Table S33 in SI) and asymmetry
parameters (0.00–0.09 vs 0.15–0.28) than a molecular calculation with ADF at the SR level
using the PBE functional. A similar situation is also present in Ref. 66, where a plane
wave calculation (CASTEP) yields significantly larger span for the 195Pt NMR spectrum of
cisplatin, as compared to a molecular calculation (ADF). The span is generally not equivalent
to the CSA parameter calculated here, but for spectra where δxx is close to δyy, while δzz
is far removed from both, they are somewhat similar quantities. As the authors have not
specified whether spin–orbit coupling is accounted for in the molecular calculation or not,
and since only pure DFT functionals have been used,66 it is difficult to assess the reason
behind the discrepancy between the experimental and calculated results.
To avoid such very large errors as found here, if the crystal lattice correction is taken as
18
ACS Paragon Plus Environment

Page 19 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
the difference between a solid-state and a molecular calculation, both need to be performed
within CASTEP to make use of error cancellation. The problem likely stems from the
core–valence partitioning of the heavy element pseudopotential: a large core is preferred due
to reduced computational cost, but may lead to an unrealistic polarization of the sub-valence
region and hence a discrepancy in comparison with an all-electron calculation. The disparity
is more likely to be visible in the anisotropic CS parameters, which do not benefit from
error cancellation similarly as the isotropic chemical shift. A rigorous approach to correct
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the issue would be the development of an optimized pseudopotential with a smaller core, as
74
demonstrated by Truflandier et al. for the CSA parameter of vanadium in VOCl3
,
but it is
outside the scope of the present study.
Hybrid DFT functional. The inclusion of exact exchange into the DFT functional yields
positive contributions to the CSA parameter, and negative ones to the CS and asymmetry
parameter in all complexes, as shown in Figure 5. The largest effect is the increase of 195Pt
δaniso in all compounds by ca. 700 ppm (ca. 30% of the total value), which is over 1.5 times as
large as the absolute value of the next largest contribution, the SO coupling. In compounds IV
and V, the contribution to CS (relative to compound I) is also larger than those due to
either the crystal lattice or the SO-coupling. The absolute contribution to
η is small, up to
−0.
1, but the relative contribution (17–28%) is still quite large due to a small value of the
parameter.
Spin–orbit correction. SO-coupling decreases (increases) the 195Pt CSA (asymmetry)
parameters in all compounds, acting in the opposite direction as compared to the use of
hybrid functional and, in most cases, the crystal lattice. The effect is ca.
−20% for the CSA
parameter in all complexes, and 120–140% (the result being ca. 0.25 in all complexes) for the
asymmetry. Hence, the incorporation of SO coupling is essential for obtaining reasonable
195Pt CSA and asymmetry parameters in compounds I–V.
195Pt–31P spin–spin J-coupling
195Pt
–
31P), calculated at the molecular SO-ZORA/PBE0
2
J
The spin–spin coupling constants
(
level of theory, are listed in Table 3. They are invariably determined to have the negative
sign, with the absolute values in the range 401–428 Hz, close to the experimental values in
the range 431–456 Hz. Relative to this, the differences between the complexes are small,
down to only few Hz.
Geometry optimization. Geometry optimization of hydrogen had only a negligible
effect of about +3 Hz (compound V) or less to the spin–spin coupling constants (see Table S35
in SI). Full optimization of ions and cell parameters caused slightly larger, but still small
changes to the parameter, circa +15 Hz (compound IV A) or less.
Hybrid DFT functional. The use of a hybrid functional was found to be vital for
obtaining values of
columns PBE and PBE0 in Table S35 in SI). Pure DFT (PBE) coupling constants were only
of the order of 250 Hz at both SR and SO ZORA level of theory, i.e., the inclusion of 25%
J that are even remotely close to experimental observations (see the
−
of exact exchange in PBE0 resulted in a 63–78% increase of the magnitude of the coupling,
depending on the complex.
Relativistic effects and spin–orbit correction. The SO coupling yields negligible
corrections (only few Hz) to the spin–spin coupling constants (see Table S35 in SI), as
19
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 30
1
2
Table 3: Calculated and experimental J(195Pt–31P) spin–spin J-coupling con-
stants in compounds I–V. Values in Hz.
3
4
5
2
6
7
8
9
Compounda Calc.b
Exp.c
444.4
I
−428
II
−401 431.1 0.7
−410 443.2 0.5
−416
−407
−403
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
III A
III B
IV A
IV B
V
ca. 456
−411 433.2 0.5
aPt-dtp-R, R: ethyl (I), iso-propyl (II), iso-butyl (III), sec-butyl (IV), and cyclo-hexyl (V).
A and B refer to calculations on two different conformations of the same complex.
bADF SO-ZORA calculation with the PBE0 hybrid functional, using optimized geometry
(ions and cell parameters, CASTEP).
cPreviously reported values. See Ref. 24 (I and IV), 22 (II and V), and 21 and 23 (III).
compared to the other contributions. Scalar relativistic effects are much larger, of the order
of 100 Hz, compared to the NR values.
Orientation and shape of the CS tensors
In addition to the CS tensor parameters, also the principal axis directions of the tensors
are obtained from calculations. A graphical example, in arbitrary scale, is represented in
Figure 7 for complex V. For complexes I–V, the full shielding tensors are given in SI (see
Tables S36–S42), from which the principal axes are obtained by diagonalizing the symmetric
part of the tensor. In practice, the 195Pt shielding tensor resembles an axial disc in the nearly
planar central region of the molecule formed by the sulfur atoms, with the least-shielded
component roughly perpendicular to that plane. The tensor retains its general shape and
orientation in the present complexes; the least-shielded direction remains practically invariant
in the molecular frame of reference, and the two most shielded components remain nearly
equal in size to each other.
On the other hand, the 31P tensor is rhombic, with the most-shielded component nearly
perpendicular to the two P–O bonds in most compounds, with the two other principal
components lying roughly along one of the P–O bonds, the P–Pt direction, or between these,
depending on the complex and the level of calculation. A slight exception to this is the X-ray
structure of compound IV (both conformations A and B), where the most-shielded direction
of the phosphorus shielding tensor is considerably tilted towards the far end of a C4H9 ligand.
However, after the geometry optimization of all ions, this tilt angle is noticeably reduced,
and the tensor orientation is brought closer to that in the other compounds. By combining
that with the information on the CSA parameter before and after the ion optimization, it
allows one to estimate that an optimization is required to obtain agreement with experiment
for the shielding tensor.
The SO and hybrid functional corrections introduce only small changes to the orientation
20
ACS Paragon Plus Environment

Page 21 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 30
1
2
3
4
for heavy elements using a pseudopotential approach.
The platinum CSA and asymmetry parameters were found to be highly similar in all the
systems, with computations certifying the experimental observation of an axial tensor shape.
The accuracy of the computational model, combined with its high sensitivity to the molecular
structure, means that it can be used to detect potential structural distortions caused by, e.g.,
ligands other than the ones studied here. The directional data indicate that the smallest
195Pt shielding component is nearly perpendicular to the central plane consisting of sulfur
atoms in all the complexes. The computational phosphorus shielding tensors were found to be
in a good agreement with experimental observations, replicating the nearly rhombic character
of the tensors as well as the CSA and asymmetry parameters. Qualitative agreement was
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
found between the calculated and experimental spin–spin
J-coupling constants, with an
unanimously negative sign of the parameter visible in all calculations.
Scalar relativistic effects were estimated using the X-ray structures, and were found to be
important in all the systems, yielding (depending on the complex) absolute changes of 2–5%
(14–34%) to the CSA parameter of 195Pt 31P), and of 7–50% (46–600%) to the asymmetry
(
parameter, correspondingly. The inclusion of SO-coupling and the use of a hybrid functional
were also found important, especially for the 195Pt CS parameters. The crystal lattice
effect was relatively more important for the phosphorus-31 CS parameters than those of the
platinum-195, but the absolute contribution was non-negligible for both nuclei. The hybrid
functional was also vital for the spin–spin coupling constant
J, accounting for circa 63–78%
of the total value of the parameter, more than the SR effects.
Based on the results, the shape and orientation of the 195Pt shielding tensor is not
much affected by the ligands in the studied complexes. In contrast to that, the phosphorus
shielding tensor displays a subtle dependence on the ligands, allowing an unambiguous
assignment of the complexes from one another and, through combining the computational
and experimental data, confirming the experimentally determined structure. The results
clearly display the necessity of considering several significant factors in the NMR calculations
of heavy metal-containing molecular crystals, including crystal lattice and relativistic effects,
as well as the exact exchange component in the DFT functional. Lacking all the necessary
functionality in any single program, a tensor component-level correction method was applied
to combine the different contributions, resulting in 195Pt and 31P shielding tensors that
agree qualitatively with experimental results. This approach enables detailed analysis of
heavy-element compounds by solid-state NMR, helping to increase our understanding of such
systems.
Acknowledgement
JR thanks the Finnish Cultural Foundation for funding this research, with further support
received from the National Doctoral Programme in Nanoscience (NGS-NANO), the Tauno
Tönning Foundation, and University of Oulu. PL was supported by the Academy of Finland
projects number 125316, 218191, 255641, and 285666, and the work of JV by the directed
programme in Computational Science of the Academy of Finland, project number 258565.
JVH thanks the EPSRC and the University of Warwick for partial funding of the solid-state
NMR infrastructure at Warwick, and acknowledges additional support for this infrastructure
22
ACS Paragon Plus Environment

Page 23 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
obtained through Birmingham Science City: Innovative Uses for Advanced Materials in
the Modern World (West Midlands Centre for Advanced Materials Projects 1 and 2), with
support from Advantage West Midlands (AWM) and partial funding by the European Regional
Development Fund (ERDF). CSC – IT Center for Science and the Finnish Grid Initiative
project provided the computational resources.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information Available
Cartesian coordinates for X-ray and geometry-optimized stuctures, structure files for fully
optimized geometries, full shielding tensors, calculated shielding tensor parameters, and
spin–spin coupling constants. This material is available free of charge via the Internet at
http://pubs.acs.org/.
References
(1) Hrobárik, P.; Hrobárikova, V.; Greif, A. H.; Kaupp, M. Giant Spin-Orbit Effects on
NMR Shifts in Diamagnetic Actinide Complexes: Guiding the Search of Uranium(VI)
Hydride Complexes in the Correct Spectral Range. Angew. Chem. Int. Ed. 2012, 51,
10884–10888.
(2) Vícha, J.; Demo, G.; Marek, R. Platinum-Modified Adenines: Unprecedented Proto-
nation Behavior Revealed by NMR Spectroscopy and Relativistic Density-Functional
Theory Calculations. Inorg. Chem. 2012, 51, 1371–1379.
(3) Vícha, J.; Patzschke, M.; Marek, R. A Relativistic DFT Methodology for Calculating the
Structures and NMR Chemical Shifts of Octahedral Platinum and Iridium Complexes.
Phys. Chem. Chem. Phys. 2013, 15, 7740–7754.
(4) Vícha, J.; Straka, M.; Munzarová, M. L.; Marek, R. Mechanism of Spin–Orbit Effects
on the Ligand NMR Chemical Shift in Transition-Metal Complexes: Linking NMR to
EPR. J. Chem. Theory Comput. 2014, 10, 1489–1499.
(5) Vícha, J.; Foroutan-Nejad, C.; Pawlak, T.; Munzarová, M. L.; Straka, M.; Marek, R.
Understanding the Electronic Factors Responsible for Ligand Spin–Orbit NMR Shielding
in Transition-Metal Complexes. J. Chem. Theory Comput. 2015, 11, 1509–1517.
(6) Hrobárik, P.; Hrobárikova, V.; Meier, F.; Repiský, M.; Komorovský, S.; Kaupp, M. Rela-
1
tivistic Four-Component DFT Calculations of H NMR Chemical Shifts in Transition-
Metal Hydride Complexes: Unusual High-Field Shifts Beyond the Buckingham–Stephens
Model. J. Phys. Chem. A 2011, 115, 5654–5659.
(7) Vícha, J.; Novotný, J.; Straka, M.; Repisky, M.; Ruud, K.; Komorovsky, S.; Marek, R.
Structure, Solvent, and Relativistic Effects on the NMR Chemical Shifts in Square-
Planar Transition-Metal Complexes: Assessment of DFT Approaches. Phys. Chem.
Chem. Phys. 2015, 17, 24944–24955.
23
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 30
1
2
3
4
5
6
7
8
9
13
29
(8) Vícha, J.; Marek, R.; Straka, M. High-Frequency C and Si NMR Chemical Shifts
in Diamagnetic Low-Valence Compounds of TlI and PbII: Decisive Role of Relativistic
Effects. Inorg. Chem. 2016, 55, 1770–1781.
129
(9) Lantto, P.; Standara, S.; Riedel, S.; Vaara, J.; Straka, M. Exploring New Xe Chemical
Shift Ranges In HXeY Compounds: Hydrogen More Relativistic Than Xenon. Phys.
Chem. Chem. Phys. 2012, 14, 10944–10952.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
129
(10) Standara, S.; Kulhánek, P.; Marek, R.; Straka, M. Xe NMR Chemical Shift in Xe@C60
Calculated at Experimental Conditions: Essential Role of the Relativity, Dynamics, and
Explicit Solvent. J. Comput. Chem. 2013, 34, 1890–1898.
(11) Vaara, J.; Hanni, M.; Jokisaari, J. Nuclear Spin-Spin Coupling in a Van Der Waals-
Bonded System: Xenon Dimer. J. Chem. Phys. 2013, 138, 104313.
(12) Greif, A. H.; Hrobárik, P.; Hrobáriková, V.; Arbuznikov, A. V.; Autschbach, J.; Kaupp, M.
1
A Relativistic Quantum-Chemical Analysis of the Trans Influence on H NMR Hydride
Shifts in Square-Planar Platinum(II) Complexes. Inorg. Chem. 2015, 54, 7199–7208.
(13) Rosenberg, B.; Vancamp, L. Platinum Compounds: a New Class of Potent Antitumour
Agents. Nature 1969, 222, 385–386.
(14) Rosenberg, B. Anticancer Activity of Cis-Dichlorodiammineplatinum(II) and Some
Relevant Chemistry. Cancer Treat. Rep. 1979, 63, 1433–1438.
(15) Harrap, K. R. Platinum Co-ordination Complexes in Cancer Chemotherapy. Platinum
Metals Rev. 1984, 28, 14–19.
(16) Bleiberg, H. Oxaliplatin (L-OHP): a New Reality in Colorectal Cancer. British J. Cancer
1998, 77, 1–3.
(17) Burger, H.; Loos, W. J.; Eechoute, K.; Verweij, J.; Mathijssen, R. H.; Wiemer, E. A.
Drug Transporters of Platinum-based Anticancer Agents and Their Clinical Significance.
Drug Resistance Updates 2011, 14, 22–34.
(18) Gianini, M.; Caseri, W. R.; Gramlich, V.; Suter, U. W. Synthesis and Characterization
of Liquid Platinum Compounds. Inorg. Chim. Acta 2000, 299, 199–208.
′
(19) Tkachev, V. V.; Atovmyan, L. O. Crystal and Molecular Structure of Bis (0
Dithiophosphato) Platinum (II). Koord. Khim. 1982, 8, 215–217.
, 0 -diisopropyl
′
(20) Kuan, F. S.; Tiekink, E. R. T. Bis(O, O -dicyclohexyl Dithiophosphato-
′
κ2S, S )platinum(II). Acta Crystallogr. Sec. E: Struct. Rep. Online 2004, 60, m1619–
m1620.
(21) Ivanov, A. V.; Rodina, T. A.; Ivanov, M. A.; Gerasimenko, A. V.; Antzutkin, O. N.
′
Crystalline Platinum(II) O, O -di-iso-butyldithiophosphate as a Model of the Structural
13
31
State of Platinum in Cooperite: Synthesis, Structure, and Multinuclear C, P, and
195
Pt MAS NMR. Dokl. Phys. Chem. 2008, 423, 311–316.
24
ACS Paragon Plus Environment

Page 25 of 30
The Journal of Physical Chemistry
1
2
3
4
5
(22) Ivanov, A. V.; Lutsenko, I. A.; Ivanov, M. A.; Gerasimenko, A. V.; Antzutkin, O. N.
13
31
195
Synthesis, Structural and Multinuclear Natural Abundance ( C, P, Pt) CP/MAS
′
6
7
8
9
NMR Studies of Crystalline O, O -dialkyldithiophosphate Platinum(II) Complexes. Russ.
J. Coord. Chem. 2008, 34, 584–593.
′
(23) Rodina, T. A.; Lutsenko, I. A.; Gerasimenko, A. V.; Ivanov, A. V. Platinum(II) O, O -
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
diisobutyl Dithiophosphate: Synthesis, Structure, Thermal Properties, and Solid-State
C, P, and Pt CP/MAS NMR Spectra. Model of the Structural State of Platinum
in Cooperite. Russ. J. Coord. Chem. 2009, 35, 534–540.
13
31
195
(24) Rodina, T. A.; Ivanov, A. V.; Gerasimenko, A. V.; Lutsenko, I. A.; Ivanov, M. A.;
′
′
Hanna, J. V.; Antzutkin, O. N.; Sergienko, V. I. Crystalline O, O -di-sec-butyl and O, O
-
13
31
diethyl Dithiophosphate Platinum(II) Complexes: Synthesis, C and P CP/MAS
NMR, Single Crystal X-ray Diffraction Studies and Thermal Behaviour. Polyhedron
2011, 30, 2210–2217.
(25) Ivanov, A. V.; Palazhchenko, V. I.; Strikha, V. E.; Antzutkin, O. N.; Forsling, W.
195
Structural Features and NMR Pt Spectroscopy of Cooperite Monocrystals (Natural
PtS). Dokl. Earth. Sci. 2006, 410, 1141–1144.
(26) Rozhdestvina, V. I.; Ivanov, A. V.; Zaremba, M. A.; Antzutkin, O. N.; Forsling, W.
Single-Crystalline Sooperite (PtS): Crystal-Chemical characterization, ESR spectroscopy,
195
and Pt NMR spectroscopy. Crystallogr. Reports 2008, 53, 391–397.
(27) Rees, G. J.; Orr, S. T.; Barrett, L. O.; Fisher, J. M.; Houghton, J.; Spikes, G. H.;
Theobald, B. R.; Thompsett, D.; Smith, M. E.; Hanna, J. V. Characterisation of
195
Platinum-Based Fuel Cell Catalyst Materials Using Pt Wideline Solid State NMR.
Phys. Chem. Chem. Phys. 2013, 15, 17195–17207.
(28) Sindorf, D. W.; Bartuska, V. J. Wide-line NMR Spectroscopy in Solids Using Variable
Frequency Pulses. J. Magn. Reson. 1989, 85, 581–585.
(29) Fu, L.-j.; Vaara, J. Nuclear-Spin-Induced Cotton–Mouton Effect in a Strong External
Magnetic Field. ChemPhysChem 2014, 15, 2337–2350.
(30) Fedorov, S. V.; Rusakov, Y. Y.; Krivdin, L. B. Towards the Versatile DFT and MP2
31
Computational Schemes for P NMR Chemical Shifts Taking Into Account Relativistic
Corrections. Magn. Reson. Chem. 2014, 52, 699–710.
(31) Orr, S. T. Multinuclear Solid-State NMR of Fuel Cell Materials. Ph.D. Dissertation;
University of Warwick, UK, 2010.
(32) Massiot, D.; Fayon, F.; Capron, M.; King, I.; Le Calvé, S.; Alonso, B.; Durand, J.-O.;
Bujoli, B.; Gan, Z.; Hoatson, G. Modelling One- and Two-Dimensional Solid-State NMR
Spectra. Magn. Reson. Chem. 2002, 40, 70–76.
(33) Turner, G. L.; Smith, K. A.; Kirkpatrick, R. J.; Oldfieldt, E. Structure and Cation
Effects on Phosphorus-31 NMR Chemical Shifts and Chemical-Shift Anisotropies of
Orthophosphates. J. Magn. Reson. 1986, 70, 408–415.
25
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 30
1
2
3
4
5
6
13
15
(34) Antzutkin, O. N.; Lee, Y.; Levitt, M. C and N—Chemical Shift Anisotropy of
Ampicillin and Penicillin-V Studied by 2D-PASS and CP/MAS NMR. J. Magn. Reson.
1998, 135, 144–155.
7
8
9
(35) Mathematica, version 4.1.2; software for technical computation; Wolfram Research:
Champaign, IL, 1999.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(36) Wolfram, S. The Mathematica Book; Cambridge University Press, Cambridge, 1999.
(37) Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert, M. J.; Refson, K.;
Payne, M. C. First Principles Methods Using CASTEP. Z. Kristallogr. 2005, 220,
567–570.
(38) Pickard, C. J.; Mauri, F. All-Electron Magnetic Response With Pseudopotentials: NMR
Chemical Shifts. Phys. Rev. B 2001, 63, 245101.
(39) Yates, J. R.; Pickard, C. J.; Payne, M. C.; Mauri, F. Relativistic Nuclear Magnetic
Resonance Chemical Shifts of Heavy Nuclei With Pseudopotentials and the Zeroth-Order
Regular Approximation. J. Chem. Phys. 2003, 118, 5746–5753.
(40) Yates, J. R.; Pickard, C. J.; Mauri, F. Calculation of NMR Chemical Shifts for Extended
Systems Using Ultrasoft Pseudopotentials. Phys. Rev. B 2007, 76, 024401.
(41) Schreckenbach, G.; Ziegler, T. Calculation of NMR Shielding Tensors Using Gauge-
Including Atomic Orbitals and Modern Density Functional Theory. J. Phys. Chem.
1995, 99, 606–611.
(42) ADF, version 2013.01; DFT software for modeling chemistry; Software for Chemistry
and Materials (SCM): Amsterdam, Netherlands, 2013.
(43) ADF, version 2014.01; DFT software for modeling chemistry; Software for Chemistry
and Materials (SCM): Amsterdam, Netherlands, 2014.
(44) Krykunov, M.; Ziegler, T.; van Lenthe, E. Hybrid Density Functional Calculations of
Nuclear Magnetic Shieldings Using Slater-Type Orbitals and the Zeroth-Order Regular
Approximation. Int. J. Quantum Chem. 2009, 109, 1676–1683.
(45) Wolff, S. K.; Ziegler, T. Calculation of DFT-GIAO NMR Shifts With the Inclusion of
Spin-Orbit Coupling. J. Chem. Phys. 1998, 109, 895–905.
(46) Wolff, S. K.; Ziegler, T.; van Lenthe, E.; Baerends, E. J. Density Functional Calculations
of Nuclear Magnetic Shieldings Using the Zeroth-Order Regular Approximation (ZORA)
for Relativistic Effects: ZORA Nuclear Magnetic Resonance. J. Chem. Phys. 1999, 110,
7689–7698.
(47) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. Relativistic Regular Two-component
Hamiltonians. J. Chem. Phys. 1993, 99, 4597–4610.
(48) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. Relativistic Total Energy Using Regular
Approximations. J. Chem. Phys. 1994, 101, 9783–9792.
26
ACS Paragon Plus Environment

Page 27 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
(49) Bonhomme, C.; Gervais, C.; Babonneau, F.; Coelho, C.; Pourpoint, F.; Azaïs, T.;
Ashbrook, S. E.; Griffin, J. M.; Yates, J. R.; Mauri, F. et al. First-Principles Calculation
of NMR Parameters Using the Gauge Including Projector Augmented Wave Method: A
Chemist’s Point of View. Chem. Rev. 2012, 112, 5733–5779.
9
(50) Repisky, M.; Komorovsky, S.; Malkin, V. G.; Malkina, O. L.; Kaupp, M.; Ruud
K., With contributions from Bast, R.; Ekström, U.; Knecht S.; Malkin Ondik, I.;
Malkin, E. ReSpect, version 3.4.1 (beta); Relativistic Spectroscopy DFT program (see
http://rel-qchem.sav.sk); Bratislava, Slovakia, 2014.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(51) Komorovsky, S.; Repisky, M.; Malkina, O. L.; Malkin, V. G.; Malkin Ondík, I.;
Kaupp, M. A Fully Relativistic Method for Calculation of Nuclear Magnetic Shielding
Tensors With a Restricted Magnetically Balanced Basis in the Framework of the Matrix
Dirac–Kohn–Sham Equation. J. Chem. Phys. 2008, 128, 104101.
(52) Komorovsky, S.; Repisky, M.; Malkina, O. L.; Malkin, V. G. Fully Relativistic Calcu-
lations of NMR Shielding Tensors Using Restricted Magnetically Balanced Basis and
Gauge Including Atomic Orbitals. J. Chem. Phys. 2010, 132, 154101.
(53) van Lenthe, E.; Baerends, E. J. Optimized Slater-Type Basis Sets for the Elements
1–118. J. Comput. Chem. 2003, 24, 1142–1156.
(54) Dyall, K. G. Relativistic Double-Zeta, Triple-Zeta, and Quadruple-Zeta Basis Sets for
the 5d Elements Hf–Hg. Theor. Chem. Acc. 2004, 112, 403–409, basis sets are available
from the Dirac web site, http://dirac.chem.sdu.dk (accessed Sep 29, 2016).
(55) Dyall, K. G.; Gomes, A. S. P. Revised Relativistic Basis Sets for the 5d Elements Hf–Hg.
Theor. Chem. Acc. 2010, 125, 97–100, basis sets are available from the Dirac web site,
http://dirac.chem.sdu.dk (accessed Sep 29, 2016).
(56) Dyall, K. G. Core Correlating Basis Functions for Elements 31–118. Theor. Chem. Acc.
2012, 131, 1217, basis sets are available from the Dirac web site, http://dirac.chem.
sdu.dk (accessed Sep 29, 2016).
(57) K. G. Dyall, unpublished. Basis sets are available from the Dirac web site, http:
//dirac.chem.sdu.dk (accessed Sep 29, 2016).
(58) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made
Simple. Phys. Rev. Lett. 1996, 77, 3865–3868.
(59) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made
Simple [Phys. Rev. Lett. 77, 3865 (1996)]. Phys. Rev. Lett. 1997, 78, 1396–1396.
(60) Ernzerhof, M.; Scuseria, G. E. Assessment of the Perdew–Burke–Ernzerhof Exchange-
Correlation Functional. J. Chem. Phys. 1999, 110, 5029–5036.
(61) Adamo, C.; Barone, V. Toward Reliable Density Functional Methods Without Adjustable
Parameters: The PBE0 Model. J. Chem. Phys. 1999, 110, 6158–6170.
27
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 30
1
2
3
4
5
(62) Vanderbilt, D. Soft Self-Consistent Pseudopotentials in a Generalized Eigenvalue For-
malism. Phys. Rev. B 1990, 41, 7892–7895.
6
7
8
9
129
(63) Straka, M.; Lantto, P.; Vaara, J. Toward Calculations of the Xe Chemical Shift in
Xe@C60 at Experimental Conditions: Relativity, Correlation, and Dynamics. J. Phys.
Chem. A 2008, 112, 2658–2668.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(64) Roukala, J.; Maldonado, A. F.; Vaara, J.; Aucar, G. A.; Lantto, P. Relativistic Effects on
Group-12 Metal Nuclear Shieldings. Phys. Chem. Chem. Phys. 2011, 13, 21016–21025.
(65) Mastrorilli, P.; Todisco, S.; Bagno, A.; Gallo, V.; Latronico, M.; Fortuño, C.; Gudat, D.
Multinuclear Solid-State NMR and DFT Studies on Phosphanido-Bridged Diplatinum
Complexes. Inorg. Chem. 2015, 54, 5855–5863.
(66) Lucier, B. E. G.; Reidel, A. R.; Schurko, R. W. Multinuclear Solid-State NMR of
Square-Planar Platinum Complexes — Cisplatin and Related Systems. Can. J. Chem.
2011, 89, 919–937.
(67) Pecul, M.; Urbańczyk, M.; Wodyński, A.; Jaszuński, M. DFT Calculations of 31P
Spin–spin Coupling Constants and Chemical Shift in Dioxaphosphorinanes. Magn.
Reson. Chem. 2011, 49, 399–404.
(68) Tkatchenko, A.; Scheffler, M. Accurate Molecular Van Der Waals Interactions from
Ground-State Electron Density and Free-Atom Reference Data. Phys. Rev. Lett. 2009
102, 073005.
,
(69) Larsson, A.-C.; Ivanov, A. V.; Forsling, W.; Antzutkin, O. N.; Abraham, A. E.;
31
de Dios, A. C. Correlations Between P Chemical Shift Anisotropy and Molecular Struc-
′
ture in Polycrystalline O, O -dialkyldithiophosphate Zinc(II) and Nickel(II) Complexes:
31
P CP/MAS NMR and Ab Initio Quantum Mechanical Calculation Studies. J. Am.
Chem. Soc. 2005, 127, 2218–2230.
31
(70) Larsson, A.-C.; Ivanov, A. V.; Pike, K. J.; Forsling, W.; Antzutkin, O. N. Solid State
P
207
′
and Pb MAS NMR Studies on Polycrystalline O, O -dialkyldithiophosphate Lead(II)
Complexes. J. Magn. Reson. 2005, 177, 56–66.
(71) Pyykkö, P.; Görling, A.; Rösch, N. A Transparent Interpretation of the Relativistic
Contribution to the N.M.R. ’Heavy Atom Chemical Shift’. Mol. Phys. 1987, 61, 195–205.
(72) Vaara, J.; Ruud, K.; Vahtras, O.; Ågren, H.; Jokisaari, J. Quadratic Response Calcula-
tions of the Electronic Spin-Orbit Contribution to Nuclear Shielding Tensors. J. Chem.
Phys. 1998, 109, 1212–1222.
(73) Kaupp, M.; Malkina, O. L.; Malkin, V. G.; Pyykkö, P. How Do Spin–Orbit-Induced
Heavy-Atom Effects on NMR Chemical Shifts Function? Validation of a Simple Analogy
to Spin–Spin Coupling by Density Functional Theory (DFT) Calculations on Some Iodo
Compounds. Chem. Eur. J. 1998, 4, 118–126.
28
ACS Paragon Plus Environment

Page 29 of 30
The Journal of Physical Chemistry
1
2
3
4
5
6
(74) Truflandier, L.; Paris, M.; Boucher, F. Density Functional Theory Investigation of 3d
Transition Metal NMR Shielding Tensors in Diamagnetic Systems Using the Gauge-
Including Projector Augmented-Wave Method. Phys. Rev. B 2007, 76, 035102.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment