On-surface Reaction between Tetracarbonitrile Functionalized
Molecules and Copper Atoms
Elena Nardi*,†, Long Chen‡,§, Sylvain Clair†, Mathieu Koudia†, Luca Giovanelli†, Xinliang Feng‡,
Klaus Müllen‡, Mathieu Abel*,‡
†
‡
§
Aix Marseille Université, CNRS, IM2NP UMR 7334, 13397, Marseille, France
Max-Planck Institute for Polymer Research, Ackermannweg 10, 55128, Mainz, Germany
Present Adress : Department of Chemistry, School of Science, Tianjin University, 300072, Tianjin,
People’s Republic of China
* [email protected] [email protected]
Supplementary Information
Table of Contents
S-1. Synthesis
1
S-2. DFT Simulations
2
S-3. Metal Organic Coordination Network 2
3
S-4. CuPyc Polymeric Networks
4
S-5. X-Ray Photoemission Spectroscopy
5
S-1. PPCN Synthesis
a. Materials and Methods
Pyrene, RuCl3-3H2O, sodium periodate, diaminomaleonitrile were purchased from Aldrich. 1H NMR
spectra were recorded in deuterated solvents on a Bruker DPX 250. MALDI-TOF mass spectra were
recorded on a Bruker Reflex II-TOF Spectrometer using a 337 nm nitrogen laser with TCNQ as
matrix.
b. Synthesis
Figure S1. Synthesis of the pyrazine-pyrene based tetracyanides 1: (i) RuCl3-3H2O, NaIO4, CH2Cl2/CH3CN;
(ii) HOAc, EtOH, reflux.
Pyrene-4,5,9,10-tetraone (3):
To a solution of pyrene 2 (2.02 g, 10 mmol) in CH2Cl2 (40 mL) and CH3CN (40 mL) were added
NaIO4 (17.1 g, 80 mmol), H2O (50 mL), and RuCl3-3H2O (0.31 g, 1.2 mmol). The brown suspension
was slightly heated up to 40 oC and stirred overnight. The reaction mixture was poured into 200 mL
of water, and the solid was removed by filtration washed by water. The organic phase was separated,
and the aqueous phase was extracted with CH2Cl2 three times. All the organic phase was combined
and washed with brine, dried over MgSO4. The solvent was removed under reduced pressure and the
residue was purified by column chromatography (hexane/EtOAc = 3/1) to give compound 3 as orange
crystals (0.91 g) in a yield of 35%. MALDI-TOF MS: m/z = 262.1 (100%, M+), 1H-NMR (250 MHz,
DMSO-d6): δ = 8.32 (d, 4H), 7.71 (t, 2H); 13C-NMR (solubility too low).
Pyrazino[2',3':9,10]phenanthro[4,5-fgh]quinoxaline-5,6,12,13-tetracarbonitrile (1):
To a solution of diaminomaleonitrile (250 mg, 2.3 mmol) in ethanol (40 ml) was added pyrene4,5,9,10-tetraone 3 (200 mg, 0.76 mmol) under Argon stream. Acetic acid (3.0 ml) was added, and the
reaction mixture was refluxed for 24 hours. The resulted brown reaction mixture was cooled to r.t.
The precipitate was collected by filtration, washed with EtOH, HOAc to afford a black solid. The
crude product was purified by boiling with 30% HNO3 and further Soxhlet extraction with CH3CN
and THF to afford a yellow solid in a yield of 71%. MALDI-TOF MS: m/z = 406.3 (100%, M+), 1H
NMR (d6-DMSO): δ (ppm) 9.40 (d, J = 7.5 Hz, 4H), 8.36 (t, J = 7.5 Hz, 2H); 13C NMR (d6-DMSO): δ
(ppm) 141.51, 131.72, 129.72, 129.55, 126.60, 114.54. Anal. Calc. for C24H6N8: C, 70.94; H, 1.49; N,
27.58. Found: C, 70.88; H, 1.53; N, 27.53%.
1
S-2. DFT Simulations
Electronic structure calculations were performed within the framework of density functional theory
using SIESTA Spanish Initiative for Electronic Simulations with Thousands of Atoms [1] and [2]. The
exchange-correlation energy is treated within the generalized gradient approximation GGA using a
parameterization proposed by Perdew, Burke and Ernzerhof [3]. The wave function of the valence
electrons is expanded in a localized basis set consisting of finite-range pseudoatomic orbitals [4]: a
double-zeta plus polarization basis set was used for each atom. The core electrons are treated within
the frozen core approximation with norm-conserving Troullier-Martins pseudopotentials [5]. The unit
cell used for k-sampling grid is a (2 x 2 x1) unit cell. The molecular networks were relaxed until the
forces acting on each atom were smaller than 0.04 eV Å-1 Both pseudopotential and basis set were
previously tested by comparing calculated molecular geometry to x-ray geometry characterization of
phthalocyanine compounds such as CuPc.
[1] Sánchez-Portal, D.; Ordejón, P.; Artacho, E.; Soler, J. M. Density-functional method for very large systems with
LCAO basis set. Int. J. Quantum Chem. 1997, 65, 453-461.
[2] Soler, J. M.; Artacho, E.; Gale, J. D.; García, A.; Junquera, J.; Ordejón, P.; Sánchez-Portal, D. The SIESTA method for
ab initio order-N materials simulation. J. Phys. Condens. Matter 2002, 14, 2745-2779.
[3] Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77,
3865-3868.
[4] Junquera J.; Paz, O.; Sánchez-Portal, D.; Artacho, E. Numerical atomic orbitals for linear-scaling calculations. Phys.
Rev. B 2001, 64, 235111.
[5] Troullier, N.; Martins, J. L. Efficient pseudopotentials for plane-wave calculations. Phys. Rev. B 1991, 43, 1993-2006
(1991).
2
S-3. Metal Organic Coordination Network 2
The MOCN2 network (Fig. S2a), is characterized by parallel rows, linked by two PPCN molecules.
From the analysis of the STM images we propose a model in which every copper atom is coordinated
to three PPCNs: two PPCN of the rows and one spacer PPCN perpendicular to the row. The analysis
of the STM images yields for the rhombic unit cell the parameters: a1=1.73±0.04 nm, a2=3.27 ±0.07
nm, and θ~82°. According to this model, the Cu:PPCN ratio is 2:3.
Figure S2. Metal Organic Coordination Network 2. (a) STM images of the MOCN2 network. Inset: zoom
(1.2x1.2 nm2). (b) Tentative model of the MOCN2.
3
S-4. CuPyc Polymeric Networks
Figure S3. CuPyc Polymeric Networks. (a),(b) representative STM images of CuPyc polymeric chains
obtained by annealing at 540 K. (c),(d) representative STM images of CuPyc polymeric grid obtained by
annealing at 675 K.
4
S-5. X-RayPhotoemission spectroscopy
For the C1s and Cu2p3/2 spectra, an integral background was subtracted; for the N1s spectra a cubic
background was subtracted. The C1s and N1s core level spectra were modelled with Gaussian line
shapes. The Cu2p3/2 spectra of copper on Au(111), of MOCN1, and of the CuPyc coordination
network were modelled with a Voigt lineshape for metallic copper, and a Gaussian lineshape for the
Cu(II). For the polymeric grid a Gaussian lineshape was a better fit also for metallic copper, probably
due to the emerging of a slightly shifted component of copper bonded with polymeric structures at the
border of the domains. All the spectra were normalized with the appropriate analyser transmission
factor [1].
[1] Ruffieux, P.; Schwaller, P.; Gröning, O.; Schlapbach, L.; Gröning, P.; Herd, Q. C.; Funneman, D.; Westermann, J.
Experimental determination of the transmission factor for the Omicron EA125 electron analyzer. Rev. Sci. Instrum. 2000,
71, 3634-3639.
5
a. N1s Photoemission Spectra
The N1s spectra of the networks are shown in Figure S4. The N1s spectrum of the self-assembled
PPCN network has been decomposed with a Gaussian peak centered at about 399.4 eV and a satellite
feature. The N1s spectrum of the MOCN1 network doesn’t present any changes, except for a slight
rigid shift towards higher binding energies. The spectrum of the CuPyc coordination network and of
the polymeric networks can still be fitted with a single peak and its satellite, but the width of the main
peak is broadened. According to previous reports, the binding energy of the pyrrolic nitrogens (the
inner ones bonded with two carbon atoms and the central Cu metal) lies at slightly different values
(0.4 eV) than the binding energy of the aza-bridging nitrogens (the outer nitrogens bonded with two
carbon atoms) [1,2]. The broadening of the N1s peak can then be explained by the emerging of the
new component N-Cu.
N1s
a)
Intensity (a.u.)
b)
c)
d)
e)
402
401
400
399
398
397
Binding Energy (eV)
Figure S4. N1s X-ray photoemission spectroscopy as a function of the annealing temperature. N1s spectra of
1) Self assembled PPCN network, 2) MOCN1, 3) CuPyc Coordination network (475 K annealing), 4) CuPyc
polymeric chains (540 K annealing), 5) CuPyc polymeric grid (675 K annealing). All the spectra were
normalized to the same total area. Circular markers represent the experimental data, solid lines the fit. The
individual components of the fit are represented with red lines.
[1] Ottaviano, L.; Lozzi, L.; Ramondo, F.; Picozzi, P.; Santucci, S. Copper hexadecafluoro phthalocyanine and
naphthalocyanine: The role of shake up excitations in the interpretation and electronic distinction of high-resolution X-ray
photoelectron spectroscopy measurements. J. Electron. Spectrosc. Relat. Phenom. 1999, 105, 145-154.
[2] Gulyaev, R. V.; Kryuchkova, N. A.; Mazalov, L. N.; Boronin, A.I.; Basova, T. V.; Plyashkevich, V. A. et al. X-Ray
Photoelectron Investigation of Charge Distribution in Copper(II) Phthalocyanine Complexes. J. Surf. Investig-X-Ra 2011,
5, 48-56.
6
b. Fitting Parameters
C1s
N1s
C1s
N1s
Cu2p3/2
C1s
N1s
Cu2p3/2
C1s
N1s
Cu2p
C1s
N1s
Cu2p
PPCN Self Assembled Network
Binding Energy (eV)
FWMH (eV)
284.6 (aromatic carbons)
1.24
286.39 (carbons bonded with nitrogen)
1.41
288.44 (shake up)
1.96
399.44
1.27
401.1 (shake up)
1.37
MOCN1
284.87 (aromatic carbons)
1.36
286.71 (carbons bonded with nitrogen)
1.59
289 (shake up)
1.9
399.85
1.3
401.4 (shake up)
1.32
932.54
1.23
CuPyc coordination network (475 K)
284.69 (aromatic carbons)
1.38
286.36 (carbons bonded with nitrogen)
1.71
289 (shake up)
2.32
399.37
1.68
400.96 (shake up)
1.47
932.53 (metallic copper)
1.3
935.41 (CuII)
2.01
CuPyc Polymeric Chains (540 K)
284.52 (aromatic carbons)
1.38
286.08 (carbons bonded with nitrogen)
2.2
289.11 (shake up)
1.31
399.07
1.68
401.39 (shake up)
2.04
932.47 (metallic copper)
1.37
935.23 (CuII)
1.60
CuPyc Polymeric Grid (675 K)
284.44 (aromatic carbons)
1.37
285.63 (carbons bonded with nitrogen)
1.85
288.03 (shake up)
2.14
398.94
1.68
400.76
1.37
932.5 (metallic copper)
1.31
935.26 (CuII)
2
Table S1. Fitting Parameters
7
AREA (peak intensity %)
48.6
45.9
5.5
85.7
14.3
51.8
43.4
4.8
89.5
10.5
100
54.7
36.9
8.4
83.2
16.8
87
13
48.2
48.7
3.1
86.9
13.1
67.9
32.1
46
46.6
7.4
83.4
16.6
47.1
52.9