Adsorption and dehydrogenation of tetrahydroxybenzene on Cu(111)

Article abstract of DOI:10.1039/c3cc45052j

Adsorption of tetrahydroxybenzene (THB) on Cu(111) and Au(111) surfaces is studied using a combination of STM, XPS, and DFT. THB is deposited intact, but on Cu(111) it undergoes gradual dehydrogenation of the hydroxyl groups as a function of substrate temperature, yielding a pure dihydroxy-benzoquinone phase at 370 K. Subtle changes to the adsorption structure upon dehydrogenation are explained from differences in molecule-surface bonding.

Full text of DOI:10.1039/c3cc45052j

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Adsorption and dehydrogenation of
tetrahydroxybenzene on Cu(111)†
Fabian Bebensee,a Katrine Svane,a Christian Bombis,a Federico Masini,a
Svetlana Klyatskaya,b Flemming Besenbacher,a Mario Ruben,bc Bjørk Hammera and
Trolle Linderoth*a
Cite this: Chem. Commun., 2013,
49, 9308
Received 5th July 2013,
Accepted 9th August 2013
DOI: 10.1039/c3cc45052j
www.rsc.org/chemcomm
Adsorption of tetrahydroxybenzene (THB) on Cu(111) and Au(111) changes to the adsorption structures. Adsorption of THB and
surfaces is studied using a combination of STM, XPS, and DFT. THB is related molecules was studied previously in relation to metal
deposited intact, but on Cu(111) it undergoes gradual dehydrogena- coordination11 or covalent networks12 on surfaces. From X-ray
tion of the hydroxyl groups as a function of substrate temperature, Photoelectron Spectroscopy (XPS) we show that THB is deposited
yielding a pure dihydroxy-benzoquinone phase at 370 K. Subtle intact on Au(111), but on Cu(111) it gradually undergoes thermally
changes to the adsorption structure upon dehydrogenation are activated dehydrogenation of its hydroxyl groups, eventually yielding
explained from differences in molecule–surface bonding.
dihydroxy-benzoquinone at 370 K. The energetics of the dehydrogena-
tion process is elucidated from Density Functional Theory (DFT)
The fabrication of supramolecular structures on surfaces1 through calculations emphasizing the importance of the entropic contribution
non-covalent interactions2 has recently received tremendous inter- of desorbed hydrogen. The subtle changes to the adsorption structure
est. To rationalise and exploit this approach, it is essential that the observed in STM images upon dehydrogenation are explained from
chemical state of the molecular building blocks on the surface is differences in molecule–surface bonding.
known, which can be complicated by degradation/reaction during
The experiments were conducted in a UHV-setup equipped with
deposition or after adsorption on the surface.3 Dehydrogenation an Aarhus STM and in situ facilities for XPS (see ESI† for details).
reactions are particularly interesting in this respect since they can Fig. 1a depicts the adsorption structure observed after deposition of
e.g. remove or create the possibility to establish hydrogen bonds or
change the bonding of molecules to the substrate.4 Dehydrogena-
tion is well known to occur for carboxylic acids5 including amino
acids,6 but has also been observed for less acidic hydroxyl groups,
e.g. for alcohols7 or even terminal acetylenes.8
In general, there is a need to expand the knowledge base for
molecular dehydrogenation reactions on surfaces and improve our
understanding of the conditions under which dehydrogenation
occurs. Dehydrogenation and similar chemical changes are often
linked to pronounced structural changes in the supramolecular
assemblies which can be observed using Scanning Tunneling Micro-
scopy (STM).1a,4c,9 Here we investigate a case of dehydrogenation for a
comparatively simple molecule 1,2,4,5-tetrahydroxybenzene10 (THB)
where pronounced dehydrogenation is only accompanied by subtle
a Sino-Danish Center for Molecular Nanostructures on Surfaces, Interdisciplinary
Nanoscience Center (iNANO) and Department of Physics and Astronomy,
Aarhus University, Aarhus C, Denmark. E-mail: trolle@inano.au.dk
b
¨
Fig. 1 (a and b) STM images obtained after deposition of THB on Cu(111).
(a) As-deposited at 300 K and (b) after annealing to 370 K (IT 0.35 nA,
UT ꢀ1.25/ꢀ1.4 V). The inset in (b) shows a model for the annealed structure
(cf. Fig. S5, ESI†). (c–e) XP spectra of the O 1s region after THB deposition. (c) After
deposition at 115 K (acquired at this temperature), (d) after deposition at RT and
(e) after annealing to 370 K (d and e acquired at 320 K).
Karlsruhe Institute of Technology (KIT), Institut fur Nanotechnologie,
Karslruhe, Germany
c
´
IPCMS-CNRS, Universite de Strasbourg, Strasbourg, France
† Electronic supplementary information (ESI) available: Experimental and com-
putational details, STM and XPS data for DHBQ on Cu(111) and the synchrotron
based XP spectrum of THB/Cu(111) at RT. See DOI: 10.1039/c3cc45052j
c
9308 Chem. Commun., 2013, 49, 9308--9310
This journal is The Royal Society of Chemistry 2013

View Article Online
Communication
ChemComm
THB on Cu(111) at room temperature (RT, 300 K). The structure (Fig. 2b) exhibits a single peak at a BE of 532.4 eV, while DHBQ
consists of protrusions in a hexagonal arrangement described by a (Fig. 2e) features two peaks of almost equal intensity at BEs of
unit cell with side lengths of 7.5 Å oriented at angles of ꢁ(12 ꢁ 21) 532.7 eV and 531.1 eV, respectively. The two spectra correlate well
with respect to the high symmetry h10ꢀ1i directions of the Cu with the molecular structures, establishing the assignment of the
substrate. Interestingly, the protrusions show different STM contrast high/low BE features to the hydroxyl/carbonyl groups on Au(111) and
with a variation in the apparent height of 0.28 ꢁ 0.07 Å (a few point corroborating the assignment done above on Cu(111). The corre-
defects are visible which lie outside this range). Annealing to 370 K sponding C 1s spectra shown in Fig. 2c and f are almost identical
results in the adsorption structure shown in Fig. 1b, which has the and exhibit two features at BEs of 285.2 eV and 283.8 eV (THB)/
same hexagonal unit cell size (7.5 Å) but a slightly different orienta- 283.9 eV (DHBQ) with an intensity ratio of B2:1 in favour of the
tion with respect to the substrate of ꢁ(19 ꢁ 21) (in rare cases 0 ꢁ 21). high BE feature. The high intensity feature is assigned to the 4
Most notably, the structure is now smooth with all protrusions carbons attached to oxygens in hydroxyl or carbonyl groups, which
showing the same apparent height. Two questions arise from these are expected to have similar BE’s, and the low intensity feature to the
STM images: what is the origin of the inhomogeneity in the 2 carbons not connected to oxygen (see ESI† for comparison to
as-deposited structure (Fig. 1a) and what is the molecular level effect calculated core level shifts). The control experiments thus establish
of the annealing which removes this inhomogeneity? To probe the that THB is sublimated intact onto the surface (also X-ray beam
chemical state of the molecules in the respective situations, we damage does not occur). The spectrum of DHBQ on Au(111) (Fig. 2e)
employed XPS. Fig. 1c–e presents spectra of the O 1s region recorded bears very close resemblance to that of annealed THB on Cu(111)
after deposition of THB on a Cu(111) substrate held at 115 K (c) as (Fig. 1e). Further control experiments with DHBQ adsorbed on
well as at RT (d) and after annealing to 370 K (e). All three spectra Cu(111) are also in good agreement with annealed THB regarding
can be fitted with two contributions at binding energies (BEs) of STM monolayer structure and XPS spectra (see ESI†). We therefore
532.6 ꢁ 0.1 eV and 530.8 ꢁ 0.1 eV, respectively. As the four oxygen conclude that annealing THB on Cu(111) to 370 K results in
atoms of THB are chemically equivalent, the observation of two formation of DHBQ via dehydrogenation. Based on literature
features in the O 1s region suggests partial dehydrogenation. reports, the abstracted hydrogen is thought to desorb recombina-
Hence, the high BE feature is ascribed to oxygen in hydroxyl tively from the surface.13
groups and the low BE feature to oxygen in carbonyl groups in
To clarify the driving forces for the observed dehydrogenation and
agreement with the literature.9c The sequence of spectra shows to establish the origin of the different contrasts for as-deposited THB,
an increasing contribution from the low BE component which we performed DFT calculations. The DFT calculations included
increases from 22% (c) over 37% (d) to 45% (e). The level of dispersive effects by using the meta-GGA functional M06-L (cf. ESI†
dehydrogenation thus increases with increasing substrate tem- for details).14 The resulting optimised adsorption geometries for THB
perature, suggesting a thermally activated process.
and DHBQ on Cu(111) are shown in Fig. 3a and b and the
As control experiments, we deposited both THB and DHBQ corresponding energy landscape in Fig. 3d. The energy calculations
(2,5-dihydroxy-benzoquinone; a related compound in which two demonstrate that the dehydrogenation results in a net increase in
hydroxyls are replaced by carbonyl groups in the para position) on enthalpy (black bars in Fig. 3d). In many cases, a strong enthalpic
an inert Au(111) substrate (see ESI† concerning the possibility of an driving force for dehydrogenation renders it unnecessary to consider
o-benzoquinone product). The STM images in Fig. 2a and d reveal entropy effects.15 Here, however, the entropy of the desorbed hydro-
virtually identical structures for THB and DHBQ adsorbed on gen molecules abstracted from THB plays an important role in the
Au(111) at RT: both form hexagonal structures with a unit cell size thermodynamics. A hydrogen molecule desorbed from the surface
of 7.5 Å oriented at ꢁ(20 ꢁ 21) with respect to the substrate’s high after dehydrogenation has high entropy, in particular at the low
symmetry directions. The XP spectrum of the O 1s region of THB pressure in the UHV chamber. A conservative estimate for the
pressure during our experiments is 10ꢀ9 mbar, corresponding to an
entropy contribution of ꢀ1.11 eV per H2 to the free energy at RT.
Another contribution that is often neglected is the change in vibra-
tional zero point energy resulting from a reaction. Here, we estimate
this correction to be ꢀ0.17 eV (cf. ESI†). Adding these two contribu-
tions to establish the free energy landscape (red bars) shows that the
system moves down in free energy by dehydrogenation.
In its optimised adsorption geometry, THB adsorbs with the
plane of the carbon ring parallel to the surface, at a distance of 2.9 Å
above the surface, and with the centre of the ring located above a
bridge site. DHBQ also adsorbs at the bridge site, but the distance
between the carbon ring and the surface is shortened to 2.3 Å. The
calculations show that the carbonyl bonds in DHBQ are elongated
from 1.23 Å (typical of a double bond) when the molecule is in the
gas phase, to 1.31 Å (an intermediate between single and double
bonds), when it is adsorbed on the surface. This bond lengthening
Fig. 2 Left: STM images of THB (a) and DHBQ (d) adsorbed on Au(111) at RT
(IT 0.35/0.37 nA, UT ꢀ1.36/ꢀ1.4 V). Insets show molecular models of the respective
compounds. The observed height modulation is caused by the Au(111) herring-
indicates that the bond between the carbonyl oxygen of DHBQ and
the surface plays the role of the O–H bond found in THB. In this way
bone reconstruction. Right: corresponding XP spectra of the O 1s and C 1s regions
for THB (b and c) and DHBQ (e and f).
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 9308--9310 9309

View Article Online
ChemComm
Communication
rotational orientation between the mixed DHBQ–THB and pure
DHBQ overlayers cannot be quantitatively assessed due to the many
inequivalent bonding situations in the large computational unit
cell, but we speculate that the rotation occurs to optimise the
molecule–substrate interaction arising from an increased number
of Cu–carbonyl bonds.
We acknowledge support from the Danish National Research
Foundation, the Danish Council for Independent Research, the
Marie-Curie networks SMALL and MONET, the Lundbeck Founda-
tion, DCSC, and the Alexander von Humboldt-Foundation.
Notes and references
1 (a) J. A. A. W. Elemans, S. B. Lei and S. De Feyter, Angew. Chem., Int.
Ed., 2009, 48, 7298–7332; (b) G. Franc and A. Gourdon, Phys. Chem.
Chem. Phys., 2011, 13, 14283–14292.
2 J. V. Barth, Annu. Rev. Phys. Chem., 2007, 58, 375–407.
3 (a) F. Bebensee, C. Bombis, S.-R. Vadapoo, J. R. Cramer, F. Besenbacher,
K. V. Gothelf and T. R. Linderoth, J. Am. Chem. Soc., 2013, 135, 2136–2139;
(b) M. Kittelmann, P. Rahe, A. Gourdon and A. Kuhnle, ACS Nano, 2012, 6,
7406–7411.
4 (a) L. Bartels, Nat. Chem., 2010, 2, 87–95; (b) A. M. Beatty, Coord.
Chem. Rev., 2003, 246, 131–143; (c) A. C. Papageorgiou, S. Fischer,
J. Reichert, K. Diller, F. Blobner, F. Klappenberger, F. Allegretti,
A. P. Seitsonen and J. V. Barth, ACS Nano, 2012, 6, 2477–2486.
5 (a) M. E. Canas-Ventura, F. Klappenberger, S. Clair, S. Pons, K. Kern,
H. Brune, T. Strunskus, C. Woll, R. Fasel and J. V. Barth, J. Chem.
Phys., 2006, 125; (b) M. N. Faraggi, C. Rogero, A. Arnau, M. Trelka,
D. Ecija, C. Isvoranu, J. Schnadt, C. Marti-Gastaldo, E. Coronado,
J. M. Gallego, R. Otero and R. Miranda, J. Phys. Chem. C, 2011, 115,
21177–21182; (c) L. Kanninen, N. Jokinen, H. Ali-Loytty, P. Jussila,
K. Lahtonen, M. Hirsimaki, M. Valden, M. Kuzmin, R. Parna and
E. Nommiste, Surf. Sci., 2011, 605, 1968–1978; (d) N. Lin,
A. Dmitriev, J. Weckesser, J. V. Barth and K. Kern, Angew. Chem.,
Int. Ed., 2002, 41, 4779–4783; (e) M. Matena, M. Stohr, T. Riehm,
J. Bjork, S. Martens, M. S. Dyer, M. Persson, J. Lobo-Checa,
K. Muller, M. Enache, H. Wadepohl, J. Zegenhagen, T. A. Jung and
L. H. Gade, Chem.–Eur. J., 2010, 16, 2079–2091.
6 (a) T. Eralp, A. Shavorskiy, Z. V. Zheleva, G. Held, N. Kalashnyk,
Y. X. Ning and T. R. Linderoth, Langmuir, 2010, 26, 18841–18851;
(b) V. Humblot, C. Methivier, R. Raval and C. M. Pradier, Surf. Sci.,
2007, 601, 4189–4194.
7 (a) R. Pawlak, S. Clair, V. Oison, M. Abel, O. Ourdjini, N. A. A.
Zwaneveld, D. Gigmes, D. Bertin, L. Nony and L. Porte, ChemPhysChem,
2009, 10, 1032–1035; (b) S. Polmann, A. Bayer, C. Ammon and
H. P. Steinruck, Z. Phys. Chem., 2004, 218, 957–971; (c) C. Ammon,
A. Bayer, G. Held, B. Richter, T. Schmidt and H. P. Steinruck, Surf. Sci.,
2002, 507, 845–850.
Fig. 3 Calculated adsorption geometry for THB (left) and DHBQ (right) on
Cu(111) in side (a) and top (b) view, simulated STM images (c) and the energy
landscape (d) relative to THB in the gas phase. Black bars represent enthalpies
and red bars represent free energies at RT.
the conjugated p-system of the carbon ring in THB can remain
relatively intact when the molecule is dehydrogenated on the surface.
A strong interaction between the carbonyl oxygen and the Cu surface
is supported by the XPS measurements showing that the O 1s BE in
the carbonyl group is measurably higher on Au(111) (531.1 eV)
compared to that on Cu(111) (530.8 eV), while the hydroxyl O 1s
BE is the same. Overall, the stronger interaction between the carbonyl
oxygen and the surface, compared to the hydroxyl–surface interaction,
brings the dehydrogenated DHBQ molecule closer to the surface
compared to THB. Simulated STM images based on the Tersoff–
Hamann model (Fig. 3c) show a lower apparent height for DHBQ
than for THB reflecting mainly these different adsorption heights.
The change in the STM images observed upon annealing can
now be rationalised: at RT, a mixture of THB and DHBQ exists, as
implied by the appearance of features with different apparent height
in the STM images. Quantitative analysis of STM images in Fig. 1a
shows that the protrusions exhibit a variation in apparent heights
rather than two distinct levels, possibly due to locally varying
bonding configurations. We estimate the fraction of DHBQ (dim
protrusions) to lie in the range 0.5–0.75. This range corresponds to
dehydrogenation of 31 ꢁ 7% of the hydroxyl groups consistent with
XPS (Fig. 1b and ESI† Section 3). The DHBQ and THB species are
thoroughly intermixed indicating a homogenous level of dehydro-
8 Y. Q. Zhang, N. Kepcija, M. Kleinschrodt, K. Diller, S. Fischer, A. C.
Papageorgiou, F. Allegretti, J. Bjork, S. Klyatskaya, F. Klappenberger,
M. Ruben and J. V. Barth, Nat. Commun., 2012, 3, 1286.
9 (a) N. Lin, D. Payer, A. Dmitriev, T. Strunskus, C. Woll, J. V. Barth and
K. Kern, Angew. Chem., Int. Ed., 2005, 44, 1488–1491; (b) M. Ruben,
D. Payer, A. Landa, A. Comisso, C. Gattinoni, N. Lin, J. P. Collin,
J. P. Sauvage, A. De Vita and K. Kern, J. Am. Chem. Soc., 2006, 128,
15644–15651; (c) S. Stepanow, T. Strunskus, M. Lingenfelder,
A. Dmitriev, H. Spillmann, N. Lin, J. V. Barth, C. Woll and K. Kern,
J. Phys. Chem. B, 2004, 108, 19392–19397.
10 C. A. Hansen and J. W. Frost, J. Am. Chem. Soc., 2002, 124, 5926–5927.
genation across the surface. Annealing the sample to 370 K leads to 11 H. M. Zhang, W. Zhao, Z. X. Xie, L. S. Long, B. W. Mao, X. Xu and
L. S. Zheng, J. Phys. Chem. C, 2007, 111, 7570–7573.
12 (a) L. M. Lanni, R. W. Tilford, M. Bharathy and J. J. Lavigne, J. Am.
further dehydrogenation, converting all THB molecules into DHBQ.
A suggested structural model for the DHBQ overlayer is shown in
Chem. Soc., 2011, 133, 13975–13983; (b) N. A. A. Zwaneveld,
Fig. 1c and Fig. S5 (ESI†). The model is based on the optimal
hydrogen bonding pattern found from DFT calculations in the gas
phase using the experimentally observed unit cell. This structure was
R. Pawlak, M. Abel, D. Catalin, D. Gigmes, D. Bertin and L. Porte,
J. Am. Chem. Soc., 2008, 130, 6678–6679.
13 T. Kammler and J. Kuppers, J. Chem. Phys., 1999, 111, 8115–8123.
14 L. Ferrighi, G. K. H. Madsen and B. Hammer, J. Chem. Phys., 2011, 135.
subsequently transferred to a Cu substrate and relaxed, allowing 15 (a) S. Clair, S. Pons, S. Fabris, S. Baroni, H. Brune, K. Kern and
J. V. Barth, J. Phys. Chem. B, 2006, 110, 5627–5632; (b) T. Classen,
M. Lingenfelder, Y. Wang, R. Chopra, C. Virojanadara, U. Starke,
G. Costantini, G. Fratesi, S. Fabris, S. de Gironcoli, S. Baroni, S. Haq,
formation of Cu–carbonyl bonds. The optimised structure is com-
prised of parallel rows of molecules which are inter-linked within the
rows via double hydrogen bonds. The origins of the slight change in
9310 Chem. Commun., 2013, 49, 9308--9310
R. Raval and K. Kern, J. Phys. Chem. A, 2007, 111, 12589–12603.
c
This journal is The Royal Society of Chemistry 2013