Straightforward synthesis of substituted p-Quinones: Isolation of a key intermediate and use as a bridging ligand in a diruthenium complex

Article abstract of DOI:10.1002/chem.200903409

(Figure Presented) Keep it simple: A new transamination method is reported for the synthesis of diamino-substituted p-quinones (see scheme). Asymmetrically substituted p-quinones are shown to be key intermediates in this reaction. The redox properties of these ligands are discussed and their utility in coordination and redox chemistry is shown with a representative example.

Full text of DOI:10.1002/chem.200903409

COMMUNICATION
DOI: 10.1002/chem.200903409
Straightforward Synthesis of Substituted p-Quinones: Isolation of a Key
Intermediate and Use as a Bridging Ligand in a Diruthenium Complex
Hari Sankar Das,[a] Fritz Weisser,[a] David Schweinfurth,[a] Cheng-Yong Su,[b]
Lapo Bogani,[c] Jan Fiedler,[d] and Biprajit Sarkar*[a]
Quinones are naturally occurring redox active molecules
that function in vital electron-transport processes, often in
conjugation with a transition-metal center.[1] p-Quinones,
such as vitamin K derivatives, ubiquinones or plastoqui-
nones, play important roles in photosynthesis, respiration,
and information-transfer processes.[2,3] Substituted p-qui-
nones have been extensively used as ligands in coordination
chemistry in recent years.[4–6] The remarkable properties that
such ligands impart to their metal complexes make such
compounds useful in a variety of fields, such as homogenous
catalysis,[7] supramolecular chemistry,[5,8] coordination poly-
mers,[9,10] and as bridging ligands in combination with redox
active metal centers such as ruthenium.[11–14] The last field
has gained tremendous attention in recent years because of
ambiguities arising in oxidation-state formulations also with
“organometallic-type” non-innocent ligands.[15] In addition,
the creation of quinone-based molecular magnets is a lively
field of research that has produced many interesting re-
sults.[16–20]
rarely straightforward one-pot reactions.[21,24] Some years
back Braunstein et al. reported straightforward and
“green” synthesis of a new class of mole-
a
cules that occurs through transamina-
tion.[25,26] Such molecules, which are iso-
mers of 1 and its derivatives, are best de-
scribed as zwitterions, 6.[25,26] Inspired by
this process we looked for an elegant syn-
thesis for the parent compound 1 and for possible intermedi-
ates formed during the transamination process. Herein we
report a simple one-pot synthesis of 1 and its mono- (2 and
3) and dialkyl (4 and 5) derivatives (Scheme 1). To elucidate
2,5-Diamino-1,4-benzoquinone 1[21] and its substituted de-
rivatives[22,23] have been known for decades. The synthesis of
1 has previously been reported and such syntheses are
[a] H. S. Das, F. Weisser, D. Schweinfurth, Dr. B. Sarkar
Institut fꢀr Anorganische Chemie, Universitꢁt Stuttgart
Pfaffenwaldring 55, 70550 Stuttgart (Germany)
Fax : (+49)711-68564165
E-mail: sarkar@iac.uni-stuttgart.de
[b] Prof. C.-Y. Su
Scheme 1.
School of Chemistry and Chemical Engineering
Sun Yat-Sen University, Guangzhou, 510275 (China)
[c] Dr. L. Bogani
1 Physikalisches Institut, Universitꢁt Stuttgart
Pfaffenwaldring 57, 70550 Stuttgart (Germany)
the use of such ligands in coordination and redox chemistry
we have prepared a diruthenium complex with the doubly
(m-4ꢀ2H)] (7; acacꢀ =
deprotonated form of 4: [{RuACHTNUGTRENNUG(acac)2}2ACHTUTGNRENNUGN
[d] Dr. J. Fiedler
´
J. Heyrovsky Institute of Physical Chemistry
acetylacetonato). Redox processes involving multinuclear
ruthenium complexes are known to display phenomena such
as charge-transfer isomerism[27] and valence ambiguity when
in combination with potentially non-innocent ligands.[11,13]
v.v.i., Academy of Sciences of the Czech Republic
ˇ
Dolejskova 3, 18223 Prague (Czech Republic)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903409.
Chem. Eur. J. 2010, 16, 2977 – 2981
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2977

mation) shows that there is extensive delocalization within
the O1-C1-C2-C3’-N1’ and N1-C3-C2’-C1’-O1’ sections of
ꢀ
the molecule and these are connected by authentic C C
ꢀ
single bonds with C1 C3 distances of 1.519(2) ꢃ (see
Table S2 in the Supporting Information). These molecules
can thus be considered as two merocyanine units connected
[29]
ꢀ
by C C single bonds as has been observed before. Cyclic
voltammetry experiments on 2–5 in CH2Cl2 with 0.1m
Bu4NPF6 showed one oxidation and two reduction processes
(see Figure S2 in the Supporting Information). While the
first reduction is reversible, the oxidation and second reduc-
tion are irreversible processes. The influence of double-
alkyl, rather than single-alkyl, substitution is shown by
slightly more negative reduction potentials for 4 and 5 com-
pared with 2 and 3, respectively. The influence of n-butyl
versus isopropyl groups on the reduction potentials is negli-
gible (Table 1).
Structural, electrochemical, magnetic, and UV/Vis-NIR and
EPR spectroelectrochemical results are reported below.
Reaction of 2,5-diamino-1,4-dihydroxybenzene in water
with air leads to the formation of 1. This is a one-pot reac-
tion and, in stark contrast to the synthetic procedures
known in the literature for 1, does not require any further
purification.[21,24]
When the same reaction was carried out in the presence
of isopropylamine or n-butylamine we observed “double”
transamination leading to the formation of 4 or 5, respec-
tively. Such transamination reactions have been observed
previously, but this is only the second example of transami-
nation reactions in quinonoid chemistry.[25] However, in con-
trast to reference [25] the yields of 4 and 5 were unsatisfac-
tory. Careful examination of the reaction mixture showed
that the formation of the parent compound 1, which is
poorly soluble in water, is extremely fast and its precipita-
tion limits the yield of the transamination. To circumvent
this problem, we carried out the transamination reactions in
a mixture of water, dichloromethane, and THF with Bu4NCl
as a phase-transfer catalyst (see Supporting Information).
These reaction conditions not only led to higher yields, but
also to the formation of a second product. Analysis of the
products showed it to be a mixture of the monosubstituted
products, 2 or 3, and the disubstituted products, 4 or 5. Thus,
the monosubstituted compounds 2 and 3 are key intermedi-
ates in the transamination reactions leading to the formation
of the disubstituted products 4 and 5, respectively, as has
been proposed for the formation of compounds of type 6.
However, in these cases they could not be isolated due to
the extremely fast second transamination reaction leading to
the disubstituted products.[25,26] To the best of our knowl-
edge, examples of asymmetrically substituted p-quinones,
such as 2 and 3, are extremely rare in the literature.[28]
Table 1. Redox potentials of the ligands 2–5 and complex 7.[a]
ox1
1=2
ox2
1=2
red1
1=2
E
(DEp)[b]
E
(DEp)[b]
E
(DEp)[b]
E
(DEp)[b]
red2
1=2
2
3
4
5
7
1.04[c]
n.o.[e]
n.o.
n.o.
n.o.
0.76 (65)
ꢀ1.43 (70)
ꢀ1.48 (65)
ꢀ1.50 (73)
ꢀ1.56 (75)
ꢀ1.09 (65)
ꢀ1.88[d]
ꢀ1.92[d]
ꢀ2.10[d]
ꢀ2.05[d]
ꢀ1.72 (62)
1.01[c]
1.00[c]
0.95[c]
0.11 (60)
[a] Electrochemical potentials in V from cyclic voltammetry in CH2Cl2
with 0.1m Bu4NPF6 at 298 K; scan rate: 100 mVsꢀ1; ferrocene/ferroceni-
um was used as the internal standard. [b] DEp =difference between peak
potentials in mV. [c] Anodic peak potential for irreversible oxidation.
[d] Cathodic peak potential for irreversible reduction. [e] n.o.=not ob-
served.
To test the utility of such ligands in coordination and
redox chemistry we prepared a dinuclear complex, 7, with
the doubly deprotonated form of 4. It was discovered that
compound 7 is paramagnetic, with two RuIII centers. This
compound was characterized by elemental analysis and
mass spectrometry. The RuIII complexes of this family are
known to show temperature-independent paramagnetism
(TIP).[13,30] The magnetic susceptibility of compound 7 was
measured with a SQUID setup (see Figure S3 in the Sup-
porting Information). As in previously reported cases,[11,13,30]
the spectrum was fitted by considering the system as com-
posed of dimers with a small fraction, P, of defective para-
magnetic sites[11,31] and using g, TIP, and the magnetic ex-
change, J, as free parameters; this yielded TIP=0.01ꢁ
0.01 cm3 molꢀ1, g=2.1ꢁ0.1, and J=ꢀ19ꢁ3 cmꢀ1. This anti-
ferromagnetic behavior is in agreement with previous re-
ports[11,13] and is also visible from the nonsaturated value at-
tained by the low-temperature M versus H curves. The ob-
tained value is comparable to those previously reported for
similar compounds and its slightly diminished magnitude
can probably be attributed to the reduced twisting induced
by the isopropyl substituents on the imino nitrogen atoms.
The discrepancy from the fit at lower temperatures can
probably be attributed to the presence of either the onset of
interdimer interactions or a zero-field splitting effect.[32]
1
Compounds 1–5 were characterized by H- and 13C NMR
spectroscopy, elemental analysis, and mass spectrometry.
ꢀ
Whereas compounds 4 and 5 showed only one N H signal
in their H NMR spectra, compounds 2 and 3 showed two
1
ꢀ
N H signals, which was the first evidence for the formation
of these monosubstituted species. To elucidate the bonding
situation within these compounds, we determined, through
X-ray diffraction, the crystal structure of 4 (see Figure S1 in
the Supporting Information). Even though the crystal struc-
ture of 4 has been reported previously, no detailed analysis
of the bonding situation, with respect to localization and de-
localization of electrons, was presented.[23] A look at the
bond lengths within 4 (see Figure S1 in the Supporting Infor-
2978
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2977 – 2981

Diamino-Substituted p-Quinones
COMMUNICATION
While the latter case cannot be excluded, our attempt at a
fit with the same parameters as above, but adding interdim-
er interactions, J’, obtained satisfactory agreement at low T
when J’=ꢀ4ꢁ1 cmꢀ1. The magnetic interaction pathway for
J’ is not purely dipolar in origin, since its magnitude is
higher than that expected for a dipolar interaction transmit-
ted by the interdimer RuIII-RuIII distance of 7.649(1) ꢃ.
However this can be considered an indicative value, since
single-crystal measurements would be needed for a com-
plete determination of the local anisotropic axes of RuIII
and an accurate evaluation of the interdimer interactions.
The EPR spectra of all species showed no signal at room
crystals for compound 7 and determined its X-ray structure.
A perfectly statistic disorder is present, with all acacꢀ and
bridging ligands occupying two sets of positions around the
same Ru center and 50% occupancy of each position was
used in the refinements (see Experimental Section). None-
theless the connectivity within the molecule has been unam-
biguously determined to be the meso-isomer (Figures 2 and
3). The Ru centers are in a distorted octahedral environ-
2
temperature. This can be attributed to the T2g state of RuIII,
which leads, as in low-spin FeIII, to three closely spaced
Kramers doublets.[19] Thus, fast relaxation makes the system
EPR silent until the solution is frozen. The observation of
the signal only in a frozen matrix can perhaps be linked to
the anisotropy expected for RuIII[19,31,32] with tumbling pro-
cesses incapable of merging the very different g values. At
110 K the EPR spectrum of 7 in CH2Cl2 showed a weak
signal (Figure 1a) and could be simulated using g1 =2.33,
Figure 2. ORTEP view of 7 (occupancy of each set of ligands assigned to
50%).
Figure 1. Left: EPR investigation of the compounds, with the upper scale
referring to spectra a) and b) and the lower scale to spectrum c). All
spectra have been vertically shifted for clarity and the gray lines are the
corresponding simulations (see text). a) EPR spectrum of 7 at frequency
f=9.6722 GHz in CH2Cl2 at 110 K; b) EPR spectrum of 7+ at f=
9.6949 GHz, in CH2Cl2 with 0.1m Bu4NPF6 at 110 K; c) EPR spectrum of
7ꢀ at f=9.4741 GHz in CH2Cl2 with 0.1m Bu4NPF6 at 110 K, d) Cyclic
voltammogram of 7 in CH2Cl2 with 0.1m Bu4NPF6 at 295 K. Ferrocene/
ferrocenium was used as the internal standard.
Figure 3. ORTEP view of 7’ (occupancy of each set of ligands assigned to
50%).
ment as it is coordinated by one oxygen and one nitrogen
atom from the bridging ligand and four oxygen atoms from
the two acacꢀ ligands. The bonding situation within the
bridging ligand in 7 shows higher localization of the double
bonds compared with free ligand 4 (see Table S2 in the Sup-
porting Information), possibly as a result of the aforemen-
g2 =2.24 and g3 =1.91 (gav =2.16) in overall agreement with
the extracted susceptibility value. Hyperfine interaction with
the 99Ru and 101Ru isotopes (natural abundances of 13 and
17%, respectively) was considered for the simulation and
gave Ak =260 MHz and A? =150 MHz, similar to reported
literature values.[33] The weakness of the signal may indicate
the detection of the fraction, P, of single RuIII centers in de-
fective dimers that would otherwise be silent due to antifer-
romagnetic coupling (as in the case of Cu acetate).[34] It is
worth noting that g1 is slightly lower than expected for RuIII
centers in this type of environment, a fact that can be attrib-
uted to orbital reduction factors <1, that is, to partial elec-
tronic delocalization onto the ligand.[19] We also obtained
tioned delocalization of RuIII electrons.[19] The C1 C3 bond,
ꢀ
at 1.48(3) (1.46(2)) ꢃ, remains an authentic single bond as
in the case of the free ligand 4. Such localization of bonds
within a substituted p-quinone ligand has been observed
previously in the doubly deprotonated form of 1.[13] Thus,
the bridging ligand binds to the Ru centers through Oꢀ and
imino nitrogen donors (see image of 7). The Ru–Ru dis-
tance in 7 is 7.840(1) ꢃ.
Cyclic voltammetry of 7 in CH2Cl2 with 0.1m Bu4NPF6 re-
veals two oxidation and two reduction processes (Figure 1d)
that are all completely reversible. This is also authenticated
by spectroelectrochemical measurements (vide infra). The
Chem. Eur. J. 2010, 16, 2977 – 2981
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2979

B. Sarkar et al.
difference between the two oxidation potentials of 650 mV
translates to a comproportionation constant Kc (Kc =10DE/59
at 298 K)[35] of the order of 1011 for the one-electron oxi-
dized species. The corresponding difference in the reduction
potentials is 630 mV (Table 1) and the Kc value is of the
order of 1010.
ligand library, tuning of the redox properties, and the use of
asymmetric ligands such as 2 and 3 in coordination chemis-
try.
Acknowledgements
The RuIII–RuIII species 7 shows an intense LMCT (4ꢀ2H
!
RuIII) transition at 534 nm (e=19900mꢀ1 cmꢀ1; see Figure S4
in the Supporting Information). In addition, there are fur-
ther ligand-centered transitions in the UV region. In the
one-electron oxidized form 7+, the LMCT band is red-shift-
ed to 708 nm with the intensity remaining almost un-
changed. Furthermore, a new band grows in the near IR
region at 1515 nm with e of 5000mꢀ1 cmꢀ1 that disappears on
second oxidation (see Figures S4 and S5 in the Supporting
Information). Thus, the UV/Vis/NIR measurements of 7+
point to a mixed-valent situation leading to the formation of
a RuIII–RuIV species on one-electron oxidation. The EPR
signal of 7+ at 110 K (Figure 1b) is very similar to that of 7,
strengthening the hypothesis of the detection of defective
dimers for 7. The spectrum can be simulated well with the
same g and A parameters by using a narrower linewidth.
The Dn1/2 of the IVCT (intervalence charge transfer) band
We are much indebted to Prof. D. Gatteschi for enlightening discussions.
We acknowledge the German DFG, the Ministry of Education of the
Czech Republic (grant COST OC 140) and the Landesstiftung Baden-
Wꢀrttemberg (Eliteprogram for Postdocs) for financial support.
Keywords: cyclic voltammetry · quinones · ruthenium ·
spectroelectrochemistry · synthetic methods
[1] S. D. Thomson, Naturally Occuring Quinones: IV, Springer, Amster-
dam, 1996.
[2] Y. Izumi, H. Sawada, N. Sakka, N. Yamamoto, T. Kume, H. Katsuki,
S. Shimohama, A. Akaike, J. Neurosci. Res. 2005, 79, 849–860.
[3] M. He, P. J. Sheldon, D. H. Sherman, Proc. Natl. Acad. Sci. USA
2001, 98, 926–931.
[4] S. Scheuermann, T. Kretz, H. Vitze, J. W. Bats, M. Bolte, H.-W.
Lerner, M. Wagner, Chem. Eur. J. 2008, 14, 2590–2601.
[5] W.-G. Jia, Y.-F. Han, Y.-J. Lin, L.-H. Weng, G.-X. Jin, Organometal-
lics 2009, 28, 3459–3464.
[6] J. S. Miller, K. S. Min, Angew. Chem. 2009, 121, 268–278; Angew.
Chem. Int. Ed. 2009, 48, 262–272.
at 1515 nm is measured at about 1475 cmꢀ1. The Hush for-
1/2
mulation gives Dn1/2 =(2310nIVCT
)
ꢂ3905 cmꢀ1.[36] This
points to delocalization in the mixed-valent form and hence
7+ belongs to the strongly-coupled Class III system. The
one-electron reduced form 7ꢀ shows a very strong band at
750 nm (e=31700mꢀ1 cmꢀ1). In addition, this species shows
absorptions in the near IR region at 1725 nm (e=
1100mꢀ1 cmꢀ1). Since the starting RuIII–RuIII compound can,
in principle, be reduced to the RuIII–RuII form, these near
IR bands can be taken as evidence for a mixed-valent spe-
cies. However, the EPR spectrum of 7ꢀ at 110 K (Figure 1c)
shows a strong quinone signal that shadows the much
weaker RuIII line. The spectrum is well reproduced by using
an isotropic signal with g=2.004, and considering hyperfine
coupling to the N nuclei. This points to a reduction of the
bridging ligand leading to a three-spin situation, RuIII-
[7] D. Zhang, G.-X. Jin, Organometallics 2003, 22, 2851–2854.
[8] J. Mattsson, P. Govindaswamy, A. K. Renfrew, P. J. Dyson, P. Step-
nicka, G. S. -Fink, B. Therrien, Organometallics 2009, 28, 4350–4357.
[9] S. Kitagawa, S. Kawata, Coord. Chem. Rev. 2002, 224, 11–34.
[10] G. Margraf, T. Kretz, F. Fabrizi de Biani, F. Laschi, S. Losi, P. Zanel-
lo, J. W. Bats, J. Wolf, K. Removic-Langer, M. Lang, A. Prokofiev,
W. Assmus, H.-W. Lerner, M. Wagner, Inorg. Chem. 2005, 44, 1277–
1288.
[11] D. Kumbhakar, B. Sarkar, S. Maji, S. M. Mobin, J. Fiedler, R. J.
Aparicio, W. Kaim, G. K. Lahiri, J. Am. Chem. Soc. 2008, 130,
17575–17583.
[12] T. E. Keyes, R. J. Forster, P. M. Jayaweera, C. G. Coates, J. J. McGar-
vey, J. G. Vos, Inorg. Chem. 1998, 37, 5925–5932.
[13] S. Kar, B. Sarkar, S. Ghumaan, D. Janardanan, J. von Slageren, J.
Fiedler, V. G. Puranik, R. B. Sunoj, W. Kaim, G. K. Lahiri, Chem.
Eur. J. 2005, 11, 4901–4911.
[14] H. Masui, A. L. Freda, M. C. Zerner, A. B. P. Lever, Inorg. Chem.
2000, 39, 141–152.
[15] J. Maurer, R. F. Winter, B. Sarkar, J. Fiedler, S. Zalis, Chem.
Commun. 2004, 1900–1901.
[16] A. Caneschi, A. Dei, Angew. Chem. 1998, 110, 3220–3222; Angew.
Chem. Int. Ed. 1998, 37, 3005–3007.
ACHTUNGTRENNUNG
(4ꢀ2H·ꢀ)RuIII (›,›,fl), with antiferromagnetic coupling be-
tween the RuIII spins. This contrasts with what has been ob-
served for the reduced state of the compound, analogous to
7ꢀ, with the doubly deprotonated form of 1 as the bridging
ligand. In that case the metal-centered spin was observed by
EPR spectroscopy.[13] This shows the subtle changes in or
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[17] P. L. Gentili, L. Bussotti, R. Righini, A. Beni, L. Bogani, A. Dei,
Chem. Phys. 2005, 314, 9–17.
[18] D. M. Adams, A. Dei, A. L. Rheingold, D. N. Hendrickson, J. Am.
Chem. Soc. 1993, 115, 8221–8229.
In summary, we have reported a new, one-pot, synthetic
route to 1 and its transaminated analogues 4 and 5. We also
detected compounds 2 and 3 as key intermediates in the
transamination reaction step. Asymmetrically substituted p-
quinones, such as 2 and 3, are rare in the literature. Com-
pound 4 was used then to prepare the diruthenium complex
7. Oxidation of this compound leads to a RuIII–RuIV-contain-
ing mixed-valent species and its reduction leads to a quino-
noid radical-containing RuIII–RuIII species. Future efforts
will be directed towards a target-oriented expansion of the
[19] A. Dei, D. Gatteschi, L. Pardi, Inorg. Chem. 1990, 29, 1442–1444.
[20] A. Dei, D. Gatteschi, C. Sangregorio, L. Sorace, Acc. Chem. Res.
2004, 37, 827–835.
[21] A.-M. Osman, J. Am. Chem. Soc. 1957, 79, 966–968.
[22] R. N. Harger, J. Am. Chem. Soc. 1924, 46, 2540–2551.
[23] S. Bayen, N. Barooah, R. J. Sarma, T. K. Sen, A. Karmakar, J. B.
Baruah, Dyes Pigm. 2007, 75, 770–775.
[24] H. B. Mereyala, M. V. Chary, S. Kantevari, Synthesis 2007, 187–189.
[25] Q.-Z. Yang, O. Siri, P. Braunstein, Chem. Commun. 2005, 2660–
2662.
[26] Q.-Z. Yang, O. Siri, P. Braunstein, Chem. Eur. J. 2005, 11, 7237–
7246.
2980
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2977 – 2981

Diamino-Substituted p-Quinones
COMMUNICATION
[27] J. C. Salsman, C. P. Kubiak, T. Ito, J. Am. Chem. Soc. 2005, 127,
2382–2383.
[28] C. Xing, E. B. Skibo, Biochemistry 2000, 39, 10770–10780.
[29] S. Dꢁhne, D. Leupold, Angew. Chem. 1966, 78, 1029–1039; Angew.
Chem. Int. Ed. 1966, 5, 984–993.
[33] I. A. Miller, E. R. Offenbacher, Phys. Rev. 1968, 166, 269–278.
[34] A. Bencini, D. Gatteschi, EPR of Exchange Coupled Systems,
Springer, Heidelberg, 1990.
[35] C. Creutz, Prog. Inorg. Chem. 1983, 30, 1–73.
[36] N. S. Hush, Prog. Inorg. Chem. 1967, 8, 391.
[30] S. Ghumaan, B. Sarkar, S. Patra, K. Parimal, J. V. Slageren, J. Fie-
dler, W. Kaim, G. K. Lahiri, Dalton Trans. 2005, 706.
[31] O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993.
[32] F. Fabrizi de Biani, A. Dei, C. Sangregorio, L. Sorace, Dalton Trans.
2005, 3868–3873.
Received: December 11, 2009
Published online: February 15, 2010
Chem. Eur. J. 2010, 16, 2977 – 2981
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2981