An effective non-chromatographic method for the purification of phenanthrolines and related ligands

Article abstract of DOI:10.1016/j.tetlet.2020.152080

1,10-Phenanthrolines are widely employed as ligands, but their use on a large scale is constrained by their difficult purification, which usually requires lengthy chromatographic separations. Herein, we describe a purification strategy that takes advantage of the high stability and low solubility of phenanthroline complexes to separate them from the by products of their synthesis. The formation of ZnCl₂ complexes was employed, from which the free ligand can be recovered by reaction with aqueous NH₃ in a biphasic CH₂Cl₂/H₂O system. The same strategy was also successfully employed to purify related quinolino-guanidine ligands, demonstrating that the procedure is of general applicability.

Full text of DOI:10.1016/j.tetlet.2020.152080

Tetrahedron Letters xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
An effective non-chromatographic method for the purification
of phenanthrolines and related ligands
⇑
Francesco Ferretti, Fabio Ragaini
Dipartimento di Chimica, Università degli Studi di Milano, via C. Golgi 19, 20133 Milano Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
1,10-Phenanthrolines are widely employed as ligands, but their use on a large scale is constrained by
their difficult purification, which usually requires lengthy chromatographic separations. Herein, we
describe a purification strategy that takes advantage of the high stability and low solubility of phenan-
Received 17 April 2020
Revised 18 May 2020
Accepted 22 May 2020
Available online xxxx
throline complexes to separate them from the by products of their synthesis. The formation of ZnCl
plexes was employed, from which the free ligand can be recovered by reaction with aqueous NH
biphasic CH Cl /H O system. The same strategy was also successfully employed to purify related quino-
lino-guanidine ligands, demonstrating that the procedure is of general applicability.
2
com-
3
in a
2
2
2
Keywords:
Phenanthrolines
Nitrogen ligands
Purification
Ó 2020 Elsevier Ltd. All rights reserved.
Guanidines
Introduction
of an effective purification strategy because such complexes are
2
very insoluble, whereas ZnCl complexes of monodentate nitrogen
Phenanthrolines are amongst the most useful and investigated
nitrogen ligands, with applications spanning from material chem-
istry to homogeneous catalysis [1–7]. However, few substituted
phenanthrolines are commercially available. We have been
involved in their use as ligands for many years [8–13] and have
needed to synthesize some substituted phenanthrolines that are
either not commercially available or not previously reported in
the literature [14–19]. We, as other researchers working with
phenanthroline ligands [20–30], aim to develop catalytic reactions
that could potentially be employed at an industrial scale. Thus, we
are interested in synthetic strategies that only require inexpensive
reagents. Various synthetic pathways have been reported for the
preparation of phenanthrolines, but when cost and availability of
the reagents are a concern, Skraup type condensation reactions
are invariably employed. These reactions always produce many
byproducts that still contain basic nitrogen atoms and cannot be
eliminated by a simple acidic extraction/back extraction. A simple
crystallization is most of the time insufficient to obtain an analyt-
ically pure compound. On the other hand, chromatographic purifi-
cation of phenanthrolines is notoriously difficult and strongly
limits the scale of the preparation.
ligands are much more soluble. In the case of Ar-BIAN and even
Alk-BIAN ligands [35–37], the procedure we developed to remove
the coordinated zinc, that is treating a CH
2 2
Cl solution or suspen-
sion of the ZnCl /ligand complex with an aqueous solution of
2
potassium oxalate, is very efficient and has become the standard
procedure employed in the literature to effect this decomplexation.
The ligand remains in the organic phase, whereas zinc either forms
insoluble ZnC
2
O
4
, or remains in the aqueous phase as a mixture of
2
À
4À
[Zn(C
2
O
4
)
2
]
and [Zn(C
2
O
4
)
3
]
depending on the oxalate concen-
tration. Thus, we decided to examine if a complexation/decom-
plexation procedure may be effective in purifying
phenanthrolines without having to resort to column
chromatography.
Results and discussion
In order to synthesize the skeleton of phenanthroline ligands, in
our laboratories we commonly employ the procedure described by
Bernhard [38] and co-workers. This involves the reaction of an
unsaturated aldehyde with a suitably substituted 8-aminoquino-
line in 70% H SO , employing sodium iodide as a catalyst. This pro-
a-b
2
4
Based on our previous experience with Ar-BIAN (Ar-BIAN = bis-
aryliminoacenaphthene) type ligands [31–34], we considered that
the formation of a (phenanthroline)ZnCl complex could form part
2
cedure is in turn adapted from one previously reported for the
synthesis of quinolines [39,40] and has the advantage of avoiding
highly toxic arsenic compounds as in the classical procedure. In
our hands, the procedure reliably worked in many cases, but the
crude compound is always dark brown and a chromatographic
purification is needed to obtain a colorless compound, as also
⇑
Corresponding author.
E-mail address: fabio.ragaini@unimi.it (F. Ragaini).
https://doi.org/10.1016/j.tetlet.2020.152080
040-4039/Ó 2020 Elsevier Ltd. All rights reserved.
0
Please cite this article as: F. Ferretti and F. Ragaini, An effective non-chromatographic method for the purification of phenanthrolines and related ligands,
Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2020.152080

2
F. Ferretti, F. Ragaini / Tetrahedron Letters xxx (xxxx) xxx
reported in the original paper. As an alternative, in one case we
also employed m-nitrobenzenesulfonic acid as the oxidant.
[
15,41] This is also an effective reagent, but its use does not solve
the problem of the insufficient purity of the product obtained by a
simple acidic extraction/back-extraction procedure (Scheme 1).
In this work, we employed the synthesis of 4-methyl-1,10-
1
2
3
phenanthroline (4-MePhen, R = Me, R = R = R = H in Scheme 1)
1
to optimize our purification strategy. The H NMR spectrum of the
product obtained by employing NaI as a reagent, after the standard
acidic extraction/back extraction procedure, is shown in Fig. 1a.
Apart from the main peaks, due to the desired product, many smal-
ler intensity peaks are clearly visible.
The first aspect that needed attention is the choice of the sol-
vent for the complexation step. It should be polar enough to dis-
2
solve ZnCl , but should also not coordinate strongly to zinc or
dissolve the final complex. It should also readily dissolve organic
impurities. Ethylene glycol was found to give best results, with
acetic acid also being employable (methanol and other alcohols
2
do not dissolve ZnCl efficiently even when hot, Scheme 2).
We performed the complexation step under different condi-
tions in order to maximize the yield of the complex, while leaving
most of the impurities in solution. The best strategy involved mix-
ing ethylene glycol solutions of the phenanthroline and of ZnCl
2
at
~50 °C and then heating at 100 °C to complete the coordination
step, followed by slow cooling. The solid complex was separated
by filtration on a Buchner funnel at room temperature and sub-
jected to a further purification step by heating it in fresh ethylene
glycol. This step cannot be considered a true recrystallization
because the complex is poorly soluble even in the hot solvent,
but still succeeded in removing some additional impurities, as
inferred from the color of the solvent after the procedure.
2
To remove the ZnCl moiety from the phenanthroline ligand, we
initially tested the same conditions employed by us for the analo-
gous step in the synthesis of Ar-BIAN and Alk-BIAN ligands. How-
ever, when the same protocol was applied to the present case, an
even less soluble precipitate and no free ligand was obtained.
The identity of the precipitate was not investigated in detail,
Fig. 1. (a) 1H NMR spectrum of 4-MePhen as obtained from the reaction employing
NaI, after acidic workup. (b) H NMR spectrum after the purification procedure
described in this paper.
1
2 4
although the formation of [Zn(Phen)(C O )] is a reasonable possi-
bility. In any case, oxalate is clearly not suitable to remove zinc
in this case.
Complete decomplexation of zinc from the phenanthroline
could on the other hand be obtained by substituting the oxalate
solution with a concentrated aqueous ammonia solution, where
2
+
zinc cations are present in the form of [Zn(NH
3
)
4
]
[42]
Figure Scheme 3 (Scheme 3).
1
Scheme 2. Formation of the zinc complex.
The H NMR spectrum of the so obtained product is shown in
Fig. 1b. It is immediately evident that the previously observable
extra signals are no longer present. The compound is also analyti-
cally pure, but is not completely colorless. The solid is off-white,
but its solutions still show a tan color that may constitute a prob-
lem if the optical or optoelectronic properties of its complexes
must be investigated. For synthetic use, however, the purity of
the compound is clearly sufficient [43].
The global yield of the synthesis + purification was only 60%. In
order to test if the moderate yield of the purified compound is due
to a lack of selectivity in the synthesis or due to losses in the purifi-
Scheme 3. Decomplexation procedure.
cation, the efficiency of each step of our protocol was tested on
pure 1,10-phenanthroline, where the yield may possibly be 100%.
The [ZnCl (Phen)] complex was obtained in 98% yield following
2
the same procedure employed for 4-MePhen. The ‘‘recrystalliza-
tion” step allowed recovery of 95% of the complex, whereas the
decomplexation step proceeded in essentially quantitative yield.
Thus, the overall efficiency of the procedure was 93%, with the
recrystallization step being the weaker point of the protocol.
Scheme 1. General synthetic strategy employed for the synthesis of
phenanthrolines.
Please cite this article as: F. Ferretti and F. Ragaini, An effective non-chromatographic method for the purification of phenanthrolines and related ligands,
Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2020.152080

F. Ferretti, F. Ragaini / Tetrahedron Letters xxx (xxxx) xxx
3
ring. However, most complex ligands are usually prepared by mul-
tistep procedures requiring the intermediate formation of simpler
derivatives and the protocol here described could still be useful at
this stage.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgments
Scheme 4. Synthesis of quinoline-guanidine ligands.
This work was supported by the Ministero dell’Università e
della Ricerca (MIUR) (PRIN 20154X9ATP).
The synthesis of 4-MePhen was also performed employing m-
nitrobenzenesulfonic acid as co-oxidant. The corresponding spec-
tra are reported in the ESI. Again, many impurities were present
in the product after the acidic work-up, which could be eliminated
References
[1] A. Bencini, V. Lippolis, Coord. Chem. Rev. 254 (2010) 2096–2180.
[
[
[
2] E. Schoffers, Eur. J. Org. Chem. (2003) 1145–1152.
3] G. Chelucci, R.P. Thummel, Chem. Rev. 102 (2002) 3129–3170.
4] W.W. Brandt, F.P. Dwyer, E.D. Gyarfas, Chem. Rev. 54 (1954) 959–1017.
2 3
by the ZnCl /NH treatment. The global yield was very close to that
obtained with the previously described method (61%), but the pro-
duct was of a lighter color at the beginning and was almost color-
less after the purification strategy. Although this is an advantage, it
should be noted that m-nitrobenzenesulfonic acid and the prod-
ucts derived from its reactions have a strong tendency to generate
foams during the extraction/back-extraction steps, making the
separation more difficult [44].
The same protocol was then applied to the synthesis of 3,4-
2
dimethylphenanthroline (3,4-Me Phen) with good results (52%
overall yield of the purified material).
To test if the protocol may be useful in the case of other chelat-
ing ligands, it was employed for the purification of two quinoline-
guanidine hybrid ligands. In the original article, the authors
reported that these ligands were obtained in a pure form without
any chromatographic workup [45]. Although this may be due to
high purity of the reagents used, when we performed the same
synthesis (Scheme 4) with commercially available 8-aminoquino-
line, a greenish product was obtained that required a chromato-
graphic purification step to give the pure yellow compound [15].
The only variation in our procedure is that the intermediate chlo-
roformamidiunium salt was prepared by the reaction of 1,3-
dimethyl-2-imidazolidinone or tetramethylurea with oxalyl chlo-
ride [46], instead of employing highly toxic phosgene as reported
in the previously published account.
[5] A. Togni, L.M. Venanzi, Angew. Chem., Int. Ed. Engl. 33 (1994) 497–526.
[6] P.G. Sammes, G. Yahioglu, Chem. Soc. Rev. 23 (1994) 327–334.
[
[
[
7] D. Wang, A.B. Weinstein, P.B. White, S.S. Stahl, Chem. Rev. 118 (2018) 2636–
679.
8] E. Gallo, F. Ragaini, S. Cenini, F. Demartin, J. Organomet. Chem. 586 (1999) 190–
2
195.
9] M. Gasperini, F. Ragaini, C. Cazzaniga, S. Cenini, Adv. Synth. Catal. 347 (2005)
105–120.
[
[
[
10] M. Gasperini, F. Ragaini, C. Remondini, A. Caselli, S. Cenini, J. Organomet. Chem.
690 (2005) 4517–4529.
11] M.A. EL-Atawy, D. Formenti, F. Ferretti, F. Ragaini, ChemCatChem 10 (2018)
4707–4717.
12] D. Formenti, F. Ferretti, F. Ragaini, ChemCatChem 10 (2018) 148–152.
[13] M. Rimoldi, F. Ragaini, E. Gallo, F. Ferretti, P. Macchi, N. Casati, Dalton Trans. 41
2012) 3648–3658.
14] F. Ferretti, F. Ragaini, R. Lariccia, E. Gallo, S. Cenini, Organometallics 29 (2010)
465–1471.
[15] F. Ferretti, E. Gallo, F. Ragaini, J. Organomet. Chem. 771 (2014) 59–67.
(
[
1
[16] F. Ferretti, E. Gallo, F. Ragaini, ChemCatChem 7 (2015) 2241–2247.
[17] M.A. El-Atawy, F. Ferretti, F. Ragaini, Eur. J. Org. Chem. (2017) 1902–1910.
[18] M.A. EL-Atawy, F. Ferretti, F. Ragaini, Eur. J. Org. Chem. (2018) 4818–4825.
[19] M.A. EL-Atawy, F. Ferretti, F. Ragaini, Eur. J. Org. Chem. (2019) 4968.
[20] W.C. Ho, K. Chung, A.J. Ingram, R.M. Waymouth, J. Am. Chem. Soc. 140 (2018)
7
48–757.
21] A.J. Ingram, K.L. Walker, R.N. Zare, R.M. Waymouth, J. Am. Chem. Soc. 137
2015) 13632–13646.
[22] D.M. Pearson, N.R. Conley, R.M. Waymouth, Organometallics 30 (2011) 1445–
453.
[
(
1
[
23] A. Meduri, D. Cozzula, A. D’Amora, E. Zangrando, S. Gladiali, B. Milani, Dalton
Trans. 41 (2012) 7474–7484.
The purification protocol described herein worked well even for
these ligands, indicating that it is not limited to phenanthrolines
and may be of more general use. It should be noted that TMGqu
is an oil when non-coordinated. Thus it could not be purified by
[24] J. Durand, E. Zangrando, M. Stener, G. Fronzoni, C. Carfagna, B. Binotti, P.C.J.
Kamer, C. Muller, M. Caporali, P.W.N.M. van Leeuwen, D. Vogt, B. Milani, Chem.
Eur. J. 12 (2006) 7639–7651.
[
[
[
[
25] A. Scarel, B. Milani, E. Zangrando, M. Stener, S. Furlan, G. Fronzoni, G. Mestroni,
S. Gladiali, C. Carfagna, L. Mosca, Organometallics 23 (2004) 5593–5605.
26] B. Milani, A. Anzilutti, L. Vicentini, A.S. Santi, E. Zangrando, S. Geremia, G.
Mestroni, Organometallics 16 (1997) 5064–5075.
2
crystallization, whereas its ZnCl complex is a crystalline solid.
27] P. Wehman, P.C.J. Kamer, P.W.N.M. van Leeuwen, Chem. Commun. (1996)
217–218.
Conclusion
28] P. Wehman, V.E. Kaasjager, W.G.J. Delange, F. Hartl, P.C.J. Kamer, P.W.N.M. van
Leeuwen, J. Fraanje, K. Goubitz, Organometallics 14 (1995) 3751–3761.
We have described a purification protocol for phenanthrolines
and related chelating ligands that takes advantage of the low solu-
bility of their complexes with respect to that of the non-chelating
or weakly chelating byproducts, which would not be separated by
an acidic work-up. The procedure can be applied to multi-gram
preparations only requiring inexpensive and non-toxic reagents
and limited amounts of solvents. The method also worked for
quinoline-guanidine hybrid ligands, showing that it can be of more
general applicability. There are obviously also limitations. The
method is not applicable to phenanthrolines containing hydrolyti-
cally sensitive groups (we could not apply it to 4-methoxyphenan-
throline) and is probably of lesser use when the synthesis of the
phenanthroline involves the functionalization of a pre-existing
phenanthroline rather than the creation of a new heteroaromatic
[29] B.V. Popp, J.L. Thorman, S.S. Stahl, J. Mol. Catal. A: Chem. 251 (2006) 2–7.
[
30] S.S. Stahl, J.L. Thorman, R.C. Nelson, M.A. Kozee, J. Am. Chem. Soc. 123 (2001)
188–7189.
31] M. Gasperini, F. Ragaini, S. Cenini, Organometallics 21 (2002) 2950–2957.
7
[
[32] M. Gasperini, F. Ragaini, Organometallics 23 (2004) 995–1001.
[
33] M. Gasperini, F. Ragaini, E. Gazzola, A. Caselli, P. Macchi, Dalton Trans. (2004)
376–3382.
34] F. Ragaini, S. Cenini, S. Tollari, J. Mol. Catal. 85 (1993) L1–L5.
3
[
[35] F. Ragaini, M. Gasperini, P. Parma, E. Gallo, N. Casati, P. Macchi, New J. Chem.
30 (2006) 1046–1057.
[
36] F. Ragaini, M. Gasperini, E. Gallo, P. Macchi, Chem. Commun. (2005) 1031–
033.
1
[
37] M. Viganò, F. Ferretti, F. Ragaini, P. Macchi, Inorg. Chim. Acta 483 (2018) 305–
309.
[
[
[
38] P. Belser, S. Bernhard, U. Guerig, Tetrahedron 52 (1996) 2937–2944.
39] C. O’Murchu, Synthesis (1989) 880–882.
40] J.T. Hewitt, T.F. West, Stafford Allen and Sons, Ltd., US, 1944.
[41] W.P. Utermohlen, J. Org. Chem. 8 (1943) 544–549.
Please cite this article as: F. Ferretti and F. Ragaini, An effective non-chromatographic method for the purification of phenanthrolines and related ligands,
Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2020.152080

4
F. Ferretti, F. Ragaini / Tetrahedron Letters xxx (xxxx) xxx
[
[
42] Use of ammonia is not suggested for standard Schiff bases, since basic work to optimize the reagents amounts for this synthesis and the same ratios
conditions may cause hydrolysis of the ligand.
and reaction time employed for the NaI-based procedure were employed.
However, it is clear that margin for improvement exists.
[45] A. Hoffmann, J. Börner, U. Flörke, S. Herres-Pawlis, Inorg. Chim. Acta 362
(2009) 1185–1193.
43] Use of decolorizing carbon or reduction by an hydridic reagent (NaH was best
in our hands) allowed an even clearer compound to be obtained at the expense
of some product loss, but we did not optimize these strategies.
[
44] Some unreacted 8-aminoquinoline was present among the ‘‘impurities” when
m-nitrobenzenesulphonic acid was employed. It was not the purpose of this
[46] M.A. Bailén, R. Chinchilla, D.J. Dodsworth, C. Nájera, J. Org. Chem. 64 (1999)
8936–8939.
Please cite this article as: F. Ferretti and F. Ragaini, An effective non-chromatographic method for the purification of phenanthrolines and related ligands,
Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2020.152080