Pyridine-derived tetrapodal ligands with NO4 and NN4 donor sets

Article abstract of DOI:10.1515/znb-1998-0903

The Mg²⁺ assisted synthesis of a pyridine-derived tetraalcohol ligand with an NO₄ donor set is described. 2,6-Diethylpyridine reacts cleanly with aqueous formaldehyde solution in the presence of 1 equivalent of MgSO₄ hydrate in a pressurised vessel to give the quadruply hydroxymethylated product 2,6-C₅H₃N[CMe(CH₂OH)₂]₂ (1) as a crystalline solid. Two alkali/alkaline earth metal perchlorate adducts of 1 have been structurally characterised, viz. [(1)₂ ? LiClO₄] (6) and [(1)₂ ? Ba(ClO₄)₂] (7). The ligand adopts a bridging coordination mode in both 6 (distorted tetrahedral coordination of Li⁺) and 7 (square prismatic coordination of Ba²⁺). The further derivatization of 1 leads to the tetratosylate (2) and the tetraazide (3), both of which have been obtained in pure form for the first time. Reduction of 3 gives the pentaamine ligand 2,6-C₅H₃N[CMe(CH₂NH₂) ₂]₂ (4), isolated as the tetrakis(hydrobromide) salt 4 ? 4 HBr. The presence of four ammonio substituents and an unprotonated pyridine nitrogen atom in the solid state has been unequivocally established by an X-ray structural analysis.

Full text of DOI:10.1515/znb-1998-0903

Pyridine-Derived Tetrapodal Ligands with N 04 and NN4 Donor Sets
Stefan Schmidt3, Laurent Omnèst,a, Frank W. Heinemann^3, Jörg KuhnigkI,b,
Carl Krügeri,b, Andreas Grohmann*’3
a Institut für Anorganische Chemie, Universität Erlangen-Nürnberg,
Egerlandstraße 1, D-91058 Erlangen, Germany
b Max-Planck-Institut für Kohlenforschung,
Kaiser-Wilhelm-Platz 1, D-45470 Mülheim an der Ruhr, Germany
Z. Naturforsch. 53 b, 946-954 (1998); received May 8 , 1998
Template Reaction, Tetrapodal Pentadentate Ligand, Pentaamine, Hydrobromide
The Mg2+ assisted synthesis of a pyridine-derived tetraalcohol ligand with an NO4 donor
set is described. 2,6-Diethylpyridine reacts cleanly with aqueous formaldehyde solution in
the presence of 1 equivalent of MgSCU hydrate in a pressurised vessel to give the quadruply
hydroxymethylated product 2 ,6 -C5H3N[CMe(CH2 0 H)2]2 (1) as a crystalline solid. Two al-
kali/alkaline earth metal perchlorate adducts of 1 have been structurally characterised, viz. [(1)2
• LiClCU] (6) and [(1)2 • Ba(C1 0 4 )2] (7). The ligand adopts a bridging coordination mode in
both 6 (distorted tetrahedral coordination of Li+) and 7 (square prismatic coordination of Ba2+).
The further derivatization of 1 leads to the tetratosylate (2) and the tetraazide (3), both of which
have been obtained in pure form for the first time. Reduction of 3 gives the pentaamine ligand
2,6-C5H3N[CMe(CH2NH2)2]2 (4), isolated as the tetrakis(hydrobromide) salt 4 * 4 HBr. The
presence of four ammonio substituents and an unprotonated pyridine nitrogen atom in the solid
state has been unequivocally established by an X-ray structural analysis.
Introduction
X = OH:
1
X - OTos: 2
With a view to creating a concave ligand en-
vironment for octahedrally coordinating transition
metal ions, we recently introduced a tetrapodal pen-
tadentate ligand with an NN4 donor set (4) and pre-
sented crystal structures of a series of cobalt(III) and
nickel(II) complexes in which the ligand provides a
square pyramidal coordination cap [1]. This ligand
represents a novel structural motif in coordination
chemistry in that one central donor atom is sur-
rounded by four equivalent pendent donor groups,
the former being set up to occupy the apical posi-
tion of a coordination octahedron, while the latter
take the equatorial positions [2]. A different type of
square-pyramidally coordinating pentadentate lig-
and of Ci symmetry involves a 'superstructured'
porphyrin derivative of considerable architectural
complexity [3, 4].
X = N3:
3
4
X = NH2:
The synthesis of 4 ('N(NH2)4') requires the four-
fold functionalization of 2 ,6 -diethylpyridine and
proceeds in four stages. The initial step exploits
the known reactivity of ortho alkylpyridines to-
wards formaldehyde to furnish a tetraalcohol pre-
cursor ('N(OH)4', 1) which is then transformed
into the pentaamine ligand by fourfold tosylation
('N(OTos)4', 2), tosylate-azide exchange ('N(N3)4',
3), and reduction [la]. Problems inherent in a
polyfunctionalization of this kind are associated
chiefly with the yields of the individual steps and
the purity of the intermediates. While polytosy-
lates [5, 6 ] and polyazides [7, 8] may be prepared
in satisfactory yields from polyalcohols by litera-
ture methods, exhaustive hydroxymethylations of
ortho alkylpyridines using formaldehyde (cf step
1 above) generally give low yields and mixtures
of products [9 - 12]. Our elaboration of the above
reaction sequence therefore concentrated on the
first step, in addition to preparing pure samples
of every intermediate. We now report an improved
* Reprint requests to Dr. A. Grohmann;
E-mail: grohmann@anorganik.chemie.uni-erlangen.de;
1Visiting from Universite de Rennes, Chimie, Campus
de Beaulieu, 35042 Rennes Cedex, France;
1 X-ray structure analysis.
K
0932-0776/98/0900-0946 $ 06.00 © 1998 Verlag der Zeitschrift für Naturforschung, Tübingen • www.znaturforsch.com
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
947
synthesis of the tetraalcohol 1 , the crystal struc-
tures of its lithium and barium perchlorate adducts,
and structural details of the methanol solvate of the
pentaamine tetrakis(hydrobromide), 4 * 4 HBr •
CH3OH.
5
could not be induced to crystallise. Purification and
isolation of a crystalline solid were achieved by
HPLC (cf Experimental) but did not appear to be
practical for larger quantities of material.
Results and Discussion
The q hydrogen atoms of 2-alkyl and/or 4-alkyl
substituents on a pyridine ring are sufficiently acidic
to allow replacement with hydroxymethyl groups
through the action of formaldehyde under appropri-
ate conditions [13, 14]. The tetraalcohol 'N(OH)4'
1 may thus be prepared from 2 ,6-diethylpyridine in
a pressurised vessel at 135 °C. The reaction is es-
sentially a Knoevenagel reaction where aldehydes
or ketones R-C(=0)-R' (R' = H, alkyl) are treated
with compounds of the form Z-CH2-Z' to give
compounds of the type R-C(OH)R'-CHZZ' (R' =
H, alkyl). These may subsequently eliminate wa-
ter with concomitant formation of a double bond
[15]. In our case, elimination of water from partially
hydroxymethylated intermediates constitutes a de-
composition reaction which competes with product
formation, leading to undesired side products such
as la [10b], which may be one of the species giving
rise to olefinic resonances in the 1H NMR spectra
of crude 1. The reaction is complicated further by
the fact that conversion is incomplete under stan-
dard conditions, giving a mixture of products with
varying degrees of hydroxymethylation in the side
chains [12]. We therefore set out to optimise the
synthesis of the tetraalcohol 1 .
In view of the extensive coordination chemistry
of carbohydrates [21 - 23] and other polyalcohols
with alkali and alkaline earth cations [24], complex
formation of 1 with suitable metal salts suggested
itself as an alternative means of purification. A se-
ries of experiments employing salts such as LiClÜ4,
Na2S 04, N aN 03, CH3COOK, M gS04, Mg(C104)2,
Ca(N0 3)2, Ba(C104)2 or Ba(SCN)2 were un-
successful, giving resinous materials throughout
[25 - 27], However, solutions of crude 1 in CHC13
which had been treated with Na2S0 4 showed better
resolved signals in their 1H NMR spectra. This ob-
servation suggested a certain degree of complex for-
mation and led us to explore the possibility of using
metal ions to preorganise the reactants in the prepa-
ration of the tetraalcohol 1. In addition to a possible
template effect, we expected the presence of a metal
cation of suitable charge/radius ratio to increase the
electrophilic character of the aldehyde component
through complexation. At the same time, the anion
was to be unreactive under the given conditions. For
these reasons, we chose M gS04 as the metal salt
and ran a series of hydroxymethylation reactions
in the presence of varying amounts of this mate-
rial. Optimal results were obtained with a 1 : 1 ra-
tio of 2,6-diethylpyridine and MgSÜ4: The reaction
gave virtually pure tetraalcohol 1 , and yields were
comparable to those of previous reactions without
added metal salt where the reaction time had been
OH OH
1a
Polyalcohols such as 1 cannot be distilled without more than twice as long, at the expense of introduc-
decomposition owing to their high boiling points ing considerable amounts of impurities. Whether
or not the reaction between 2 ,6 -diethylpyridine and
but may be purified by chromatographic methods
[16, 17] usually giving highly viscous liquids or 4 equivalents of formaldehyde in the presence of
syrups [18]. In order to obtain a derivative of 1 1 equivalent of MgS04 in fact proceeds through
which would enable purification by recrystalliza- an intermediate where the magnesium cation co-
tion, we prepared the bis(acetal) 5 from crude 1 ordinates to the pyridine nitrogen atom and helps
and two equivalents of 4-nitrobenzaldehyde with to direct the incoming aldehyde moieties must re-
azeotropic removal of water [19, 20]. Contrary to main an open question at this stage. It should be
our expectations, the material thus obtained was it- noted, however, that such action of an alkaline earth
self an oil (due to the presence of impurities) and cation in the formation of polyalcohols would not be
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

948
St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
Fig. 2. Partial view of the solid state structure of the
pyridine tetraalcohol/barium perchlorate adduct 7. For
clarity, a ball-and-stick representation is shown where
hydrogen atoms are omitted.
Fig. 1. Partial view of the solid state structure of the
pyridine tetraalcohol/lithium perchlorate adduct 6. For
clarity, a ball-and-stick representation is shown where
hydrogen atoms are omitted.
act as a podand in this structure [29]; rather, the
lithium ion asserts its strong preference for tetrahe-
dral coordination, and the overall bonding situation
is reminiscent of the lithium iodide methanol sol-
vate [Li(CH3OH)4]+r (Li-O: 1.919(5) A) [30].
entirely unprecedented, an example being the base-
catalysed condensation of formaldehyde in the pres-
ence of Ca(OH)2 (the “formose reaction”) where the
coordination of intermediates to the calcium ion ap-
pears to be essential [28].
Complexes of barium (M) with aminoalcohols
(L) have been isolated both as 1 : 1 and as 1 :
The specificity with which the tetraalcohol 1 is
produced in the presence of MgSC>4 allows its sub- 2 (M : L) species [18, 26]. In the X-ray structure
sequent isolation in a pure form. After aqueous of the 1 : 2 complex 7, the tetraalcohol 1 again
acts as a bridging ligand (albeit with partial chela-
tion) rather than a podand (Fig. 2). The coordina-
tion geometry of the barium cation is square pris-
matic (coordination number 8). One of the faces
of the prism is formed by the hydroxyl oxygen
atoms of two chelating 1,3-diol units of two dif-
ferent tetraalcohol molecules (Ol, 02, 01 A, 02A).
work-up, a methylene chloride solution of 1 was
treated with 4 A molecular sieves to remove water,
leading to the formation of a white crystalline pre-
cipitate which was identified as the hemihydrate 1 •
0.5 H2O by elemental analysis. The 'H NMR spec-
trum shows excellent resolution of signals including
a three-line pattern for the hydroxyl protons.
The availability of pure 1 led us to reinvesti- The opposite face is made up of two diagonally
gate its complex formation with a variety of alkali disposed hydroxyl oxygen atoms of two further
tetraalcohol molecules (04B, 04C) and two oxy-
gen atoms of two terminally coordinated perchlo-
rate anions (06, O6 A). Each tetraalcohol molecule
chelates one barium ion through one of its 1,3-diol
units (Ol, 02) and coordinates to a second bar-
ium ion through one hydroxyl group of its other
1,3-diol unit (04), which also forms a hydrogen
bond to one of the non-coordinated perchlorate oxy-
gen atoms (07B; 0 4 07B = 2.83(1) A; 04 07B
corresponds to 04B 07). The second hydroxyl
and alkaline earth metal salts, such as LiC104 • 3
H20, Na2S 0 4, Mg(C104)2 • 6 H20 , Ca(N 03)2 • 4
H20, Ba(SCN)2 • 3 H26 , and Ba(C104)2; ZnBr2
and NiCl2 • 6 H2O were also tested. Solid state
structures have so far been determined for [(1)2
• LiC104] (6) and [(1)2 • Ba(C104)2] (7). In the
case of 6 (Fig. 1), the lithium ions are coordinated
in a distorted tetrahedral fashion by four hydroxyl
groups, each belonging to a different tetraalcohol
molecule (Li-O: 1.88(1) - 2.02(2) Ä). In each tetraal-
cohol molecule, one hydroxyl group of each 1,3- group (03) remains uncoordinated but forms a bi-
furcated hydrogen bond, recipients being a barium-
coordinated hydroxyl oxygen atom within the same
tetraalcohol molecule (0 2 ) and one of the hydroxyl
oxygen atoms of the chelating 1,3-diol functionality
diol functionality is coordinated to a lithium ion
while the other hydroxyl group remains uncoordi-
nated, as does the pyridine nitrogen atom; hydrogen
bonds are not formed. Consequently, 1 does not
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

949
St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
atom in the 4-position of the pyridine ring lying
on the two-fold rotation axis. In each 1,3-ditosylate
sidearm the tosyl groups point away from each other
in order to minimise steric congestion. Two tosy-
late phenyl rings on diagonally opposite oxygen
atoms (with respect to the N 04 set of heteroatoms)
of another tetraalcohol molecule (02B; 0 2 03 =
2.929(4) A; 02B 03 = 2.896(4) Ä). The pyridine
nitrogen atom is uncoordinated. The barium ions
and tetraalcohol molecules form rings of the type
M2L2, which are linked into infinite arrays at the
barium centers. The Ba-OH distances (2.654(3) -
2.823(3) A) fall within the range observed for other approach the pyridine ring in face-to-face fashion
alcohol complexes of barium perchlorate [26, 31], from either side, which suggests the presence of
weak intramolecular tt-tt stacking interactions.
Nucleophilic displacement of the tosyl groups
in 2 by azide ion leads to the tetraazide 'N(N3)4'
3 which may be obtained as a pure material (by
!H NMR) when starting from pure tetratosylate
2. Subsequent reduction of 3 with triphenylphos-
phine and aqueous workup gives the pentaamine
ligand 4 which is conveniently isolated as the hy-
drochloride salt [1]. The use of pure tetraazide
3 gives virtually pure reduction product 4. The
tetrakis(hydrochloride) adduct 4 * 4 HC1 is ob-
while the comparatively short distance Ba-0 C103
(2.790(5) A) indicates stronger binding of the per-
chlorate anions in 7 than, e.g., in the barium perchlo-
rate complex of triethanolamine (2.941(3) A and
2.989(3) A) [26], The 'H NMR spectra of 7 show no
complexation-induced shifts suggesting weak coor-
dination (if any) of the tetraalcohol ligand to the
barium ion in solution. Pending an X-ray structural
study, however, podand-like coordination of 1 can-
not be ruled out for smaller alkaline earth cations
such as magnesium.
The further derivatization of 1 requires the in- tained by treatment of an ethanolic solution of 4
troduction of tosyl groups to allow nucleophilic with concentrated aqueous HC1, while prolonged
reaction with HC1 leads to the formation of the
and pentakis(hydrochloride) salt 4 * 5 HC1. In both
4 equivalents of toluene-4-sulfonyl chloride in pyri-
dine gives good yields provided that the reaction is by rhodanometric titration. This fractional precip-
carried out at sufficiently low temperatures (- 5 to 0
displacement in a subsequent step. The prepara-
tion of the tetratosylate 'N(OTos)4' from
2
1
cases, the chloride equivalents were determined
itation is reminiscent of tris(2 -aminoethyl)amine
°C). Higher temperatures have been found to lead to (tren) where the hydrochloride salts tren • 3 HC1
reduced yields in related reactions, and it has been and tren • 4 HC1 are also separately accessible [32,
suggested that this is due to the formation of quater- 33]. In both tren and the pentaamine ligand 4, the
nary pyridinium tosylates [5], If pure tetraalcohol 1 central nitrogen atom is expected to be less basic
is used in the synthesis, the resulting tetratosylate in aqueous solution than the primary amino func-
2 is of high purity, whereas employment of crude tionalities, due to the inability of a tertiary amine to
1 introduces considerable impurities into the prod- disperse charge to the solvent by hydrogen bonding
uct which persist during work-up. Recrystallization
[34]. In the case of 4 • 4 HC1, the pyridine ni-
from ethanol is feasible but works well only with trogen atom should therefore remain unprotonated.
small quantities of impure 2 , as larger amounts tend While no suitable single crystals could be obtained
to lead to the separation of syrupy phases. Since of this adduct, an X-ray structural analysis of the
we were interested in the spatial orientation of the tetrakis(hydrobromide) salt methanol solvate 4 * 4
HBr • CH3OH supports this conclusion.
1,3-difunctionalised sidearms of the molecule, we
undertook an X-ray structural study of the tetrato-
sylate. Single crystals were obtained from a mix-
ture of ethanol, diethyl ether, methylene chloride,
and water (40 : 40 : 15 : 5), but were found to be
of moderate quality and to diffract poorly. While
the obtained data set did allow the solution of the
structure, refinement could not be achieved beyond
values of R\ ~ 0.10 and xvR2 « 0.30 which are
unacceptably high. The preliminary data show the
molecule to have C2 symmetry in the solid state,
with the pyridine nitrogen atom and the carbon
The structure contains
[2,6-C5H3N{CMe(CH2NH3)2}2]4+ cations, four
Br- counterions, and one molecule of methanol per
formula unit as lattice solvent. The cation (Fig. 3)
has no crystallographically imposed symmetry. Its
conformation with respect to the orientation of the
ammonio substituents, while irregular, minimises
the electrostatic repulsion between these groups
[35]. While the pyridine ring is largely undistorted,
the angles subtended by the ortho substituents at
C l 1 and C15 deviate more markedly from the ideal
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

950
St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
Experimental Section
CAUTION! Although no problems were encountered
in this work, organic polyazides are potentially explosive
and should be handled with due precautions.
Materials and instrumentation.
All manipulations were performed under dry dinitro-
gen using standard Schlenk techniques unless indicated
otherwise. Solvents were dried and distilled before use.
Reagents were AR grade or better and were purchased
from Merck, Fluka, and Aldrich. HPLC was carried out
using a Knauer instrument (HPLC pump 64; UV/VIS
filter photometer, A = 254 nm; column: Nucleosil 120
C l8 , 250 x 32 mm, 5 /im). NMR and IR spectra were
recorded on JEOL JNM-EX 270, JEOL JNM-LA 400,
and Perkin Elmer 16PC FT-IR instruments, respectively.
Mass spectra were obtained on a JEOL MSTATION 700
spectrometer. Elemental analysis was carried out using a
Carlo Erba Elemental Analyzer 1106.
Fig. 3. Molecular structure of the cation
[2 ,6 -CsH3N{CMe(CH2NH3)2 }2]4+ in the tetrakis(hydro-
bromide) salt 4 * 4 HBr • CH3OH (thermal ellipsoids at
the 50 % probability level). See text for the treatment of
the hydrogen atoms.
value of 120°, resulting in a slight displacement of
the l,3-diammonio-2-methyl-prop-2-yl groups to-
wards the pyridine nitrogen atom. The sidearms
show the usual conformation of a paraffinic chain
with unexceptional bond distances (d(C-N) = 1.49 -
1.51 A; d(C-C) = 1.52 - 1.56 A) and angles (105.4 -
114.3°) [36 - 38]. All hydrogen atoms were identi-
fied as residual electron density in the correct tetra-
hedral disposition around the carbon and, in par-
ticular, the primary amino nitrogen atoms. In order
to prevent unreasonably short C-H and N-H dis-
tances, however, the H atoms were allowed for as
riding atoms in geometrically optimised positions.
No electron density maximum in a reasonable posi-
tion could be located at the pyridine nitrogen atom
(cf Experimental). These findings, together with
the number of counterions, provide clear evidence
for the aromatic nitrogen atom to be unprotonated.
Further, the structure is characterised by a complex
network of hydrogen-bonded N Br contacts (3.30
- 3.37 A; sum of the van der Waals radii for N
and Br: 3.40 A [39]) emanating from the ammo-
nio but not from the pyridine nitrogen atoms. The
methanol molecule also participates in hydrogen
bonding to the ammonio groups and the bromide
counterions. From the present structure determina-
tion of 4 • 4 HBr it is obvious that the amino-
2,
6
-C
5
H3N{CMe(CH
2
OH)2} 2 (1)
The preparation was carried out largely in air. An au-
toclave was charged with 2,6-diethylpyridine (34 g, 0.25
mol), MgS0 4 hydrate (42.3 g, 0.25 mol), and aqueous
formaldehyde solution (37 %, 203 ml, 2.7 mol), and the
mixture heated to 135 °C for 20 h. After cooling to room
temperature, the yellow solution was decanted from solid
MgS0 4 and filtered. Volatiles were removed in an oil
pump vacuum ( 1 2 h, 1 0 0 °C) and the remaining yellow
syrup taken up in chloroform (300 ml). Residual MgS0 4
was removed by filtration, and the filtrate extracted with
water (4 x 200 ml). The aqueous extract was taken to
dryness on a rotary evaporator, the residue dissolved in
methylene chloride (150 ml), the solution dried briefly
over Na2 S0 4 , filtered, and stirred with molecular sieves
(4 A, residue/mol. sieves « 1 :1 v/v) under nitrogen. Af-
ter ca. 30 min, the solution began to precipitate copious
amounts of a white solid. Stirring was continued for 12 h,
after which time the suspension was decanted onto a glass
frit (porosity G4). The solid was collected, washed with
a small amount of cold methylene chloride, and dried in
vacuo to give 1 as a white, crystalline, very hygroscopic
material (5.5 g). Evaporation of the filtrate gave more 1 as
a yellow syrup containing minor impurities (by 1H NMR;
substituted sidearms of the pentaamine 4 are reori- 22.8 g; combined yield: 28.3 g, 44 %).
ented considerably upon coordination to a single
metal ion [1].
C13H 22NO4.5 (1 *0.5 H20 )
Calcd C 59.07 H 8.39 N 5.30 %,
Found C 58.67 H 8.34 N 5.22 %.
The synthetic procedure described above thus al-
lows the preparation of uncontaminated pentaamine
ligand 4, and the synthesis of multifunctional lig-
ands on the basis of 4 is currently in progress.
iW c m “ 1 3348 s, 2944s, 2875s, 1588s, 1577s, 1456s,
1018s, 814m, 752m (KBr). <5H(DMSO-J6) 7.63 (AB2 spin
system, 3 lines, 1 H, H4), 7.21 (AB2, 2 lines, 2 H, H?5),
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
951
4.66 (3 lines, 4 H, CH20//), 3.63 (4 lines, C//2OH, 8 H), (s(br), 12 H, NH3+), 7.87 (AB2 spin system, 3 lines,
1.18 (s, CH3, 6 H). 6c (DMSO-J6; all singlets) 163.12 (py- 1 H, H4), 7.50 (AB2, 2 lines, 2 H, H3'5), 3.59 (d(br),
C2/6), 136.47 (py-C4), 119.10 (py-C3/5), 66.53 (-CH2-), 27(HH) = 8.1 Hz, -C//H-NH2, 4 H), 3.09 (d(br), l/(HH)
= 8.1 Hz, -CH//-NH2, 4 H), 1.56 (s, CH3, 6 H). An at-
tempt to grow single crystals from methanol led to the
formation of a 1 : 1 methanol solvate. 6c (DMSO-^;
all singlets) 158.04 (py-C2/6), 139.38 (py-C4), 121.84
(py-C3/5), 48.66 (CH3OH), 45.72 (>C<), 42.89 (-CH2-),
20.39 (-CH3). Prolonged action of HC1 resulted in the
formation of the pentakis(hydrochloride) salt 4 • 5 HC1.
47.07 (>C<), 19.74 (-CH3).
2.6 -C5 HjN { CMe(CH2OTs)2} 2 (2)
This compound was prepared as described previously
[1], Careful adjustment of the temperature during the
addition of toluene-4-sulfonyl chloride gave improved
yields of up to 68 % [5]. A pure sample of 2 was ob-
tained by recrystallization from ethanol.
C,3H30N5C15 • H20 (4 * 5 HC1 • H20)
Calcd C 34.57 H7.14 N 15.50%,
Found C 34.67 H 6.98 N 15.42%.
C 4 ,H 45N O ,2 S 4
Calcd C 56.47 H 5.20 N 1.61 S 14.71 %,
Found C 56.50 H 5.24 N 1.63 S 14.56 %.
Rhodanometric titration gave 5 eq. of chloride. The
tetrakis(hydrobromide) salt methanol solvate 4 * 4 HBr
• CH3OH was prepared from 4 by reaction with concen-
trated aqueous hydrogen bromide and recrystallisation
from methanol.
i/max/cm-1 1597m, 1456m, 1359s, 1175s, 818s, 553m
(KBr). ÖH (DMSO-dfi) 7.61 (AA'XX', 8 H, C6H2//2S 03),
7.54 (AB2, 3 lines, 1 H, H4), 7.40 (AA’XX’, 8 H,
C6H2//2CH3), 7.10 (AB2, 2 lines, 2 H, H3’5), 4.13 (3 lines,
-CH2-, 8 H), 2.39 (s, C6H4C //3, 12 H), 1.07 (s, CH3, 6
H). 6c (DMSO-d6\ all singlets) 157.94 (py-C2/6), 145.05
(Cl =C-S03-CH2-), 137.76 (py-C4), 131.69 (C4), 130.13
(C3/5), 127.49 (C2/6), 119.50 (py-C3/5), 72.47 (-CH2-),
44.52 (>C<), 21.09 (C6H4CH3), 18.91 (-CH3).
C,3H29N5Br4 • CH3OH (4 * 4 HBr • CH3OH)
Calcd C 27.70 H 5.48 N 11.54 %,
Found C 28.17 H 5.79 N 11.53%.
Rhodanometric titration gave 4 eq. of bromide.
^max/cm-1 3403m (NH str); 3024s (br); 1631m, 1612s
(NH3+ asym def); 1585s, 1554m, 1515s, 1489s (C=C or
C=N str/NH3+ sym def); 1451m (CH2 scissor); 1102m,
1074m (NH3+ rock); 790s, 733m (CH oop def for 2,6-
disubstituted pyridine) (KBr disc). NMR data are for ma-
terial from which methanol was removed by drying in
vacuo. 6h (DMSO-d6) 7.79 (s(br), 12 H, NH3+), 7.96
(AB2 spin system, 3 lines, 1 H, H4), 7.55 (AB2, 2 lines,
2 H, H3-5), 3.53 (m, -C//H-NH2, 4 H), 3.20 (m, -CH//-
NH2, 4 H), 1.56 (s, CH3, 6 H). 6C (DMSO-db\ all sin-
glets) 158.05 (py-C2/6), 139.50 (py-C4), 121.71 (py-
C3/5), 45.51 (>C<), 42.61 (-CH2-), 19.63 (-CH3).
2.6-C5H3N{CMe(CH2N3)2}2 (3)
This compound was prepared as described previously
[1]. Starting from pure tetratosylate 2, the tetraazide could
be obtained nearly quantitatively as a light yellow oil
which was pure by ’H NMR. 2/max /cm-1 2965m, 2102s,
1576m, 1463m, 1263m, 1045m,813m(KBr).<5H(DMSO-
de) 7.77 (AB2spin system, 3 lines, 1H, H4), 7.35 (AB2, 2
lines, 2 H, H3’5), 3.82 (d, 2/(HH) = 12.10 Hz, -C//H-N3,
4 H), 3.68 (d, 27(HH) = 12.10 Hz, -CH//-N3, 4 H), 1.32
(s, CH3, 6 H). 6c (DMSO-<4; all singlets): 6 160.44 (py-
C2/6), 137.60 (py-C4), 119.56 (py-C3/5), 57.37 (-CH2-),
46.16 (>C<), 20.67 (-CH3).
2,6-C5H3N {CMeCH2OCH(C6H4-4-NÖ2)OCH2}2 (5).
2.6-C5H3N{CMe(CH2NH2)2} 2 (4)
Crude tetraol 1 (5.9 g, < 23 mmol), 4-nitrobenzalde-
hyde (6.9 g, 46 mmol), and 4-toluenesulfonic acid hydrate
(10.5 g, 55 mmol) were mixed with benzene (160 ml), and
the mixture refluxed for 4 h while water was azeotropi-
cally removed with a Dean-Stark trap. While still warm,
the resulting two-phase mixture was poured into a so-
lution of K2C 03 (15 g, 109 mmol) in water (200 ml)
with vigorous stirring. The phases were separated, the
organic phase washed with water (2 x 50 ml), and the
aqueous phase extracted with chloroform (2 x 40 ml).
The combined organic phases were dried over MgS0 4 ,
filtered, and taken to dryness to leave a dark brown oil (3
This compound was prepared in pure form from pure
tetraazide 3 by reduction with triphenylphosphine in 85 -
90 % yield [1]. 4 was obtained as a light yellow viscous
oil which was taken up in ethanol and treated with con-
centrated HC1 to precipitate the tetrakis(hydrochloride)
salt 4 • 4 HC1 as a light yellow solid in 70 % yield.
C13H29N5CI4 • 0.75 H20 (4*4 HC1 • 0.75 H20)
Calcd C 38.02 H 7.49 N 17.05 %,
Found C 38.43 H 7.66 N 17.06%.
Rhodanometric titration gave 4 eq. of chloride.
i'max/cm-1 2916s, 1590m, 1510m, 1456m (KBr). The g). Approx. 0.5 g of this material was purified by HPLC
material was pure by H NMR. 6\\ (DMSO-^ö) 8.19 (CH3CN/H20 = 5 : 1). Five fractions were collected, of
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

952
St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
Table I. Crystallographic data for [2 ,6 -C«;I-hN{CMe(CH2NI-h)2 }2](Br)4 • CH3OH ( 4 * 4 HBr • CH3OH), [2,6-
C5 H3N{CMe(CH2OH)2 }2]2 • LiC104 (6 ), and [2 ,6 -C5H3N{CMe(CH2OH)2 }2]2 • Ba(C104 ) 2 (7).
4 • 4 HBr • CH3OH
7
6
Formula
Formula mass [g m oL 1]
Space group
Ci4H33Br4N50
607.09
C26H4iClLiN20p
617.01
C26H42BaCl2N20,6
846.79
P2,/c (no. 14)
monoclinic
10.306(1)
14.845(2)
14.389(2)
90
94.502(9)
90
2194.6(5)
4
P2\2\2 (no. 18)
orthorhombic
12.715(3)
22.168(7)
10.304(4)
90
90
90
2904(2)
4
C2/c (no. 15)
monoclinic
18.612(2)
13.2073(7)
14.864(1)
90
107.083(7)
90
3492.5(5)
Crystal system
a [Ä]
b [A]
c [A]
a [°]
ß[°]
7 [°J
V [A ]
Z
4
2 0 0 (2 )
293(2)
293(2)
H K j
Aa [A]
0.710 73
1.837
7.348
0.0625
0.710 73
1.411
0.198
0.0836
0.710 69
1.611
1.362
0.0447
Dc [g cm
ß [mm ’]
R r
]
wR2c
0.2507
0.1867
0.1336
0.0739/14.792
m/nc
0.1018/0
0 .1 0 0 0 /0
aMoKa, graphite monochromator; bR 1= £ llF 0l- IFcll/VlF0lfor F> 4 a{F )\cwR2 = ( O w(F02- Fc2)2]/^[w (F02)2])0-5,
where w = 1/[ct2(F02) + (mP) 2 + nP] and P = (F02 + 2 Fc~)/3 for all data.
which one yielded pure 5 as a faintly yellow crystalline
solid (145 mg) upon evaporation of the solvent.
had formed was removed by filtration and dried in vacuo.
C 26 H 4 2 L iC lN 2 0 12 • H 2 O
Calcd C 49.18 H 6.98 N4.41%,
Found C 49.30 H 7.06 N4.16%.
C27FI27N3O8
Calcd C 62.18 H 5.22 N 8.06 %,
Found C 62.43 H 5.37 N 7.83 %.
More ether (60 ml) was added to the filtrate and the
solution set aside at room temperature to give single crys-
tals suitable for an X-ray structural analysis after several
days.
^max/cm" 1 1612m, 1534s, 1349s, 1115s, 1090s, 1012s,
853m, 751m, 552w. 6H (CDC13) 8.12 (AA'XX', 4 H,
-C.(CH)2 .(C//)2.C-N02), 7.65 (AB2, 4 lines, 1 H, H4),
7.53, (AA’XX', 4 H, -C.(C//)2.(CH)2.C-N02), 7.50 (AB2,
2 lines, 2 H, H3'5), 5.63 (s, -0-C//Ar-0-), 4.73 (d, 2/(HH)
= 11.7 Hz, -C//H-), 3.94 (d, 2/(HH) =11.7 Hz, -CH//-),
1.16 (s, -CH3). 6 c (CD2CI2 ; all singlets): 6 162.71 (py-
C2/6), 148.47 (Cl = -C-N02), 145.35 (C4), 136.75 (py-
C4), 127.66 (C3/5), 123.63 (C2/6), 119.67 (py-C3/5),
100.48 (-O-CHAr-O-), 75.96 (-CH2-), 40.55 (>C<), 21.85
(-CH3). MS (FD): m/z (relative intensity) 521 (100), M+;
1042(12) [M2]+.
[2,6-C5H3N{CMe(CH2OH)2} 2]2 • Ba(Cl04)2 (7).
A solution of pure tetraol 2 (1.25 g, 4.90 mmol) in ace-
tonitrile (25 ml) was added dropwise to a stirred solution
of Ba(C1 0 4 )2 (0.84 g, 2.50 mmol) in acetonitrile ( 2 0 ml).
The mixture was filtered to remove a slight cloudiness
and stirred at room temperature for 18 h. Ether (50 ml)
was added and the mixture kept at - 18 °C for 48 h after
which time a fine colourless precipitate had formed which
was removed by filtration and dried in vacuo.
[2,6-C5H3N{CMe(CH2OH)2}2]2 • LiCl04 (6).
C26H42BaCl2N20 ,6
A solution of pure tetraol 1 (1.25 g, 4.90 mmol) in a
mixture of CH2CI2 (30 ml) and THF (4 ml) was added
dropwise to a stirred solution of LiClCH • 3 H2O (0.87 g,
5.42 mmol) in a mixture of CH2CI2 (60 ml) and THF (30
ml). The mixture was stirred at room temperature for 48
h. Ether (50 ml) was added and the mixture kept at - 18
Calcd C 36.88 H 5.00 N3.31%,
Found C 36.70 H 5.42 N 2.87 %.
More ether (25 ml) was added to the filtrate and the
solution set aside at room temperature to give single crys-
tals suitable for an X-ray structural analysis after several
°C for several days. The fine colourless precipitate which days.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
953
Table II. Selected bond lengths (A) and angles (°) for [2,6-
C5H3N{CMe(CFbNFh)2 }2](Br)4 • CH^OH(4 • 4 HBr •
CH3OH), [2 ,6 -C5H,N{CMe(CH2OH)2 }2]2 • LiC104 (6 ),
and [2 ,6 -C5H^N{CMe(CH7OH)2 }2]2 • Ba(C104 )2 (7).
coefficient and in order to avoid unreasonably short C-H
and N-H distances, the hydrogen atoms were incorporated
in the final calculation as riding atoms in geometrically
optimised positions. The residual electron density was +
0.92/- 0.88 eA-3. Crystal data and data collection pa-
rameters are listed in Table I, and selected distances and
angles are given in Table II. Data for 6 were measured
on a Nonius MACH 3 diffractometer (MoKa radiation,
graphite monochromator, rotating anode). The structure
was solved by direct methods and refined on F2 using
the program package SHELXTL 5.03. The non-hydrogen
atoms were refined anisotropically. All hydrogen atoms
were positioned geometrically except for the hydroxyl H
atoms which were not included in the calculation. The
residual electron density was + 0.37/- 0.33 eA-3. Crystal
data, data collection parameters, and selected distances
and angles are listed in Tables I and II, respectively. Data
for 7 were measured on an Enraf Nonius CAD4 diffrac-
tometer (MoKa radiation, graphite monochromator). The
structure was solved by direct methods (SHELXS 8 6 )
and refined on F2 using SHELXL 93. The non-hydrogen
atoms were refined anisotropically. All hydrogen atoms
were located in a difference Fourier map, and the residual
electron density was + 1.13/- 1.14 eA- 3. Crystal data and
data collection parameters as well as selected distances
and angles for 4 • 4 HBr • CH3OH, 6 and 7 are given
in Tables I and II, respectively. Crystallographic data (ex-
cluding structure factors) for the structures reported in
this paper have been deposited with the Cambridge Crys-
tallographic Data Centre as supplementary publication
nos. CCDC-101 563 (4 * 4 HBr • CH3OH), CCDC-101
564 (6 ), and CCDC-101 565 (7). Copies of the data can
be obtained free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK [Fax: int. code
+44 (1223) 336-033; E-mail: deposit@ccdc.cam.ac.uk],
Compound 4 * 4 HBr • CH
3
OH:
1.39(1)
1.54(1)
1.55(1)
1.54(1)
1.49(1)
N (ll)-C (ll)
N(ll)-C(15)
C(11)-C( 12)
C(12)-C(13)
C(13)-C(14)
1.35(1) C(14)-C(15)
1.33(1)
1.39(1)
1.38(1)
1.38(1)
C(11)-C( 18)
C(18)-C(19)
C(16)-C(18)
C(16)-N(12)
122.0(9) N(11)-C(15)-C(22) 115.1(7)
119.1(9) C(14)-C(15)-C(22)
N(11)-C( 11)-C( 12)
C(11)-C( 12)-C( 13)
C(12)-C(13)-C(14)
C(13)-C(14)-C(15)
C(14)-C(15)-N(l 1)
C(15)-N(l 1)-C( 11)
123.8(8)
118.2(8)
120.2(9) C(11)-C(18)-C(19) 111.2(7)
C(11)-C( 18)-C( 16)
1
1
1
.
1
(8 )
1
2
1
.0
(8 )
C(16)-C(18)-C(17) 109.2(7)
119.5(8) C(16)-C(18)-C(19)
1
1
0
.0
(8 )
N(11)-C( 11)-C( 18) 117.3(8)
C(12)-C(l 1)-C( 18)
( )
C(17)-C(18)-C(19) 105.4(8)
C(18)-C(16)-N(12) 112.3(8)
1
2
0
.6
8
Compound
6
:
L id)-O (ll)
Li(l)-0(21 A)
1.90(2)
Li(2)-0(13)
Li(2)-0(23)
1
.
8
8
2
(
(
1
1
)
)
2
.0
1
(
1
)
2 .0
0(13)-Li(2)-0(23) 105.1(4) N(2)-C(21)-C(27) 115.5(9)
0(13A)-Li(2)-0(23) 103.0(4) C(22)-C(21)-C(27)
0( 13)-Li(2)-0( 13A) 132.1(13) N(2)-C(21)-C(22)
0(23A)-Li(2)-0(23) 106.6(11)
123.7(10)
1
2
0
.
6
(
1
0
)
Compound 7:
Ba-O(l)
Ba-0(2)
Ba-0(4B)
Ba-0(6)
Cl-0(5)
Cl-0(6)
2.655(3) Cl-0(7)
2.823(3) Cl-
( )
2.705(4) 0(2)- •-0(3)
1.412(9)
1.32(1)
2.929(4)
2.896(4)
2.83(1)
0
8
2.789(5) 0(2B)- • 0(3)
0(4B)- • 0(7)
1.45(1)
1.362(5)
0(2)-Ba-0(l)
65.0(1)
73.2(1)
97.7(1)
114.4(1) C(10)-C(9)-N
123.2(1)
0(6A)-Ba-0(6)
C(2)-C(9)-N
C(10)-C(9)-C(2)
115.5(2)
117.8(4)
121.6(4)
120.6(4)
0(2A)-Ba-0(1)
O(IA)-Ba-Od)
0(2A)-Ba-0(2)
0(4B)-Ba-0(4C)
Acknowledgements.
Crystallography.
We thank Professor D. Sellmann for generous support
of this work. We also thank Dr. L. H. Gade (Universität
Würzburg) for communicating details of a related ligand
synthesis prior to publication (ref. [10a]), Dr. F. Hampel
for determining the crystallographic data set of 6 , and
Dr. W. Schindler for HPLC purification of 5. Further,
a research fellowship from the Fonds der Chemischen
Industrie (to A. G.) and a scholarship under the Euro-
pean exchange program ERASMUS (PIC 95-F-1060/13,
to L. O.) are gratefully acknowledged.
Data for 4 • 4 HBr • CH3OH were collected on a
Siemens P4 four-circle diffractometer (MoKa radiation,
graphite monochromator). The structure was solved by di-
rect methods and refined on F2 using the program package
SHELXTL 5.03. The non-hydrogen atoms were refined
anisotropically. All hydrogen atoms were located in the
difference Fourier map, and no reasonable maximum of
electron density could be found at the pyridine nitrogen
atom (the highest 1 0 0 peaks of residual electron density
were considered). In view of the relatively high absorption
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM

954
St. Schmidt et al. • Pyridine-Derived Tetrapodal Ligands with NO4 and NN4 Donor Sets
[1] a) A. Grohmann. F. Knoch, Inorg. Chem. 35, 7932
b) F. Vögtle, H. Sieger, W. M. Müller, J. Chem. Res.
(M) 1978, 4846.
(1996);
b) C. Dietz, F. W. Heinemann, J. Kuhnigk, C.
Krüger, M. Gerdan, A. X. Trautwein, A. Grohmann,
Eur. J. Inorg. Chem. 1998. 1041;
c) A. Grohmann, F. W. Heinemann, P Kofod, Inorg.
Chim. Acta, in press;
d) T. Poth, H. Paulus, H. Elias, R. van Eldik, A.
Grohmann. J. Chem. Soc., Dalton Trans., submit-
ted.
[19] A. R. Katritzky, W.-Q. Fan, Q.-L. Li, Tetrahedron
Lett. 28, 1195 (1987).
[20] T. W. Greene, P. G. M. Wuts, Protective Groups in
Organic Synthesis; 2nd ed., p. 135, Wiley, New York
(1991).
[21] J. A. Rendleman (Jr.), Adv. Carbohyd. Chem. 21,
209(1966).
[22] S. J. Angyal, Pure Appl. Chem. 35, 131 (1973).
[23] D. M. Whitfield, S. Stojkovski, B. Sarkar, Coord.
Chem. Rev. 122, 171 (1993).
[2] M. E. de Vries, R. M. La Crois, G. Roelfes, H.
Kooijman, A. L. Spek, R. Hage, B. L. Feringa, J.
Chem. Soc., Chem. Commun. 1997, 1549.
[3] a) D. H. Busch, Chem. Rev. 93, 847 (1993);
b) D. H. Busch in Proceedings of the First Han-
ford Separation Science Workshop; p. II.9, Battelle
Memorial Institute: Richland, WA (1993).
[4] C.-H. Lee, B. Garcia, T. C. Bruice, J. Am. Chem.
Soc. 112, 6434(1990).
[5] R. Riemschneider, O. Göhring, P. Groß, A. Rook, K.
Brendel, C. Faria, Monatsh. Chem. 96, 147 (1965).
[6] E. Buchta, W. Merk, Liebigs Ann. Chem. 694, 1
(1966).
[24] a) N. S. Poonia, A. V. Bajaj, Chem. Rev. 79, 389
(1979);
b) U. Olsher, R. M. Izatt, J. S. Bradshaw, N. K.
Dailey, Chem. Rev. 91, 137 (1991).
[25] A. J. Watters, R. C. Hockett, C. S. Hudson, J. Am.
Chem. Soc. 56, 2199(1935).
[26] A. A. Naiini, J. Pinkas, W. Plass, V. G. Young (Jr.),
J. G. Verkade, Inorg. Chem. 33, 2137 (1994).
[27] A. A. Naiini, V. Young, J. G. Verkade, Polyhedron
14, 393 (1995).
[28] Y. Shigemasa, O. Nagae, C. Sakazawa, R.
Nakashima, T. Matsuura, J. Am. Chem. Soc. 100,
1309(1978).
[7] E. B. Fleischer, A. E. Gebala, A. Levey, P. A. Tasker,
J.Org. Chem. 36, 3042(1971).
[8] K. Henrick, M. McPartlin, S. Munjoma, P. G. Ow-
ston, R. Peters, S. A. Sangokoya, P. A. Tasker, J.
Chem. Soc., Dalton Trans. 1982, 225.
[29] V. M. Padmanabhan, V. S. Jakkal, N. S. Poonia, Acta
Crystallogr. Sect. C 43, 1061 (1987).
[30] W. Weppner, W. Welzel, R. Kniep, A. Rabenau,
Angew. Chem., Int. Ed. Engl. 25, 1087 (1986).
[31] H. Adams, N. A. Bailey, D. E. Fenton, R. J. Good,
R. Moody, C. O. Rodriguez de Barbarin, J. Chem.
Soc., Dalton Trans. 1987, 207.
[9] A. Lipp, E. Zirngibl, Ber. Dtsch. Chem. Ges. 39,
1045 (1906).
[10] a) S. Friedrich, M. Schubart, L. H. Gade, I. J.
Scowen, A. J. Edwards, M. McPartlin, Chem.
Ber./Recueil 130, 1751 (1997);
b) K. Löffler, A. Grosse, Ber. Dtsch. Chem. Ges. 40,
1325 (1907).
[11] S. Ohki, Y. Noike, Yakugaku Zasshi, J. Pharm. Soc.
Jpn. 72, 490(1952).
[12] R. Bodalski, J. Michalski, K. Studniarski, Roczniki
Chem. 38, 1337(1964).
[32] L. J. Wilson, N. J. Rose, J. Am. Chem. Soc. 90, 6041
(1968).
[33] J. H. Forsberg, T. M. Kubik, T. Moeller, K. Gucwa,
Inorg. Chem^O, 2656(1971).
[34] A. E. Martell, R. D. Hancock, Metal Complexes in
Aqueous Solution, p. 55, Plenum Press, New York,
(1996).
[13] W. Koenigs, G. Happe, Ber. Dtsch. Chem. Ges. 36,
[35] S. Subramanian, M. J. Zaworotko, Coord. Chem.
2904(1903).
Rev. 137, 357 (1994).
[14] H. Würziger, S. 36, Kontakte (Darmstadt) (1988).
[15] J. March, Advanced Organic Chemistry. Reactions,
Mechanisms, and Structure; 4th ed. p. 945, Wiley,
New York, (1992).
[16] B. P. Kremer, J. Chromatogr. 110, 171 (1975).
[17] M. Ghebregzabher, S. Rufini, B. Monaldi, M. Lato,
J. Chromatogr. 127, 133 (1976).
[18] a) F. Vögtle, H. Sieger, W. M. Müller, J. Chem. Res.
(S) 1978", 398;
[36] Mingyi Wei, R. D. Willett, K. W. Hipps, Inorg.
Chem. 35, 5300(1996).
[37] R. Grönbsek Hazell, S. E. Rasmussen, Acta Chem.
Scand. 22, 348 (1968).
[38] T. S. Cameron, W. J. Chute, O. Knop, Can. J. Chem.
63, 586(1985).
[39] A. Bondi, J. Phys. Chem. 68, 441 (1964).
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 7/18/15 6:53 AM