Solid-state reactivity of the hydrazine-hydroquinone complex

Article abstract of DOI:10.1002/1099-1395(200007)13:7<388::AID-POC278>3.0.CO;2-H

The solid-state reactivities of the hydrazine-hydroquinone 1:1 complex and of hydrazine hydrochloride with solid aldehydes, ketones, carboxylic acids, thiohydantoin and 4-nitrophenyl isothiocyanate were investigated. Only the hydrazine complex provides quantitative additions, condensations, ring openings and ring closures. The solid-state mechanisms were investigated by atomic force microscopy (AFM) and the far-reaching anisotropic molecular movements are correlated with the crystal packing, both on the hydrazine complex surface and on the surface of two benzaldehydes. The hydrazine moves into the aldehyde crystals for chemical reaction without melting. Characteristic surface features are created by the common phase rebuilding and phase transformation on both the hydrazine-donating and -accepting crystals. Copyright

Full text of DOI:10.1002/1099-1395(200007)13:7<388::AID-POC278>3.0.CO;2-H

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2000; 13: 388–394
Solid-state reactivity of the hydrazine±hydroquinone
complex
Gerd Kaupp* and Jens Schmeyers
Organische Chemie I, UniversitaÈ t Oldenburg, Postfach 2503, D-26111 Oldenburg, Germany
Received 10 January 2000; accepted 4 April 2000
ABSTRACT: The solid-state reactivities of the hydrazine–hydroquinone 1:1 complex and of hydrazine hydrochloride
epoc
with solid aldehydes, ketones, carboxylic acids, thiohydantoin and 4-nitrophenyl isothiocyanate were investigated.
Only the hydrazine complex provides quantitative additions, condensations, ring openings and ring closures. The
solid-state mechanisms were investigated by atomic force microscopy (AFM) and the far-reaching anisotropic
molecular movements are correlated with the crystal packing, both on the hydrazine complex surface and on the
surface of two benzaldehydes. The hydrazine moves into the aldehyde crystals for chemical reaction without melting.
Characteristic surface features are created by the common phase rebuilding and phase transformation on both the
hydrazine-donating and -accepting crystals. Copyright 2000 John Wiley & Sons, Ltd.
Additional material for this paper is available from the epoc website at http://www.wiley.com/epoc
KEYWORDS: solid hydrazine; solid-state reactivity; anisotropic molecular movements; atomic force microscopy;
crystal packing; carbonyl compounds; sustainable
INTRODUCTION
provides quantitatively solid mixtures of the azines 3
and hydroquinone 4 when ball-milled at room tempera-
ture. The previous host 4 is easily extracted with water at
room temperature and the extract is concentrated by
evaporation for precipitation of 2 with aqueous hydrazine
hydrate. The solid-state double condensation is quantita-
tive and more effective than the reported solution
reactions [3a, no data; 3b, 94% (1 h, room temperature);5
3c, 92% (1 h, room temperature);6 3d, 91%7]. On the
other hand, solid-state reactions of 1 with hydrazine
monohydrochloride remain largely incomplete. All of the
aldehydes 1a–c underwent complete reaction with 2 in
1 h, whereas 4-nitroacetophenone required a longer
milling time for complete reaction even though dry
powders were obtained in all cases. The water of reaction
was fully included by the product crystals, most likely by
4. It has been shown that 2.0 mmol of 4 take up 4.0 mmol
of water in the ball-mill to form a dry, crystalline
complex at room temperature (G. Kaupp and J.
Schmeyers, unpublished work).
Hydrazine is a versatile though poisonous reagent, and
quantitative reactions in the solid state1–3 will facilitate
its use. Unfortunately, solid salts of hydrazine such as the
monohydrochloride (m.p. 89°C) are of low reactivity in
solid–solid reactions. However, the easily accessible 1:1
complex of hydrazine with hydroquinone deserves a
closer look, as Toda et al. were able to use it for the
conversion of solid esters into hydrazides in good yields.4
We therefore tried a number of quantitative reactions
with solid anhydrous hydrazine and solid carbonyl
compounds and studied the mechanism in order to
unravel the reasons for the success of the quantitative
additions, condensations, ring openings and ring closures
that emerged and may be conducted ‘waste-free’3 if the
host hydroquinone is properly recycled.
RESULTS AND DISCUSSION
Azines
AFM investigation
The reaction of solid aldehydes or ketones 1 with the
hydrazine–hydroquinone complex 2 in a 2:1 ratio
The solid-state mechanism was elucidated by atomic
force microscopy (AFM) for the hydrogen-bridged 1a
and the non-bridged 1c in order to assess the directions of
molecular movements in view of the known fact that the
hydrazine is exhaustively bridged in 2.4
*Correspondence to: G. Kaupp, Organische Chemie I, Universita¨t
Oldenburg, Postfach 2503, D-26111 Oldenburg, Germany.
E-mail: kaupp@kaupp.chemie.uni-oldenburg.de
Contract/grant sponsor: Fonds der Chemischen Industrie.
Contract/grant sponsor: Deutsche Forschungsgemeinschaft.
The AFM analyses of the reaction of hydrazine–
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

SOLID-STATE REACTIVITY OF THE HYDRAZINE–HYDROQUINONE COMPLEX
389
hydroquinone complex 2 and 4-hydroxybenzaldehyde
(1a) is summarized in Figs 1 and 2. The reaction is rather
slow. Almost no change of a smooth surface is observed
in Fig. 1(b) at a 1 mm distance from the edge of the
crystal 1a that was placed on 2. After a further 1 h the
minute craters (<2 nm) were changed into skew floes that
are about 10 nm in height at their steps in Fig. 1(c). The
craters in Fig. 1(d) after 1 h at a 0.1 mm distance are
typically 7 nm deep. Later, these became deeper (down to
50 nm) and additionally many small (20–30 nm) and
some large hills (up to 100 nm) are present in Fig. 1(e).
After 4 h, that site was too rough to be measured with the
standard AFM but we recorded isolated volcano-like hills
with heights of up to 220 nm at a 0.5 mm distance from
the contact edge that are depicted in Fig. 1(f). Clearly,
Fig. 1 indicates molecular movements above the (100)
Figure 1. 10 mm AFM topographies of 2 (C2/c; plates) on (100) at 1, 0.5 and 0.1 mm distance from a crystal of 1a that was
placed on top of it. The reaction times are indicated. The z-scale is 100 nm in (a), (b), (c) and (d), 200 nm in (e) and 500 nm in (f).
High-and low-resolution VRML images of (a)±(f) are available at the epoc website at http://www.wiley.com/epoc
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

390
G. KAUPP AND J. SCHMEYERS
Figure 2. 10 mm AFM topographies of 1a (P21/c; plates) on (010) at 0.1 mm distance from a crystal of 2 that was placed on top
of it: (a) before placement; (b) after 5 min; (c) after 2 h. The z-scale is 100 nm in (a), 200 nm in (b) and 300 nm in (c). High-and
low-resolution VRML images of (a)±(c) are available at the epoc website at http://www.wiley.com/epoc
face of 2 in two stages at a fairly close distance, which
indicates phase rebuilding8 [Fig. 1(e)] and finally phase
transformation8 [Fig. 1(f)] to give the host lattice
[presumably as a water clathrate (G. Kaupp and J.
Schmeyers, unpublished work)] on the surface. However,
hydrazine moves away into the crystal of 1a, which
becomes yellow on formation of the solid azine 3a. It was
therefore also important to study the AFM features on a
single crystal of 1a with a crystal of 2 on it at a 0.1 mm
distance where the product 3a was formed. The result is
shown in Figure 2.
A slighty corrugated initial surface (Rms = 4.56 nm)
created uniform ‘egg trays’ (416 craters and 663
volcanoes; Rms = 14.73 nm) after 5 min of reaction.
These had bottom to peak distances of about 80 nm [Fig.
2(b)] and they collapsed to give 157 and 191 broader and
higher features after 2 h of reaction [Fig. 2(c)]. These
features did not grow significantly until crystal disin-
tegration occurred upon further reaction. Again, there
was anisotropic molecular transport above the (010) face
while leaving craters. No clear further direction of
preference is discernable: the steep features in Fig. 2(c)
are slightly distorted.
It was also important to study the reaction of crystals of
non-hydrogen-bridged 4-dimethylaminobenzaldehyde
(1c) on 2. They withdrew the hydrazine molecules more
rapidly from 2. The initially corrugated surface of 2 in
Fig. 3(a) (Rms = 4.92 nm) formed craters (down to
300 nm deep) and volcanoes (up to 90 nm high) in
5 min at a 1 mm distance from the contact edge [Fig.
3(b)]. After 3 h the first signs of crystal disintegration
were seen at that distance: the craters in Fig. 3(c) are
down to 220 nm deep and start to form valleys. Some
volcanoes are still present on the large floes. As in Fig.
1(e) and (f) there was molecular transport above the
surface but it was more rapid and thus gave a different
appearance of the features than in Fig. 1. The formation
of the yellow azine 3c is documented in Fig. 3(d)–(f)
starting from a smooth surface (Rms = 4.78) of 1c with a
crystal of 2 on it. Here flat covers [step height 50–150 nm
in Fig. 3(e)] were rapidly formed (5 min) and in the next
step huge islands appeared with heights of up to 970 nm,
the slopes of which were frequently larger than 45°, so
they appear somewhat broadened in the two 2-dimen-
sional projection of Fig. 3(f). As expected, both phase
rebuilding8 and phase transformation8 are totally differ-
ent from the reaction of 1a in Fig. 2.
Mechanistic discussion
The interpretation of the AFM results requires the
consideration of the available crystal structures. The
phase rebuilding on (010) of reacting 1a has been
described to give craters and volcanoes in the gas–solid
reaction with methylamine.2 Again, the reason is the
steep packing of the molecules 1a, half of them
exposing their carbonyl function, the other half having
it down in the crystal where it bridges to a phenolic OH
group of the next layer. It is therefore not surprising to
obtain the ‘egg trays’ [shown in Fig. 2(b), inasmuch as
the molecules must move above the surface if their H-
bonds are broken after reaction with hydrazine.1,2
However, the second step, the phase transformation into
the product lattice of 3a, has nothing in common with
the formation of extended craters in the referenced
reaction.2 In the present case we may admit some
remaining guidance of the crystal packing of 1a while
the lattice of 3a is formed in Fig. 2(c). The strongly
different efficiency in the reaction of 1c with 2 supports
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

SOLID-STATE REACTIVITY OF THE HYDRAZINE–HYDROQUINONE COMPLEX
391
Figure 3. 10 mm AFM topographies of 2 on (100) with crystals of 1c on top at a 1 mm distance: (a) fresh; (b) 5 min; (c) 3 h. 10 mm
AFM topographies of 1c (plates) on its main surface with crystals of 2 on top at a 0.1 mm distance: (d) fresh; (e) 5 min; (f) 43 min.
The z-scales are 100 nm in (a) and (d), 200 nm in (b), 400 nm in (c) and (e) and 1400 nm in (f). High-and low-resolution VRML
images of (a)±(f) are available at the epoc website at http://www.wiley.com/epoc
the view that hydrazine molecules move continuously
from the complex through the contact points into the
surface region of 1a where they react to produce the
intermediate hydrazone and finally the azine 3a and two
molecules of water that may move back to the host.
Clearly, the condensation reaction spreads not only
down into the bulk but also laterally, permitting its
detection by AFM at remote sites. Vaporization of
hydrazine along the surface of the reactant appears
unlikely, as thermogravimetric analyses (TGA) did not
show significant weight losses (at heating rates of 10°C
min 1) below 80–100°C. However, reactions of 2 at
>100°C4 or even the reaction of 7 at 80°C (see below)
may be gas–solid reactions if the thermal exit of
hydrazine gas from the lattice of 2 is fast enough. The
crystal structure of 1c is not known, but the flat covers9
in Fig. 3(e) are certainly characteristic and the high and
steep islands in Fig. 3(f) are probably the result of
crystal characteristics of 3c and do not require guidance
by the initial lattice of 1c.
in the bulk of the host lattice of 2 form six hydrogen
bonds to six adjacent hydroxyl groups of six hydro-
quinone molecules (Fig. 4). The latter are bridged by
hydrazine and form alternating layers. The NH
O
N
ÁÁÁ
˚
distances are 3.131 and 3.161 A and the OH
ÁÁÁ
˚
distances are 2.732 A. All free electron pairs and acidic
hydrogens are involved in hydrogen bridges, which
explains the high m.p. of 158°C and stability of 2
(heating rate 10°C min 1; hydroquinone, m.p. 173–
174°C). It may be envisioned that the hydrazine
molecules at the outer surfaces [e.g. at (100) at the top
of Fig. 4] are released upon initial reaction and that a
drain of hydrazine ensues in the layer parallel to (100).
After higher dilution of the hydrazine in the outer layer,
the layered hydroquinone molecules start to dangle and
form hydrogen bonds with each other and most likely
pick up the water of reaction in the condensations. By
doing so, free space will be created for a drain from the
next layer of hydrazine molecules, etc. Anyhow, we were
not surprised that the solid-state reactions of 2 required
longer milling times for complete reaction than the
previous syntheses.1,10 Similarly, Toda et al.’s acid
hydrazide synthesis at 100–125°C was not rapid (25
The reactivity of the solid hydrazine–hydroquinone
complex 2 was not foreseen, as the hydrazine is
exhaustively hydrogen bridged: the hydrazine molecules
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

392
G. KAUPP AND J. SCHMEYERS
Figure 4. Stereoscopic view of the packing of the hydrazine±hydroquinone complex 24 on (001) showing exhaustive hydrogen
bridging. Large circles are N and O, small circles are H bound to N or O; thick hatches are OH N and thin hatches NH
hydrogen bridges. The top face is (100)
O
ÁÁÁ
ÁÁÁ
h).4 Anyhow, the hydrogen bonds that are lost after exit
of hydrazine may be replaced in a more and more
distorted lattice of 2 that ends with the stable lattice of
hydroquinone 4 or its water clathrate (G. Kaupp and J.
Schmeyers, unpublished work). On the other hand, the
low reactivity of solid hydrazine monohydrochloride is
understandable from the crystal packing11 (Fig. 5). Here
both ionic and hydrogen bonds are present. The chloride
anions are almost symmetrically located between two
ammonium cation centers and the hydrogen bridges
Applications
The mechanistic information was applied to the use of
‘solid hydrazine’ 2 for additions to the isothiocyanate 5 to
give the thiosemicarbazide 6, for ring opening of
thiohydantoin 7 to give the thiourea hydrazide 8 and for
cyclizations of 9 and 11 to give 10 and 12. As, apparently,
hydrazine is extracted from the host lattice by the
reagents, we obtained quantitative yields of the highly
functionalized products that are not available by milling
the same starting crystals with hydrazine monohydro-
chloride at room temperature owing to its unsuitable
crystal packing. The yields are generally lower in
solution and higher temperatures are required (6, 80°C,
90%;12 8, 80°C, quantitative;13 10, 25°C, 88%;14 12,
140°C, 43%15). The present solid-state syntheses are a
sustainable improvement and the need for separation and
recovery of hydroquinone 4 is tolerable, as the products
are insoluble in water. It is to be expected that numerous
further syntheses will be successful with 2, as the crystal
structures of the various carbonyl partners do not seem to
have much influence on the reactivity of the small and
highly reactive hydrazine that escapes its cage for
reaction in the ball-mill despite its six hydrogen bridges.
‡
‡
˚
between NH3 and NH centers (N —H N, 2.921 A)
ÁÁÁ
2
build a three-dimensional network which is further
stabilized by short N—H Cl bridges (2.080, 2.242
ÁÁÁ
˚
and 2.347 A) and does not allow the exit of the
interlocked ions from the crystal because every hydrazi-
nium cation is held in place by two strong ionic
interactions and two strong hydrogen bonds. The
energetics may be not so bad because the hydrochloride
salts formed would experience ionic and hydrogen-bond
interactions. However, there are kinetic obstacles that
explain the failure of crystalline N H HCl. At present, it
4Á
2
cannot be foreseen if further salts of hydrazine might
pack more favorably for exhibiting improved solid-state
reactivity.
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

SOLID-STATE REACTIVITY OF THE HYDRAZINE–HYDROQUINONE COMPLEX
393
Figure 5. Stereoscopic view of the packing of N H HCl (rhombic, Fdd2)11 skew on (001). Large circles are Cl, medium circles N
Á
2
4
‡
and small circles H atoms; short ionic interactions (Cl
highlighted for clarity
N ) and strong H-bridges are dotted; NÐH Cl bridges are not
ÁÁÁ
ÁÁÁ
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394

394
G. KAUPP AND J. SCHMEYERS
The reactions are not violent and it is likely that scale-ups
in large mills can be easily controlled.
6-Oxo-1,4,5,6-tetrahydropyridazine-3-carbonic acid
(10). Yield 98%; m.p. 194°C (lit.14 194–195°C).
3-Phenyl-1,4,5,6-tetrahydro[1,2]diazepin-7-one (12).
Yield 99%; m.p. 156°C (lit.15 157°C).
EXPERIMENTAL
Cantilever contact AFM with non-scraping16 standard
Si3N4 tips8 at 10–30 nN force, the imaging techniques8
and the possible tip–sample artifacts on rough surfaces17
have been described. All AFM surface data are given in
fully interactive VRML files at the epoc website at http://
www.wiley.com/epoc.
Solid–solid techniques with AFM and ball-mills have
been described.2,8,10,18 A 4.00 mmol amount of the
aldehyde/ketone 1 or 2.00 mmol of the (thio)carbonyl
compound 5, 7, 9 or 11 were ball-milled with 2.00 mmol
of 2 at 25–30°C (70–80°C in the case of 7) for 1 h (3 h in
the case of 1d). The yields were quantitative in all cases.
Spectroscopically pure (IR, 1H- and 13C NMR) mixtures
of 4 (or its water clathrate) with the products 3, 6, 8, 10 or
12 were obtained. The previous host 4 was removed by
5 min of trituration with 20 ml of water, filtration and
three washings with 2 ml of water each (at room
temperature 10 ml of water dissolves 700 mg of 4 and
250 mg of 2), the residue dried in a vacuum, weighed and
again spectroscopically characterized.
EPOC material
The supplementary material available at the epoc website
at http://www.wiley.com/epoc contains color images in
GIF format and interactively usable full original 3D data
of the images in VRML format in high and low resolution
in order to allow for critical evaluation of the AFM
results, their analytical use with suitable imaging soft-
ware and future data mining. For example, the impor-
tance of unavoidable minor geometric artifacts at very
steep features, that are not easily recognized in the
common two-dimensional projections, can be judged in
some of the three-dimensional views from various sides.
However, it is safely concluded that these do not affect
the conclusions derived from the stable scans (at least five
consecutive scans) that also exclude nanoliquid phases in
the reactions studied. The interactive analysis is possible
with public domain software, e.g. Cosmo-Player or
Blaxxun.
The hydrazine–hydroquinone complex 2 was prepared
according to Toda et al.4 or obtained by recycling of the
hydroquinone 4 from the aqueous washings of the
reported solid-state syntheses (e.g. the reaction of 1a
and 2) by evaporation, addition of 1.0 g of 80% hydrazine
hydrate in water per 4.00 mmol of initially reacted 2 and
recrystallization to give a 91% yield (520 mg) of
spectroscopically pure 2, m.p. 158°C, after filtration,
washing with water and drying.
REFERENCES
1. Schmeyers J, Toda F, Boy J, Kaupp G. J. Chem. Soc. Perkin Trans.
2 1998; 989.
2. Kaupp G, Schmeyers J, Boy J. Tetrahedron, in press.
3. Kaupp G, Schmeyers J, Boy J. Chemosphere in press.
4. Toda F, Hyoda S, Okada K, Hirotsu K. J. Chem. Soc., Chem.
Commun. 1995; 1531.
5. Mlochowski J, Giurg M. Pol. J. Chem. 1994; 68: 2333.
6. Vickers S, Stuart EK. Anal. Chem. 1974; 46: 138.
7. Miyatake K. Yakugaku Zasshi 1953; 73: 455; Chem. Abstr. 1954;
48: 5144.
8. Kaupp G. In Comprehensive Supramolecular Chemistry, vol. 8,
Davies JED (ed). Elsevier: Oxford, 1996; 381–423 ‡ 21 color
plates; Kaupp G. Chem. Unserer Zeit 1997; 31: 129; Engl.
translation at http://kaupp.chemie.uni-oldenburg.de.
9. Kaupp G. Mol. Cryst. Liq. Cryst. 1992; 211: 1.
10. Kaupp G, Boy J, Schmeyers J. J. Prakt. Chem./Chem.-Ztg. 1998;
340: 346.
4-Hydroxybenzaldazine (3a). Yield 98%; m.p. 270°C
(lit.19 270°C); IR (KBr), 3338(OH), 1608 (C
=
N), 1592,
1
1515 cm
.
4-Chlorobenzaldazine (3b). Yield 99%; m.p. 207°C
(lit.5 208–210°C).
11. Chekhlov AN, Martynov IV. Kristallografiya 1988; 33: 1010.
12. Lieber E, Ramachandran J. Can. J. Chem. 1959; 37: 101.
13. Hideaki S, Tamatsu Y. Yakugaku Zasshi, 1967; 87: 128; Chem.
Abstr. 1967; 67: 32379e.
4-Dimethylaminobenzaldazine (3c). Yield 99%; m.p.
273°C (lit.6 274°C).
14. Kline GB, Cox SH. J. Org. Chem. 1961; 26: 1854.
15. Wermuth CG, Ko¨nig JJ. Angew. Chem. Int. Ed. Engl. 1972; 11:
152.
16. Kaupp G. Mol. Cryst. Liq. Cryst. 1994; 252: 259; Kaupp G. J. Vac.
Sci. Technol. B 1994; 12: 1952; Kaupp G, Plagmann M. J.
Photochem. Photobiol. A 1994; 80: 399.
4-Nitroacetophenone azine (3d). Yield 98%; m.p.
201°C (lit.7 204°C).
4-(4-Nitrophenyl)thiosemicarbazide (6). Yield 99%;
m.p. 190°C (lit.12 190°C).
17. Kaupp G, Herrmann A, Haak M. J. Phys. Org. Chem. 1999; 12:
797.
18. Kaupp G, Schmeyers J, Kuse A, Atfeh A. Angew. Chem., Int. Ed.
Engl. 1999; 38: 2896.
19. Okafor EC. Spectrochim. Acta, Part A 1980; 36: 207.
Thiohydantoic acid hydrazide (8). Yield 100%; m.p.
202°C (lit.13 204°C).
Copyright 2000 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2000; 13: 388–394