Understanding "on-water" catalysis of organic reactions. Effects of H+ and Li+ ions in the aqueous phase and nonreacting competitor H-bond acceptors in the organic phase: On H2O versus on D2O for Huisgen cycloadditions

Article abstract of DOI:10.1021/jo502732y

For a typical Huisgen cycloaddition, carried out on water, the behavior of water molecules at the oil-water interface depended on the properties of the reactants. With weakly basic reactants, a small quantity of added H⁺ (HClO₄, 0.0001-0.01 M) present in the aqueous phase had negligible effects, but larger quantities of H⁺ (HClO₄, 0.1-3.0 M) increased the catalytic effect and caused protons to cross the water-organic interface and affect the products. Added Li⁺ ions (LiClO₄, 0.1-3.0 M) had no effect for on-water reactions but enhanced the rates and endo products for in-water reactions. For these cycloaddition reactions, the product endo:exo ratios, when compared to those in organic solvents, can be used to distinguish between the on-water and in-water modes. Comparisons of organic reactions on H₂O and on D₂O indicate that on-water catalysis ranges from weak to strong trans-phase H-bonding for reactants with basic pK_a < ca. -6 and to interfacial proton transfer for reactants with higher basic pK_a > ca. 2 (pK_a of conjugate acid). Water shows a chameleon-type response to organic molecules at hydrophobic surfaces.

Full text of DOI:10.1021/jo502732y

Article
pubs.acs.org/joc
Understanding “On-Water” Catalysis of Organic Reactions. Effects of
H+ and Li+ Ions in the Aqueous Phase and Nonreacting Competitor
H‑Bond Acceptors in the Organic Phase: On H2O versus on D2O for
Huisgen Cycloadditions
Richard N. Butler*,† and Anthony G. Coyne*,‡
†School of Chemistry, National University of Ireland, Galway, Ireland
‡Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, United Kingdom
S
* Supporting Information
ABSTRACT: For a typical Huisgen cycloaddition, carried out
on water, the behavior of water molecules at the oil−water
interface depended on the properties of the reactants. With
weakly basic reactants, a small quantity of added H+ (HClO4,
0.0001−0.01 M) present in the aqueous phase had negligible
effects, but larger quantities of H+ (HClO4, 0.1−3.0 M)
increased the catalytic effect and caused protons to cross the
water−organic interface and affect the products. Added Li+
ions (LiClO4, 0.1−3.0 M) had no effect for on-water reactions
but enhanced the rates and endo products for in-water reactions. For these cycloaddition reactions, the product endo:exo ratios,
when compared to those in organic solvents, can be used to distinguish between the on-water and in-water modes. Comparisons
of organic reactions on H2O and on D2O indicate that on-water catalysis ranges from weak to strong trans-phase H-bonding for
reactants with basic pKa < ca. −6 and to interfacial proton transfer for reactants with higher basic pKa > ca. 2 (pKa of conjugate
acid). Water shows a chameleon-type response to organic molecules at hydrophobic surfaces.
certain preferential orientations or increased strengths of H-
bonds leading to faster reactions at the water−organic interface.
Experimental results also give rise to varied opinions. Sela and
Vigalok reported that neat conditions or other “non-solvents”
are better than on-water reactions for some Passerini reactions
and amine epoxide ring openings.21 Zhang, Wang, and co-
workers report highly efficient on-water reactions with
substrates containing C−F bonds where trans-facial H-bonding
by OHfree groups to fluorine atoms is important.22 Zuo and Qu
have emphasized that small levels of water solubility for organic
reactants are a necessary requirement in order to achieve on-
water reactions.23 Among the theoretical studies there has been
a considerable focus on the cycloaddition reaction of
quadricyclane (1) with dimethyl azodicarboxylate (2) (Figure
1). This reaction is a complicated two-stage mechanism with a
large polar character in the gas phase, and it is likely to be more
complicated under on-water conditions.24 In our opinion, the
literature has overlooked an important result reported by
Sharpless, Fokin, and their co-workers, namely, the difference
for on H2O and on D2O, concerning which these researchers
stated, “Interestingly a significant solvent isotope effect was also
observed; the reaction slowed noticeably when D2O was used
in place of water.”9 The Beattie and McErlean research groups
have proposed that on-water catalysis involves a proton transfer
INTRODUCTION
■
Following the pioneering work of Breslow on the influence of
water on the Diels−Alder reactions and his recognition and
exploration of the hydrophobic effect as central to the
understanding of organic reactions in water, there has been
an ever-growing interest in using water as a medium for organic
reactions.1−8 The subsequent recognition that water-insoluble
organic compounds could also be induced to undergo reactions
in the water medium by Sharpless, Fokin and their co-workers,
named the “on-water” phenomenon, further increased the drive
toward extensive use of water as a medium for organic
synthesis.9−16 The on-water catalytic effect, however, is not well
understood, despite the fact that there have been extensive
theoretical studies focusing on the oil−water interface. Trans-
phase H-bonding from dangling OHfree groups at the interface
to H-bond acceptor sites on the organic reactants was proposed
by Jung and Marcus to be the key contribution to the catalytic
effect.17 Experimental kinetic and thermodynamic studies by
Manna and Kumar supported these conclusions.18 Using
quantum mechanical/molecular mechanical modeling, Jorgen-
sen and co-workers concluded that the Diels−Alder reactions
were less accelerated on the surface of water than in the bulk
water.19 Kuhne and co-workers, using extended Car−Parrinello
̈
molecular dynamics, concluded that H-bonds from OHfree
groups at the aqueous−organic interface play a much smaller
role in on-water catalysis than has been suggested.20 However,
they also state that their results do not exclude the possibility of
Received: December 3, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Herein we examine Huisgen cycloadditions in different
environments: (a) neat (no solvent), (b) in an organic solvent
(MeCN), (c) in water (dissolved in bulk water), (d) on water
(highly water insoluble reactants), and (e) using water with
very small (0.0001 M) to large (3.0 M) quantities of H+ and Li+
(0.1 and 3.0 M) present and also adding non-reacting,
competing H-bond acceptors to the organic phase. Previously,
we have established an experimental distinction between in-
water and on-water reactions for aqueous suspensions of
organic reactants.27 Diels−Alder and Huisgen 1,3-dipolar
cycloadditions that occur in water show enhanced endo:exo
product ratios arising from the Breslow hydrophobic effect,
which favors endo-transition states in bulk water. Reactions that
occur in aqueous suspensions by the on-water mode do not
show enhancements of product endo:exo ratios because they
take place in the organic phase under trans-phase on-water
catalysis and are not experiencing the Breslow hydrophobic
effect of bulk water. In choosing substrates for the on-water
reactions herein, attention was paid to the sensitivity of
isomeric endo:exo mixtures to acid. Acid may isomerize endo
Figure 1. Cycloaddition reaction of quadricyclane and dimethyl
azodicarboxylate under neat conditions, H2O, and D2O.9
across the organic−water interface, giving rise to acid
catalysis.25,26 They have provided compelling evidence for
this with strong D/H isotope effects measured for on-water
Claisen rearrangements of allyl aryl amines.26
Table 1. (A) In-Water Reaction of Phthalazinium-2-dicyanomethanide (s ≤ 5 × 10−6 mol L−1) (4) with 2-Cyclopenten-1-one
(5) and Methyl Acrylate (6) and (B) On-Water Reaction of Phthalazinium-2-dicyanomethanide (4) with 1-Octen-3-one (9),
Phenyl Acrylate (10), and Diphenylacetylene (13)
a
Added corresponding unreactive H-bonding competitor, cyclopentaneone, methyl propionate, octane-3-one, and phenyl propionate, respectively.
b
c
The term neat refers to the identical reaction as with MeCN, H2O, or D2O however with no solvent present. The completion time is the time to
d
maximum yield when no further products are appearing. See Figure 3 for the transition from conversion yield to completion yield for this reaction.
B
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Scheme 1. Huisgen 1,3-Dipolar Cycloaddition Reaction with a Range of Dipolarophiles
a
These reactions are normal-electron-demand 4π-HOMO-controlled cycloadditions with catalytic H-bonding at the 2π dipolarophiles 5, 6, 9, and
10. The reaction with diphenylacetylene (13) is an inverse-electron-demand 4π-LUMO-controlled cycloaddition with H-bonding at the CN groups
of the 1,3-dipole 4.15
isomers to the thermodynamic exo-forms, thereby providing a
marker for protons crossing the interface to the organic phase.
42 in water (Table 1, entries 2, 3), while the endo enhancement
for reactants with little hydrophobic surface area, such as
methyl acrylate (6), show smaller endo:exo enhancements, e.g.,
8.3 in MeCN and 9.8 in water (Table 1). Reactions that occur
by the on-water mode show no endo enhancement because
they occur at the organic side of the organic−water interface
and do not experience the hydrophobic effect of bulk water.
Hence, the product endo:exo ratios for the reactions of 4 with
1-octen-3-one (9) (8.4 0.2) and phenyl acrylate (10) (7.3
0.5) are the same in MeCN and water (Table 1).
Also shown in Table 1 are the “neat” reactions of 4 with the
dipolarophiles concerning the points raised by Sela and
Vigalok.21 For these reactants the neat reactions are faster
than the in-water cases but the on-water reactions are slightly
faster or similar to the neat reactions. The term “neat” here
means a necessary 10 molar quantity of liquid 2π-reactant to
one mole of the insoluble solid 4π-reactant 4 with separation
and workup procedures required to isolate the products. Such
neat reactions are not more efficacious or efficient than on-
water reactions.
If on-water catalysis arises from H-bonding across the water−
organic interface from dangling OHfree groups to H-bond
acceptor sites on the reactants, it follows that if an excess of a
competitor H-bond acceptor molecule were introduced, which
did not interfere with the reaction, then it should inhibit the on-
water reaction. Having explored possible competitors for H-
bond acceptance, it became clear that the requirement of not
interfering with the reactants or the reaction was essential, and
hence, the saturated versions of the alkene dipolarophiles were
chosen as the most suitable competitors for trans-interphase H-
bonds. Thus, octan-3-one and phenyl propionate were
introduced to compete with the on-water reactions of 1-
octen-3-one (9) and phenyl acrylate (10). In Table 1 (entries
RESULTS AND DISCUSSION
■
i. On-Water vs In-Water Reactions; Competitor H-
Bond Acceptors in the Organic Phase. Phthalazinium-2-
dicyanomethanide (4) is a high-melting yellow solid (mp 252−
254 °C dec) that is highly insoluble in water (s ≤ 5 × 10−6 mol
L−1 at 37 °C). The nature of its Huisgen cycloaddition
reactions with liquid 2π-dipolarophiles in water depends on the
water solubility of the dipolarophile. For dipolarophiles with
water solubilities >0.1 mol L−1, the reactions occur in water.
The dissolved 2π-reactant facilitates an equilibrium transport of
the 1,3-dipole through the solution prior to the separation of
the highly insoluble products. In physical appearance the yellow
suspension of reactants gradually changes to a sticky white
suspension of products (see Supporting Information, Figure
S5). Two examples of such reactions of 4, with 2-cyclopenten-
1-one (5) and methyl acrylate (6), are shown in Table 1 and
Scheme 1. When the water solubility of the 2π-reactant is an
order of magnitude lower, ca. 0.01 mol L−1, the reactions occur
by the on-water mode, although the physical appearance in the
reaction flask is no different (see Supporting Information,
Figure S5).27 Three examples of such on-water reactions, with
1-octen-3-one (9), phenyl acrylate (10), and diphenylacetylene
(13) are shown in Table 1 and Scheme 1. Reactions that occur
in water, where the transition states experience the bulk water
environment, display the result of the Breslow hydrophobic
effect.2,3 This favors endo-transition states and enhances the
endo:exo ratio of the cycloaddition products. Reactants with
large hydrophobic surfaces, such as 2-cyclopenten-1-one (5)
show large enhancements of the endo:exo product ratio in
water relative to organic solvents or neat, e.g. 5.1 in MeCN and
C
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. Effects of H+ and Li+ in the Aqueous Phase on the Reaction of 1,3-Dipole 4 with (A) 1-Octen-3-one (9) and (B) Phenyl
Acrylate (10) at 20 °C
11 and 12) are shown the effect of adding these to the organic
phase. Up to 10 molar excesses of the additives caused a 10−
12% reduction in the yields of products for the cycloadditions
with compound 4. As the molar excesses of the additives were
increased into the range of 15−50 mol, the additives took on
the role of an organic solvent, and further inhibitions of the
reactions were not observed.
D
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 3. Effects of H+ and Li+ in the Aqueous Phase on the Reaction of 1,3-Dipole 4 with (A) Methyl Acrylate (6) and (B) 2-
Cyclopenten-1-one (5) at 20 °C
ii. Effects of H+ and Li+ in the Aqueous Phase. a. On-
Water Reactions. There have been extensive studies of the
catalytic effects of Li+ ions on the Diels−Alder [4 + 2]
cycloadditions as well as [2 + 2] cycloadditions and a range of
other important reactions, including aldol reactions, Michael
reactions, Alder−ene reactions, sigmatropic rearrangements,
and Mannich reactions,28−39 Lithium perchlorate in diethyl
ether has been widely used but other salts, such as lithium
trifluoromethanesulfonate and trifluoromethane sulfonimide in
solvents such as acetone and acetonitrile, have also been
explored.30,40 In the context of on-water chemistry, it is of
particular interest to explore the presence of H+ and the related
charge intensive Li+ ion as additives in the aqueous layer. In
Table 2 are shown the results of adding low to high
concentrations of HClO4 and LiClO4 to the on-water and in-
water reactions in Scheme 1. The results for on-water reactions
are significantly different from the in-water cases. For the on-
water reactions with 1-octen-3-one (9) and phenyl acrylate
E
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(10), small quantities of added acid, 0.0001−0.01 M, had
negligible or small effects on the reactions (Table 2, entries 1−4
and 12−16). Hence, the presence of small quantities of protons
in the bulk water did not influence events at the water−organic
interface. However, as the quantity of added acid was increased,
significant effects were encountered. For the reaction with 1-
octen-3-one (9), increasing the concentration of HClO4 in the
water layer from 0.1 to 3.0 M reduced the reaction completion
times, eventually to a few minutes, accompanied by a decline
and eventual collapse of the endo:exo ratio of the products
(Table 2, entries 5−8). A similar effect was observed for the
reaction of phenyl acrylate (10) for 1.0−3.0 M concentrations
of HClO4 in the aqueous layer, (Table 2, entries 17−19).
Separate control exposure of the normal isolated endo:exo
product mixtures to aqueous HClO4 solutions showed that the
product endo-isomer for the 1-octen-3-one cycloadduct 11 was
rapidly isomerized to the thermodynamic exo isomer in >0.1 M
HClO4 and more slowly for the phenyl acrylate product endo-
12 in >1.0 M HClO4. The 1-octen-3-one endo cycloadduct 11
was particularly prone to acid isomerization, in a clean reaction
with no more than 5% decomposition (Table S2, Supporting
Information). These results indicate that with 0.1 M HClO4 in
the water layer, hydrated protons are present at the water−
organic interface and give rise to stronger on-water catalysis.
When the concentration of HClO4 in water surpasses 1.0 M,
these protons are crossing the water−organic interface,
reducing the reaction completion times and causing isomer-
ization of the endo-isomers (Table 2). In contrast to the in-
water reactions, introduction of LiClO4 (0.1−3.0 M) to the
aqueous phase of the on-water reactions had no effect. The
presence of Li+ ions in the bulk water did not influence the
reactions occurring at the water−organic interface for these
systems (Table 2).
Normal kinetic measurements show that there is no D/H
isotope effect for the in-water reactions, for example, between
phthalazinium-2-dicyanomethanide 4 and water-soluble methyl
vinyl ketone (Figure 2 and Table S3, Supporting Information).
Figure 2. Comparison of the rate of cycloaddition reaction of
phthalazinium-2-dicyanomethanide (4) and methyl vinyl ketone in
varying mole fractions of H2O (black) or D2O (red) in acetonitrile at
37 °C.
This is expected, since there are no bonds involving H atoms
active in the transition state. The on-water reactions are taking
place in multiphase sticky mixtures of solids (products and
reactants), organic oil, and water (Figure S5, Supporting
Information). Normal solution kinetic measurements cannot be
made, and extracts suitable for quantitative NMR analysis
cannot be acquired.
Also the comparative rates of appearance of the insoluble
products cannot be measured spectroscopically. In Figure 3 we
b. In-Water Reactions. For the in-water reactions of methyl
acrylate (6), the presence of small quantities of HClO4,
(0.0001−0.1 M) caused no increase in the rate of the reaction
with a small increase in the product endo:exo ratio (Table 3,
entries 2−5). Increasing the concentration of HClO4 from 0.1
to 3.0 M in the solution gave significantly shorter reaction
completion times and further increases in the product endo:exo
ratios (Table 3, entries 6−8). The presence of LiClO4, 0.1 and
3.0 M, in the solution gave similar reduction in the reaction
completion times (Table 3, entries 9−11). The in-water
cycloaddition of 4 with 2-cyclopenten-1-one (5) showed small
reductions in the times to completion for 0.0001−0.1 M
HClO4 (Table 3, entries 13−16) in the solution, but for higher
concentrations of 1.0−3.0 M, the reaction times were again
shortened considerably (Table 3, entries 17−19). The presence
of LiClO4 (0.1−3.0 M) also significantly shortened the reaction
completion times (Table 3, entries 20−22). For these in-water
reactions, the product endo:exo ratio is influenced by the bulk
water hydrophobic effect, particularly for compound 5, and the
effect of the added salt on this ratio is complicated by the
Figure 3. Comparison of the cycloaddition reaction of phthalazinium-
2-dicyanomethanide (4) and 1-octen-3-one (9) in both H2O (black)
or D2O (red) over the reaction time period of 60 min at 20 °C (see
Table 1, entry 9).
show a plot of the rate of growth of products from the
cycloaddition reaction of phthalazinium-2-dicyanomethanide
(4) with 1-octen-3-one (9) on H2O and on D2O. The results in
Figure 3 were acquired by setting up many reactions that were
identical in every respect but with H2O and D2O as the
aqueous phase (see Table S4, Supporting Information). The
reactions were allowed to run for the times shown and worked
up, and the products were isolated and quantified.
−
contrast in the effect of the ions (Li+, pro-hydrophobic; ClO4
anti-hydrophobic). These results are consistent with an
expected complexation of the hydrates H+·nH2O and Li+·
nH2O with O atoms of the 2π-organic molecules in the in-water
transition states.41 Importantly, these effects did not arise for
the on-water reactions, where the ions are confined to the water
phase and not in contact with the reaction transition states.
iii. On H2O versus on D2O. The synthetic results for the
on-water reactions in Table 1 suggest that there is no difference
between on H2O and on D2O for these cycloaddition reactions.
The product endo:exo ratios were constant (8.0
0.5)
across all the runs. The results showed conclusively that there is
no D/H isotope effect in these on-water reactions. If the small
experimental error range were to be considered an isotope
effect, the maximum H/D rate effect would be 1.08. This
laborious approach was necessary. It is also reflective of the
single result quoted by Sharpless, Fokin, and co-workers for the
F
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
of 1,3-dipole 4 in water was ≤5 × 10−6 mol L−1 from UV spectra of
saturated water solution at λmax of 413 nm using the extinction
coefficient of homogeneous solutions in H2O:MeCN (9:1 v/v). The
water solubilites of dipolarophiles 5, 6, 9, 10, and 13 were obtained
computationally from the chemical search engine SciFinder, where the
solubility of organic compounds for any searchable compound within
the CAS database are calculated using the ACD lab software.
Synthesis of Tetracyanoethylene Oxide.45 A solution of
tetracyanoethylene (3.0 g, 34 mmol) in acetonitrile (22 mL) was
cooled to −5 °C in an acetone−ice bath. Hydrogen peroxide (30%)
(2.66 mL, 34 mmol) was added dropwise at such a rate that the
temperature of the reaction remained between 10 and 12 °C. When
the addition was complete, the reaction mixture was stirred for a
further 5 min and then diluted with ice-cold water (150 mL). The
precipitated solid was collected by filtration and washed with water.
The solid was left to dry on a suction pump for 1 h and then used
immediately. The product was obtained as a white solid (3.62 g, 74%):
mp 177−179 °C (sealed tube) (lit. mp 177−178 °C). Found C, 50.0;
N, 38.7. C6N4O requires C, 50.0; N, 38.9.
on-water reaction for dimethyl azodicarboxylate (Figure 1) and
for the plots presented by Beare and McErlean for the on-water
Claisen rearrangements of allylarylamines mentioned above,
where strong D/H isotope effects were discovered.9,26
Alkylarylamines are significant bases with basic pKa values of
3−5 (pKa of conjugate acid).42 Azo compounds in general have
a basic pKa values of ca. 2.42,43 For these reactants, on-water
catalysis involves proton transfer across the water−organic
interface. The reactants herein, ketones and esters, have basic
pKa values of −6 to −7 and are 8−10 orders of magnitude
weaker bases.42 On-water catalysis of their reactions involve
trans-phase H-bonding and no proton transfer. These results
show that on-water catalysis responds to the nature and
properties of the organic reactants. Basic and nucleophilic sites
in organic molecules can facilitate proton transfer across the
water organic interface, and they should also facilitate the
appearance of dangling OHfree groups prior to any potential
proton transfer. It has recently been shown by Ben-Amotz and
co-workers that for water in contact with hydrophobic surfaces
the extent of OHfree groups depends on the surface area and the
electrostatic nature of the surface, with a higher probability of
dangling OH groups near regions of negative polarity.44 The
results of Beare and McErlean and Sharpless, Fokin, and co-
workers combined with these herein point in the same
direction.9,26 They suggest a chameleon-type of behavior for
water at hydrophobic surfaces. This has significant implications
for the interactions of water with hydrophobic regions in
biological systems, such as protein surfaces, grooves in DNA,
and ion-transport mechanisms.
Caution: All operations must be carried out in a fumehood.
Phthalazinium-2-dicyanomethanide 1,3-Dipole (4).45 A sol-
ution of phthalazine (0.91 g, 7.0 mmol) in ethyl acetate (40 mL) was
cooled to below 0 °C in an ice bath. This was treated dropwise with a
cooled ethyl acetate solution (5 mL) of TCNEO (1.0 g, 7.0 mmol).
The yellow product precipitated immediately and was collected by
1
filtration (1.25 g, 92%): mp 263−265 °C (acetonitrile); H NMR δH
(400 MHz, DMSO-d6, 80 °C) 7.92−7.96 (m, 1H), 8.02−8.06 (m,
1H), 8.18−8.24 (m, 2H), 9.40 (s, 1H), 9.60 (s, 1H); 13C NMR δC
(100 MHz, DMSO-d6, 80 °C) 63.5, 117.2, 122.8, 126.4, 128.0, 129.8,
132.8, 135.5, 150.9, 153.9; IR νmax (mull)/cm−1 2191, 2159 (CN).
Found C, 67.9, H, 3.1; N, 28.7. C11H6N4 requires C, 68.0, H, 3.1, N,
28.9.
endo-3,3-Dicyano-1,2-cyclopentano-5-one-1,2,3,10b-
tetrahydropyrrolo[2,1-a]phthalazine and exo-3,3-Dicyano-1,2-
cyclopentano-5-one-1,2,3,10b-tetrahydropyrrolo[2,1-a]-
phthalazine (7).27 A suspension of compound 4 (0.010 g, 0.051
mmol) in water (0.66 mL) was treated with an excess of 2-
cyclopenten-1-one 5 (0.021 mL, 0.257 mmol) and stirred at room
temperature to give a white suspension. The reaction was extracted
into dichloromethane (3 × 10 mL), and the organic layer was dried
over Na2SO4. The solvent was removed under reduced pressure, and
the residue was placed on a flash column of silica gel (230−400 mesh
ASTM) and eluted with a petroleum spirit (bp 40−60 °C)−
dichloromethane mixture in the gradient from 1:1 to 0:1. The
products were eluted from the column as follows.
CONCLUSION
■
In the realm of on-water chemistry, “on-water catalysis” is an
interfacial interaction across the water−organic interface. The
interaction varies in intensity with the properties of the
molecules in the organic phase and can range from weak to
strong H-bonds or proton transfer. Water at the hydrophobic
organic surface displays chameleon-type behavior by respond-
ing to the properties of the organic reactant molecules.
EXPERIMENTAL SECTION
■
Melting points were measured on an electrothermal apparatus. IR
spectra were measured using an FT-IR instrument. All the NMR
spectra were measured on either a 400 or 500 MHz instrument for 1H
NMR and 100 or 125 MHz for 13C NMR. The NMR spectra were
measured with tetramethylsilane as an internal reference and either
CDCl3 or DMSO-d6 as a solvent. The structures were also examined
using COSY, NOEDS, and DEPT. J values are given in hertz (Hz).
Water used for synthesis was ultrapure grade. The stereochemistries of
the endo products and their exo-isomers were established from NOE
difference spectra (NOEDS), which showed strong (7−10%)
enhancements from H-10b to the cis-H-1 in the endo compounds
and the absence of a through-space enhancement for the exo products.
1
Endo-isomer: white solid, mp 228−229 °C (ethanol); H NMR δH
(400 MHz, DMSO-d6) 2.08−2.58 (m, 4H), 3.55 (dd, 1H), 3.80−3.82
(m, 1H), 4.72 (d, J 6.3, 1H), 7.37−7.55 (m, 3H), 7.56 (d, J 7.3, 1H),
7.82 (s, 1H); 13C NMR δC (100 MHz, DMSO-d6) 24.5, 38.2, 38.5,
47.6, 60.5, 112.8, 113.5, 124.4, 126.6, 127.1, 130.7, 131.3, 131.2, 146.6,
213.6; IR νmax/cm−1 (Nujol mull) 1748 (CO). Found C, 69.8; H,
4.1; N, 19.8. C16H12N4O requires C, 69.5; H, 4.4; N, 20.2.
1
Exo-isomer: gum (recolumned crude sample); H NMR δH (400
MHz, DMSO-d6) 2.20−2.52 (m, 4H), 3.49 (dd, 1H), 3.83−3.85 (m,
1H), 4.19 (d, J 9.2, 1H), 7.34−7.78 (m, 4H), 7.89 (s, 1H); 13C NMR
δC (100 MHz, DMSO-d6) 22.9, 37.8, 51.1, 57.0, 60.4, 112.2, 114.3,
124.5, 126.3, 128.9, 132.5, 133.2, 146.7, 211.2; IR νmax/cm−1 (CCl4
liquid cell) 1734 (CO).
1
Endo:exo isomer ratios for product were determined by H NMR
analysis using integration of the H-10b signals, and the error for these
measurements was typically 3% to 5%. For reactions in the water
medium, the water-insoluble product mixtures were separated and
each product isolated as described after prior NMR estimation of the
isomer ratio, unless otherwise stated. Throughout the work the
isolated product yields were accurate to 1% to 3%, with most being
at the lower end of this range. Because of the low water solubilities of
the reactants, the reaction milieu appeared as insoluble multiphase
mixtures and suspensions in water (Figure S5, Supporting
Information). All reactions were carried out in identical glassware,
with a stirring speed of 1200 rpm. This stirring speed was necessary to
ensure that there was sufficient mixing of the reactants. The solubility
endo-1-Methoxycarbonyl-3,3-dicyano-1,2,3,10b-
tetrahydropyrollo[2,1-a]phthalazine and exo-1-Methoxycar-
bonyl Isomer and endo-2-Methoxycarbonyl-3,3-dicyano-
1,2,3,10b-tetrahydropyrrolo[2,1-a]phthalazine (8).27 A suspen-
sion of the compound 4 (0.010 g, 0.051 mmol) in water (0.66 mL)
was treated with methyl acrylate 6 (0.023 mL, 0.257 mmol) and stirred
at ambient temperature, giving a pale yellow solution. The reaction
was extracted into dichloromethane (3 × 10 mL), and the organic
layer was dried over Na2SO4 The solvent was removed under reduced
pressure, and the residue was placed on a flash column of silica gel
(230−400 mesh ASTM) and eluted with a petroleum spirit (bp 40−60
G
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
°C)−dichloromethane mixture in the gradient from 1:1 to 0:1. The
products from the column were isolated in the following order.
2-endo Isomer: gum (recolumned crude sample); 1H NMR δH (400
MHz, CDCl3) 2.53−2.98 (m, 2H), 3.96 (s, 1H), 3.90−3.96 (m, 1H),
4.34 (dd, J 8.6, 8.5, 1H), 7.13−7.53 (m, 4H), 7.81 (s, 1H); IR νmax
(CCl4 liquid cell)/cm−1, 1742 (CO).
(d, J 8.6, 1H), 7.09−7.48 (m, 6H), 7.74 (s, 1H); IR νmax(neat)/cm−1
1750 (CO).
1
1-endo Isomer: off-white solid, mp 126−128 °C (ethanol); H NMR
δH (500 MHz, CDCl3) 3.11 (dd, J 13.7, 8.1, 1H), 3.24 (dd, J 13.7, 3.1,
1H), 3.82−3.86 (m, 1H), 4.97 (d, 1H, J 6.7, 1H), 6.60 (d, J 7.8, 2H),
7.16 (d, J 7.3, 1H), 7.23−7.26 (m, 2H), 7.31−7.33 (m, 1H), 7.40−7.48
(m, 3H), 7.60 (s, 1H); 13C NMR δC (125 MHz, CDCl3) 39.6, 43.7,
55.3, 59.5, 113.1, 113.9, 120.9, 124.8, 125.8, 126.4, 126.9, 129.4, 129.9,
131.7, 144.1, 149.9, 169.0; IR νmax(neat)/cm−1 1748 (CO). Found
C, 70.2; H, 4.2; N, 16.5. C20H14N4O2 requires C, 70.1; H, 4.1; N, 16.4.
1,2-Diphenyl-3-cyanopyrrolo[2,1-a]phthalazine (14).27 A sus-
pension of compound 4 (0.010 g, 0.051 mmol) in water (0.66 mL)
was treated with diphenylacetylene (13) (0.091 g, 0.51 mmol) and
stirred under reflux for 24 h. The reaction was extracted into
dichloromethane (3 × 10 mL), and the organic layer was dried over
Na2SO4. The solvent was removed under reduced pressure, and the
residue was placed on a flash column of silica gel (230−400 mesh
ASTM) and eluted with a petroleum spirit (bp 40−60 °C)−
dichloromethane mixture in the gradient from 1:1 to 0:1: off-white
solid, mp 213−214 °C (acetonitrile); 1H NMR δH (400 MHz, DMSO-
This isomer was formed in a yield of 2% and was not included in the
endo:exo ratios that are mentioned in the tables.
1
1-exo Isomer: gum (recolumned crude sample); H NMR δH (400
MHz, CDCl3) 3.00−3.22 (m, 2H), 3.86 (s, 3H), 3.54−3.59 (m, 1H),
4.46 (d, J 8.8, 1H), 7.40−7.61 (m, 4H), 7.89 (s, 1H); 13C NMR δC
(100 MHz, CDCl3) 42.3, 58.7, 113.2, 113.5, 123.4, 124.9, 126.0, 128.8,
131.5, 145.8; IR νmax(CCl4 liquid cell)/cm−1, 1751 (CO).
1-endo Isomer: white crystalline solid, mp 132−133 °C (ethanol);
1H NMR δH (400 MHz, CDCl3) 2.99−3.05 (m, 1H), 3.12−3.16 (m,
1H), 3.55 (s, 3H), 3.63−3.68 (m, 1H), 4.82 (d, J 6.6, 1H), 7.26−7.45
(m, 4H), 7.66 (s, 1H); 13C NMR δC (100 MHz, CDCl3) 39.2, 42.9,
52.4, 55.8, 59.2 113.3, 113.9, 124.7, 125.1,126.1, 127.1, 129.7, 130.2,
144.4, 170.6; IR νmax(mull)/cm−1 1742 (CO). Found: C, 63.9; H,
4.3; N, 19.9. C15H12N4O2 requires C, 64.3; H, 4.3; N, 19.9.
d6, 60 °C) 7.31−7.64 (m, 13H), 8.06 (d, J 7.3, 1H), 8.95 (s, 1H); 13
C
1-endo-Hexanoyl-3,3-dicyano-1,2,3,10b-tetrahydropyrrolo-
[2,1-a]phthalazine and 1-exo-Hexanoyl-3,3-dicyano-1,2,3,10b-
tetrahydropyrrolo[2,1-a]phthalazine (11).27 A suspension of
compound 4 (0.010 g, 0.051 mmol) in water (0.66 mL) was treated
with an excess of 1-octen-3-one 9 (0.038 mL, 0.0255 mmol) and
stirred at ambient temperature to give a pale yellow solution. The
reaction was extracted into dichloromethane (3 × 10 mL), and the
organic layer was dried over Na2SO4. The solvent was removed under
reduced pressure, and the residue was placed on a flash column of
silica gel (230−400 mesh ASTM) and eluted with a petroleum spirit
(bp 40−60 °C)−dichloromethane mixture in the gradient from 1:1 to
0:1. The endo:exo isomers proved difficult to isolate. The character-
ization given is for the mixture of the endo and exo isomers and the
ratio of endo:exo isomers was determined through integration of the
H-10b signals.
NMR δC (100 MHz, DMSO-d6, 60 °C) 112.5, 116.3, 120.9, 121.2,
123.1, 126.0, 127.6, 127.9, 128.6, 128.7, 129.0, 130.6, 130.8, 132.1
(some overlap of signals), 146.1; IR νmax /cm−1 (Nujol mull) 2216
(CN); HRMS (ESI) calcd for C24H15N3 (M + H)+ 346.1345, found
346.1349.
Caution: All operations must be carried out in a fumehood.
Kinetics. The kinetics were measured by recording the
disappearance of phthalazinium-2-dicyanomethanide (4) at 420 nm
using its UV−Vis spectrum. Spectra were measured using a Hewlett-
Packard Agilent Technologies 8453 UV−vis spectrophotometer
featuring an automatic changer for up to eight glass cuvettes of path
length 1 cm. The temperature was maintained at 37 °C. The reaction
was monitored under pseudo-first-order conditions. The 1,3-dipole 4
was recrystallized twice before use. The solvents used were HPLC
grade and the water was Millipore grade. The initial concentration of
the 1,3-dipole 4 was 3.2 × 10−5 M and the dipolarophiles were used in
excess ranging from 100 to 700 times. The reactions were monitored
using the π−π* transition of the 1,3-dipole 4 at 420 and 413 nm for
0.80 and 0.90 mole fraction water−acetonitrile. Kinetic runs were
performed at three different concentrations of dipolarophiles and
repeated a minimum of three times. The rate constants were
reproducible to 2%. The solutions changed from yellow to colorless
as the rates progressed, and typical run times were on the order 5 min
to 1 h, depending on the dipolarophile concentration. The results of
the kinetics are shown in Figure 2, and the rate data are given in the
Supporting Information (Table S3).
1-endo And 1-exo isomers: gum; 1H NMR δH (500 MHz, CDCl3) 0.75
(t, 3H, endo), 0.87−0.98 (m, 2H, endo, 3H, exo), 1.06−1.14 (m, 2H,
endo), 1.22−1.36 (m, 2H, endo, 2H, exo), 1.64−1.73 (m, 2H, exo),
2.25−2.30 (m, 2H, exo), 2.29−2.32 (m, 2H, endo, 2H, exo), 2.88−
2.92 (m, 1H endo; 1H exo), 3.03 (dd, J 13.8, 9.1, 1H, endo), 3.10 (dd,
J 13.8, 11.2, 1H, exo), 3.55−3.60 (m, 1H, endo), 3.64−3.68 (m, 1H,
exo), 4.50 (d, J 9.4, 1H, exo), 4.81 (d, 1H, J 7.5, 1H, endo), 6.97 (d, J
7.8, 1H, exo), 7.09 (d, J 7.4, 1H, endo), 7.26−7.53 (m, 3H, endo; 3H,
exo), 7.57 (s, 1H, endo), 7.73 (s, 1H, exo); 13C NMR δC (125 MHz,
CDCl3) 13.8 (endo), 13.9 (exo), 22.1 (endo), 22.7 (exo), 23.1 (exo),
22.3 (endo), 30.8 (exo), 30.9 (endo), 38.2 (exo), 38.8 (endo), 42.5
(endo), 42.3 (exo), 49.2 (exo), 50.0 (endo), 55.9 (endo), 58.0 (exo),
58.9 (exo), 112.9, 113.3 (exo), 113.1, 113.6 (endo), 123.6 (exo), 124.6
(endo), 125.3 (endo), 126.2 (exo), 127.0 (endo), 128.9 (exo), 129.3
(endo), 130.5 (endo and exo), 131.8 (endo), 132.0 (exo), 144.5
(endo), 146.3 (exo), 207.1 (exo), 208.0 (endo); HRMS (ESI) calcd
for C19H20N4O (M + H)+ 321.1716, found 321.1725.
ASSOCIATED CONTENT
* Supporting Information
■
S
Influence of the concentration of HClO4 (Table S1), exploring
the stability of the cycloaddition products (Table S2), details on
the rates of reactions from Figure 2 (Table S3) and the yields
obtained that are shown in Figure 3 (Table S4), graphs
showing the influence of the concentration of HClO4 on the
various reactions (Figures S1−S4), photos showing the reaction
The terms endo and exo refer to the isomers to which the signal
belongs. Some of the exo isomer peaks are missing in the 13C NMR
due to overlap with the major endo isomer.
endo-1-Phenoxycarbonyl-3,3-dicyano-1,2,3,10b-
tetrahydropyrollo[2,1-a]phthalazine and exo-1-Phenoxycar-
bonyl-3,3-dicyano-1,2,3,10b-tetrahydropyrollo[2,1-a]-
phthalazine (12).27 A suspension of compound 4 (0.010 g, 0.051
mmol) in water (0.66 mL) was treated with an excess of phenyl
acrylate (10) (0.035 mL, 0.257 mmol) and stirred at ambient
temperature to give a pale yellow solution. The reaction was extracted
into dichloromethane (3 × 10 mL), and the organic layer was dried
over Na2SO4. The solvent was removed under reduced pressure, and
the residue was placed on a flash column of silica gel (230−400 mesh
ASTM) and eluted with a petroleum spirit (bp 40−60 °C)−
dichloromethane mixture in the gradient from 1:1 to 0:1. The
products from the column were isolated in the following order.
1
progress (Figure S5), H and 13C NMR spectra for 11 and 12,
and selected NOEDS enhancements for compounds 7 and 8.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*R.N.B. e-mail: r.debuitleir@nuigalway.ie.
*A.G.C. e-mail: anthony.g.coyne@gmail.com.
■
1
Notes
1-exo Isomer: gum; H NMR δH (500 MHz, CDCl3) 3.14 (m, 1H),
3.27 (dd, J 14.2, 5.3, 1H), 3.68−3.74 (m, 1H), 4.55 (d, J 9.2, 1H), 6.74
The authors declare no competing financial interest.
H
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Chem. Rev. 1957, 57, 1−45. (c) Handbook of Chemistry and Physics,
55th ed.; CRC: Cleveland, OH, 1974−1975; pp D126−D128.
(43) (a) Hine, J. Physical Organic Chemistry, 2nd ed.; McGraw-Hill
Book Co.: New York, 1962; pp 54−55. (b) Albert, A. Heterocyclic
Chemistry, 2nd ed.; Athlone Press: London, 1968; pp 438−440.
(44) Davis, J. G.; Rankin, B. M.; Gierszal, K. P.; Ben-Amotz, D. Nat.
Chem. 2013, 5, 796−802.
REFERENCES
■
(1) Rideout, D. C.; Breslow, R. J. Am. Chem. Soc. 1980, 102, 7816−
7817.
(2) Breslow, R. Acc. Chem. Res. 1991, 24, 159−164.
(3) Breslow, R. Acc. Chem. Res. 2004, 37, 471−478.
(4) Garner, P. P. Organic Synthesis in Water; Grieco, P. A., Ed.;
Blackie Academic and Professional: London, 1998; p 1−41.
(5) Li, C.-J. Chem. Rev. 2005, 105, 3095−3165.
(45) Linn, W. J.; Webster, O. W.; Benson, R. E. J. Am. Chem. Soc.
1965, 87, 3651−3656.
(6) Lindstrom, U. M. Chem. Rev. 2002, 102, 2751−2771.
̈
(7) Otto, S.; Engberts, J. B. F. N. Pure Appl. Chem. 2000, 72, 1365−
1372.
(8) Engberts, J. B. F. N.; Blandamer, M. J. Chem. Commun. 2001,
1701−1708.
(9) Narayan, S.; Muldoon, J.; Finn, M. G.; Fokin, V. V.; Kolb, H. C.;
Sharpless, K. B. Angew. Chem., Int. Ed. 2005, 44, 3275−3279.
(10) Klijn, J. E.; Engberts, J. B. F. N. Nature 2005, 435, 746−747.
(11) Molteni, G. Heterocycles 2006, 68, 2177−2202.
(12) Pirrung, M. C. Chem.Eur. J. 2006, 12, 1312−1317.
(13) Lindstrom, W. M. Organic Reactions in Water: Principles,
̈
Strategies and Applications, 1st ed.; Blackwell Publishing: Oxford, 2007.
(14) Chanda, A.; Fokin, V. V. Chem. Rev. 2009, 109, 725−748.
(15) Butler, R. N.; Coyne, A. G. Chem. Rev. 2010, 110, 6302−6337.
(16) Kobayashi, S. Water in Organic Synthesis. Science of Synthesis
Workbench Edition, Thieme: Stuttgart, 2012.
(17) Jung, Y. S.; Marcus, R. A. J. Am. Chem. Soc. 2007, 129, 5492−
5502.
(18) Manna, A.; Kumar, A. J. Phys. Chem. A 2013, 117, 2446−2454.
(19) Thomas, L. L.; Tirado-Rives, J.; Jorgensen, W. L. J. Am. Chem.
Soc. 2010, 132, 3097−3104.
(20) Karhan, K.; Khaliullin, R. Z.; Kuhne, T. D. arXiv.org 2014,
̈
arXiv:1408.5161v1 [physics.chem-ph] (accessed Oct 2014).
(21) Sela, T.; Vigalok, A. Org. Lett. 2014, 16, 1964−1967.
(22) Yu, J.-S.; Liu, Y.-L.; Tang, J.; Wang, X.; Zhou, J. Angew. Chem.,
Int. Ed. 2014, 53, 9512−9516.
(23) Zuo, Y.-J.; Qu, J. J. Org. Chem. 2014, 79, 6832−6839.
́ ́ ́
(24) Domingo, L. R.; Saez, J. A.; Zaragoza, R. J.; Arno, M. J. Org.
Chem. 2008, 73, 8791−8799.
(25) Beattie, J. K.; McErlean, C. S. P.; Phippen, C. B. W. Chem.Eur.
J. 2010, 16, 8972−8974.
(26) Beare, K. D.; McErlean, C. S. P. Org. Biomol. Chem. 2013, 11,
2452−2459.
(27) Butler, R. N.; Coyne, A. G.; Cunningham, W. J.; Moloney, E. M.
J. Org. Chem. 2013, 78, 3276−3291.
(28) Braun, R.; Sauer, J. Chem. Ber. 1986, 119, 1269−1274.
(29) Grieco, P. A.; Nunes, J. J.; Gaul, M. D. J. Am. Chem. Soc. 1990,
112, 4595−4596.
(30) Auge,
877−879.
́
J.; Gil, R.; Kalsey, S.; Nubin-Germain, N. Synlett 2000,
(31) Desimoni, G.; Faita, G.; Righetti, P. P.; Taccori, G. Tetrahedron
1991, 47, 8399−8406.
(32) Shtyrlin, Y. G.; Murzin, D. G.; Luzanova, N. A.; Iskhakova, G.
G.; Kieselev, V. D. Tetrahedron 1998, 54, 2631−2646.
(33) Heard, N. E.; Turner, J. J. Org. Chem. 1995, 60, 4302−4304.
(34) Srisiri, W.; Padias, A. B.; Hall, H. K., Jr. J. Org. Chem. 1993, 58,
4185−4186.
(35) Reetz, M. T.; Fox, D. N. A. Tetrahedron Lett. 1993, 34, 1119−
1122.
(36) Grieco, P. A.; Cooke, R. J.; Henry, K. J.; Vanderroest, J. M.
Tetrahedron Lett. 1991, 32, 4665−4668.
(37) Davies, A. G.; Kinart, W. J. J. Chem. Soc. Perkin Trans. 2 1993,
2281−2284.
(38) Grieco, P. A.; Clark, J. D.; Jagoe, C. T. J. Am. Chem. Soc. 1991,
113, 5488−5489.
(39) Saidi, M. R.; Heydari, A.; Ipakschi, J. Chem. Ber. 1994, 127,
1761−1764.
́
(40) Auge, J.; Leroy, F. Tetrahedron Lett. 1996, 37, 7715−7716.
(41) Jenner, G.; Salem, R. B. Tetrahedron 1997, 53, 4637−4648.
(42) (a) March, J. Advanced Organic Chemistry, 4th ed.; John Wiley &
Sons Inc: New York, 1992; pp 250−252. (b) Paul, M. A.; Long, F. A.
I
DOI: 10.1021/jo502732y
J. Org. Chem. XXXX, XXX, XXX−XXX