Investigation of 1,3,5-triaza-7-phosphaadamantane-stabilized silver nanoparticles as catalysts for the hydration of benzonitriles and acetone cyanohydrin

Article abstract of DOI:10.1021/cs500830s

A straightforward synthesis of water-soluble silver nanoparticles stabilized by PTA (1,3,5-triaza-7-phosphaadamantane, a water-soluble phosphine ligand) ligands was developed. The nanoparticles were thoroughly characterized by ultraviolet-visible spectroscopy, ³¹P nuclear magnetic resonance spectroscopy, transmission electron microscopy, and energy dispersive X-ray spectroscopy. The effectiveness of the Ag-PTA nanoparticles as catalysts for the hydration of nitriles to amides in water under mild conditions was explored using a series of substituted benzonitriles and cyanohydrins. In comparison to all previously investigated homogeneous catalysts, the Ag-PTA system excels at cyanohydrin hydration, including acetone cyanohydrin hydration. Cyanohydrins are in equilibrium with small amounts of cyanide, and experiments revealed that the Ag-PTA nanoparticles disassemble in the presence of cyanide. The catalyst solution, which is proposed to contain a soluble Ag(CN)_n^1-ncomplex (with n likely equal to 2), remained unpoisoned even in the presence of 10 equiv of cyanide. It is suggested that no cyanide poisoning occurs because the Ag(I) complex is labile. Overall, the Ag-PTA catalyst system (a) is not poisoned by cyanide, (b) catalyzes hydration reactions under mild conditions (in air and at relatively low temperatures), (c) is easily synthesized from cheap starting materials, and (d) can hydrate heteroaromatics in good yields. The recognition of the importance of labile metal cyanide bonding represents an important step forward in catalyst design for improving the catalytic hydration of acetone cyanohydrin. (Chemical Equation Presented).

Full text of DOI:10.1021/cs500830s

Research Article
pubs.acs.org/acscatalysis
Investigation of 1,3,5-Triaza-7-phosphaadamantane-Stabilized Silver
Nanoparticles as Catalysts for the Hydration of Benzonitriles and
Acetone Cyanohydrin
Tobias J. Sherbow,† Emma L. Downs,† Richard I. Sayler,† Joshua J. Razink,‡ J. Jerrick Juliette,§
and David R. Tyler*,†
†Department of Chemistry, University of Oregon, Eugene, Oregon 97403, United States
‡Center for Advanced Materials Characterization in Oregon, University of Oregon, Eugene, Oregon 97403, United States
§Shell Oil Products LLC, Shell Deer Park Site, Deer Park, Texas 77536, United States
S
* Supporting Information
ABSTRACT: A straightforward synthesis of water-soluble
silver nanoparticles stabilized by PTA (1,3,5-triaza-7-phos-
phaadamantane, a water-soluble phosphine ligand) ligands was
developed. The nanoparticles were thoroughly characterized
by ultraviolet−visible spectroscopy, 31P nuclear magnetic
resonance spectroscopy, transmission electron microscopy,
and energy dispersive X-ray spectroscopy. The effectiveness of
the Ag−PTA nanoparticles as catalysts for the hydration of
nitriles to amides in water under mild conditions was explored
using a series of substituted benzonitriles and cyanohydrins. In
comparison to all previously investigated homogeneous
catalysts, the Ag−PTA system excels at cyanohydrin hydration,
including acetone cyanohydrin hydration. Cyanohydrins are in equilibrium with small amounts of cyanide, and experiments
revealed that the Ag−PTA nanoparticles disassemble in the presence of cyanide. The catalyst solution, which is proposed to
1−n
contain a soluble Ag(CN)n complex (with n likely equal to 2), remained unpoisoned even in the presence of 10 equiv of
cyanide. It is suggested that no cyanide poisoning occurs because the Ag(I) complex is labile. Overall, the Ag−PTA catalyst
system (a) is not poisoned by cyanide, (b) catalyzes hydration reactions under mild conditions (in air and at relatively low
temperatures), (c) is easily synthesized from cheap starting materials, and (d) can hydrate heteroaromatics in good yields. The
recognition of the importance of labile metal cyanide bonding represents an important step forward in catalyst design for
improving the catalytic hydration of acetone cyanohydrin.
KEYWORDS: nitrile hydration, cyanohydrin hydration, nanoparticle catalysis, silver nanoparticles, PTA ligand,
water-soluble nanoparticles, cyanide-resistant catalyst
Our research group and others have shown that nitriles can
be successfully hydrated to their corresponding amides in high
yields and under mild conditions with various transition metal
catalysts.1,4,5 Accordingly, we tested many of these same
homogeneous catalysts to see if they could be used for the
hydration of acetone cyanohydrin. Although the hydration of
cyanohydrins would seem to be relatively straightforward, there
were complications. Cyanohydrins are in equilibrium with
hydrogen cyanide and the corresponding aldehyde or ketone
(Scheme 2). When cyanide is present, it can poison a
homogeneous catalyst by irreversibly binding to one or more
active sites.6 We previously reported on cyanohydrin hydration
with a number of known nitrile hydration catalysts, but only
INTRODUCTION
■
In an industrial setting, the hydration of nitriles to amides is
typically performed under extreme conditions.1 For example,
the industrial route for the hydration of acetone cyanohydrin
[α-hydroxyisobutyronitrile (ACH), the industrial precursor to
methyl methacrylate (MMA)] uses concentrated sulfuric acid
(Scheme 1). A major byproduct of this reaction is ammonium
hydrogen sulfate (AHS), 2.5 kg of which is formed for every
kilogram of MMA produced.2 This byproduct poses a
significant environmental problem and economic disadvantage.
To dispose of the 7 million metric tons of AHS produced
annually, it is pyrolyzed at 1000 °C to re-form sulfuric acid
(Scheme 1).2,3 This process is energy intensive, and alternate
synthetic routes have therefore been sought.1 A transition
metal-catalyzed hydration reaction run under mild conditions
would potentially be more economically and environmentally
favorable.
Received: February 12, 2014
Revised: July 24, 2014
© XXXX American Chemical Society
3096
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
Scheme 1. Current Chemical Process Used for the Industrial Hydration of Cyanohydrins
a
Scheme 2. Equilibrium between a Cyanohydrin and the Corresponding Aldehyde/Ketone and Hydrogen Cyanide
a
When cyanide is present, it typically binds irreversibly to the metal catalyst, poisoning it.
minimal hydration was observed because of cyanide poison-
ing.6,7
silver nanoparticles are stabilized for a short time by excess
borohydride that is adsorbed to the surface of the particles.24
However, over a period of 30 min, the borohydride
decomposes24 and the nanoparticles precipitate from the
solution as bulk metal. To stabilize the particles in this study,
PTA was added to the solution immediately after all of the
AgNO3 solution had been added. Experiments showed that a
1.1:1 ratio (PTA:AgNO3) gave the maximal nanoparticle
stability with respect to decomposition or unwanted precip-
In an effort to find cyanide-resistant catalysts, we turned our
attention to nanocatalysts. Nanoparticles are capable of
catalyzing a wide variety of reactions,8 including hydrogenation,
dehalogenation, oxidation, and photocatalytic reactions. How-
ever, only a handful of investigations have focused on hydration
reactions and specifically on nitrile hydration.9−19 These studies
show that nanoparticles have activity comparable to that of
homogeneous transition metal catalysts for nitrile hydration.
However, no nanoparticle catalysts have been tested for
cyanohydrin hydration. In this paper, we report on our
investigation of nitrile and cyanohydrin hydration using silver
nanoparticles. Because hydration reactions are typically
conducted in an aqueous solution, the nanoparticles in this
study incorporated a water-soluble stabilizing ligand shell of
1,3,5-triaza-7-phosphaadamantane (PTA), which has been used
previously to produce water-soluble nanoparticle catalysts of Pt
and Ru.20
itation.25 In contrast to the BH4 -stabilized particles, the
−
nanoparticles stabilized by PTA remained dispersed in an
aqueous solution for several weeks. The PTA-stabilized silver
nanoparticles are a slightly darker orange color than the yellow
borohydride-stabilized particles. The ultraviolet−visible (UV−
vis) spectrum showed a broadening and a decrease in the
intensity of the LSPR peak at λmax = 398 nm in the PTA-
stabilized particles (Figure 1) compared to that of the
RESULTS AND DISCUSSION
■
Figure 1. UV−vis spectrum of the localized surface plasmon resonance
band of the Ag−PTA nanoparticles before (blue) and after (red) the
addition of 1.1 equiv of PTA ligand.
Synthesis and Characterization of Ag−PTA Nano-
particles. The Ag−PTA nanoparticles were synthesized by the
reduction of silver nitrate with sodium borohydride in cold
water (eq 1).
borohydride-stabilized particles. In other studies, the broad-
ening and the decrease in the intensity of the LSPR peak have
been attributed to modification of the electronic properties of
the nanoparticle surface by the ligand shell.21,22 A similar
explanation is likely applicable here.
H2O,0°C
AgNO + NaBH4 ⎯⎯⎯⎯⎯⎯⎯⎯→⎯ Ag0 + 1/2B2H6 + NaNO3
3
(1)
As the AgNO3 was slowly added to the NaBH4 over 10 min, the
solution in the reaction flask turned from clear to yellow. The
yellow color is attributed to the localized surface plasmon
resonance (LSPR) of the nanoparticles (λmax = 398 nm) and is
evidence that a silver nanoparticle surface is present.21−23 In the
absence of an added ligand, Solomon et al. showed that the
The 31P nuclear magnetic resonance (NMR) spectrum of the
Ag−PTA nanoparticles showed a peak at −85.6 ppm, assigned
to the coordinated PTA ligand. This resonance is shifted
downfield and broadened compared to that of uncoordinated
PTA, which has a sharp peak at −98.3 ppm (Figure 2). The
3097
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
showed an average diameter of 3.5 2 nm (Figure 4). The
smaller size of the nanoparticles reported here is attributed to
Figure 2. 31P NMR spectrum of (a) the free PTA ligand at −98.3 ppm
and (b) the bound PTA ligand on the surface of the Ag−PTA
nanoparticles at −85.6 ppm.
Figure 4. Size distribution histogram of the Ag−PTA nanoparticles
(1175 nanoparticles were analyzed).
the PTA ligand. In the preparation described above, PTA was
added after the addition of all the AgNO3 to the NaBH4. Prior
literature preparations did not use a stabilizing ligand other
than excess borohydride. Borohydride does not stabilize the
nanoparticles as well as PTA, and in its absence, the
nanoparticles precipitate as bulk metal.29,30 In the case of
added PTA, further growth is prevented by coordination of the
stronger-binding PTA ligand.
In addition to TEM characterization, the NPs were also
analyzed by energy dispersive X-ray spectroscopy (EDS). This
technique confirmed that the NPs were composed of Ag (see
Figure S1 of the Supporting Information).
Catalysis Studies. A variety of activated and deactivated
benzonitriles were hydrated to test the catalytic properties of
the Ag−PTA nanoparticles (Table 1). Benzonitrile, a common
substrate for testing nitrile hydration,12,15 was hydrated to
benzamide in 90% yield. Initial turnover frequencies (iTOFs)
for a range of para-substituted benzonitriles (calculated after
reaction for 26 h) followed the trend that hydration becomes
faster as the electron-withdrawing ability of the substituent
increases, as indicated by the Hammett plot in Figure 5.6 [The
positive slope (ρ = 1.8) of the Hammett plot indicates that
electron-withdrawing groups facilitate the reaction.] As shown
in prior studies,1,6,31 the correlation between a faster initial
turnover frequency and improved substituent electron-with-
drawing ability is attributed to the increased electrophilicity of
the carbon atom in the nitrile as the electron-withdrawing
ability improves. Overall, these results suggest that nucleophilic
attack of hydroxide or water is the rate-limiting step, as has
been shown previously with homogeneous nitrile hydration
catalysts.1,6 An additional important point to note is that the
Ag−PTA nanoparticles are catalytically active for nitrile
hydration in air and at relatively mild temperatures (90 °C).
In contrast, previously reported Ag nanoparticle catalysts
required air-free conditions and temperatures in excess of 150
°C.12,15,17
broadened peak may be indicative of rapid exchange between
between free and bound PTA. However, a number of studies
have shown that ligands bonded to nanoparticles typically
display broadened and shifted peaks upon coordinating to the
nanoparticle.20,26−28 The broadening and downfield shifts are
attributed to the induced shielding of the phosphine ligands
when they are bonded to the nanoparticles. Consistent with
this latter explanation, it is noted that no resonance for the
uncoordinated PTA ligand was observed in the 31P NMR
spectrum of the solution containing the nanoparticles, strongly
suggesting that all of the PTA is bonded to the surface.
Transmission Electron Microscopy (TEM) Character-
ization. The size distribution of the Ag−PTA nanoparticles
was analyzed by TEM and HRTEM (high-resolution trans-
mission electron microscopy). Prior research showed that
NaBH4 reduction of AgNO3 yielded nanoparticles with
diameters between 10 and 14 nm.24 In contrast, TEM images
of the Ag−PTA nanoparticles prepared in this study (Figure 3)
Hydration of several aliphatic nitriles was also examined with
the Ag−PTA nanoparticles. Acetonitrile, propionitrile, and
methoxyacetonitrile were not hydrated appreciably. It is
hypothesized that this lack of activity is due to the increased
electron density of the nitrile carbon in these aliphatic systems
compared to that of the benzonitriles, making the nucleophilic
attack on the nanoparticle-bound substrate less favorable.
Recycling Studies. To test the longevity of the catalyst and
to determine if it could be recycled after use, catalyst recycling
Figure 3. TEM images of (a and b) 2 nm Ag−PTA nanoparticles and
(c) 4−10 nm particles and (d) HRTEM image of 4−10 nm particles.
3098
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
a
Table 1. Selected Nitrile Hydration Results Using Ag−PTA Nanoparticles
a
All reactions were conducted at 90 °C. The catalyst concentration was 0.21 M for all trials except for entry 1, for which it was 0.22 M.
1) to the Ag−PTA nanoparticle solution produced an
unexpected result: in <30 s, the orange nanoparticle solution
became colorless. The UV−vis spectrum of the colorless
solution showed the disappearance of the LSPR band at 398
nm, indicating that the nanoparticles had dissolved. Although
the Ag−PTA nanoparticles were no longer present, the
resulting solution was monitored to see if hydration would
occur.
After 504 h, 9.7% of the ACH was hydrated to 2-
hydroxyisobutyramide (HIBAM) (Table 2, entry 9). No further
conversion was possible because the ACH had decomposed
completely to acetone and HCN. For comparison, the best
homogeneous catalyst for hydration of ACH, [Ru(η6-cymene)-
Cl2P(NMe2)3], showed a maximum of 15% conversion of ACH
to HIBAM in 22 h.31 (The catalyst was completely poisoned at
this point so no further conversion was possible.)
Figure 5. Hammett plot for hydration of para-substituted benzoni-
triles.
Although all other homogeneous catalysts become inactive
because of cyanide poisoning after ACH degradation, the
colorless solution generated from the dissolution of the Ag
nanoparticles remained active. Addition of more ACH to the
solution (after the first 504 h of reaction) resulted in the
conversion of the added ACH to HIBAM, indicating that the
solution was still catalytically active despite the presence of
cyanide. As with the original solution, not all of the additional
ACH was hydrated because some of it degraded to acetone and
HCN. The solution, however, remained catalytically active
because addition of yet a third aliquot of ACH led to more
HIBAM. This is a very exciting result because all previous
catalysts tested for cyanohydrin hydration have been poisoned
after one trial.
studies were conducted for the hydration of 4-nitrobenzonitrile.
After the initial batch of substrate had all reacted, the reaction
solution was cooled to 0 °C, causing the 4-nitrobenzamide to
crystallize. The product was removed by filtration, and an
additional aliquot of 4-nitrobenzonitrile was added to the
catalyst solution. The new aliquot of 4-nitrobenzonitrile was
also completely converted to 4-nitrobenzamide. The catalyst
was reused yet again, without any loss of activity. After
recycling, the catalyst was examined by STEM (Figure 6),
which showed the nanoparticles were still present and had
approximately the same size and shape that they did before they
were used in the catalytic reaction.
Cyanohydrin Hydrations. The catalytic activity of the
Ag−PTA nanoparticles toward cyanohydrin hydration resulted
in several observations. Addition of either acetone cyanohydrin
(structure I in Chart 1) or glycolonitrile (structure II in Chart
Ag−PTA Nanoparticle Dissolution Studies. Studies
were next conducted to determine what caused the dissolution
of the nanoparticle in the presence of cyanohydrins. As
3099
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
Figure 6. STEM images of PTA-stabilized Ag NPs before (a) and after (b) catalyzing the hydration of benzonitrile. The particles do not undergo
drastic changes under catalytic conditions.
laboratory, several homogeneous ruthenium nitrile hydration
catalysts were found to increase in activity with small amounts
(<1 equiv) of added cyanide.7,31,35 This change was attributed
to the electron-withdrawing ability of coordinated cyanide, a
feature that facilitates nucleophilic attack of water or hydroxide
ion on the coordinated nitrile in a complex with two active
sites.31 However, as the amount of added cyanide increased
above 1 equiv, the rate of hydration significantly decreased
because cyanide irreversibly bonded to both active sites in the
complex. The result of adding cyanide to the nanoparticles did
not follow this behavior because excess cyanide did not inhibit
the catalytic reactivity. For example, when 0.17 equiv of cyanide
was added to the reaction mixture, the rate of hydration
increased compared to the case in which no cyanide was added
(Table 2, entry 10). However, as further cyanide was added to
the reaction mixture (up to 9.3 equiv), the rate of hydration was
not affected.
Catalysis with Silver Cyanide Complexes. A control
reaction showed that a solution of AgNO3 and KCN hydrated
ACH, which suggests that a Ag(CN)n1−n-type complex is
capable of hydrating cyanohydrins. Thus, ACH was hydrated to
HIBAM in 6.5% yield (Table 3). This result supports the
hypothesis that, when the Ag−PTA nanoparticles become
oxidized in the presence of oxygen and cyanide, they form a
Ag(CN)n1−n-type complex that is capable of hydrating nitriles.
Further studies were conducted to determine the specific
identity of the active catalyst.
Chart 1. Molecular Structures of Acetone Cyanohydrin (I)
and Glycolonitrile (II)
mentioned previously, some cyanide is present in these
cyanohydrin substrate solutions because of the equilibrium
among the cyanohydrin, the aldehyde or ketone, and HCN
(Scheme 2).6 It has been shown in the literature that silver
nanoparticles can disassemble to form Ag(CN)n-type com-
plexes in the presence of cyanide and dissolved oxygen.23,34 It is
proposed that this reactivity is also present in the Ag−PTA
nanoparticle system. Addition of as little as 0.17 equiv of
cyanide to the Ag−PTA nanoparticle solution resulted in the
solution becoming colorless almost immediately, accompanied
by the disappearance of the LSPR band from the electronic
spectrum. Under an inert atmosphere, the nanoparticles
persisted for several hours but dissolution eventually occurred.
In an inert atmosphere, hydration of ACH proceeded at a rate
similar to those of trials conducted in air.
Cyanohydrin Hydration with the Silver Cyanide
Solution. Experiments showed that the colorless solution
formed by exposure of the Ag−PTA NPs to cyanide was
catalytically active. Table 2 (entries 8−13) has five separate
ACH hydration trials with varying amounts of cyanide to show
the effect of cyanide concentration on the catalyst.
AgCN is known to form insoluble linear coordination
polymers in the solid state, and this species was observed as a
Note that, unusually, the rate of hydration increased as the
amount of added cyanide increased. In previous work by our
Table 2. Hydration of Acetone Cyanohydrin (ACH) Using the Catalytically Active, Colorless Solution Resulting from the
a
Mixing of Ag−PTA Nanoparticles and Cyanohydrins
[nitrile]
(mM)
[KCN]
(mM)
[catalyst]
(mM)
catalyst loading
(mol %)
equiv of CN− to Ag
atoms
%
reaction time
(h)
iTOF
(h−1
entry
nitrile
hydration
)
8
glycolonitrile
ACH
68.8
65.3
65.2
64.6
61.8
61.5
0.0
0.27
0.27
0.27
0.27
0.25
0.24
0.40
0.41
0.41
0.41
0.41
0.39
0
4.1
9.7
336
504
504
504
843
843
9
0.0
0
0.078
0.087
0.09
10
11
12
13
ACH
0.05
0.26
1.26
2.26
0.2
1.0
5.0
9.3
13.8
17.1
14.5
17.1
ACH
ACH
0.1
ACH
0.093
a
All reactions were conducted at 90 °C.
3100
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
a
Table 3. Hydration Reactions with Silver Cyanide Complexes
entry
nitrile
ACH
catalyst
[nitrile] (M) [catalyst] (mM) catalyst loading (mol %) % hydration reaction time (h) iTOF (h−1
)
14
15
16
17
18
AgNO3(aq) and KCN(aq)
KAgCN2(aq)
0.063
0.049
0.048
0.048
0.048
2.70
0.25
0.25
0.25
0.25
4.3
6.5
6.5
499
216
216
216
216
0.01
0.14
0.7
ACH
0.52
0.52
0.52
0.52
benzonitrile
ACH
KAgCN2(aq)
50
KAgCN2(aq) and PTA(aq)
KAgCN2(aq) and KCN(aq)
6.0
6.0
0.13
0.13
ACH
a
All reactions were conducted at 90 °C.
Table 4. Control Reactions for the Hydration of p-Nitrobenzonitrile
[nitrile]
(M)
[catalyst]
(mM)
catalyst loading
(mol %)
reaction temperature
iTOF
%
reaction time
(h)
entry
catalyst
(°C)
(h−1
)
hydration
19
20
21
Ag−PTA nanoparticles
AgNO3(aq)
0.052
0.057
0.057
0.21
0.24
0.24
0.40
0.42
0.42
90
90
90
23.5
1.4
48
11
14
22
19
19
AgNO3(aq) and
4PTA(aq)
1.8
precipitate in the solution of AgNO3 and KCN.36−38 Therefore,
this polymer is unlikely to be the active catalyst. It is more likely
that a water-soluble, anionic complex such as [Ag(CN)2]−
catalyzes the hydration reaction. To test this hypothesis, we
tested the commercially available salt K[Ag(CN)2] as a catalyst
for nitrile and cyanohydrin hydration. Benzonitrile and ACH
were tested as substrates, and hydration of both species was
observed (Table 3, entries 15 and 16). Addition of PTA to the
catalyst solution did not affect the rate of hydration (Table 3,
entry 17). Furthermore, the presence of additional CN− did not
significantly change the rate (Table 3, entry 18). These
experiments allow for two conclusions. First, the species formed
in the nanoparticle solutions is likely a complex of the form
CONCLUSIONS
■
This study prepared water-soluble Ag nanoparticles with a PTA
ligand shell, in which PTA is a water-soluble phosphine ligand.
The Ag−PTA nanoparticles are nitrile hydration catalysts, as
demonstrated by their ability to hydrate a variety of aromatic
nitriles with various steric and electronic properties at rates
comparable to those of some common homogeneous catalysts.1
The Ag−PTA nanoparticle catalysts have three particularly
noteworthy properties. (1) They catalyze hydration reactions
under mild conditions (i.e., in air at relatively low temper-
atures). (2) They are easily synthesized from cheap starting
materials. (3) They can hydrate heteroaromatics in good yields.
Solutions of the Ag−PTA NPs are also catalysts for the
hydration of cyanohydrins. Remarkably, the catalyst solution
could be recycled without any loss of activity or evidence of
cyanide poisoning. Closer examination, however, showed that
Ag NPs were not the catalyst but rather a Ag(I)−cyanide
−
Ag(CN)nn−1. The Ag(CN)2 complex is suggested, but similar
complexes are possible, as well. Second, the increase in the
hydration rate in the presence of larger quantities of cyanide in
the initial trials using the nanoparticle solution as the catalyst
was not due to the cyanide itself. Rather, as more cyanide was
−
complex, likely Ag(CN)2 , was the active catalyst. This is a
notable result because cyanide poisoning in cyanohydrin
hydration reactions has been an intractable problem that has
plagued the homogeneous catalytic hydration of cyanohydrins.
In short, all homogeneous catalysts studied to date are
poisoned by cyanide. The results presented in this paper
suggest a strategy for overcoming the poisoning problem.
−
added, more of the active catalyst species [probably Ag(CN)2 ]
formed. The increased concentration of the catalyst caused the
hydration reaction to proceed at a faster rate.
−
The resistance of the Ag(CN)2 complex to cyanide
poisoning likely results from the lability of bonds to the d10
Ag(I) metal center. Thus, cyanide reversibly binds to the Ag
center. Binding sites therefore remain available for the
−
Specifically, the insensitivity of the Ag(CN)2 catalyst to
cyanide poisoning is attributed to the lability of the Ag(I) metal
center, which suggests it will be worthwhile to examine the
catalytic hydration abilities of other complexes with labile
electronic configurations. Although low TOFs are still a
problem, a judicious choice of ligands in these complexes
may help to alleviate the slow rates. Alternatively, strategies that
stabilize ACH and prevent it from forming HCN, namely a
strongly acidic medium, may prove viable. Both strategies are
currently being investigated in our laboratory.
−
substrate. The greater activity of the anionic Ag(CN)2
complex compared to that of Ag(I) alone (i.e., in the absence
of cyanide) is attributed to the electron-withdrawing ability of
the cyanide.1
Control Reactions. Control experiments showed that
aqueous Ag+ (from AgNO3) is a catalyst for nitrile hydration;
however, the rate of hydration is slow compared to that of the
Ag−PTA nanoparticles. (Compare entries 19 and 20 in Table
4.)
Solutions of AgNO3 and PTA form coordination polymers in
which the PTA ligands bind silver through both P and N sites.39
This polymer was synthesized, and it had a catalytic activity
similar to that of aqueous AgNO3.
The pH of the aqueous Ag−PTA NP solution was 8. To
ensure that the reaction was not base-catalyzed, hydration trials
with benzonitrile and ACH were conducted at this pH with no
added catalyst or ligand. No reactivity was observed over 72 h.
EXPERIMENTAL SECTION
■
Instrumentation and Procedures. Nuclear magnetic
resonance spectra were recorded on a Varian Unity/Inova
500 MHz (1H, 500.10 MHz; 31P, 202.45 MHz; 13C, 151 MHz)
spectrometer or on a Bruker Biospin 600 MHz (1H, 600.02
MHz; 31P, 242.83 MHz) spectrometer. The H chemical shifts
1
were referenced to the solvent peak or TMS (0.00 ppm), and
the 31P chemical shifts were referenced to H3PO4 (0.00 ppm).
The solvent used for all NMR trials was D2O. UV−vis spectra
3101
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
were recorded on an HP 8453 spectrometer using 1 cm quartz
cuvettes. All hydration reaction samples were prepared in 1
dram screwcap vials fitted with septum caps. (HR)TEM images
were acquired with a FEI Titan 80−300 kV transmission
electron microscope equipped with a spherical aberration (Cs)
image corrector, an EDAX energy dispersive spectrometer, and
a Tridiem 863 Gatan imaging filter and electron energy loss
spectrometer. All images were acquired at 300 kV. PTA was
synthesized via literature methods.40
Preparation of 1,3,5-Triaza-7-phosphaadamantane
(PTA)-Stabilized Silver Nanoparticles. A preparation from
the group of S. D. Solomon for uncapped silver nanoparticle
synthesis was modified.17 NaBH4 (0.0082 g, 0.217 mmol) was
dissolved in 90 mL of water and the mixture placed in an ice
bath while being magnetically stirred. AgNO3 (0.0059 g, 0.035
mmol) was dissolved in 30 mL of water and the mixture added
over 10 min to the cool NaBH4 solution. The solution turned
yellow, characteristic of silver nanoparticles. After the particles
had been formed, PTA (0.0061 g, 0.039 mmol) dissolved in 10
mL of water was added to the solution as a stabilizer. Samples
were prepared for TEM on a lacey carbon grid with an ultrathin
(3 nm) carbon support film supported by copper mesh. A
dilute nanoparticle solution was dropped on the grid and
allowed to evaporate.
resonances at 7.74 ppm (d, J = 8.6 Hz) and 7.01 ppm (d, J =
8.7 Hz) in the H NMR spectrum.
1
p-Aminobenzonitrile. The progress of the reaction was
monitored by observing the disappearance of the aromatic p-
aminobenzonitrile resonance at 7.43 ppm (d, J = 8.5 Hz) and
the appearance of the amide resonance at 7.57 ppm (d, J = 8.4
1
Hz) in the H NMR spectrum.
p-Trifluoromethylbenzonitrile. The progress of the reaction
was monitored by observing the disappearance of the aromatic
p-trifluoromethylbenzonitrile resonances at 7.89 ppm (d, J =
8.4 Hz) and 7.82 ppm (d, J = 8.0 Hz) and the appearance of the
amide resonances at 7.87 ppm (d, J = 8.3 Hz) and 7.77 ppm (d,
1
J = 8.1 Hz) in the H NMR spectrum.
Hydration of Nicotinonitrile with PTA-Stabilized Ag
NPs. Nicotinonitrile (0.0493 g, 0.474 mmol) was added to 1
mL of acetone. This solution (125 μL) was added to 1000 μL
of a 0.22 mM Ag NP solution. The mixture was heated to 90
°C while being stirred. Aliquots (100 μL) were removed
periodically using a gastight syringe and combined in an NMR
tube with 450 μL of D2O and 50 μL of a 10.86 mM NMe4PF6
internal standard solution in D2O. The progress of the reaction
was monitored by observing the disappearance of the
nicotinonitrile resonance at 8.81 ppm and the appearance of
the amide resonance at 8.73 ppm.
Hydration of ACH with PTA-Stabilized Ag NPs. ACH
(10 μL, 0.109 mmol) was added to 1750 μL of a 0.22 mM Ag
NP solution. This solution was heated to 90 °C while being
stirred. After 264 h, an additional 0.082 mmol of ACH (7.5 μL)
was added to the reaction vessel. Aliquots (100 μL) were
removed periodically using a gastight syringe and combined in
an NMR tube with 450 μL of D2O and 50 μL of a 10.86 mM
NMe4PF6 internal standard solution in D2O. The progress of
the reaction was monitored by observing the disappearance of
the methyl resonance of acetone cyanohydrin at 1.57 ppm [s,
6H, HO(CH3)2CCN] and the appearance of the amide
resonance at 1.34 ppm [s, HO(CH3)2CC(O)NH2].
General Procedure for the Hydration of Benzonitrile
with PTA-Stabilized Ag NPs. Benzonitrile (10 μL, 0.09
mmol) was added to 1000 μL of a 0.22 mM Ag NP solution
and heated to 90 °C while being stirred. Aliquots (100 μL)
were removed periodically using a gastight syringe and
combined in an NMR tube with 500 μL of a 2.28 mM
NMe4PF6 solution in D2O as an internal standard. The progress
of the reaction was monitored by observing the disappearance
of the benzonitrile resonances at 7.70 ppm (d, J = 7.48 Hz),
7.65 ppm (t, J = 7.85 Hz), and 7.49 ppm (t, J = 7.87 Hz) and
the appearance of the amide resonances at 7.78 ppm (m) and
1
7.15 ppm (t, J = 7.97 Hz) in the H NMR spectrum of the
Hydration of p-Nitrobenzonitrile with AgNO3. AgNO3
(0.006 g, 0.03 mmol) was dissolved in 10 mL of water.
Nitrobenzonitrile (0.0408 g, 0.275 mmol) was dissolved in 1
mL of acetone-d6. The p-nitrobenzonitrile solution (400 μL)
and the AgNO3 solution (130 μL) were added to 1400 μL of
water in a 1 dram screwcap vial capped with a septum. This
mixture was heated to 90 °C while being stirred. Aliquots (100
μL) were removed periodically using a gastight syringe and
combined in an NMR tube with 450 μL of D2O and 50 μL of a
10.86 mM NMe4PF6 internal standard solution in D2O. The
progress of the reaction was monitored by observing the
disappearance of the p-nitrobenzonitrile resonance at 8.32 ppm
(d, J = 8.5 Hz) and the appearance of the amide resonance at
mixture.
General Procedure for the Hydration of Para-
Substituted Benzonitriles with PTA-Stabilized Ag NPs.
The nitrile was dissolved in 2 mL of acetone and the mixture
added to the 0.22 mM catalyst solution to achieve a
concentration of approximately 50 mM. This solution was
heated to 90 °C while being stirred. Aliquots (100 μL) were
removed periodically using a gastight syringe and combined in
an NMR tube with 500 μL of a 2.28 mM solution of NMe4PF6
in D2O. Specific details for individual nitriles are as follows.
p-Fluorobenzonitrile. The progress of the reaction was
monitored by observing the disappearance of the aromatic p-
fluorobenzonitrile resonances at 7.78 ppm (m) and 7.24 ppm
(t, J = 8.8 Hz) and the appearance of the amide resonances at
1
8.27 ppm (d, J = 8.6 Hz) in the H NMR spectrum.
Hydration of p-Nitrobenzonitrile with AgNO3 and 4
equiv of PTA. Nitrobenzonitrile (0.0408 g, 0.275 mmol) was
dissolved in 1 mL of acetone-d6. AgNO3 (0.006 g, 0.0353
mmol) was dissolved in 10 mL of water. PTA (0.0088 g, 0.0566
mmol) was added to 4 mL of the AgNO3 solution. The p-
nitrobenzonitrile solution (400 μL) and the AgNO3/PTA
solution (130 μL) were added to 1400 μL of water. This
mixture was heated to 90 °C while being stirred in a 1 dram
screwcap vial capped with a septum. Aliquots (100 μL) were
removed periodically using a gastight syringe and combined in
an NMR tube with 450 μL of D2O and 50 μL of a 10.86 mM
NMe4PF6 internal standard solution in D2O. The progress of
the reaction was monitored by observing the disappearance of
1
7.79 ppm (m) and 7.17 ppm (t, J = 9.1 Hz) in the H NMR
spectrum.
p-Nitrobenzonitrile. The progress of the reaction was
monitored by observing the disappearance of the aromatic p-
nitrobenzonitrile resonances at 8.32 ppm (d, J = 8.5 Hz) and
7.96 ppm (d, J = 8.7 Hz) and the appearance of the amide
resonances at 8.27 (d, J = 8.6 Hz) and 7.92 ppm (d, J = 8.9 Hz)
1
in the H NMR spectrum.
p-Methoxybenzonitrile. The progress of the reaction was
monitored by observing the disappearance of the aromatic p-
methoxybenzonitrile resonances at 7.66 ppm (d, J = 8.7 Hz)
and 7.02 ppm (d, J = 8.8 Hz) and the appearance of the amide
3102
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
the p-nitrobenzonitrile resonance at 8.32 ppm (d, J = 8.5 Hz)
and the appearance of the amide resonance at 8.27 (d, J = 8.6
Hz) in the H NMR spectrum.
REFERENCES
■
(1) Ahmed, T. J.; Knapp, S. M. M.; Tyler, D. R. Coord. Chem. Rev.
2011, 255, 949−974.
1
Hydration of ACH with AgNO3 and KCN. ACH (10 μL,
0.109 mmol) was added to a solution containing 100 μL of
AgNO3 (58.87 mM) and varying amounts of KCN (24.5 mM)
that had been diluted to 1610 μL. This mixture was heated to
90 °C while being stirred in a 1 dram screwcap vial with a
septum cap. Aliquots (100 μL) were removed periodically using
a gastight syringe and combined in an NMR tube with 500 μL
of a 10.86 mM NMe4PF6 solution in D2O. The progress of the
reaction was monitored by observing the disappearance of the
methyl resonance of acetone cyanohydrin at 1.57 ppm [s, 6H,
HO(CH3)2CCN] and the appearance of the amide resonance
at 1.34 ppm [s, HO(CH3)2CC(O)NH2].
(2) Green, M. M.; Witcoff, H. A. Organic Chemistry Principles and
Industrial Practice; Wiley-VCH: Weinheim, Germany, 2003.
(3) Bizzari, S. Chemical Economics Handbook; SRI Consulting: Menlo
Park, CA, 2010.
(4) García-Alvarez, R.; Crochet, P.; Cadierno, V. Green Chem. 2013,
15, 46.
́
(5) Kukushkin, V. Y.; Pombeiro, A. J. L. Inorg. Chim. Acta 2005, 358,
1−21.
(6) Ahmed, T. J.; Fox, B. R.; Knapp, S. M. M.; Yelle, R. B.; Juliette, J.
J.; Tyler, D. R. Inorg. Chem. 2009, 48, 7828−7837.
(7) Knapp, S. M. M.; Sherbow, T. J.; Juliette, J. J.; Tyler, D. R.
Organometallics 2012, 31, 2941−2944.
(8) Yan, N.; Xiao, C.; Kou, Y. Coord. Chem. Rev. 2010, 254, 1179−
1218.
General Procedure for the Hydration of ACH with
KAg(CN)2. ACH (10 μL, 0.109 mmol) was added to 1000 μL
of a 5.03 mM solution of KAg(CN)2 in water. In some cases, a
solution of PTA (100 μL of a 50 mM solution in H2O) was
added to the catalyst solution. This solution was heated to 90
°C while being stirred in a 1 dram screwcap vial capped with a
septum. Aliquots (100 μL) were removed periodically using a
gastight syringe and combined in an NMR tube with 500 μL of
a 10.86 mM NMe4PF6 internal standard solution in D2O. The
progress of the reaction was monitored by observing the
disappearance of the methyl resonance of acetone cyanohydrin
at 1.57 ppm [s, 6H, HO(CH3)2CCN] and the appearance of
the amide resonance at 1.34 ppm [s, HO(CH3)2CC(O)NH2].
Hydration of Benzonitrile with KAg(CN)2. Benzonitrile
(10 μL, 0.0969 mmol) was added to 1000 μL of a 5.03 mM
solution of KAg(CN)2 in water. This mixture was heated to 90
°C while being stirred in a 1 dram screwcap vial capped with a
septum. Aliquots (100 μL) were removed periodically using a
gastight syringe and combined in an NMR tube with 500 μL of
a 10.86 mM NMe4PF6 internal standard solution in D2O. The
progress of the reaction was monitored by observing the
disappearance of the benzonitrile resonance at 7.49 ppm (t, J =
7.87 Hz) and the appearance of the amide resonance at 7.15
ppm (t, J = 7.97 Hz) in the 1H NMR spectrum of the mixture.
pH Studies. The pH of the aqueous Ag−PTA NP solution
was 8. Hydration trials at this pH (with no added catalyst or
ligand) resulted in no reactivity over 72 h.
(9) Shimizu, K.; Kubo, T.; Satsuma, A.; Kamachi, T.; Yoshizawa, K.
ACS Catal. 2012, 2, 2467−2474.
(10) Subramanian, T.; Pitchumani, K. Catal. Commun. 2012, 29,
109−113.
(11) Ishizuka, A.; Nakazaki, Y.; Oshiki, T. Chem. Lett. 2009, 38, 360−
361.
(12) Kim, A. Y.; Bae, H. S.; Park, S.; Park, S.; Park, K. H. Catal. Lett.
2011, 141, 685−690.
(13) Hirano, T.; Uehara, K.; Kamata, K.; Mizuno, N. J. Am. Chem.
Soc. 2012, 134, 6425−6433.
(14) Liu, Y.-M.; He, L.; Wang, M.-M.; Cao, Y.; He, H.-Y.; Fan, K.-N.
ChemSusChem 2012, 5, 1392−1396.
(15) Mitsudome, T.; Mikami, Y.; Mori, H.; Arita, S.; Mizugaki, T.;
Jitsukawa, K.; Kaneda, K. Chem. Commun. 2009, 3258−3260.
(16) Baig, R. B. N.; Varma, R. S. Chem. Commun. 2012, 48, 6220−
6222.
(17) Shimizu, K.; Imaiida, N.; Sawabe, K.; Satsuma, A. Appl. Catal., A
2012, 421−422, 114−120.
(18) Polshettiwar, V.; Varma, R. S. Chem.Eur. J. 2009, 15, 1582−
1586.
(19) Woo, H.; Lee, K.; Park, S.; Park, K. H. Molecules 2014, 19, 699−
712.
(20) Debouttiere, P.-J.; Coppel, Y.; Denicourt-Nowicki, A.; Roucoux,
̀
A.; Chaudret, B.; Philippot, K. Eur. J. Inorg. Chem. 2012, 2012, 1229−
1236.
(21) Hendrich, C.; Bosbach, J.; Stietz, F.; Hubenthal, F.; Vartanyan,
T.; Trager, F. Appl. Phys. B: Lasers Opt. 2003, 76, 869−875.
̈
(22) Lismont, M.; Dreesen, L. Mater. Sci. Eng., C 2012, 32, 1437−
1442.
(23) Hajizadeh, S.; Farhadi, K.; Forough, M.; Sabzi, R. E. Anal.
Methods 2011, 3, 2599−2603.
ASSOCIATED CONTENT
■
(24) Mulfinger, L.; Solomon, S. D.; Bahadory, M.; Jeyarajasingam, A.
V.; Rutkowsky, S. A.; Boritz, C. J. Chem. Educ. 2007, 84, 322.
(25) Solutions in which a <1.1:1 PTA:AgNO3 molar ratio was added
yielded unstable nanoparticles that crashed out of solution after ∼30
min. Solutions in which a >2:1 PTA:AgNO3 molar ratio was added
yielded unstable particles that caused the nanoparticle solution to turn
colorless.
S
* Supporting Information
Experimental procedure for the air-free hydration of ACH and
EDX spectrum of the Ag−PTA nanoparticles. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
̀
(26) Debouttiere, P.-J.; Martinez, V.; Philippot, K.; Chaudret, B.
Corresponding Author
Dalton Trans. 2009, 10172−10174.
*E-mail: dtyler@uoregon.edu.
Author Contributions
T.J.S. and E.L.D. contributed equally to this work.
(27) Son, S. U.; Jang, Y.; Yoon, K. Y.; Kang, E.; Hyeon, T. Nano Lett.
2004, 4, 1147−1151.
(28) Wu, L.; Li, Z.-W.; Zhang, F.; He, Y.-M.; Fan, Q.-H. Adv. Synth.
Catal. 2008, 350, 846−862.
Notes
(29) Madras, G.; McCoy, B. J. J. Chem. Phys. 2002, 117, 8042−8049.
(30) Marqusee, J. A.; Ross, J. J. Chem. Phys. 1983, 79, 373−378.
(31) Knapp, S. M. M.; Sherbow, T. J.; Yelle, R. B.; Zakharov, L. N.;
Juliette, J. J.; Tyler, D. R. Organometallics 2013, 32, 824−834.
(32) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96−103.
(33) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.
(34) Ghosh, S. K.; Kundu, S.; Pal, T. Bull. Mater. Sci. 2002, 25, 581−
582.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Rohm and Hass Chemical Co. and the National Science
Foundation (CHE-0719171) are acknowledged for the support
of this research.
3103
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104

ACS Catalysis
Research Article
(35) Knapp, S. M. M.; Sherbow, T. J.; Yelle, R. B.; Juliette, J. J.; Tyler,
D. R. Organometallics 2013, 32, 3744−3752.
(36) Bayse, C. A.; Ming, J. L.; Miller, K. M.; McCollough, S. M.; Pike,
R. D. Inorg. Chim. Acta 2011, 375, 47−52.
(37) Bryce, D. L.; Wasylishen, R. E. Inorg. Chem. 2002, 41, 4131−
4138.
(38) Bowmaker, G. A.; Kennedy, B. J.; Reid, J. C. Inorg. Chem. 1998,
37, 3968−3974.
(39) Mohr, F.; Falvello, L. R.; Laguna, M. Eur. J. Inorg. Chem. 2006,
2006, 3152−3154.
(40) Daigle, D. J. Inorg. Synth. 1998, 32, 40−45.
3104
dx.doi.org/10.1021/cs500830s | ACS Catal. 2014, 4, 3096−3104