Adam Goldstein
Copper Promoted Aromatic Acyloxylation:
The Formation of Aryl Esters through the Cross Coupling of Aryl Halides Via Comproportionation of Copper II Reactants
Thesis Submitted in Fulfillment of the Requirements of the Jay and Jeanie Schottenstein Honors Program Yeshiva College
Yeshiva University August 2016
Adam Goldstein
Mentor: Professor Fabiola Barrios-Landeros Department of Chemistry, Yeshiva College
Acknowledgements
2
As this thesis is the capstone of my undergraduate work, I feel it important to thank all of those who have guided me along the path that has led me here.
Thank you to all of my professors. You have helped me to become a better student and a better person.
Specifically, thank you to Professor Fabiola Barrios-Landeros. From start to finish, working
in your lab was a pleasure. I must thank you for the skills you helped me hone and develop,
but also for the comfort that came from working in your lab. More importantly, you showed me just how much teachers truly care about their students.
Abstract
3
Previous research in Prof. Barrios-Landeros lab has shown that Cu (I) acetate complexes
can successfully promote the cross coupling of aryl halides with carboxylates to make aryl esters. The reaction proceeds by a sequence of steps that can be accessed at different
points using different reagents. One of this entry points was the focus of this thesis work,
specifically, to explore the ability of Cu (II) to comproportionate, as Cu (II) complexes are often cheaper and more stable. It was shown that Cu (II) acetate does, in fact, drive the
copper- promoted acetoxylation of phenyl iodide in DMF. Product formation, as measured
by gas chromatography (GC), showed yields of up to 90%. Furthermore, the change in color of the copper as it transitions between the various oxidation states during the reaction
allowed for an in- depth analysis of the rate of reaction, via flow cell spectroscopy. Additionally, Cu (II) sulfate, an even cheaper alternative, was able to drive the reaction in
the presence of acetate salts, though with lower yields. Phase transfer agents, such as
tetrabutylammonium tetraflouroborate and hexaflourophosphate, increased the yield of
the Cu (II) sulfate reactions. Ultimately, it was shown that the comproportionation of Cu (II) is a viable option for adding Cu (I) complexes into cross coupling reactions.
Introduction
4
Organometallic synthesis has been at the forefront of chemistry related scientific
progress in recent years. Case in point, three Nobel Prizes in Chemistry from the past 15
years have been awarded to projects focused on organometallic synthesis (All Nobel Prizes
in Chemistry). The power of organometallic synthesis, particularly in relation to classical organic chemistry techniques, lies in the use of transition metals that ultimately allow
scientists to break and form bonds that were otherwise too difficult to perform. The
transition metals, with their multiple oxidation states, can form bonds with multiple
compounds. By bringing the reactants near each other through bonding with the transition
metal, the transition metal drives the reaction. Specifically, organometallic processes generally have lower activation energies than comparable classic processes. Also, organometallic synthesis allows for higher levels of selectivity, including chemo selectivity,
regio selectivity, and stereo selectivity. While there are many different pathways that organometallic synthesis can follow, they all share a reliance on transition metals.
In particular, cross coupling chemistry is an especially potent process. In short,
cross coupling chemistry allows for the formation of a carbon-carbon or a carbon-
heteroatom bond. Aromatic cross coupling utilizes the possibility of these bonds between aryl halides and nucleophilic compounds to form aromatic hydrocarbons, amines, amides, ethers, and alcohols. (Figure 1)
5
Figure 1 Examples of cross-coupling reactions catalyzed by transition metal complexes
Early aromatic cross coupling reactions focused on copper as the transition metal,
but eventually shifted to a focus on palladium. For example, Fritz Ullman, the earliest chemist to work on aromatic cross coupling reactions was able to show that copper metal drives bond formations that were not seen in that time. With further research, the Ullman
reactions were discovered (Figures 2). These reactions showed that carbon-carbon, carbon-oxygen, and carbon-nitrogen bonds were indeed possible in the presence of a
copper catalyst (Monnier). However, as the field of aromatic cross coupling reactions grew, the focus shifted from copper to palladium. This has been attributed to the discovery of ability for palladium to catalyze an amination reaction (Beletskaya). Figure 2 Examples of Ullman's Work in Cross Coupling
6
The general steps of an aromatic cross coupling reaction have been established. The
three important steps that must occur for aromatic cross coupling are oxidative addition,
reductive elimination, and trasmetalation. In the oxidative addition step, the metal contributes two electrons to break the bond between the aromatic ring and the halide. At the end of this step, the metal oxidizes to +2 of its original state and is attached to both the
aromatic ring and the halide. In the reductive elimination step, the metal, which is coordinated to an aromatic ring and a nucleophillic species, is reduced as the ring bonds with the nucleophilic species. In the transmetalation step, an anionic ligand is displaced from the metal in favor of another one. While both copper and palladium-catalyzed
reactions utilize each of these steps, they vary in the order used. Copper reactions follow the process: transmetalation followed by oxidative addition followed by reductive
elimination. In palladium reactions the oxidative addition step and the transmetalation step are reversed (Figure 3).
7
Figure 3 Steps for Cross Coupling Utilizing Palladium L= PR3
X'
Pd0L A
LnPdII
Reductive Elimination: Formation of C-C bond
MX
n
ArX
Ar X'
C
X
LnPdII B
Ar
X= Cl, Br, I, OTf Oxidative Addition
X
MX'
Transmetalation: Displacement of X by the X' or "coupling partner"
While the steps are the same between copper and palladium aromatic cross
coupling reactions and the focus recently has been on palladium reactions, there are many advantages to using copper for these reactions. First, while copper is relatively inexpensive, palladium can be extremely expensive. Therefore, cross coupling reactions involving
copper would not have to worry about the waste of an expensive metal. Second, whereas palladium only has two stable oxidative states at 0 and +2, copper has accessible oxidative states: 0, +1, +2, and +3 (Beletskaya). The mere existence of the odd number oxidative
states and the amount of states would allow copper to undergo reaction conditions under which palladium could not.
Even considering the advantages that copper holds in cross coupling reactions and
the work that has been done in the field, there are still relatively unexplored areas of copper based aromatic cross coupling reactions. It has been shown that the formation of
ethers works via copper reactions. However, there is little evidence of a reaction for
esterification. Furthermore, it has been shown that amidation is possible, which is similar
8
to the esterification but with a focus on nitrogen instead of oxygen (Beletasky). It is the focus of Dr. Barrios-Landeros’ laboratory to develop a method for the production of
aromatic esters from aryl halides with the use of copper based aromatic cross coupling reactions, or copper- catalyzed aromatic acyloxylation (Figure 4).
Figure 4 Aromatic acyloxylation catalyzed by copper complexes X + X= I, Br, Cl
O
Cu catalyst
R HO R= alkyl or aryl
base
R
O O
Previous work under Dr. Barrios-Landeros has shown that it is indeed possible to
use copper (I) species to perform acyloxylation. By mixing copper (I) acetate and phenyl iodide, phenyl acetate was produced. The proposed mechanism: starting with a
stoichiometric amount of copper(I) acetate as the starting material, this copper species will serve as both the copper catalyst and acetate source. Oxidative addition of the phenyl
iodide is followed by reductive elimination to produce phenyl acetate. Along with the aromatic acyloxylation reaction, there is an aromatic halide exchange reaction that occurs simultaneously (Figure 5) (BenZvi).
Figure 5 Previous Work's Mechanism for Aromatic Acyloxylation
9
While copper (I) species has been shown to drive the aromatic acyloxylation, this
work turns it focus on copper (II) species. The comproportionation- disproportionation reactions, which will allow copper (II) species to enter the reaction, are relatively
straightforward reactions. Comproportionation is a reaction in which two reactants, both containing the same element but at different oxidation numbers, can combine to form a
product with the element at a different oxidation number. This experiment utilizes copper (II) and copper metal to form copper (I). The disproportionation reaction is the opposite reaction, where a reactant is oxidized and reduced to form two products with different
oxidation states. Ultimately, the proposed mechanism introduces copper (II) and copper metal so that the reaction can continue as before (Figure 6).
Figure 6 Proposed Mechanism for the Complete Aromatic Acyloxylation with Comproportionation
10 Copper (II) species have advantages over the copper (I) species for this reaction.
First, the price of copper (II) species is generally lower than for copper (I), making the reaction even cheaper. Second, the copper (II) species is generally more soluble than that of
the same copper (I) species. This will give the reaction further access to the important
transition metal. Third, copper (II) compounds are stable in the presence of oxygen and much easier to synthesize, handle and store than the copper (I) species.
Copper (II) also has a very distinctive blue color. Focusing on the color changes of
the reaction will give further insight into the inner workings of this cross coupling reaction.
Particularly, the disappearance of the copper species will enable a study of the rate of the reaction for aromatic acyloxylation. Fourth, and most importantly, by utilizing copper (II) it
is possible to create copper (I) in situ via comproportionation reactions. As the previous
work has shown, the reaction relies on a copper (I) carboxylate as a starting material. By creating this species in situ, there is more control over the overall reaction.
11
Results Time Dependence of the Reaction with Copper (I) Acetate
To continue the exploration of the mechanism required for the cross coupling reaction to occur, the time dependence of the reaction was measured. Color changes were observed even for reactions that were heated for an hour. The final colors were all similar. However,
monitoring over time the progress of the reactions either via side-by-side reactions or aliquots of the same reaction yielded inconclusive results. Reactions with Labeled Benzene Though the formation of product could be determined via GC, monitoring the product formation over time by gas chromatography could be complicated. Therefore, the reaction
with Copper (I) acetate was performed using substituted benzene rings in attempt to ultimately study the reaction via
19F
NMR. First, 74 mg of CuOAc and 70 μL of 2-
fluoroiodobenzene were mixed in 1 mL of DMF with 10 μL of nonane. Though a color change of brown to blue/green was observed, no product was detected. However, when the
reaction was performed with 4-fluoroiodobenzene, a color change of brown to blue was
observed together with the presence of product, verified both by GC and the GC-MS. Switch to Copper (II) Acetate
To study the reaction further and to find the rate of reaction for the cross coupling of the
copper (I) acetate, different amounts of copper (I) acetate were mixed with DMF in order to
determine the λ (max) and molar absorptivity of copper (I) acetate in DMF. However, due
to the low solubility of copper (I) acetate, it was difficult to obtain solutions with reliable
12
concentrations. Instead, copper (II) complexes were explored as they have better solubility
and would allow the in situ production of copper (I). Copper (II) complexes have been shown to be viable entry points for the proposed mechanism (BenZvi). Furthermore, the
color change that was observed during the reaction would allow for a spectroscopic study of the rate of reaction (Picture 1).
Picture 1 Color Change of the Copper (II) Acetate Reaction. From left to right: Copper (II) acetate reaction before heating, after heating for 5 hours, after centrifugation, after heating for another 5 hours
In preparation for studying the copper (II) complex rate of reaction, the λ (max) and molar absorptivity were determined via conventional methods. As shown, the λ (max) was determined to be at 700 nm (Graph 1) and the molar absorptivity (Table 1) was determined to be 162 mol/L (Graph 2) or 0.87 g/L (Graph 3).
13
Graph 1. Absorption spectra of Cu(II) acetate
Visible Spectra of Copper (II) Acetate 3
2
1.5 1
Absorbance
2.5
0.5 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800
0
Wavelength (nm)
Table 1 Concentrations vs. Absorbance at 700nm of Copper (II) Acetate Concentration (g/L)
Concentration (mol/L)
Absorbance
1
0.0055
0.972
0.5 1.5 2 3
0.0028 0.008 0.011 0.016
0.4489 1.1907 1.8076 2.602
14
Graph 2 Determination of Molar Absorptivity in mol/L
Concentration (mol/L) vs. Absorbance of Copper (II) Acetate 3
y = 162.07x R² = 0.9933
2.5
2
Absorbance 1.5
1
0.5
0
0
0.005 0.01 0.015 Concentration (mol/L)
0.02
Graph 3 Determination of Molar Absorptivity in g/L
Concentration (g/L) vs. Absorbance Copper (II) Acetate 3
2.5
y = 0.873x R² = 0.9897
2
Absorbance 1.5 1
0.5
0
0
1
2 Concentration (g/L)
3
4
Using Copper (II) Bromide as an Access Point
15
Copper(II) bromide in the presence of a source of acetate ions can be used as precursor for copper acetate species. Copper (II) bromide also has a very unique color, a deep green
(Picture 2). Since it had a different color than the copper (II) acetate and the copper (I)
acetate, the absorption spectra of copper (II) bromide was explored. First, λ (max) was determined to be 600 nm (Graph 4). Then, the molar absorptivity was calculated. The stock solution was mixed at a concentration of 3 g/L, or 0.0067 mol/L. The vials of lower
concentrations were dilutions from the stock solution (Table 2). The molar absorptivity was determined to be 323.15 mol/L (Graph 5) or 1.45 g/L (Graph 6).
Picture 2 Color Comparison of Copper Acetate (Left), Copper Bromide (Right), and Mix (Middle)
16
Graph 4 Absorption spectra of Copper (II) Bromide
Visible Spectra of Copper (II) Bromide in DMF 1.6 1.4 1.2 1
0.8 Absorbance 0.6 0.4 0.2 780
730
680 630 580 Wavelength (nm)
530
480
0
Table 2 Concentrations vs. Absorbance at 600nm of Copper (II) Bromide Concentration (g/L)
Concentration (mol/L)
Absorbance
0.1875
0.0008394
0.1417
0.09375 0.375 0.75 1.5
0.0004197 0.001679 0.003358 0.006715
0.1077 0.4038 0.9851 2.273
17
Graph 5 Determination of Molar Absorptivity in mol/L
Concentration (mol/L) vs. Absorbance of Copper (II) Bromide 2.5
y = 323.15x R² = 0.9825
2
Absorbance
1.5
1
0.5
0
0
0.001
0.002
0.003 0.004 0.005 Concentration (mol/L)
0.006
0.007
0.008
Graph 6 Determination of Molar Absorptivity in g/L
Concentration (g/L) vs. Absorbance of Copper (II) Bromide 2.5
y = 1.4467x R² = 0.9825
2 Absorbance
1.5 1 0.5 0
0
0.2
0.4
0.6 0.8 1 Concentration (g/L)
1.2
1.4
1.6
A side- by- side comparison of the CuBr2 and Cu(OAc)2 was also visualized via UV/Vis. A 1
g/L solution of each CuBr2 and Cu(OAc)2 was mixed. Then, 1 ml from each of those 1 g/L
18
solutions was added to a separate vial, which created a solution that was 0.5 g/L CuBr2 and 0.5 g/L Cu(OAc)2 (Picture 2). The vials were then analyzed via UV/Vis (Graph 7). The peaks
were distinct for the solutions that had only one copper complex. However, for the solution with both copper complexes, neither peak remained distinct.
Graph 7 Side- by- Side Uv/Vis Analysis of Different Copper Complexes
Side- by- Side UV/Vis Analysis of Different Copper Complexes 1.5 1.3 1.1 0.9 0.7 0.5 0.3
1 g/L Copper (II) Acetate 1 g/L Copper (II) Bromide 0.5 g/L Copper (II) Acetate and 0.5 g/l Copper (II) Bromide
200 230 260 290 320 350 380 410 440 470 500 530 560 590 620 650 680 710 740 770 800
0.1
Absorbance
Wavelength (nm)
-0.1
The main goal was still to produce phenyl acetate. Towards that end, an experiment was set up to test the formation of phenyl acetate from CuBr2. Copper (II) bromide, Cu (0), DMF,
PhI, nonane and an acetate source were mixed. While some of these reactions did produce phenyl acetate, the percent yields were extremely low (Table 4).
19
Table 3 Acetate Screening for Copper (II) Bromide Acetate Source
Percent PhI Left
Percent Yield PhOAc
Percent Yield PhBr
Silver
6.05
5.72
18.7
Control Zinc
2.07 10.6
0
Tetrabutylammonium 0.387 Cesium
11.1
Lithium
14.3
Potassium
Sodium
0
53.1
34.5
34.6
0
4.04
26.3
3.13
19.5
6.11
10.6
59.2
45.1
1.42
39.5
At first glance, the low yield of phenyl acetate was attributed to an inability of acetate to transmetallate and replace bromide on a copper species. Since the copper species drives
the reaction according to the proposed mechanism, this would leave acetate without the
ability to become phenyl acetate. To test this hypothesis, an anion exchange test was
performed. Different copper species were mixed with silver acetate in 1 mL DMF and stirred for 15 minutes at room temperature (Table 4). Spectroscopic readings in the visible
region showed that the vial with CuBr2 and AgOAc ultimately produced the highest amount
of copper acetate, showing that the anion exchange works. Table 4 Anion Exchange Test Vial #
1
2
3
CuBr2 (mg)
-
67
-
CuBr (mg)
43
CuS04 (mg)
-
AgOAc (mg)
-
-
50
100
100
1
1
1
DMF (mL)
-
20
48
Using CuSO4 as an Access Point As the copper bromide provided low yields, the spectroscopy peaks overlapped, and there
was halide interference, copper sulfate was explored as another alternative access point of copper acetate species. Furthermore, copper sulfate has a high solubility and it is also fairly inexpensive. Again, the ultimate goal was production of phenyl acetate so the copper sulfate was screened with various acetate sources. The percent yield of phenyl acetate was still low (Table 5).
Table 5 Acetate Screening for Copper (II) Sulfate Acetate Source
Percent PhI Left
Percent Yield Phenyl Acetate
Silver
0
64.9
Control Zinc
Tetrabutylammonium Cesium
Potassium Lithium Sodium
29.8 16.7 0
28.2 13.0 11.2 13.3
0
56.3 16.8 2.12 1.98 13.5 15.5
21
Phase transfer agents were used in an attempt to increase the concentration of acetate in solution
and
raise
the
percent
yields
of
the
reaction.
Tetrabutylammonium
tetrafluoroborate and tetrabutylammonium hexafluorophosphate were chosen as the
experiment phase transfer agents since these anions are not coordinating and should not interfere with the transformation. While other side products similar to those seen by the
copper bromide reactions were produced, a comparison of the two agents between the original reactions and those with the phase transfer agents, showed that the phase transfer agents do indeed raise the percent yield of the reaction, though relatively minimally. The hexaflorophosphate has a slight advantage (Table 6).
Table 6 Percent Yield of Phenyl Acetate of Reactions of Copper Sulfate and Phase Transfer Agents
Acetate Source Control Silver Zinc
Cesium
Potassium Lithium Sodium
Without
Transfer Agent 0
64.88 56.45 2.12 1.99
13.54 15.46
Phase With
Tetrafluoroborate 0 0
76.27 0
1.29
19.16 44.13
With
Hexafluorophosphate 0 0
77.47 2.29 0
19.54 16.63
Interestingly, the silver acetate no longer produced any phenyl acetate in the presence of phase transfer agents. This was surprising, as previous experiments had concluded that
silver acetate was the best form of acetate to drive the reaction (BenZvi). In the presence of
22
the phase transfer agent, silver undergoes reduction and the vials that were used to conduct the silver acetate with phase transfer agent experiments had silver globules inside. Test of Other Copper (II) Carboxylates It was also important to determine whether the proposed mechanism would work for
other carboxylates. Therefore, the general method for the copper acetate experiments was used but with copper pivalate and copper benzoate as well (Table 7).
Table 7 Standard Reaction with Copper Carboxylates Carboxylate
Percent
Acetate
69.2
Benzoate Pivalate
Yield
Carboxylate (Attempt 1) 15.9 24.1
Aryl Percent
Yield
Carboxylate (Attempt 2)
Aryl
62.6 58.9 71.2
The production of phenyl benzoate was a positive step towards finalizing the rate of
reaction for the cross coupling reaction. Therefore, further information about the
spectroscopic ability of copper benzoate was explored. The λ (max) was determined to be
718 nm (Graph 8). Additionally, the molar absorptivity was determined (Table 8). The molar absorptivity of copper benzoate was shown to be 133.8 mol/L (Graph 9) or 0.22 g/L (Graph 10).
23
Graph 8 Lamda Max Scan of Copper Benzoate
Absorbance Scan of Copper Benzoate 1.6 1.4 1.2 1 0.8 Absorbance 0.6 0.4 0.2 0
300
400
500 600 Wavelength (nm)
700
800
Table 8 Concentrations vs. Absorbance of Copper (II) Benzoate at 718nm Concentration (g/L)
Concentration (mol/L)
Absorbance
0.5
0.001635
0.2196
1
0.25
0.125
0.003270
0.0008176 0.0004088
0.4280 0.1392
0.06856
24
Graph 9 Determination of Molar Absorptivity in mol/L
Absorbance
Concentration (mol/L) vs. Absorbance of Copper (II) Benzoate 0.5
y = 133.79x R² = 0.9839
0.4 0.3 0.2 0.1
0
0
0.0005
0.001
0.0015 0.002 0.0025 Concentration (mol/L)
0.003
0.0035
Graph 10 Determination of Molar Absorptivity in g/L
Absorbance
Concentration (g/L) vs. Absorbance of Copper (II) Benzoate
0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0
y = 0.2188x R² = 0.9839
0
0.5
1 1.5 Wavelength (nm)
2
2.5
Interaction of Copper, Iodide, and Acetate The reaction that makes phenyl acetate also yields copper iodide as product. Before starting the process of spectroscopic analysis of the in situ production of CuOAc from
Cu(OAc)2, it was important to establish that the observed peaks were a result of the copper
25
acetate complexes and not a result of copper iodide complexes. Vials containing different
amounts of copper complexes were analyzed via absorption spectroscopy (Table 9). It was shown that the CuI would not affect the readings from the copper(I) or copper(II) acetate complexes (Graph 11).
Table 9 Test to See the Interaction Between Copper, Acetate, and Iodide Vial #
Cu(OAc)2 (mg)
CuOAc (mg)
CuI (mg)
DMF (mL)
2
-
36
-
1
1
55
3
-
-
4
-
28
5
57
-
-
-
1
29
18
1
1
29
1
Graph 11 Spectroscopic Data for the Interaction between Copper, Acetate, and Iodine
Interaction of Copper Salts of Acetate and Iodide
Absorbance
2.5
2
Cu(OAc)2
1.5
CuOAc CuI
1
Cu(Oac)2 + CuI CuOAc +CuI
0.5
0
250
350
450
550
Wavelength (nm)
650
750
26
Rate of Reaction- Static Spectroscopy
To study the rate of the reaction, the original experiment designed required heating a vial with the reaction mixture and taking aliquots to acquire absorption spectroscopy. Because the reaction was heterogeneous, the vial was centrifuged in order to prevent solid particles
from interfering with the spectrophotometry data. This method originally had many
complications. Eventually, the experiment produced a graph that showed a general decline
in the copper (II) acetate peak and a general decrease in the copper (I) acetate peak (Graph 12). It was shown that spectrophotometry could be a proper method for determining the rate of reaction.
Graph 12 Spectrophotometry Data Proving UV/Vis Could be Used to Determine Rate of Reaction
Standard Reaction Microwaved at 160° C 3
2.5 2
1.5 1
0.5 200 225 250 275 300 325 350 375 400 425 450 475 500 525 550 575 600 625 650 675 700 725 750 775 800
0
Wavelength (nm)
-0.5
Absorbance 30 min 60 min 90 min
120 min
27
Rate of Reaction- Flow-Cell Spectroscopy with Copper Acetate
Though it was shown that spectroscopy was a viable method for studying the rate of the
aromatic cross coupling reaction, the static spectroscopy method was not the best way to achieve the best data for the reaction. The reaction was then studied via flow-cell spectroscopy. Following the flow-cell spectroscopy method, the reaction was analyzed.
After much trial and error, the method worked for some analysis but has not yet been
perfected. The insolubility of certain reactants in the solution was one of the largest
obstacles to consider. Therefore, reliable readings were often only obtained for early stages of the reaction because the flow cell would become clogged. In some experiments, a reasonable trend was observed for the copper species. Yet, no product was formed and the graphs showed very different results (Graph 13 and 14).
Graph 13 Copper Acetate Reaction with Stoichiometric Amount of Copper (II) Reactant
Flow- Cell Spectroscopy of Copper Acetate Reaction 0.7 0.6 Absorbance
0.5 0.4
Cu (II)
0.3
Cu (I)
0.2 0.1
0
0
20
40
60 80 Time (min)
100
120
140
28
Graph 14 Copper Acetate Reaction with Halved Amount of Copper (II) Reactant (data formatted to remove data points during which the flow cell was clogged)
Flow Cell Spectroscopy of Copper Acetate Reaction 0.7 0.6 Absorbance
0.5 0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
10
20 Time (min)
30
40
In an attempt to study the actual rate of reaction without worrying about clogs during the time of the comproportionation-disproportionation reaction, an experiment was
performed where the Cu(OAc)2 and Cu metal was left heating overnight. Then, the reaction was continued with the addition of the PhI and analyzed via flow cell spectroscopy (Graph 15).
Graph 15 Copper Acetate Flow- Cell Spoctroscopy with Product Formation
29
Flow- Cell Spectroscopy of Copper Acetate with Product Formation Observed 0.6 0.5 Absorbance
0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
5
10
15 20 Time (min)
25
30
35
Rate of Reaction- Flow-Cell Spectroscopy with Copper Benzoate As many of the issues with studying the copper acetate rate of reactions were a result of solubility, the experimental method was attempted with copper benzoate (Graph 16). Graph 16 Flow- Cell Spectroscopy of Copper Benzoate
Flow Cell Spectroscopy of Copper Benzoate 0.7 0.6 Absorbance
0.5 0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
20
40 Time (min)
60
80
Conclusion
30
Comproportionation of copper (II) complexes was shown to be a reliable avenue for in situ production of copper (I) complexes. Ultimately, this was able to drive the aromatic
acyloxylation, following the proposed mechanism. Copper (II) acetate, copper (II) bromide,
and copper (II) sulfate were all able to form phenyl acetate under the reaction conditions. Copper (II) sulfate in the presence of acetate salts and phase transfer agents yielded the
highest amount of product. It was also shown that the reaction works with other copper carboxylates, including copper pivalate and copper benzoate, to produce aromatic esters.
Even more so, the use of copper (II) complexes added an important approach for studying the reaction by causing an observable color change. The solution starts with a copper (II)
species, lending a vivid blue color. As the reaction progresses, copper (I) species are
formed, which are usually a pale beige color. Together, both copper species give the solution a green color. Static spectroscopy was not able to provide reliable data for studying the rate of reaction. This is likely due to the heterogeneous nature of the reaction mixture. Flow-cell spectroscopy, on the other hand, does indeed seem like a reliable
method for determining the rate of the reaction. However, at present time, the ideal conditions under which to perform the flow-cell spectroscopy were not determined. The
experiments performed on the copper (II) acetate for this work often stopped working as
the result of clogs. The copper (II) benzoate showed more promise. Larger tubing, larger
diameter flow cells, and reactants/products with higher absorptivity will likely lend themselves to more accurate data readings.
31 Experimental Section
Representative Method for Copper Reactions on Hot Plate Under N2 atmosphere, 0.3 mmol of copper (II) compound together with 0.3 mmol of copper (0) powder was weighed in a small vial inside a glove box. If applicable, the additional 0.6 mmol of acetate source or 0.1 mmol of transfer agent was then weighed in the vial. 1 mL of solvent, 0.6 mmol phenyl halide, and 10 uL nonane (for GC calibration) were added to the vial with a micropipette. A magnetic stirrer was added to the vial, which was then sealed with a screw cap with a Teflon septum. The vial was brought out of the glove box and the mixture was stirred at 110°C for 10 hours on a hot plate. After cooling the vial, the reaction mixture was then “worked up” via the ethyl acetate workup procedure in preparation for GC analysis.
Representative Method for Copper Reactions in Microwave Under N2 atmosphere, the 0.3 mmol of copper (II) catalyst together with the 0.3 mmol of copper (0) powder was weighed in a small vial inside a glove box. 1 mL of solvent, 0.6 mmol phenyl halide, and 10 uL nonane (for GC calibration) were added to the vial with a micropipette. An appropriate magnetic stirrer was added to the vial, which was then sealed with cap and crimper. The mixture was brought out of the glove box and was stirred and heated within the microwave. After cooling the vial, the reaction mixture was then “worked up” via the ethyl acetate workup procedure in preparation for GC analysis.
Ethyl Acetate Workup
32
Using a pipet, about 0.5 mL of the reaction mixture was transferred to a test tube under a fume
hood. 1.5-2 mL of ethyl acetate were added to the tube; the solution was stirred and then allowed to sit for five minutes. Using a pipet, the liquid layer was transferred to the top of the silica plug and eluted into a GC vial. General GC Method (JM40-220) A sample of 0.2 µL was injected into the GC at 250 ºC. The temperature of the 30-meter column was 40 ºC for the first 3 minutes. The temperature rose at a rate of 10 ºC /minute for 18 minutes until 220 ºC was reached and remained at that temperature for a final minute. The helium gas flow rate was 4 mL/min. The flame ionization detector temperature was 300 ºC.
Yield Calculations The phenyl halide conversions and phenyl carboxylate yields were calculated using the GC peak areas and slopes of calibration curves built with known samples of nonane and the substrate. In cases where there was no determined calibration factor, a ratio was taken of the area of the expected product divided by the total phenyls found in the reaction. Table 10 Selected Retention Times and Calibration Factors Compound Name
Retention Time in GC
Calibration Factor with Nonane
Phenyl Iodide
7.8
0.734
Phenyl Acetate
8.3
0.7413
Phenyl Benzoate
16.3
1.15
Phenyl Pivalate
10.9
-
λ (max) and Molar Absorptivity Determination
33
The λ (max) was determined by mixing 10 mg of Cu carboxylate in 10 mL of DMF. The molar absorptivity was determined by creating a stock solution of Cu carboxylate in DMF and then diluting to the lower concentrations. Static Spectroscopy The mixture was removed from the heat source and either allowed to cool on the bench or submerged in ice. The vial was then placed in the centrifuge. Once the solid was settled at
the bottom, 100 µL aliquot was removed by puncturing the septum with a needle from the clear, liquid solution towards the top. 25 µL of solution were syringed to the spectroscopy
cuvette which was previously filled with 1 mL of DMF under N2 and sealed with a screw top cap and septum. . 75 µL were placed in a GC vial in preparation for GC analysis. Flow-Cell Spectroscopy A peristaltic pump was used for the circulation system. A hole was made in a septum via a cork borer. The inlet tube of the pump was threaded through the hole in the septum. The
end of a Luer lock syringe with an attached syringe filter was then fitted to the end of the
tube. The outlet tube for the pump was attached to a 0.2 mm flow-cell. Then, using a needle,
the flow- cell outlet tube was then inserted through the septum. The solids and stirrer were
added to the vial before sealing it with the flow-cell system’s septum. The reaction was performed with half concentration of copper (II) carboxylate while the other reactants
remained the same. A balloon, via a needle, was attached to the septum and the system
was then vacuumed and filled with nitrogen gas three times. On the last fill with nitrogen
gas, the balloon was allowed to fill with the gas. The solvent and other liquid reagents were
added via syringe. The reaction was then heated in an oil bath as the pump continuously
34
filled the flow-cell with solution at 1.2 mL/min. Between uses, the tubes were rinsed with a 5:5:1 mixture of water, ethanol, and hydrochloric acid followed by an acetone rinse. Picture 2 Flow Cell Spectroscopy Setup
Machines Used Agilent Technologies 7820A GC System Biotage Initiator Microwave
Varian Cary 50 Bio UV- Visible Spectrophotometer Fischer Scientific Mini- Pump: Variable Flow Shimadzu GCMS-QP2010
Works Cited
35
"All Nobel Prizes in Chemistry." All Nobel Prizes in Chemistry. Nobel Prize, n.d. Web. 08 June 2016.
Beletskaya, I.; Cheprakov, A. Copper In Cross-Coupling Reactions The Post-Ullmann Chemistry. Coordination Chemistry Reviews 2004, 248, 2337–2364.
Beleskaya, I.; Cheprakov, A. The Complementary Competitors: Palladium And Copper In C−N Cross-Coupling Reactions. Organometallics 2012, 31, 7753-7808.
Ben-Zvi, Benjamin. Copper-catalyzed Aromatic Acyloxylation: The Formation of Aryl Esters through the Cross Coupling of Aryl Halides. 2015. Yeshiva University, New York.
Monnier, Florian, and Marc Taillefer. "Catalytic C-C, C-N, and C-O Ullmann-Type Coupling
Reactions: Copper Makes a Difference." Angewandte Chemie International Edition Angew. Chem. Int. Ed. 47.17 (2008): 3096-099. Web.
The Formation of Aryl Esters through the Cross Coupling of Aryl Halides Via Comproportionation of Copper II Reactants
Thesis Submitted in Fulfillment of the Requirements of the Jay and Jeanie Schottenstein Honors Program Yeshiva College
Yeshiva University August 2016
Adam Goldstein
Mentor: Professor Fabiola Barrios-Landeros Department of Chemistry, Yeshiva College
Acknowledgements
2
As this thesis is the capstone of my undergraduate work, I feel it important to thank all of those who have guided me along the path that has led me here.
Thank you to all of my professors. You have helped me to become a better student and a better person.
Specifically, thank you to Professor Fabiola Barrios-Landeros. From start to finish, working
in your lab was a pleasure. I must thank you for the skills you helped me hone and develop,
but also for the comfort that came from working in your lab. More importantly, you showed me just how much teachers truly care about their students.
Abstract
3
Previous research in Prof. Barrios-Landeros lab has shown that Cu (I) acetate complexes
can successfully promote the cross coupling of aryl halides with carboxylates to make aryl esters. The reaction proceeds by a sequence of steps that can be accessed at different
points using different reagents. One of this entry points was the focus of this thesis work,
specifically, to explore the ability of Cu (II) to comproportionate, as Cu (II) complexes are often cheaper and more stable. It was shown that Cu (II) acetate does, in fact, drive the
copper- promoted acetoxylation of phenyl iodide in DMF. Product formation, as measured
by gas chromatography (GC), showed yields of up to 90%. Furthermore, the change in color of the copper as it transitions between the various oxidation states during the reaction
allowed for an in- depth analysis of the rate of reaction, via flow cell spectroscopy. Additionally, Cu (II) sulfate, an even cheaper alternative, was able to drive the reaction in
the presence of acetate salts, though with lower yields. Phase transfer agents, such as
tetrabutylammonium tetraflouroborate and hexaflourophosphate, increased the yield of
the Cu (II) sulfate reactions. Ultimately, it was shown that the comproportionation of Cu (II) is a viable option for adding Cu (I) complexes into cross coupling reactions.
Introduction
4
Organometallic synthesis has been at the forefront of chemistry related scientific
progress in recent years. Case in point, three Nobel Prizes in Chemistry from the past 15
years have been awarded to projects focused on organometallic synthesis (All Nobel Prizes
in Chemistry). The power of organometallic synthesis, particularly in relation to classical organic chemistry techniques, lies in the use of transition metals that ultimately allow
scientists to break and form bonds that were otherwise too difficult to perform. The
transition metals, with their multiple oxidation states, can form bonds with multiple
compounds. By bringing the reactants near each other through bonding with the transition
metal, the transition metal drives the reaction. Specifically, organometallic processes generally have lower activation energies than comparable classic processes. Also, organometallic synthesis allows for higher levels of selectivity, including chemo selectivity,
regio selectivity, and stereo selectivity. While there are many different pathways that organometallic synthesis can follow, they all share a reliance on transition metals.
In particular, cross coupling chemistry is an especially potent process. In short,
cross coupling chemistry allows for the formation of a carbon-carbon or a carbon-
heteroatom bond. Aromatic cross coupling utilizes the possibility of these bonds between aryl halides and nucleophilic compounds to form aromatic hydrocarbons, amines, amides, ethers, and alcohols. (Figure 1)
5
Figure 1 Examples of cross-coupling reactions catalyzed by transition metal complexes
Early aromatic cross coupling reactions focused on copper as the transition metal,
but eventually shifted to a focus on palladium. For example, Fritz Ullman, the earliest chemist to work on aromatic cross coupling reactions was able to show that copper metal drives bond formations that were not seen in that time. With further research, the Ullman
reactions were discovered (Figures 2). These reactions showed that carbon-carbon, carbon-oxygen, and carbon-nitrogen bonds were indeed possible in the presence of a
copper catalyst (Monnier). However, as the field of aromatic cross coupling reactions grew, the focus shifted from copper to palladium. This has been attributed to the discovery of ability for palladium to catalyze an amination reaction (Beletskaya). Figure 2 Examples of Ullman's Work in Cross Coupling
6
The general steps of an aromatic cross coupling reaction have been established. The
three important steps that must occur for aromatic cross coupling are oxidative addition,
reductive elimination, and trasmetalation. In the oxidative addition step, the metal contributes two electrons to break the bond between the aromatic ring and the halide. At the end of this step, the metal oxidizes to +2 of its original state and is attached to both the
aromatic ring and the halide. In the reductive elimination step, the metal, which is coordinated to an aromatic ring and a nucleophillic species, is reduced as the ring bonds with the nucleophilic species. In the transmetalation step, an anionic ligand is displaced from the metal in favor of another one. While both copper and palladium-catalyzed
reactions utilize each of these steps, they vary in the order used. Copper reactions follow the process: transmetalation followed by oxidative addition followed by reductive
elimination. In palladium reactions the oxidative addition step and the transmetalation step are reversed (Figure 3).
7
Figure 3 Steps for Cross Coupling Utilizing Palladium L= PR3
X'
Pd0L A
LnPdII
Reductive Elimination: Formation of C-C bond
MX
n
ArX
Ar X'
C
X
LnPdII B
Ar
X= Cl, Br, I, OTf Oxidative Addition
X
MX'
Transmetalation: Displacement of X by the X' or "coupling partner"
While the steps are the same between copper and palladium aromatic cross
coupling reactions and the focus recently has been on palladium reactions, there are many advantages to using copper for these reactions. First, while copper is relatively inexpensive, palladium can be extremely expensive. Therefore, cross coupling reactions involving
copper would not have to worry about the waste of an expensive metal. Second, whereas palladium only has two stable oxidative states at 0 and +2, copper has accessible oxidative states: 0, +1, +2, and +3 (Beletskaya). The mere existence of the odd number oxidative
states and the amount of states would allow copper to undergo reaction conditions under which palladium could not.
Even considering the advantages that copper holds in cross coupling reactions and
the work that has been done in the field, there are still relatively unexplored areas of copper based aromatic cross coupling reactions. It has been shown that the formation of
ethers works via copper reactions. However, there is little evidence of a reaction for
esterification. Furthermore, it has been shown that amidation is possible, which is similar
8
to the esterification but with a focus on nitrogen instead of oxygen (Beletasky). It is the focus of Dr. Barrios-Landeros’ laboratory to develop a method for the production of
aromatic esters from aryl halides with the use of copper based aromatic cross coupling reactions, or copper- catalyzed aromatic acyloxylation (Figure 4).
Figure 4 Aromatic acyloxylation catalyzed by copper complexes X + X= I, Br, Cl
O
Cu catalyst
R HO R= alkyl or aryl
base
R
O O
Previous work under Dr. Barrios-Landeros has shown that it is indeed possible to
use copper (I) species to perform acyloxylation. By mixing copper (I) acetate and phenyl iodide, phenyl acetate was produced. The proposed mechanism: starting with a
stoichiometric amount of copper(I) acetate as the starting material, this copper species will serve as both the copper catalyst and acetate source. Oxidative addition of the phenyl
iodide is followed by reductive elimination to produce phenyl acetate. Along with the aromatic acyloxylation reaction, there is an aromatic halide exchange reaction that occurs simultaneously (Figure 5) (BenZvi).
Figure 5 Previous Work's Mechanism for Aromatic Acyloxylation
9
While copper (I) species has been shown to drive the aromatic acyloxylation, this
work turns it focus on copper (II) species. The comproportionation- disproportionation reactions, which will allow copper (II) species to enter the reaction, are relatively
straightforward reactions. Comproportionation is a reaction in which two reactants, both containing the same element but at different oxidation numbers, can combine to form a
product with the element at a different oxidation number. This experiment utilizes copper (II) and copper metal to form copper (I). The disproportionation reaction is the opposite reaction, where a reactant is oxidized and reduced to form two products with different
oxidation states. Ultimately, the proposed mechanism introduces copper (II) and copper metal so that the reaction can continue as before (Figure 6).
Figure 6 Proposed Mechanism for the Complete Aromatic Acyloxylation with Comproportionation
10 Copper (II) species have advantages over the copper (I) species for this reaction.
First, the price of copper (II) species is generally lower than for copper (I), making the reaction even cheaper. Second, the copper (II) species is generally more soluble than that of
the same copper (I) species. This will give the reaction further access to the important
transition metal. Third, copper (II) compounds are stable in the presence of oxygen and much easier to synthesize, handle and store than the copper (I) species.
Copper (II) also has a very distinctive blue color. Focusing on the color changes of
the reaction will give further insight into the inner workings of this cross coupling reaction.
Particularly, the disappearance of the copper species will enable a study of the rate of the reaction for aromatic acyloxylation. Fourth, and most importantly, by utilizing copper (II) it
is possible to create copper (I) in situ via comproportionation reactions. As the previous
work has shown, the reaction relies on a copper (I) carboxylate as a starting material. By creating this species in situ, there is more control over the overall reaction.
11
Results Time Dependence of the Reaction with Copper (I) Acetate
To continue the exploration of the mechanism required for the cross coupling reaction to occur, the time dependence of the reaction was measured. Color changes were observed even for reactions that were heated for an hour. The final colors were all similar. However,
monitoring over time the progress of the reactions either via side-by-side reactions or aliquots of the same reaction yielded inconclusive results. Reactions with Labeled Benzene Though the formation of product could be determined via GC, monitoring the product formation over time by gas chromatography could be complicated. Therefore, the reaction
with Copper (I) acetate was performed using substituted benzene rings in attempt to ultimately study the reaction via
19F
NMR. First, 74 mg of CuOAc and 70 μL of 2-
fluoroiodobenzene were mixed in 1 mL of DMF with 10 μL of nonane. Though a color change of brown to blue/green was observed, no product was detected. However, when the
reaction was performed with 4-fluoroiodobenzene, a color change of brown to blue was
observed together with the presence of product, verified both by GC and the GC-MS. Switch to Copper (II) Acetate
To study the reaction further and to find the rate of reaction for the cross coupling of the
copper (I) acetate, different amounts of copper (I) acetate were mixed with DMF in order to
determine the λ (max) and molar absorptivity of copper (I) acetate in DMF. However, due
to the low solubility of copper (I) acetate, it was difficult to obtain solutions with reliable
12
concentrations. Instead, copper (II) complexes were explored as they have better solubility
and would allow the in situ production of copper (I). Copper (II) complexes have been shown to be viable entry points for the proposed mechanism (BenZvi). Furthermore, the
color change that was observed during the reaction would allow for a spectroscopic study of the rate of reaction (Picture 1).
Picture 1 Color Change of the Copper (II) Acetate Reaction. From left to right: Copper (II) acetate reaction before heating, after heating for 5 hours, after centrifugation, after heating for another 5 hours
In preparation for studying the copper (II) complex rate of reaction, the λ (max) and molar absorptivity were determined via conventional methods. As shown, the λ (max) was determined to be at 700 nm (Graph 1) and the molar absorptivity (Table 1) was determined to be 162 mol/L (Graph 2) or 0.87 g/L (Graph 3).
13
Graph 1. Absorption spectra of Cu(II) acetate
Visible Spectra of Copper (II) Acetate 3
2
1.5 1
Absorbance
2.5
0.5 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700 720 740 760 780 800
0
Wavelength (nm)
Table 1 Concentrations vs. Absorbance at 700nm of Copper (II) Acetate Concentration (g/L)
Concentration (mol/L)
Absorbance
1
0.0055
0.972
0.5 1.5 2 3
0.0028 0.008 0.011 0.016
0.4489 1.1907 1.8076 2.602
14
Graph 2 Determination of Molar Absorptivity in mol/L
Concentration (mol/L) vs. Absorbance of Copper (II) Acetate 3
y = 162.07x R² = 0.9933
2.5
2
Absorbance 1.5
1
0.5
0
0
0.005 0.01 0.015 Concentration (mol/L)
0.02
Graph 3 Determination of Molar Absorptivity in g/L
Concentration (g/L) vs. Absorbance Copper (II) Acetate 3
2.5
y = 0.873x R² = 0.9897
2
Absorbance 1.5 1
0.5
0
0
1
2 Concentration (g/L)
3
4
Using Copper (II) Bromide as an Access Point
15
Copper(II) bromide in the presence of a source of acetate ions can be used as precursor for copper acetate species. Copper (II) bromide also has a very unique color, a deep green
(Picture 2). Since it had a different color than the copper (II) acetate and the copper (I)
acetate, the absorption spectra of copper (II) bromide was explored. First, λ (max) was determined to be 600 nm (Graph 4). Then, the molar absorptivity was calculated. The stock solution was mixed at a concentration of 3 g/L, or 0.0067 mol/L. The vials of lower
concentrations were dilutions from the stock solution (Table 2). The molar absorptivity was determined to be 323.15 mol/L (Graph 5) or 1.45 g/L (Graph 6).
Picture 2 Color Comparison of Copper Acetate (Left), Copper Bromide (Right), and Mix (Middle)
16
Graph 4 Absorption spectra of Copper (II) Bromide
Visible Spectra of Copper (II) Bromide in DMF 1.6 1.4 1.2 1
0.8 Absorbance 0.6 0.4 0.2 780
730
680 630 580 Wavelength (nm)
530
480
0
Table 2 Concentrations vs. Absorbance at 600nm of Copper (II) Bromide Concentration (g/L)
Concentration (mol/L)
Absorbance
0.1875
0.0008394
0.1417
0.09375 0.375 0.75 1.5
0.0004197 0.001679 0.003358 0.006715
0.1077 0.4038 0.9851 2.273
17
Graph 5 Determination of Molar Absorptivity in mol/L
Concentration (mol/L) vs. Absorbance of Copper (II) Bromide 2.5
y = 323.15x R² = 0.9825
2
Absorbance
1.5
1
0.5
0
0
0.001
0.002
0.003 0.004 0.005 Concentration (mol/L)
0.006
0.007
0.008
Graph 6 Determination of Molar Absorptivity in g/L
Concentration (g/L) vs. Absorbance of Copper (II) Bromide 2.5
y = 1.4467x R² = 0.9825
2 Absorbance
1.5 1 0.5 0
0
0.2
0.4
0.6 0.8 1 Concentration (g/L)
1.2
1.4
1.6
A side- by- side comparison of the CuBr2 and Cu(OAc)2 was also visualized via UV/Vis. A 1
g/L solution of each CuBr2 and Cu(OAc)2 was mixed. Then, 1 ml from each of those 1 g/L
18
solutions was added to a separate vial, which created a solution that was 0.5 g/L CuBr2 and 0.5 g/L Cu(OAc)2 (Picture 2). The vials were then analyzed via UV/Vis (Graph 7). The peaks
were distinct for the solutions that had only one copper complex. However, for the solution with both copper complexes, neither peak remained distinct.
Graph 7 Side- by- Side Uv/Vis Analysis of Different Copper Complexes
Side- by- Side UV/Vis Analysis of Different Copper Complexes 1.5 1.3 1.1 0.9 0.7 0.5 0.3
1 g/L Copper (II) Acetate 1 g/L Copper (II) Bromide 0.5 g/L Copper (II) Acetate and 0.5 g/l Copper (II) Bromide
200 230 260 290 320 350 380 410 440 470 500 530 560 590 620 650 680 710 740 770 800
0.1
Absorbance
Wavelength (nm)
-0.1
The main goal was still to produce phenyl acetate. Towards that end, an experiment was set up to test the formation of phenyl acetate from CuBr2. Copper (II) bromide, Cu (0), DMF,
PhI, nonane and an acetate source were mixed. While some of these reactions did produce phenyl acetate, the percent yields were extremely low (Table 4).
19
Table 3 Acetate Screening for Copper (II) Bromide Acetate Source
Percent PhI Left
Percent Yield PhOAc
Percent Yield PhBr
Silver
6.05
5.72
18.7
Control Zinc
2.07 10.6
0
Tetrabutylammonium 0.387 Cesium
11.1
Lithium
14.3
Potassium
Sodium
0
53.1
34.5
34.6
0
4.04
26.3
3.13
19.5
6.11
10.6
59.2
45.1
1.42
39.5
At first glance, the low yield of phenyl acetate was attributed to an inability of acetate to transmetallate and replace bromide on a copper species. Since the copper species drives
the reaction according to the proposed mechanism, this would leave acetate without the
ability to become phenyl acetate. To test this hypothesis, an anion exchange test was
performed. Different copper species were mixed with silver acetate in 1 mL DMF and stirred for 15 minutes at room temperature (Table 4). Spectroscopic readings in the visible
region showed that the vial with CuBr2 and AgOAc ultimately produced the highest amount
of copper acetate, showing that the anion exchange works. Table 4 Anion Exchange Test Vial #
1
2
3
CuBr2 (mg)
-
67
-
CuBr (mg)
43
CuS04 (mg)
-
AgOAc (mg)
-
-
50
100
100
1
1
1
DMF (mL)
-
20
48
Using CuSO4 as an Access Point As the copper bromide provided low yields, the spectroscopy peaks overlapped, and there
was halide interference, copper sulfate was explored as another alternative access point of copper acetate species. Furthermore, copper sulfate has a high solubility and it is also fairly inexpensive. Again, the ultimate goal was production of phenyl acetate so the copper sulfate was screened with various acetate sources. The percent yield of phenyl acetate was still low (Table 5).
Table 5 Acetate Screening for Copper (II) Sulfate Acetate Source
Percent PhI Left
Percent Yield Phenyl Acetate
Silver
0
64.9
Control Zinc
Tetrabutylammonium Cesium
Potassium Lithium Sodium
29.8 16.7 0
28.2 13.0 11.2 13.3
0
56.3 16.8 2.12 1.98 13.5 15.5
21
Phase transfer agents were used in an attempt to increase the concentration of acetate in solution
and
raise
the
percent
yields
of
the
reaction.
Tetrabutylammonium
tetrafluoroborate and tetrabutylammonium hexafluorophosphate were chosen as the
experiment phase transfer agents since these anions are not coordinating and should not interfere with the transformation. While other side products similar to those seen by the
copper bromide reactions were produced, a comparison of the two agents between the original reactions and those with the phase transfer agents, showed that the phase transfer agents do indeed raise the percent yield of the reaction, though relatively minimally. The hexaflorophosphate has a slight advantage (Table 6).
Table 6 Percent Yield of Phenyl Acetate of Reactions of Copper Sulfate and Phase Transfer Agents
Acetate Source Control Silver Zinc
Cesium
Potassium Lithium Sodium
Without
Transfer Agent 0
64.88 56.45 2.12 1.99
13.54 15.46
Phase With
Tetrafluoroborate 0 0
76.27 0
1.29
19.16 44.13
With
Hexafluorophosphate 0 0
77.47 2.29 0
19.54 16.63
Interestingly, the silver acetate no longer produced any phenyl acetate in the presence of phase transfer agents. This was surprising, as previous experiments had concluded that
silver acetate was the best form of acetate to drive the reaction (BenZvi). In the presence of
22
the phase transfer agent, silver undergoes reduction and the vials that were used to conduct the silver acetate with phase transfer agent experiments had silver globules inside. Test of Other Copper (II) Carboxylates It was also important to determine whether the proposed mechanism would work for
other carboxylates. Therefore, the general method for the copper acetate experiments was used but with copper pivalate and copper benzoate as well (Table 7).
Table 7 Standard Reaction with Copper Carboxylates Carboxylate
Percent
Acetate
69.2
Benzoate Pivalate
Yield
Carboxylate (Attempt 1) 15.9 24.1
Aryl Percent
Yield
Carboxylate (Attempt 2)
Aryl
62.6 58.9 71.2
The production of phenyl benzoate was a positive step towards finalizing the rate of
reaction for the cross coupling reaction. Therefore, further information about the
spectroscopic ability of copper benzoate was explored. The λ (max) was determined to be
718 nm (Graph 8). Additionally, the molar absorptivity was determined (Table 8). The molar absorptivity of copper benzoate was shown to be 133.8 mol/L (Graph 9) or 0.22 g/L (Graph 10).
23
Graph 8 Lamda Max Scan of Copper Benzoate
Absorbance Scan of Copper Benzoate 1.6 1.4 1.2 1 0.8 Absorbance 0.6 0.4 0.2 0
300
400
500 600 Wavelength (nm)
700
800
Table 8 Concentrations vs. Absorbance of Copper (II) Benzoate at 718nm Concentration (g/L)
Concentration (mol/L)
Absorbance
0.5
0.001635
0.2196
1
0.25
0.125
0.003270
0.0008176 0.0004088
0.4280 0.1392
0.06856
24
Graph 9 Determination of Molar Absorptivity in mol/L
Absorbance
Concentration (mol/L) vs. Absorbance of Copper (II) Benzoate 0.5
y = 133.79x R² = 0.9839
0.4 0.3 0.2 0.1
0
0
0.0005
0.001
0.0015 0.002 0.0025 Concentration (mol/L)
0.003
0.0035
Graph 10 Determination of Molar Absorptivity in g/L
Absorbance
Concentration (g/L) vs. Absorbance of Copper (II) Benzoate
0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0
y = 0.2188x R² = 0.9839
0
0.5
1 1.5 Wavelength (nm)
2
2.5
Interaction of Copper, Iodide, and Acetate The reaction that makes phenyl acetate also yields copper iodide as product. Before starting the process of spectroscopic analysis of the in situ production of CuOAc from
Cu(OAc)2, it was important to establish that the observed peaks were a result of the copper
25
acetate complexes and not a result of copper iodide complexes. Vials containing different
amounts of copper complexes were analyzed via absorption spectroscopy (Table 9). It was shown that the CuI would not affect the readings from the copper(I) or copper(II) acetate complexes (Graph 11).
Table 9 Test to See the Interaction Between Copper, Acetate, and Iodide Vial #
Cu(OAc)2 (mg)
CuOAc (mg)
CuI (mg)
DMF (mL)
2
-
36
-
1
1
55
3
-
-
4
-
28
5
57
-
-
-
1
29
18
1
1
29
1
Graph 11 Spectroscopic Data for the Interaction between Copper, Acetate, and Iodine
Interaction of Copper Salts of Acetate and Iodide
Absorbance
2.5
2
Cu(OAc)2
1.5
CuOAc CuI
1
Cu(Oac)2 + CuI CuOAc +CuI
0.5
0
250
350
450
550
Wavelength (nm)
650
750
26
Rate of Reaction- Static Spectroscopy
To study the rate of the reaction, the original experiment designed required heating a vial with the reaction mixture and taking aliquots to acquire absorption spectroscopy. Because the reaction was heterogeneous, the vial was centrifuged in order to prevent solid particles
from interfering with the spectrophotometry data. This method originally had many
complications. Eventually, the experiment produced a graph that showed a general decline
in the copper (II) acetate peak and a general decrease in the copper (I) acetate peak (Graph 12). It was shown that spectrophotometry could be a proper method for determining the rate of reaction.
Graph 12 Spectrophotometry Data Proving UV/Vis Could be Used to Determine Rate of Reaction
Standard Reaction Microwaved at 160° C 3
2.5 2
1.5 1
0.5 200 225 250 275 300 325 350 375 400 425 450 475 500 525 550 575 600 625 650 675 700 725 750 775 800
0
Wavelength (nm)
-0.5
Absorbance 30 min 60 min 90 min
120 min
27
Rate of Reaction- Flow-Cell Spectroscopy with Copper Acetate
Though it was shown that spectroscopy was a viable method for studying the rate of the
aromatic cross coupling reaction, the static spectroscopy method was not the best way to achieve the best data for the reaction. The reaction was then studied via flow-cell spectroscopy. Following the flow-cell spectroscopy method, the reaction was analyzed.
After much trial and error, the method worked for some analysis but has not yet been
perfected. The insolubility of certain reactants in the solution was one of the largest
obstacles to consider. Therefore, reliable readings were often only obtained for early stages of the reaction because the flow cell would become clogged. In some experiments, a reasonable trend was observed for the copper species. Yet, no product was formed and the graphs showed very different results (Graph 13 and 14).
Graph 13 Copper Acetate Reaction with Stoichiometric Amount of Copper (II) Reactant
Flow- Cell Spectroscopy of Copper Acetate Reaction 0.7 0.6 Absorbance
0.5 0.4
Cu (II)
0.3
Cu (I)
0.2 0.1
0
0
20
40
60 80 Time (min)
100
120
140
28
Graph 14 Copper Acetate Reaction with Halved Amount of Copper (II) Reactant (data formatted to remove data points during which the flow cell was clogged)
Flow Cell Spectroscopy of Copper Acetate Reaction 0.7 0.6 Absorbance
0.5 0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
10
20 Time (min)
30
40
In an attempt to study the actual rate of reaction without worrying about clogs during the time of the comproportionation-disproportionation reaction, an experiment was
performed where the Cu(OAc)2 and Cu metal was left heating overnight. Then, the reaction was continued with the addition of the PhI and analyzed via flow cell spectroscopy (Graph 15).
Graph 15 Copper Acetate Flow- Cell Spoctroscopy with Product Formation
29
Flow- Cell Spectroscopy of Copper Acetate with Product Formation Observed 0.6 0.5 Absorbance
0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
5
10
15 20 Time (min)
25
30
35
Rate of Reaction- Flow-Cell Spectroscopy with Copper Benzoate As many of the issues with studying the copper acetate rate of reactions were a result of solubility, the experimental method was attempted with copper benzoate (Graph 16). Graph 16 Flow- Cell Spectroscopy of Copper Benzoate
Flow Cell Spectroscopy of Copper Benzoate 0.7 0.6 Absorbance
0.5 0.4 0.3
Cu (II)
0.2 0.1
0
Cu (I) 0
20
40 Time (min)
60
80
Conclusion
30
Comproportionation of copper (II) complexes was shown to be a reliable avenue for in situ production of copper (I) complexes. Ultimately, this was able to drive the aromatic
acyloxylation, following the proposed mechanism. Copper (II) acetate, copper (II) bromide,
and copper (II) sulfate were all able to form phenyl acetate under the reaction conditions. Copper (II) sulfate in the presence of acetate salts and phase transfer agents yielded the
highest amount of product. It was also shown that the reaction works with other copper carboxylates, including copper pivalate and copper benzoate, to produce aromatic esters.
Even more so, the use of copper (II) complexes added an important approach for studying the reaction by causing an observable color change. The solution starts with a copper (II)
species, lending a vivid blue color. As the reaction progresses, copper (I) species are
formed, which are usually a pale beige color. Together, both copper species give the solution a green color. Static spectroscopy was not able to provide reliable data for studying the rate of reaction. This is likely due to the heterogeneous nature of the reaction mixture. Flow-cell spectroscopy, on the other hand, does indeed seem like a reliable
method for determining the rate of the reaction. However, at present time, the ideal conditions under which to perform the flow-cell spectroscopy were not determined. The
experiments performed on the copper (II) acetate for this work often stopped working as
the result of clogs. The copper (II) benzoate showed more promise. Larger tubing, larger
diameter flow cells, and reactants/products with higher absorptivity will likely lend themselves to more accurate data readings.
31 Experimental Section
Representative Method for Copper Reactions on Hot Plate Under N2 atmosphere, 0.3 mmol of copper (II) compound together with 0.3 mmol of copper (0) powder was weighed in a small vial inside a glove box. If applicable, the additional 0.6 mmol of acetate source or 0.1 mmol of transfer agent was then weighed in the vial. 1 mL of solvent, 0.6 mmol phenyl halide, and 10 uL nonane (for GC calibration) were added to the vial with a micropipette. A magnetic stirrer was added to the vial, which was then sealed with a screw cap with a Teflon septum. The vial was brought out of the glove box and the mixture was stirred at 110°C for 10 hours on a hot plate. After cooling the vial, the reaction mixture was then “worked up” via the ethyl acetate workup procedure in preparation for GC analysis.
Representative Method for Copper Reactions in Microwave Under N2 atmosphere, the 0.3 mmol of copper (II) catalyst together with the 0.3 mmol of copper (0) powder was weighed in a small vial inside a glove box. 1 mL of solvent, 0.6 mmol phenyl halide, and 10 uL nonane (for GC calibration) were added to the vial with a micropipette. An appropriate magnetic stirrer was added to the vial, which was then sealed with cap and crimper. The mixture was brought out of the glove box and was stirred and heated within the microwave. After cooling the vial, the reaction mixture was then “worked up” via the ethyl acetate workup procedure in preparation for GC analysis.
Ethyl Acetate Workup
32
Using a pipet, about 0.5 mL of the reaction mixture was transferred to a test tube under a fume
hood. 1.5-2 mL of ethyl acetate were added to the tube; the solution was stirred and then allowed to sit for five minutes. Using a pipet, the liquid layer was transferred to the top of the silica plug and eluted into a GC vial. General GC Method (JM40-220) A sample of 0.2 µL was injected into the GC at 250 ºC. The temperature of the 30-meter column was 40 ºC for the first 3 minutes. The temperature rose at a rate of 10 ºC /minute for 18 minutes until 220 ºC was reached and remained at that temperature for a final minute. The helium gas flow rate was 4 mL/min. The flame ionization detector temperature was 300 ºC.
Yield Calculations The phenyl halide conversions and phenyl carboxylate yields were calculated using the GC peak areas and slopes of calibration curves built with known samples of nonane and the substrate. In cases where there was no determined calibration factor, a ratio was taken of the area of the expected product divided by the total phenyls found in the reaction. Table 10 Selected Retention Times and Calibration Factors Compound Name
Retention Time in GC
Calibration Factor with Nonane
Phenyl Iodide
7.8
0.734
Phenyl Acetate
8.3
0.7413
Phenyl Benzoate
16.3
1.15
Phenyl Pivalate
10.9
-
λ (max) and Molar Absorptivity Determination
33
The λ (max) was determined by mixing 10 mg of Cu carboxylate in 10 mL of DMF. The molar absorptivity was determined by creating a stock solution of Cu carboxylate in DMF and then diluting to the lower concentrations. Static Spectroscopy The mixture was removed from the heat source and either allowed to cool on the bench or submerged in ice. The vial was then placed in the centrifuge. Once the solid was settled at
the bottom, 100 µL aliquot was removed by puncturing the septum with a needle from the clear, liquid solution towards the top. 25 µL of solution were syringed to the spectroscopy
cuvette which was previously filled with 1 mL of DMF under N2 and sealed with a screw top cap and septum. . 75 µL were placed in a GC vial in preparation for GC analysis. Flow-Cell Spectroscopy A peristaltic pump was used for the circulation system. A hole was made in a septum via a cork borer. The inlet tube of the pump was threaded through the hole in the septum. The
end of a Luer lock syringe with an attached syringe filter was then fitted to the end of the
tube. The outlet tube for the pump was attached to a 0.2 mm flow-cell. Then, using a needle,
the flow- cell outlet tube was then inserted through the septum. The solids and stirrer were
added to the vial before sealing it with the flow-cell system’s septum. The reaction was performed with half concentration of copper (II) carboxylate while the other reactants
remained the same. A balloon, via a needle, was attached to the septum and the system
was then vacuumed and filled with nitrogen gas three times. On the last fill with nitrogen
gas, the balloon was allowed to fill with the gas. The solvent and other liquid reagents were
added via syringe. The reaction was then heated in an oil bath as the pump continuously
34
filled the flow-cell with solution at 1.2 mL/min. Between uses, the tubes were rinsed with a 5:5:1 mixture of water, ethanol, and hydrochloric acid followed by an acetone rinse. Picture 2 Flow Cell Spectroscopy Setup
Machines Used Agilent Technologies 7820A GC System Biotage Initiator Microwave
Varian Cary 50 Bio UV- Visible Spectrophotometer Fischer Scientific Mini- Pump: Variable Flow Shimadzu GCMS-QP2010
Works Cited
35
"All Nobel Prizes in Chemistry." All Nobel Prizes in Chemistry. Nobel Prize, n.d. Web. 08 June 2016.
Beletskaya, I.; Cheprakov, A. Copper In Cross-Coupling Reactions The Post-Ullmann Chemistry. Coordination Chemistry Reviews 2004, 248, 2337–2364.
Beleskaya, I.; Cheprakov, A. The Complementary Competitors: Palladium And Copper In C−N Cross-Coupling Reactions. Organometallics 2012, 31, 7753-7808.
Ben-Zvi, Benjamin. Copper-catalyzed Aromatic Acyloxylation: The Formation of Aryl Esters through the Cross Coupling of Aryl Halides. 2015. Yeshiva University, New York.
Monnier, Florian, and Marc Taillefer. "Catalytic C-C, C-N, and C-O Ullmann-Type Coupling
Reactions: Copper Makes a Difference." Angewandte Chemie International Edition Angew. Chem. Int. Ed. 47.17 (2008): 3096-099. Web.
