Reduction with Sodium Borohydride and Stereoscopic Determinations

Document Sample
Reduction with Sodium Borohydride and Stereoscopic Determinations Powered By Docstoc
					                                                                             Avanti Deshpande 1

       Reduction with Sodium Borohydride and
            Stereoscopic Determinations
Avanti Deshpande
Lab III Section A
15 November 2008

This experiment was designed to reduce benzoin into hydrobenzoin with the reducing agent
sodium borohyrdride. After forming hydrobenzoin, the stereochemistry of the molecule was
determined by converting it to a cyclic acetal, using 2,2-dimethoxypropane and the catalyst
p-tolusulfonic acid. This cyclic acetal form would be make it easier to determine which
enantiomer of the product was formed. The initial reduction reaction was monitored using
thin layer chromatography (TLC) and was verified using Infrared spectroscopy (IR) after
purification via recrystallization. The conversion to the cyclic acetal was verified using
proton nuclear magnetic resonance (H-NMR) and was subsequently purified via
recrystallization as well. The overall reactions are show below with the reduction in Equation
1a and the cyclic acetal formation in Equation 1b.

                                                                                        Eqn. 1a

                                                                                        Eqn. 1b

One of the major purposes, as mentioned above, was to determine the stereochemistry of the
product. This refers to the orientation of the atoms in the molecule in space and the different
physical properties associated with each form of the molecule. In order to understand the
concepts discussed later, it is important to define a few key terms1.

First, a pair of stereoisomers is two molecules that have the same formula and connectivity
but are arranged differently in space. In a molecule, a particular atom can be a stereocenter
    1- Definitions adapted from: Patrick, Graham L. Organic Chemistry: Second Edition. New York:
                Bios Scientific Publishers. 2003. Pgs 47-51.
                                                                            Avanti Deshpande 2

when the changing of positions of its connections creates a separate stereoisomer. For
example, in this experiment, two of the carbons in hydrobenzoin can be considered
stereocenters because interchanging the hydroxyl groups with the proton on those carbons
would create a different stereoisomer. Stereocenters can lead to chirality, meaning the mirror
image of the molecule cannot be superimposed on itself. This property makes the molecules
optically active. Achiral molecules, on the other hand, can be superimposed on their mirror
images and thus are not optically active, even though they have stereocenters. Often, achiral
compounds are meso compounds (discussed in the next paragraph).

Molecules that have stereocenters can be described as diastereomoers. Diastereomers are
divided into two classes: meso compounds and +/- enantiomers. Meso compounds have a
plane of symmetry within the molecule that allows them to be superimposable on their mirror
images. This property makes them optically inactive and thus achiral. Diastereomers that are
chiral, and thus optically active, can be called enantiomers, and usually described as + or –
enantiomers, depending on the orientation of the substituents on the stereocenter. Usually,
the plus and minus enantiomers also have similar physical properties and thus cannot be
distinguished from one another in this experiment. These properties, however, are
significantly different from the meso diastereomer of the same molecule.

This difference is important in this experiment since the purpose is to determine which
diasteriomer is the major product of the reduction. Specifically, in this experiment, the
conversion of hydrobenzoin to the cyclic acetal will facilitate the determination of which
diastereomer is formed because the H-NMR of the cyclic acetal will have different peaks for
meso versus the +/- diastereomer. In addition, the physical properties, such as melting point,
will be different for each diastereomer. Understanding these differences and key terms helps
in determining which product is obtained.

   1- Definitions adapted from: Patrick, Graham L. Organic Chemistry: Second Edition. New York:
               Bios Scientific Publishers. 2003. Pgs 47-51.
                                                                            Avanti Deshpande 3

In this experiment, benzoin was reduced to hydrobenzoin using sodium borohydride. It was
then reacted with 2,2-dimethoxypropane to convert it into a cyclic acetal. The crude
hydrobenzoin was purified using a mixed solvent recrystallization with reagent grade acetone
and hexane. The physical data for hydrobenzoin is presented below in Table 1.

                         Table 1: Physical Data for Hydrobenzoin
                      Property                Observation
                      Appearance              Powdery
                      State                   Solid
                      Color                   White
                      Melting Point- Crude 134-137 C
                      Melting Point- Pure     134-136 C

This first step of the experiment had a step-wise yield of 28.2%. Equation set 2 below shows
the calculations done to obtain this value.

                                                                                    Eqn. Set 2

This initial reaction was monitored using thin layer chromatography (TLC). The stationary
phase was the silica gel plate and the mobile phase was 10:1 n-hexane/ether. There were two
compounds spotted on the plate: 1) the reaction mixture and, 2) benzoin dissolved in the
mobile phase. The TLC plate was spotted15 minutes after starting the reaction. Figure 1
below is a sketch of the plate and Table 2 below corresponds to the Rf values on the plate for
each compound.

 Table 2: TLC Analysis for Initial Reaction
Spot           Rf value
Rxn Mixture    0.44
Benzoin        0.56

                                                               Figure 1: TLC Plate Analysis
                                                                   from Initial Reaction
                                                                            Avanti Deshpande 4

Part of this crude product was used in the second part of the experiment. The rest was
recrystalized using a mixed solvent method with reagent grade acetone and hexane. This
purified product was used in IR analysis to verify that the carbonyl had been reduced to a
hydroxide. The presence of a strong OH stretch and the lack of a C=O stretch confirmed this
conversion. Table 3 below summarizes the IR data obtained from the purified hydrobonzoin.
The sample preparation method was nujol mull.

                       Table 3: IR Data for Pure Hydrobenzoin
                    Observed Peak          Assignment
                    3373/3308              OH Stretch
                    1454                   C=C Aromatic
                    1279                   C–O Stretch
                    754/699                CH Monosub. oop bend

The next part of the experiment was to convert the crude hydrobenzoin crystals into a cyclic
acetal, a form that would aid in determining the stereospeceficity of the product formed. This
was done by reacting the hydrobenzoin with 2,2-dimethoxypropane. After NMR analysis
(data discussed below), the cyclic acetal was recrystalized with petroleum ether. The
physical data for the cyclic acetal formed is shown below in Table 4.

                     Table 4: Physical Data for Cyclic Acetal Product
                      Property                Observation
                      Appearance              Shiny, small crystals
                      State                   Solid
                      Color                   Cream/Light Yellow
                      Melting Point- Crude 49-57 C
                      Melting Point- Pure     54-56 C

This second part of the experiment had a stepwise yield of 35.8%. The overall experimental
yield was 10.1%. The calculations done to obtain these values in Equation Set 3 are shown

                                                                                    Eqn. Set 3
                                                                            Avanti Deshpande 5

The crude cyclic acetal formed was used to do 1H-NMR analysis because it retained the same
stereoscopic features as the hydrobenzoin formed. The crystals were dissolved in CDCl3 +
1% TMS and analyzed in an instrument with 60 Hz field strength. The data obtained is given
below in Table 5. Figure 2 below that corresponds to the assignments made in the table.

                     Table 5: 1H-NMR Data for Crude Cyclic Acetal
                   Assignment       Shift (ppm)           Splitting (#H)
                   A                7.02                  S (10 H)
                   B                5.52                  S (2 H)
                   C                1.82                  S (3 H)
                   D                1.62                  S (3 H)
                                S– Singlet splitting pattern

                              Figure 2: Cyclic Acetal Product

Due to the two separate singlet peaks in the 1.6-1.8 range, it can be determined that the meso
form of hydrobenzoin was formed. Because there is no evidence of a peak at 1.7 ppm (which
corresponds to the +/- enantiomers), it can be concluded that the product was 100% in the
meso form.
                                                                               Avanti Deshpande 6

In this experiment, the first task was to reduce benzoin to hydrobenzoin using sodium
borohydride. TLC was used to determine when the initial reaction was complete. The TLC
plate, which was spotted with the reaction mixture and a benzoin standard, had a mobile
phase 40/60 ethyl acetate/petroleum ether. The reaction was determined to be complete
because the reaction mixture spot traveled less on the TLC plate than the standard benzoin.
This proves that the reaction was taking place because the hydrobenzoin being formed would
be more polar than the initial benzoin reactant because hydrobenzoin has an additional
hydroxyl group, meaning it has an extra hydrogen-bond donor. In addition, the reaction
mixture did not have any spot at the same Rf value as the benzoin standard meaning there
was no benzoin left in the reaction mixture. The combination of these two pieces of
information suggested that the reaction was complete.

After purification via mixed solvent recrystallization of the hydrobenzoin crystals, an IR
spectrum was also obtained. A key feature in the IR spectrum was the lack of a carbonyl
stretch in the 1600 cm-1 to 1750 cm-1 region. There was no strong signal there, suggesting
that the carbonyl had effectively been reduced. An additional feature of the spectrum that
suggested the product was present was the strong hydroxyl stretch. Although this would have
also been present in the spectrum of the hydrobenzoin, it is important to realize that it is still
in the spectrum and thus the correct functional groups are on the product. Finally, a melting
point analysis was done on both the crude and the purified crystals. For hydrobenzoin, the
crude crystals had a melting point of 134 C – 137 C while the pure crystals had a melting
point of 135 C – 137 C. The mechanism for the reduction of benzoin is shown below in
Figure 3.

                            Figure 3: Reduction Mechanism with
                                    Sodium Borohydride

In order for the reaction to proceed successfully, it is important to add hydrochloric acid and
water. This is because the sodium borohydride only reduces the carbonyl to a carbon-oxygen
single bond. The BH3+ is still attached to the oxygen atom. In order for it to disassociate, the
                                                                              Avanti Deshpande 7

hydrochloric acid and water are added to the solution, creating H3O+ in the solution. The
protons on this ion are very electrophilic and are thus easily attacked by the lone pairs on the
oxygen atom, even though it is still attached to BH3+. This attack facilitates the disassociation
of the BH3+ and leaves the oxygen attached to the proton from the H3O+ ion, creating

The addition of hydrochloric acid and water also causes intense foaming in the reaction
mixture. This is because as the acid and water are being added, various salts are formed from
the sodium borohydride ions. These salts are essentially formed when H3B-OH goes to
B(OH)3 and O-B(OH)2.

The second task in this experiment was to convert the crude hydrobenzoin into a cyclic acetal
so that its stereoscopic features could be determined. Although the reaction was not
monitored, the crude crystals were analyzed using H-NMR. The spectrum showed 10
equivalent protons in the aromatic region at 7.02 ppm. These ten protons correspond to the
two equivalent monosubstituted aromatic rings, each with five protons. In addition, there is
another singlet at 5.52 ppm which corresponds to the two equivalent protons on the carbons
that are part of the cyclic acetal that is formed. They are equivalent because they are both
attached to the equivalent phenyl rings and the equivalent oxygen atoms. Finally, there are
two singlets between 1.6 ppm and 1.8 ppm. Each of these singlets corresponds to a methyl
group on the carbon attached to two oxygen atoms as part of the cyclic acetal structure. The
fact that the two groups of protons are not equivalent in the spectrum suggests that they are
placed in different environments. This diastereomer is meso form because the structure
suggests that one set of protons is more deshielded than the other. The mechanism for the
cyclic acetal formation is shown below in Figure 4.

                     Figure 4: Mechanism for Cyclic Acetal Formation

As mentioned above, the major product was determined to be the meso compound. This was
determined based on the data obtained from the H-NMR spectrum. Because there were two
                                                                             Avanti Deshpande 8

singlets at 1.62 ppm and 1.82 ppm, each with three protons, it suggests that the two methyl
groups were not in the same environment. In the meso diastereomer, the two phenyl groups
are pointing in the same direction in the same plane. In addition, one of the terminal methyl
groups is in pointing in that direction. The fact that this methyl group is in the same direction
means that there is a significant amount of steric hindrance for those protons. This prevents
the protons from shifting as much. On the other hand, the second methyl group in the meso
diastereomer is not in the same plane as any phenyl groups, meaning there is little steric
hindrance, allowing the protons to shift much more downfield. In the +/- diastereomers, each
phenyl group shares a plane with a methyl group, creating the same environment for all of the
methyl protons. This is why in the enantiomers, there would only be one singlet at 1.7 ppm
but it would integrate to six protons. The two cyclic acetal structures described above can be
more clearly seen in Figure 5 (Meso) and Figure 6 (enantiomers).

            Figure 5: Meso                                 Figure 6: Enantiomers

Both the Cram model and the Felkin-Ahn models predict this diastereomer to be the major
product. These models, however, predict the major product from the benzoin reduction
instead of the cyclic acetal formation. First, the Cram model predicts the attack of the hydride
is favored from the side of the proton on the alpha carbon to the carbonyl. This would create
R,S stereocenters on the hydrobenzoin; in other words, it would result in the meso
diastereomer. The minor product would be obtained if the attack is from the side of the
hydroxyl on the carbonyl, which is a larger group, and thus slightly more sterically hindered.
The results of the Cram model can be seen below in Figure 7.

Benzoin                          Major (Meso)                           Minor (enantiomer)

                             Figure 7: Cram Model Predictions

This same result can be predicted from the Felkin-Ahn model. This model predicts the meso
product as well because the attack is from the side of the proton on the alpha carbon as well,
not the side of the hydroxyl group. Because the phenyl group is perpendicular to the
carbonyl, the steric hindrance is minimized. A depiction of the Felkin-Ahn model can be seen
below in Figure 8.
                                                                           Avanti Deshpande 9

                         Figure 8: Felkin-Ahn Model Predictions

Since both the Cram and Felkin-Ahn models predict the meso diastereomer, the correct
product was obtained in this experiment. Because of the difference in shifts of the two methyl
groups, it is clear that the R,S meso diastereomer was formed. The results are in fact


In this experiment, benzoin was successfully reduced to hydrobenzoin using sodium
borohydride as the reducing agent. Confirmation of this can be seen in the TLC assay, which
showed a lack of any benzoin remaining in the reaction mixture as well as having a lower Rf
value due to its increased hydrogen bonding capabilities in comparison to benzoin. In
addition, the IR spectrum obtained showed a lack of a carbonyl stretch, but a strong hydroxyl
stretch, indicating that the correct functional groups were present on the molecule. The
melting point, although not particularly indicative of a successful reaction, did show that
purification was successful because the range of the melting point decreased. The cyclic
acetal formation from the crude hydrobenzoin was also successful. This is supported by the
H-NMR data obtained. The spectrum showed no hydroxyl protons but rather four
independent singlets that corresponded to aromatic, aliphatic, and 2 non-equivalent methyl
groups. The H-NMR spectrum also proved that the meso enantiomer was the major product
because two separate singlets were present at 1.6 ppm – 1.8 ppm, instead of only one singlet
at 1.7 ppm for the +/- enantiomers. In addition to this structure verified by H-NMR, the
melting point of the crystals dropped significantly from 135 C – 137 C for the hydrobenzoin
to 54 C – 56 C for the cyclic acetal product, indicating that a significantly different
compound had been formed. The percent yield for the hydrobenzoin formation (Step 1) was
28.2% and the percent yield for the cyclic acetal formation (Step 2) was 35.8%. This gave an
overall experimental percent yield of 10.1%.
                                                                            Avanti Deshpande 10

Reduction of Benzoin
Benzoin (1.041 g, 0.00488 mol), sodium borohydride (0.271 g, .00716 mol) and 15 mL of
ether were placed in a dry, clean flask with a stir bar and began stirring. While stirring began,
a TLC chamber was prepared with the mobile phase 40/60 (v/v) ethyl acetate/petroleum
ether. At 15 minutes, a silica gel plate was spotted with the reaction mixture and a benzoin
standard (prepared by dissolving 10 mg benzoin in the mobile phase). The plate showed that
no benzoin remained in the reaction mixture so stirring was stopped. The mixture looked
cloudy white. The reaction flask was placed over ice and continued to stir. 15 mL of water
was added to the mixture and the mixture went clear. 2 mL of 6 M hydrochloric acid was
added drop-wise to the reaction mixture and then 5 mL water were added again. The mixture
continued to stir for 17 minutes during which white precipitate formed. The precipitate was
vacuum filtered and washed with 25 mL of cold water while still under vacuum pressure. The
crude precipitate (0.613 g, 58.6% yield, MP 134 C – 137 C) was placed in an evaporating
dish to dry overnight. The precipitate, hydrobenzoin (0.173 g) was recrystallized via a mixed
solvent method using 7 mL of reagent grade acetone, 6 drops hexanes and then 4 drops
acetone again. Solution was placed in an ice bath and in the fridge overnight during which
crystals formed and subsequently vacuum filtered. This gave purified hydrobenzoin (0.0834
g, 48.2% yield, MP 135 C – 135 C). Overall Step 1 yield: 28.2%. IR (nujol mull): 3373/3308
cm-1 (OH stretch), 1279 cm-1 (C–O stretch), 754/699 cm-1 (CH monosubstituted OOP bend).

Cyclic Acetal Formation
Crude hydrobenzoin (0.410 g, .00191 mol), 2,2-dimethoxypropane (5 mL, 4.24 g, .0406
mol), and 0.012 g p-tolusulfonic acid were placed in a 25 mL round-bottom flask and stirred.
After one hour of stirring, the mixture was transferred to a separatory funnel with 50 mL
methyline chloride. The aqueous layer was removed and the organic layer was washed with
25 mL water, 25 mL saturated sodium bicarbonate and 25 mL of water again. The organic
layer was then dried with anhydrous magnesium chloride. This mixture stood for 15 minutes
after which the solution was gravity filtered to remove the drying agent. The solvent was then
evaporated from the solution by aiming a slow stream of air at the neck of the flask and
partially submerging the flask in a warm water bath. After the solvent was removed from
product, crude cyclic acetal crystals (0.238 g, 48.9% yield, MP 49 C – 57 C) remained and
were vacuum filtered. After reserving a few crystals for H-NMR, the rest were recrystalized
using 7 mL of petroleum ether. The pure crystals (0.0611g, 73.2% recovery, MP 54 C – 56
C) were then vacuum filtered and dried. Step 2 Yield: 35.8%. Full Experimental Yield:
10.1%. H-NMR (CDCl3 + 1% TMS): 7.02 ppm (singlet, 10 H), 5.52 ppm (singlet, 2 H), 1.82
ppm (singlet, 3H), 1.62 ppm (singlet, 3H).

Shared By: