Experiment 3 Exp 3-1

Experiment 3: Part A. Preparation of Trans-9-(2

phenylethenyl)anthracene with the
Horner-Wadsworth-Emmons Reaction.

References:
CGWW, pp 814-818

Background:

The Wittig reaction constitutes a valuable method for the preparation of alkenes.
The major advantages of this approach are that (1) there is no ambiguity in the
location of the C=C generated by the reaction, as there can be in many elimination
reactions and (2), there is no potential for rearrangements, as there can be in E1
elimination reactions.

An ylide is a neutral species whose Lewis structure contains opposite charges on
adjacent atoms. The atoms involved are carbon and an element from either group
15 (VA) or 16 (VIA) of the period table, such as N, P, or S. The Wittig reaction
uses phosphorus yildes, which are obtained by deprotonation of a phosphonium
salt with a strong base. Phosphorus ylides are relatively stable, but reactive
species, for which the following resonance structures may be written; it has been
suggested that the phosphorus atom can exceed an octet by accommodating
electron donation into its 3d orbital.



The phosphonium salts are available through a nucleophilic displacement reaction
of triphenylphosphine with various alkyl halides. Triphenylphosphine is a good
nucleophile, and thus most phosphonium salts are easily prepared.







Experiment 3 Exp 3-2

The hydrogen atoms on the carbon attached to the resulting phosphorus cation are
somewhat acidic because they are adjacent to a positive charge, a significant
electron-withdrawing group. Thus, treatment of the phosphonium salt with a
strong base, such as butyllithium in THF or sodium hydride in DMSO, removes
one of these protons and produces the ylide.

The ylides are generally not isolated, but rather generated in situ and reacted
directly with the carbonyl compound.

So-called instant ylides that are solid-phase mixtures of a phosphonium salt with
sodium amide (NaNH
2
, a strong base and the conjugate base of ammonia) and are
commercially available. In the solid phase, no reaction between the strong base
and the phosphonium salt occurs. To generate the desired ylide, the instant ylide
mixture need only be placed in a suitable solvent. This process is a marked
advantage over the usual methods used to obtain these species. Ylides, because of
the significant negative charge density on carbon, are nucleophilic enough to attack
electrophiles as reactive as the carbon of a carbonyl group. When the ylide is
reacted with an aldehyde or a ketone, an intermediate oxaphosphetane (a four-
membered ring containing an oxygen and a phosphorus atom) is formed, which
decomposes to give the alkene product.

The Wittig reaction is very general. The significant limitation is that the carbonyl
compound cannot contain acidic hydrogen atoms, such as in an alcohol or carboxyl
group, or electrophiles more reactive than the aldehyde or ketone itself. The Wittig
reaction can be quite stereoselective for the formation of either Z or E alkenes, but
the factors involved in such stereoselectivity are sometimes difficult to predict and
may be difficult to explain. A reasonably general rule states that the Z alkene
predominates if a non-resonance stabilized yilide is used in the reaction. The E
alkene is preferred when a stabilized ylide is used as is the case with the Horner-
Wadsworth-Emmons reaction described below.

An important modification (often called the Horner-Wadsworth-Emmons reaction)
of the Wittig reaction makes use of phosphonate esters, RPO(OR)
2
. It is highly
stereoselective for the formation of E-alkenes. The reaction and mechanism are
depicted below for the preparation of (E)-stilbene. Instead of using a phosphorus
cation to stabilize the negative charge, as in the phosphonium ylide above, a
phosphonate ester group is used to stabilize an adjacent carbanion. Most important
Experiment 3 Exp 3-3

is the fact that the sodium dialkylphosphate is soluble in water unlike
triphenylphosphine oxide, so the workup is much easier when the phosphonate
route is used.


Trans-9-(2-phenylethenyl)anthracene:

Reagents and Equipment. Weigh and place 106 mg (0.51 mmol) of 9-
anthraldehyde in a 10-mL round-bottom flask containing a magnetic stirrer. Add 2
drops of Aliquat 336 (tricaprylmethylammonium chloride), 120 L (0.58 nmol) of
diethyl benzylphosphonate, 2.0 mL of methylcyclohexane and 2 mL of 40%
sodium hydroxide solution. Attach the vial to a reflux condenser.

Note: The diethyl benzylphosphonate, methylcyclohexane, and NaOH solution are
dispensed in the hood using automatic delivery pipets. The Aliquat 336 is very
viscous. A medicine dropper is used to dispense this material.

Reaction Conditions. Heat the two-phase mixture, on a sand bath at a temperature
of about 125-130oC, for 45 min. Stir the reaction mixture vigorously during this
period; the upper (organic) layer turns deep red. Allow the mixture top cool to
room temperature.

Isolation of Product. Transfer the two-phase solution to a 15-mL glass centrifuge
tube using a Pasteur filter pipet. Rinse the flask with 1mL of methylene chloride
and transfer this rinse to the centrifuge tube. Carefully remove the aqueous layer
using a Pasteur filter pipet, and save it in a vial until you have successfully isolated
and characterized the product. Wash the organic layer with two 2-mL portions of
water. Stir the mixture with a small glass rod after each addition and then remove
the water layer and save it in the same vial as before. Dry the methylene chloride
solution by adding sodium sulfate. Using a Pasteur filter pipet, transfer the dried
solution to a 10-mL Erlenmeyer flask containing a boiling stone. Wash the sodium
sulfate remaining in the centrifuge tube with two 1-mL portions of methylene
chloride. Also transfer these washings to the Erlenmeyer flask.
Experiment 3 Exp 3-4

Concentrate the solution, which contains the desired product, almost to dryness on
a sand bath under a slow stream of nitrogen gas. Now add approximately 1-2 mL
portions of 2-propanol (isopropanol) to the flask and allow the resulting solution to
stand at room temperature for 10-15 min, and then place it in an ice bath to
complete the crystallization of the product.

Purification and Characterization. Collect the yellow crystals by vacuum
filtration, using a Hirsch funnel, and wash the filter cake with two 1-mL portions of
cold methanol. Main the vacuum for an additional 10 min to partially dry the
crystalline product. Then place the material on a clay plate, or on filter paper, and
allow it to dry thoroughly.

The purity of the product may be checked using thin-layer chromatography.
Dissolve a small amount of the starting alhehyde and the product in ethanol and
apply a sample to the TLC plate. Use toluene as the elution solvent, silica gel
(containing a fluorescent indicator) as the stationary phase, and UV light for
visualization. Typical R
f
values are 0.92 for trans-9-(2-phenylethenyl)anthracene
and 0.50 for 9-anthraldehyde, 0.50.

Weigh the product and calculate the percent yield. A portion of the material may
be further purified by recrystallization from 2-propanol using the Craig tube.

Determine the melting point and compare it with the literature value. Obtain the IR
spectrum of the alkene product. The material should have a strong absorption band
at 962 cm
-1
, which confirms the presence of a trans double bond.

This material has a characteristic UV spectrum in methanol. Obtain a UV
spectrum at a concentration in the region of 3 x 10
-6
M. Be sure to record the final
concentration of your solution and calculate the molar absorptivity
max
for
max
at
255 and 383 nm of the UV spectrum. Include the values in your report.




Experiment 3 Exp 3-5

Questions:

1. Complete each of the following reactions by giving a suitable structure for
the species represented by the letters. Is the ylide generated in (a) stabilized by
resonance? If your answer is YES, write the resonance canonical forms. Give a
suitable name for compound B in each reaction.


2. Show how you would synthesize the products B of (a) and (b) with the
Horner-Wadsworth-Emmons reaction.




Experiment 3 Exp 3-6

Experiment 3: PART B. Molecular Modeling and UV
Spectrum Simulation of Trans-9-(2-
phenylethenyl)anthracene with
HYPERCHEM

The purpose of this portion of the lab is to simulate ultraviolet (UV) spectra of
the trans-9-(2-phenylethenyl)anthracene. To record an ultraviolet (UV) spectrum
experimentally, a sample in solution is irradiated with electromagnetic radiation
(UV light in this case) and the wavelengths that are absorbed or transmitted are
recorded as a spectrum. UV spectra are obtained when UV photons are absorbed
and excite a molecule from its ground state to an excited state; an electron is
excited from one occupied orbital to an unoccupied virtual orbital as in the case
of a * or * excitation. Since ultraviolet spectroscopy is an electronic
phenomenon, it is necessary to use quantum mechanical (QM) calculations to
model the electronic structures of the molecules. Due to the computational
restraints imposed by the relatively large systems studied in this case, a semi-
empirical QM method rather than a more rigorous ab initio QM calculation is used
to calculate/simulate the UV spectra.
While the equilibrium optimized geometry of trans-9-(2-
phenylethenyl)anthracene can be obtained with molecular mechanics (MM)
calculations, the computers in ABB and BSB are sufficiently powerful so the
optimized geometry can be obtained with a semi-empirical QM method like PM3.
The electronic structure of the excited states will be obtained by starting with a
Restricted Hartree-Fock (RHF) wavefunction and applying the Singly excited
Configuration Interaction Method (RCIS). This means that in the ground state the
occupied molecular orbitals (MOs) are filled with paired electrons (RHF) and the
excited states are factored in by allowing single-electron excitations (CIS). The
ZINDO/S method that is parameterized to give UV spectra will be used with a so-
called mininal basis set (1s, 2s, 2p
x
, 2p
y
and 2p
z
) of orbitals to calculate the MOs of
the two states that are required to simulate the UV spectrum. So the terminology
that describes this method appears as, RCIS/ZINDO/S//PM3; the single point
method appears first followed by a separator (//) and the method used to obtain the
optimize the geometry.
HyperChem has a user-friendly graphical-user interface (GUI). A number
of the toolbar buttons allow you to draw the structure and use the Model Builder
to obtain a suitable input geometry for the PM3 optimization. Remember
computational methods should not replace your chemical intuition, only
Experiment 3 Exp 3-7

complement it. Does the structure look good? Does it make sense that your
structure has a planar or non-planar geometry? When you are satisfied that you
have a good optimized geometry (use a RMS gradient setting of 0.05 for your
optimizations in this case) carry out the single-point ZINDO/S calculation for
each isomeric butadiene to simulate the UV spectra. Remember to save your work
(optimized geometries) on a diskette as you go on!! Youll save time should your
computer crash and a reboot is required or if you dont complete the modeling in
one sitting. Well start this tutorial by using HYPERCHEM to model the
two conformations of cyclohexane and then move on to trans-9-(2-
phenylethenyl)anthracene.

INTRODUCTION TO MOLECULAR MODELLING
Computational chemistry is a new discipline whose advent and popularity
have paralleled improvements in computing power (now calculations on fairly
large molecules can be carried out on PCs) and software over the last decade. As
with other disciplines in chemistry, computational chemistry involves the use of
techniques and tools to gain understanding of chemical reactions, to enhance
chemical knowledge, and to make predictions about the equilibrium geometrical
structures of molecules and their reactions. The challenges of computational
chemistry are to characterise and predict the structure and stability of chemical
systems, to estimate the energy difference between different states, and to explain
reaction pathways and mechanisms at the atomic level. Ideally it is desirable for
users to understand how the techniques and tools are implemented and the nature
of the databases used to parameterise the methods.
Most modelling/computational packages provide for:

(a) building and displaying molecules and reading structural parameters such as
bond lengths, bond angles and torsional angles - the so called internal
coordinates of molecules
(b) obtaining optimised equilibrium geometrical structures
(c) generating and viewing molecular orbitals and electronic properties such as
the electrostatic potential and the total spin density of a radical - a species
which has an unpaired electron
(d) calculating the geometrical structures and energies of transition structures
(states) of chemical reactions
(c) calculating and displaying spectral properties (UV and IR) and displaying
the vibrations of molecules

Experiment 3 Exp 3-8

In general, most high-level computational chemistry packages allow users to model
molecules and chemical reactions at the three levels of theory.

Molecular Mechanics (Force Field) Methods
Molecular mechanical (MM) force fields use the equations of classical
mechanics to describe the potential energy surfaces and physical properties of
molecules. A molecule is described as a collection of atoms whose interactions
can be described by simple empirical analytical functions which require specific
parameters which depend on the types of functional groups present in a molecule.
Consequently, reliable molecular mechanics calculations can only be carried out if
the parameters for a specific connections of atoms (functional groups) are available
in the parameter database. The collection of 5 or 6 analytical functions describes a
force field. One component of a force field is the energy arising from compression
and stretching of a bond This component is often approximated as a harmonic
oscillator (the bond between atoms is analogous to a spring connecting two
particles) and is calculated with Hooke's Law where K
r
is the force constant for the
spring and r and r
o
are the displaced and equilibrium

V
spring
= 1/2 K
r
(r - r
o
)
2

distances between the particles (atoms). PCMODEL uses harmonic functions to
calculate

V
stretch
=
bond
K
r
(r - r
o
)
2
; V
bend
=
angle
K

( -
o
)
2

potentials for bonds and bond angles. K
r
and K

are the parameters for different

types of groups included in the PCMODEL molecular mechanics parameter files.
Because a molecular mechanics calculation is fully parameterized (the calculation
can only be carried out when the parameters are available), it is known as an
empirical computational method. Because of the simple nature of the analytical
functions and the parameterization, MM calculations are fast even for large
molecules. This is the reason why they are used in calculations on large
biomolecules such as proteins and polysaccharides. AMBER and CHARMM are
the force fields currently used in calculations of this type.
Experiment 3 Exp 3-9

Ab initio and Semi-empirical Quantum Mechanics Methods
Electrons, which are more than three orders of magnitude lighter than nuclei,
are too small for classical mechanics calculations. Electronic energy must be
computed by solving the quantum mechanical Schrodinger equation. The lowest
energy solution to this equation is called the ground-state wave function.
Molecular orbital (MO) methods provide approximate solutions of the Schrodinger
equation, the groundstate wave function, and certain low energy excited states.
Quantum mechanics methods have several attractive features: (a) unlike MM
calculations, they require no information about the location or geometry of bonds
in a molecular system and (b) parameters for elements known as basis functions in
ab initio (ab initio means from scratch) methods are independent of the chemical
environment. In contrast, parameters used in molecular mechanics methods often
depend on the chemical environment, that is the nature of the functional group(s).
Computational packages like HyperChem and GAUSSIAN 98 can perform
quantum mechanics MO calculations on molecules containing 100 or more atoms.
While there is no real restriction on the number of atoms, larger structures studied
at high levels of theory require excessive computing times and computer main
memory. Even though calculations of this type are usually run on parallel
supercomputers such as the Cray, workstations based on the current Pentium and
RISC processors now have the computing power to carry out full ab initio
calculations on relatively large molecules, even trans-9-(2-
phenylethenyl)anthracene.

MOLECULAR MODELLING WITH HYPERCHEM
Chemists use molecular models (plastic or metal) in an attempt to get a 3-
dimensional image of molecules. A 3-D image is extremely important as it aids in
understanding the reactivity and reaction mechanisms of molecules. It also allows
one to predict, by changing (twisting) dihedral angles, which conformers are likely
to be the most stable. A number of difficulties are encountered in using molecular
models: i) models may give incorrect predictions of stable conformations and inter-
atomic distances because they use 'generic' bond lengths and angles; ii) no
quantitative information about the relative stabilities of different structures or
conformations can be obtained and transition states cannot be located; (iii) no
information is gained about electronic structure and bonding. Another less
obvious difficulty is deciding what to do about big molecules like proteins, for
gravity tends to break the partial structure apart before the molecule can be
constructed.
Experiment 3 Exp 3-10

In this tutorial you will use the modelling package HYPERCHEM to
calculate the relative energies of a) the chair and twist-boat conformations of
cyclohexane and b) calculate the UV spectrum of trans-9-(2-
phenylethenyl)anthracene.

PROCEDURE
The program is available through the network. The procedure below works
on the computers of the cluster in ABB 417. You may run HYPERCHEM at any
time when the room is free and use it in other courses. It is HIGHLY
RECOMMENDED that you bring a diskette so you can save your optimised
geometrical structures along the way to avoid having to go back and start from
scratch if you can't finish your work in one sitting or the computer crashes.

Getting Started

1) Re-boot the computer or turn it on and wait for the login window to appear,
then type in your McMaster ID and your password (the same as you e-mail
passwd). Leave the last entry unchanged and wait for the login to be completed.

2) Double-click (from this point on I will use the acronym DC for Double-
Click) on the Course Folders icon. This will take you to a new window where you
DC on the Chemistry icon. When you next DC on the 2BA3/2BB3 icon you will
move to the window from which you will run HYPERCHEM by DCing on the icon.

PART 1: Two Conformers of Cyclohexane

In HYPERCHEM an activity can be selected by clicking on one of the
drag-down menus (DDM) at the top of the input window. This hand-out
shows the File, Edit, Build, Select, Display, Setup, and Compute DDMs.
Your TA will explain the meaning of some of the important highlighted and
non-highlighted selections. There are also a number of tools that you must
use to draw and manipulate your structure on the screen. They are
identified/labelled on the hand out sheet and your TA will demonstrate how
they are to be used. The draw tool is used to generate a crude structure. To
say that is a crude structure is an understatement. You will see that that all
the carbon atoms of the cyclohexane (hexagon) ring that you
Experiment 3 Exp 3-11

generate initially lie in the plane of the screen ( the XY plane). This is the case for
all structures you draw. The first order of business is to left-click (L-CLK) on the
Build DDM, position the cursor on Default Element and release the button to
ascertain whether carbon is the default element. When you L-CLK on Properties a
list of useful properties of the element are displayed. Exit the element table and
draw the planar carbon framework of cyclohexane by dragging the cursor around
and releasing the left mouse button at each point where you want an atom try to
draw a symmetrical hexagon. To label the atoms, L-CLK on Display, select Labels
and tick Symbol. To generate a double bond, move the draw cursor between two
atoms and L-CLK; for a triple bond L-CLK twice. To delete a bond, position the
draw cursor between two atoms and right-click (R-CLK). To change the default
element, go to Build/Default Element and CLK on the oxygen symbol, go back to
the input screen, and L-CLK on one of the carbon atoms. You have effectively
converted cyclohexane to oxacyclohexane. Note the change in colour as the change
is made. Now go back to the cyclohexane. By selecting the x,y,z rotation tool,
positioning its cursor anywhere on the input screen, and moving the mouse you can
rotate the THINGY you thought you generated in 3-D and confirm that it is planar.
To dramatically improve this structure, use the MODEL BUILDER. Go to the
Build DDM and select Add H & Model Build and voila you have generated the
chair conformation of cyclohexane. CONGRATULATIONS!!!! However, it is not
an equilibrium optimised geometry because standard geometrical parameters of a
database were used to generate an improved molecular geometry on the basis of the
valency of the atoms of your initial structure. To optimize the geometry, go to
the Setup DDM, L-CLK on Semi-empirical, and release the button. This takes you
to the DDM that allows you to select the quantum mechanical method that is used
for the optimization. In this case use PM3 because it has been shown that this
method statistically yields the best heats of formation for organic molecules. Now
L-CLK on Options to access the DDM that allows you to set the charge and
multiplicity of your molecule. For cyclohexane the charge is 0 and the multiplicity
is 1 (this means the molecule is in its singlet state with all the electrons paired). L-
CLK on OK to move back to the input screen. Now L-CLK on Compute and select
Geometry Optimization. This DDM allows you to select an optimizer the
algorithim which walks the structure over the molecular potential energy (PE)
surface to locate a minimum. Select Polak-Ribiere that is the default optimizer.
Also select the appropriate RMS gradient (in this case use 0.1) which is the
threshold that is used to determine at what point a geometry is considered to be
optimized and to halt the calculation. For high quality data (for publications) one
would use a RMSG of 0.001. You may wish to play around with
Experiment 3 Exp 3-12

the value of the RMSG. When you L-CLK on OK the calculation will start and a
running dialogue is displayed at the bottom of the input screen. When the RMSG
drops below your selection on a consistent basis, the calculation stops. The terms N
cycles and N2 points mean that N steps were taken on the PE surface with the
energies of N2 points (geometries) being calculated in order to locate the
minimum. The calculation was fast because the model builder provided you with a
reasonably good starting geometry. Whether the calculation converges and how
fast it converges depends on how crude the starting geometry is and what
computational method is used in the calculation. It is clear that a good knowledge
of structure and stereochemistry of organic (and inorganic) molecules is crucial in
recognising whether a starting geometry is acceptable and when the results of a
calculation are meaningful. Now save the geometry of your chair cyclohexane to
your diskette by L-CLKing on File and selecting Save As. Type in the correct path
to the diskette drive along with a suitably descriptive file name and save the
geometry in the MOPAC Z-MATRIX format. You can read the optimized
geometry from the diskette if the need arises. Now reset the path back to the C:\
drive. To obtain the heat of formation for the chair and write the information to
your diskette, L-CLK on File and select Start Log. Type in a filename, set the
Quantum Print Level to 1, and L-CLK on OK. Now L-CLK on Compute and select
Single Point. The calculation starts and data are written to your diskette. Be sure to
use Stop Log otherwise the results of all your following HYPERCHEM
calculations will be written to your diskette. Note that HYPERCHEM assigns
the 3D3 point group to the chair cyclohexane. Is this correct?

Now you will use the selection tool to select a plane through which you
will reflect a methylene group to generate a the boat conformation. L-CLK on this
tool then L-CLK on the four carbon atoms of the seat of the chair. Pull down the
Select menu, go to Name Selection and tick PLANE. Now go back to the input
screen. Now go back to the input screen and L-CLK on the carbon and the
hydrogens of the methylene group you wish to flip (reflect through the plane).
Voila you have generated the boat form. Deselect the atoms (see above) and carry
out a geometry optimisation with PM3. Note that it remains in the boat form.
Occasionally a molecule gets stuck in a symmetrical form when one starts from a
symmetrical input geometry even though the symmetrical species is not a
minimum. Your TA will explain why this is so. To nudge the boat into the twist-
Experiment 3 Exp 3-13

boat, delete one of the C-C bonds, change the torsional angle (you cant alter a
torsional angle in HYPERCHEM when it is in a ring), connect the atoms (although
you dont really have to do so; WHY?) and re-optimise the geometry. You should
end up with the twist-boat if you tweaked the torsional angle sufficiently. Now
open a log file and run a single point calculation to obtain the heat of formation.
Compare the values for the chair, the boat, and the twist-boat. To obtain ball and
cylinder models, go to the Display DDM, select Rendering and tick on Balls and
cylinders. You may also wish to view the overlapping sphere models as well.
These space-filling pictures are based on the van der Waals radii of the atoms.
To change the input window colour to white, go to the File DDM and select
Preferences. Change the Window colour from Black to White. This reduces the
amount of ink/toner that is used when you obtain plots.

PART 2: The UV Spectrum of trans-9-(2-phenylethenyl)anthracene.

Now use HYPERCHEM to generate and optimise trans-9-(2-
phenylethenyl)anthracene and calculate its UV spectrum. Some helpful hints on
generating the trans-9-(2-phenylethenyl)anthracene follow. Generate and optimize
anthracene. Construct/attach the trans-2-phenylethenyl group at C9 of your
anthracene. It is advantageous to use the draw tool (draw the benzene ring as
accurately as possible) with carbon and add the hydrogens manually or use add
hydrogens of the builder. DONT USE add hydrogens and model build after
you draw/attach the trans-2-phenylethenyl group at C9 because the model builder
overlaps the phenyl groups and makes the optimization difficult and time
consumingwho knows why!!!! It may be advantageous to twist the 2-
phenylethenyl group out of planarity with the anthracene ring before you attempt
the final geometry optimization. The optimization will take less CPU time.

When you have obtained an optimised geometry go to Setup, select Semi-
empirical, choose the ZINDO/S method and select options. Your alkene trans-9-
(2-phenylethenyl)anthracene that is neutral and has a singlet ground state. Set the
total charge to 0 and the spin multiplicity to 1. Under spin pairing select RHF
since the ground state will have all the spins paired in the first approximation.
Under state select lowest since excitations from the ground state will be examined.
Click on Configuration Interaction and select singly excited. Now choose
Orbital Criterion. This option selects how many MOs will be involved in the CI
(Configuration Intereaction) expansion. Select 3 occupied and 3
Experiment 3 Exp 3-14

unoccupied orbitals. This means the calculation will consider all the possible
single-electron excitations from the 3 highest occupied molecular orbitals to the 3
lowest unoccupied orbitals. This is an important part of the calculation and is
reflected in the name RCIS(6,6)/ZINDO/S//PM3. The (6,6) stands for the number
of electrons and orbitals included in the expansion, respectively. Now click on OK
for all three boxes you have opened. Start/open a Log file and under Compute
select single point; the electronic configuration will be optimized (the MOs
calculated) fairly quickly. BE SURE TO STOP THE LOG FILE. ALSO
REMEMBER TO TOGGLE OFF THE CI CALCULATION IF YOU GO ON TO
ANOTHER OPTIMIZATION.

Now lets examine the simulated UV spectra. Under Compute select
Electronic Spectrum. The simulated spectrum will appear. The lines along the
bottom of the spectrum correspond to the wavelength in nm. Click on the two
most intense band between 200 and 400 nm and record the values. Why do they
appear as single lines instead of the broad bands observed experimentally? Click
on the copy button to copy make sure the background colour is white the
spectrum to the clipboard. The simulated spectrum can now be pasted into WORD
(it has to be opened first) for inclusion into your report. Include a copy(figure) of
your optimized geometry in your report.
Be sure to compare the
max
values calculated by ZINDO/S with the experimental
values for trans-9-(2-phenylethenyl)anthracene in your report.

Obtain plots of the HOMO and the LUMO of trans-9-(2-
phenylethenyl)anthracene To accomplish this, go to the Compute DDM and select
Orbitals. On the screen you will see the colour-coded energy levels of the occupied
and unoccupied MOs. Use the cursor to select a MO (or use the dialogue box) for
viewing. Go to Options to set the Grid Mesh Size to Fine, the Orbital
contour value to 0.03 and Rendering to Gouraud shaded surface. Over what ring
system(s) is the HOMO delocalised? Over what ring system(s) is the LUMO
delocalised?

To get a picture of the total electron density of trans-9-(2-
phenylethenyl)anthracene go to the Compute DDM and select Plot Molecular
Properties and select Total Charge Density. Make sure to the set the Grid Mesh
Size to Fine and the Total Charge Density contour value to 0.01 in Isosurface
Rendering. This contour value includes 99% of the electrons of the molecule.

HAVE FUN !!!!
Experiment 4 Exp 4-1

Experiment 4. Preparation of 2,4,5-triphenylimidazole
(lophine) and Conversion into its Piezo-
chromic and Photochromic Dimers

The preparation of 2,4,5-triphenylimidazole is simply carried out by refluxing a
solution of benzyl, benzaldehyde and ammonium acetate in ethanoic acid to give
the condensation product. The oxidative dimerization has been described using
either sodium hypochlorite or potassium ferricyanide and use of the latter oxidant
is described here.







O Ph
Ph O
KOH, K
3
Fe(CN)
6
aqueous EtOH
PhCHO, NH
4
OAc
AcOH, reflux
Ph
Ph
N
N
Ph
H Ph
N
N
Ph
Ph
N
N
Ph
Ph
Ph



Materials
1. Preparation of 2,4,5-triphenylimidazole
benzaldehyde (FW 106.1) 2.65g, 2.5mL (25mmol) toxic
benzil (FW 210.2) 5.25g (25mmol) irritant
ammonium ethanoate (FW 77.1) ca.10g (130mmol)
glacial ethanoic acid (100%) CARE !!, corrosive
ethanol (95%) flammable, toxic
ammonium hydroxide (ca. 30% NH3) corrosive, irritant

2. Preparation of piezochromic and photochromic dimmers
2,4,5-triphenylimidazole
(from part 1) (FW 296.4) 1.5g (5mmol)
potassium ferricyanide 4.5g
potassium hydroxide pellets 12.0g corrosive,
hygroscopic
ethanol (95%) flammable, toxic
light petroleum (bp 4060C) flammable
toluene flammable

Experiment 4 Exp 4-2

Procedure:

CARRY OUT THE REACTIONS AT ONE-HALF SCALE.

1. Preparation of 2,4,5-triphenylimidazole
Dissolve the benzyl, benzaldehyde and ammonium ethanoate in 100 mL glacial
ethanoic acid in a 250 mL round-bottomed flask containing a magnetic stirrer bar,
and heat the mixture to reflux for 1h with stirring (the solution must be stirred).
After this time, cool the mixture to room temperature and filter to remove any
precipitate which may be present. Add 300 mL of water to the reaction filtrate and




If the precipiate is gooey, dont do the filtration and simply decant the solution,
add the water to the decanted solution, and proceed with the neutalization.
collect the precipitate by filtration with suction. Neutralize the filtrate with
ammonium hydroxide and collect the second crop of solid Combine the two crops
of precipitate and recrystallize from aqueous ethanol. Record the yield and mp of
your purified material.

2. Preparation of piezochromic and photochromic dimers
Dissolve the potassium hydroxide in 100 mL of 95% ethanol with heating and to
this add 1.5 g of 2,4,5-triphenylimidazole. When fully dissolved add the mixture to
a 1 L beaker containing a large magnetic stirrer bar and cool to 5C in an ice-water
bath with stirring. Prepare a solution of the potassium ferricyanide in 450 mL of
water and place this in an addition funnel clamped securely over the beaker. Add
the potassium ferricyanide solution to the vigorously stirred ethanolic mixture at
such a rate that the temperature of the reaction does not exceed 10C. (This should
require about 1 h.) Near the end of addition, the rate can be increased. In the initial
stages of the addition, a violet colouration develops to be replaced by a lightly-
coloured precipitate. After addition is complete, filter off the precipitate with
suction, washing thoroughly with water (5 x 50 mL). Dry the precipitate on the
filter by rapid washing with cold 50% ethanol (10 mL) followed by light petroleum
(2 x 25 mL). Record the yield, appearance and mp of your material and store it in
the dark. Place a small quantity of your product in a mortar, grind it and observe
the colour change. Dissolve 20 mg of the material in 30 mL of toluene, stopper and
store the mixture in the dark. After the solution has become colourless or nearly so,
stand the solution in strong sunlight or irradiate it with a bright tungsten light and
note the effect. Return the solution to the dark and observe it after a day or so.
Interpret the results.
Experiment 4 Exp 4-3

Questions:

1. Propose a mechanism for the formation of 2,4,5-triphenylimidazole.
2. Discuss any chemical or physical method that might be used to demonstrate
that the violet colouration is caused by radical species.
3. Propose a mechanism by which the oxidative coupling of two molecules of
2,4,5-triphenylimidazole occurs.
4. What features of the radical favour its formation? Give other examples of
relatively long-lived radical species.

Further reading
For spurious piezochromism of dianthraquinone, see: A. Schnberg and E. Singer,
Tetrahedron Lett., 1975, 1925.

For the procedures on which this experiment is based, see: D.M. White and J .
Sonnenberg, J. Am. Chem. Soc., 88, 3825 (1966).