David Sommer
Mentors: Prof. John Farley, Ph.D.
Jack Glassman, Ph.D.
Steve Mitchell
Dr. John Farley’s research
group is attempting to achieve high-resolution spectroscopy of the methide
ion – CH3 -. My work in the REU program has been primarily centered
on this project. In this paper I will describe both the general purpose
and nature of the project as well as my specific contributions in the area
of organic chemistry.
Our group is studying the
methide ion primarily because it has not been thoroughly examined experimentally,
and there are several characteristics of the ion which make its study worthwhile.
The methide ion is electronically analogous to the ammonia molecule.
The N-H bond stretching and bending spectra in ammonia are well known.
The difference between the methide ion and ammonia is methide’s net negative
charge, which correlates to increased electron-electron interaction.
The effect of this increased electron density upon the C-H bond characteristics
is not known with certainty. Although theoretical studies of the
ion have been conducted, the results are inconclusive. We hope to
measure these effects directly.
Another important attribute
of the methide ion that is analogous to the ammonia molecule is the central
carbon’s ability to tunnel through the hydrogen plane. This molecular
inversion of configuration or fluxation causes peak splitting in the IR
vibrational spectra. By studying this splitting the group hopes to
learn about the energy barrier crossed when the ion fluxes and the frequency
with which this phenomenon occurs.
Our group is using a coaxial ion beam apparatus (which we have dubbed Big Beam) to perform the spectroscopy. The only previous spectroscopy experiment involving the methide ion was conducted using a crossed beam apparatus. That is a laser beam was crossed over a beam of ions. This technique relied on a sub-millimeter interaction region and a correspondingly short interaction time. Our experiment utilizes a laser beam that coaxially overlaps a beam of ions for an interaction region of approximately half a meter and a corresponding reaction time of several microseconds. I will now describe the structure and function of the machine.
The entire apparatus is maintained at a pressure of about 10-6 torr by two diffusion pumps and an ion pump. The ion source consists of a tungstun filament held in front of a cathode plate. The filament and cathode float at -2000V relative to ground. These are positioned in front of the anode plate which has a hole of about 1mm in diameter through which the ions pass. In the next region of the machine, the ions are accelerated to ground potential and focused. They then pass through a Wien filter allowing mass selection to occur. The mass selection provides a degree of flexibility and control in the experiment because we do not have to physically cleanse the apparatus and sample of impurities (an impossible task). After passing through the Wien filter region, the ions are turned 90 degrees to the right in a quadrapole. This right hand turn allows the laser beam to be introduced coaxially with the ion beam. The laser and ion beams then overlap each other in the interaction region. At the end of the interaction region, the ion beam is defected by an E field. The deflected ions are routed into a Faraday cup which measures the ion current. The neutrals, which have been created in the photodetachment process, continue on a straight path and collide with a metallic plate. The collisions cause secondary electron emission, and these emitted electrons are counted. This measurement of secondary electrons is the only data gathered directly from the machine for the intensity plot of the spectra.
Another feature of the machine,
which it is necessary to discuss at this time, is the control of photon
wavelength through velocity shifting of the ions. The IR source consists
of an f-center laser pumped by an argon laser. Since the IR laser
is not continuously variable in wavelength, we step the laser in increments
of about 300 MHz. The technique of spectroscopy we are using has
a resolution of about 5 MHz. To fill in the gaps in the spectrum,
we alter the acceleration voltage of the ions. This changes the velocity
of the ions in the interaction region. Because the ions “see” a red-shifted
laser, we can control the amount of this Doppler shift by slowing down
or speeding up the ions. We therefore achieve continuous control
of wavelength in the ions’ reference frame even though the control of wavelength
in the lab reference frame is roughly quantized.
The creation of CH3- is not a simple chemical reaction. In fact the ion itself, while physically stable, is extremely chemically unstable – it will not disintegrate if isolated, but reaction with any substance such as air or a container would destroy it. In order to study CH3-, we first had to create the ion. Previous research has indicated that the methide ion is formed when diazomethane – CH2N2 – is fed into the ion source. The exact reaction which produces the ion, and the answer to the question “where does the extra hydrogen come from” remain mysteries. The real purpose of our experiment lies in the physics – spectroscopy. We do not need to understand the chemistry completely. We do, however, need to do chemistry to produce our target ion. Since diazomethane has worked as a parent gas for the methide ion before, this is the gas we use.
Structure of Diazomethane
Diazomethane is a yellow gas at room temperature. It is extremely toxic (OSHA pel of 2 ppm). It is also highly unstable and can explode if exposed to sharp nucleation sites or certain chemical substances such as drying agents. Diazomethane has a boiling point of -23 degrees Celsius, so it can easily be converted into a liquid in the lab. The melting point is just below -100, so it could also be solidified with liquid nitrogen.
Diazomethane can serve as a parent for CH3- because its most important resonance structure is only one step away from dissociation into CH2- and N2+. This same propensity to form molecular nitrogen makes diazomethane a highly unstable compound. Therefore, we create the diazomethane to be used in our experiments in one-gram quantities. We store and use the gas at dry ice temperature -- - 70C. Therefore, the diazomethane is handled almost entirely in the liquid state. This allows us to avoid state changes that could cause explosions. The vapor pressure, however, is sufficient to fuel our ion source.
Originally, the group had
planned to collect the diazomethane in solution with a high boiling point
liquid – 2,2 ethoxyethoxyethanol. We discovered, however, that our
solvent adopted the consistency of glass at our intended temperature of
-70C. One of my first projects was to find a replacement solvent.
The solvent was to have ether linkages (which provide excellent solubility
for the gas) and lack carbonyl functionalities (which could react with
the gas). We failed to find a compound which met these requirements,
was readily available, had a low enough vapor pressure to not compete with
the diazomethane, and had a low enough viscosity to be functional.
We therefore deiced to collect the diazomethane neat as a liquid.
Since diazomethane is not the friendly type of gas that one would want to have sitting around the lab, we only generate enough to perform one day’s worth of spectroscopy at a time. The gas is produced through a simple one step chemical reaction of a proprietary precursor with potassium hydroxide. The precursor we use has the trade name Diazald and is chemically n-methyl, n-nitroso, paratoluenesulfonamide. This compound undergoes an internal rearrangement, a deprotonation, and a dissociation to yield the diazomethane gas. The reaction is described graphically below.
We carry out the reaction of 6g of diazald at 60C over the course of
about 15 min.
Diazomethane has been studied
and used in the world of organic chemistry for quite some time because
of its effectiveness as a methylating reagent for carboxylic acids.
This property of the gas provides us with a destruction mechanism – acetic
acid. When acetic acid comes into contact with diazomethane, the
acid is rapidly methylated yielding nitrogen gas and methyl acetate (which
is relatively harmless).
In this section, I will describe the physical apparatus we use to make the diazomethane.
We use helium as a drive
gas – a sort of molecular conveyor belt. The entire system is sealed
except for the exhaust that leaves the acetic acid bubbler. The tube
that connects the delivery funnel to the three-neck flask serves to maintain
equal pressure throughout the glassware. The He flows through the
system at a rate which produces about one bubble per second in the acetic
acid. During the reaction, the bubble rate increases slightly due
to the pressure produced from the uncondensed diazomethane.
The reaction occurs in the
three-neck flask. The flask is maintained at a temperature of 60C
with a water bath. About 6g of diazald is dissolved in about 20 mL
of decahydronapthalene with a magnetic stir bar. In the funnel above
the flask, about 2 g of KOH is dissolved in 20mL of isopropanol.
The isopropanol is dripped slowly into the diazald soln. As the gas
is formed, it travels up a tube which is cooled with a water jacket.
It then passes into a condenser which is suspended in a dry ice / isopropanol
bath. The majority of the diazomethane condenses at this point.
Any diazomethane remaining is bubbled through acetic acid and hence neutralized.
The steel jug containing the dry ice bath and the diazomethane can be closed
off with the entrance and exit valves. This unit (jug, condenser,
and valves) can be disconnected from the rest of the glassware and connected
to the ion beam.
The goal of Dr. Farley’s
group is to obtain a high-resolution spectrum for the methide ion.
This goal has not yet been achieved. However, we have made much progress
in this project during the weeks I have been working with the group.
In this time, we have had our first successful synthesis of diazomethane,
and we have made the gas on over fifteen subsequent occasions. We
have obtained a beam of anions which we believe to be methide based on
the mass spectrum (although we have no conclusive evidence). We have
not yet, however, obtained a signal produced by photodetachment with the
IR laser.
In terms of a physics experiment, the methide project is just
beginning. But the physics project could not have begun without preliminary
work in chemistry and the design of chemical apparatus and procedures.
In this area, I feel I have been able to make a worthwhile contribution
to the group. At the same time, the group and the project itself
have made very worthwhile contributions to my own science education.