Our research is primarily focused on physical organic Chemistry and its various subfields.

Typically, all research in our lab can be roughly divided into four steps. Firstly, based on our
previous experience we propose a molecular/polymeric structure that we feel is suited towards
a practical application. Secondly, we perform DFT calculations for that structure so that we can
predict its properties. If DFT calculations show promise, only then we start the actual synthesis
towards that molecular/polymeric system. Lastly, we carry out physical/chemical
characterizations and compare experimental results with DFT calculated ones. In essence, our
group is aiming towards developing basic scientific methodologies/concepts for accurately
predicting bulk/molecular properties from a simple chemical structure. It is theory guided
research efforts towards development of new functional polymers. DFT calculations are proving
extremely critical for our research without which we may end up wasting time and resources
synthesizing systems that eventually prove useless.

While our group has expertise in other areas of materials/organic/physical Chemistry, our most
recent research is aimed towards the development of electroluminescent polymers for light-
emitting field-effect transistors and blue OLEDs. While the first OLETs were developed as early
as 2003, this field experienced only modest development primarily due to limited availability of
efficient as well as stable materials. Efficient OLET materials require high and balanced
electron/hole mobility, high emission quantum yield, suitable optical bandgap, as well as an
efficient light-emission zone in the transistor channel. We propose the following guidelines in
designing our electroluminescent polymers: Firstly, our polymers, like most fluorescent
compounds, are conjugated systems. Secondly, we intend to avoid heavy atoms which may lead
to diminished luminescence quantum yield. Thirdly, large HOMO/LUMO orbital overlap favors
high luminescence quantum yields. We intend to employ DFT calculations as our primary
theoretical model to examine new polymer structures and the degree of orbital overlap.

Technical Approach and Anticipated Results

Our preliminary results have led to certain conclusions: First, not all luminescent polymers are
suited for application in OLET devices. We have tested more than ten different polymers with
strong luminescence in solution (ranging from 17 to 79%), all of which showed poor
performances. Among these, the one containing a coiled polymer backbone is the only one which
shows promise. We think that coiled structures make pi-pi stacking difficult and by extension
improve the solid-state emission. We intend to employ side-chain engineering to help to address
this issue. Secondly, a strong donor-acceptor system is detrimental to emission quantum yield
due to internal charge polarization, in which the HOMO and LUMO orbitals are not spatially
overlapped. Thirdly, polymer structures capable of further cyclization (Pdph in Figure 1) can lead
to a quick decrease in luminescent emission. Based on our previous work and the rationale
outlined above, we propose to develop strategies towards developing ambipolar OLET polymers.
We intend to perform DFT calculations to estimate the energy levels which will allow us to
evaluate the energy level match with the metal electrodes. At the same time, the HOMO/LUMO
orbital distribution will help reveal their
spatial overlap. Following this, synthetic
strategies will be proposed, and the
polymers synthesized. We also intend to
deliberately synthesize several polymer
structures with less than optimal orbital
overlap so that a comparison can be made
with those having a good overlap. Polymer
structures will be designed such that only
C, H, O and N atoms are incorporated
while the incorporation of sulfur atoms
will be minimized. After the polymers have
been synthesized, all necessary
characterizations and device studies will Figure 1. Proposed polymers and their
help develop structure/property synthesis.
relationship.

Proposed polymer structures and DFT calculation results

Figure 1 shows several structures of polymers that have been designed based on the above
rationale. To achieve ambipolar charge transport, we will need to adjust the HOMO/LUMO
energy levels. Especially when using gold as an electrode, the polymer energy levels need to be
lowered significantly. Towards this end, we introduced electron-deficient groups as side chains
attached to the electron-rich backbone. To gain insight into the electronic structure, DFT
calculations were performed with three repeating units. The results showed that these polymers,
apart from Poxdz, exhibit HOMO/LUMO energy levels that were too high when employing a gold
electrode (work function~5.1 eV). They also exhibit a minimal overlap in HOMO/LUMO molecular
orbitals with an energy gap of around 3.5 eV. Of these, polymer Poxdz seems to be interesting
because of a decent orbital overlap and a lower HOMO/LUMO energy level. As such, we chose
this polymer as the starting point for our research. This particular polymer system was recently
synthesized and exhibited blue emission with a solution fluorescence quantum yield of 57%.
Unfortunately, DFT results indicated that the polymer HOMO/LUMO energy levels are not ideal
for ambipolar charge transport unless Al electrodes are used which have a work function of
around 4.2 eV.
 Table 1. DFT calculation results.
Polymer HOMO LUMO
P2pym

P2oxz

Poxdz

P2oxdz

P3-ox-daf

P3-oxn-daf

Table 2. DFT results on energy levels and gaps.

In the view of our requirements we are requesting allocation of 1000,000 service units for the
coming year. Our molecules are often large and by extension their require large number of
service units. We need not only evaluate energy levels, but also conformations of long polymer
chains, which will affect the properties of materials. Based on our Research I allocation, we
have estimated that DFT calculation of one single molecule can consume as much as 5000
service units. We can easily perform 200 DFT calculations throughout the course of one year.
This would require the allocation of 1000,000 service units. These calculations will play
important role, not only in gaining insight of materials, but also publishable results. They will
also make our grant proposals more appealing.