index.html

---
layout: default
title: Hello
---
<div class="blurb">

	<h1>Projects</h1>
	<h2>Prediction of reorganization energy using machine learning</h2>
	<a href="https://pubs.acs.org/doi/abs/10.1021/acs.jpca.9b02733">Publication link</a>
	<p><img src="./assets/fig1_jpca.png" alt="image" width="600" /></p>
	<p>
		Organic semiconductors are an important class of (opto)electronic materials,
		with a wide range of potential applications from photovoltaics to artificial nerves.
		The goal of this project was to enable large-scale screening for high-performance
		organic semiconductors by rapid prediction of reorganization energy (RE)
		using machine-learning methods. It was featured on the <a
			href="https://pubs.acs.org/toc/jpcafh/123/36">cover</a>
		of ACS magazine JPCA.</p>

	<p> The project involved creation of a dataset using a SMILES and SMARTS
		based combinatorial molecule generation (a),
		calculation of the reorganization energy with DFT methods (b),
		and prediction of the reorganization energy with Ridge Regression, Kernel Ridge Regression and
		Deep Neural Networkds (c). (lambda represents the target reorganization energy value.)</p>

	<p>We found that deep neural networks outperform the other methods and can predict the RE with a
		coefficient of determination of 0.92 and root-mean-square error of ∼12 meV.
		This study showed that the REs of organic semiconductor molecules can be predicted
		from the molecular structures with high accuracy.</p>
</div>

<div class="blurb">

	<h2>An ETL, analysis and visualization project with credit card data</h2>
	<p><img src="./assets/capstone_overview.jpeg" alt="image" width="400" /></p>
	<p>
		I recently finished a data engineering bootcamp where
		I created an ETL pipeline and application front-end with
		Python as the <a href="https://github.com/saevrenk/dataengineering_capstone">capstone project</a>.

	<p> It was a fun project bringing together tools such as Apache Spark, Pandas, MySQL, Plotly and Tableau.</p>
</div>

<div class="blurb">

	<h2>Theoretical Characterization of Charge Transport in Zinc-Chlorin Nanotubes</h2>
	<p><img src="./assets/znp_nanotube_md.png" alt="image" width="600" /></p>
	<p>
		Some photosynthetic bacteria have nano-sized antennas called chlorosomes that
		capture light efficiently. In this project, I investigate biomimetic zinc-porphyrin supramolecular
		structures inspired by chlorosomes through atomistic simulations (GROMACS),
		and perform quantum chemical calculations to understand charge transport properties. These studies aim to
		understand
		the structure-function
		relationships and contribute to the development of new, green, and sustainable solar energy systems.
		Figure shows the construction and MD studies of the nanotubes.
	</p>
</div>

<div class="blurb">

	<h2>Structure and Binding Energies of Dimers of Zn(II)-Porphyrin Derivatives</h2>
	<a href="https://pubs.acs.org/doi/10.1021/acs.jpca.2c03692">Publication link</a>
	<p><img src="./assets/znp_jp2c03692_0003.jpeg" alt="image" width="600" /></p>
	<p><img src="./assets/znp_jp2c03692_0006.jpeg" alt="image" width="600" /></p>
	<p>
		Zinc-complexed porphyrin and chlorophyll derivatives form functional aggregates
		with remarkable photophysical and optoelectronic properties. Understanding the
		type and strength of intermolecular interactions between these molecules is
		essential for designing new materials with desired morphology and functionality. In this project
		I did a systematic study of the interaction energy (IE) in zinc-porphyrin complexes with
		increasing structural complexity so that the effect of substitutions on the dimer IEs is quantified.
		I used semiempirical, density functional, and symmetry-adapted perturbation methods to find a
		good balance of cost and accuracy (top). Moreover, the types of intermolecular interactions are
		evaluated using energy decomposition analysis based on the symmetry-adapted perturbation theory (bottom).
	</p>
</div>

<div class="blurb">
	<h2>A Quantitative structure–property study of reorganization energy</h2>
	<a href="https://pubs.rsc.org/en/content/articlelanding/2018/RA/C8RA07866A">Publication link</a>
	<p><img src="./assets/fig6_rscadvances.gif" alt="image" width="400" /></p>
	<p>
		I curated acompound set of 171, which was derived from known p-type OSCs built from moieties such as
		acenes, thiophenes, and pentalenes and studied the structure-property relationships.
		It was highlighted in the <a
			href="https://pubs.rsc.org/en/journals/articlecollectionlanding?sercode=ra&themeid=098ca36a-da87-4a2b-9182-06d87e58c5b6">
			"Celebrating recent achievements in chemical science in Turkiye"</a> themed collection.
	</p>
	<p>Here is a recent <a href="https://github.com/saevrenk/molecular_scs">notebook</a>
		where I explored clustering and
		interactive plotting with molplotly using this dataset.</p>
	<video src="https://user-images.githubusercontent.com/105816821/229763963-0742d91a-6897-4399-a04f-2e89080a1b4e.mov"
		controls="controls" style="max-width: 730px;">
	</video>
</div>