Tanner Jeffrey Bakhshi

Senior/Undergraduate Researcher

Norton Group (Department of Chemistry)

Molecular and Biological Imaging Center (BBSC 107)




Welcome! My name is Tanner Bakhshi, and I am a senior majoring in Molecular Biology and minoring in chemistry and mathematics here at Marshall University. In addition to taking classes during the fall, winter, and spring, I also conduct research part-time in Dr. Michael Norton's laboratory and the Molecular and Biological Imaging Center (MBIC), of which Dr. Norton is the director. During the summer, however, I conduct research on a full-time basis. The primary focus of Dr. Norton's lab is the study of DNA Origami and its practical applications in the field of Nanotechnology. My first summer of resarch in his lab (2012) was funded by the U.S. Army through the High School Apprenticeship Program (HSAP). My projects involved the printing ordered patterns of DNA Origami using a polydimethylsiloxane (PDMS) stamp and studying its behavior during formation in order to design a more time-efficient annealing protocol. This year (2013) I was fortunate enough to be selected for Marshall University's Summer Undergraduate Research Experience (SURE), during which I am continuing my studies of DNA Origami.

What is DNA Origami?

In 2006 Paul W. K. Rothemund, a researcher at the California Institute of Technology, published an article in Nature about something he called “DNA Origami.”1 True to the way that it sounds, DNA Origami involves the folding of single-stranded M13mp18 viral DNA into different patterns and shapes through the use of “staple strands” that tell the viral DNA where and how to fold. The type of origami used in Dr. Norton's lab is called “cross” origami.2 Just like it sounds, the shape is formed by the joining of two linked but individually-formed rectangular DNA structures, the end result’s being what can be described as a cross or plus sign. Depending on how the ends of these separate structures are designed, the crosses formed can be individual, they can stick together in one direction and form long chains, or they can stick together in two directions and form large checkerboard-like arrays. These different configurations are referred to as 0D, 1D, and 2D crosses, respectively. In recent months and years, the interest in DNA nanotechnology and in its potential physical and biotechnological applications have increased exponentially. In fact, many fascinating and groundbreaking innovations have already been accomplished. A laboratory in 2003 invented a nanoscale transistor using a carbon nanotube attached to a DNA scaffold.3 In 2004 another lab created a biped DNA walker capable of motion along a predetermined footpath.4 More recently, DNA nanotechnology in the form of DNA Origami (the topic on which Dr. Michael Norton’s lab primarily focuses) has even been used to deliver anticancer drugs to cancerous cells inducing cell apoptosis.5

Types of cross origami.



AFM scan of 0D cross origami on mica, courtesy of Dr. Masudur Rahman.







SURE Project

The current objective of my SURE research project is finding a reliable and reproducable method of adhering a small, 3-mm circle of mica to 4.4-mm silicon nitride chip. The mica needs to be bound strongly enough that it can be thinned repeatedly without being removed from the silicon nitride chip. Mica is a shiny (refractive) mineral that is slightly flexible. It can be cleaved very easily, its individual layers are atomically thin (1 nanometer), its surface is extremely flat, and it is highly positively charged. Because mica possesses these useful qualities, it is frequently used in laboratories (including Dr. Norton's) that image DNA structures using an atomic force microscope (AFM). The mica acts as a flat, blank canvas to which DNA structures bind (DNA is higly negatively charged) and upon which lay flat. Silicon nitride chips are single silicon crystals of (100) orientation with an amorphous silicon nitride film on top. Silicon nitride (SiN), known for its structural rigidity and resistance to heat, has many industrial applications.

The first method of adhering mica and SiN that I have tried involves a UV-curable optical ahesive. The adhesive remains in liquid form until exposed to UV light; it then cures and hardens.

The second method that I have explored involves bonding mica and SiN with a sort of "molecular glue." The mica is coated with 3-aminopropyltriethoxysilane (APTES) and the SiN is coated with 3-glycidoxypropyltrimethoxysilane (GPS). The two surfaces are then pressed together and heated in an oven for a short duration.6

The third method that I am have worked on is the atomic flattening of a silicon chip with identical dimensions to those of the silicon nitride chips used in previous experiments. The chips are first dipped in dilute hydrofluoric acid (HF) to remove native oxides. The chips are, then, placed inside of a horizontal furnace and heated to 900°C while Argon flows past. The literature on which this experiment was based shows that this environment (under ideal conditions) is sufficient to induce the atomic flattening of silicon.7

Progress Reports

Week 1 (5/20/13-5/24/13)

My first week of the SURE program went smoothly and was very productive. First, I found a reliable way to punch thin, 3.5-mm circles of mica out of larger, thicker, 10-mm circles of mica. These small circles will be adhered to the 4.4 mm x 4.4 mm silicon nitride chips. Second, I found a method by Dow Corning showing how to bond any two silanized surfaces together using a sort of molecular glue. The bond is formed when two compounds, 3-aminopropyltriethoxysilane (APTES) and 3-glycidoxypropyltrimethoxysilane (GPS), come in contact with each other. I currently have APTES in my lab, and I ordered the GPS from Sigma-Aldrich. Finally, I went to Dr. Day’s lab and learned how to mix, mold, and cure polydimethylsiloxane (PDMS). I will use PDMS in later experiments that (as in the APTES/GPS experiment) involve adhering mica to a silicon nitride chip. So far, I have set a nice pace for myself and am aligned with my projected timeline.

Week 2 (5/27/13-5/31/13)

In my second week of the SURE program, I first learned how to perform photolithography on silicon chips. After spin-coating photoresist onto them, a computer-generated pattern was projected onto the surface, solubilizing any photoresist exposed to light. The resulting pattern was then developed in developing solution. I also mixed and cured PDMS for the first time. The mold that I used was simply a small glass slide with sections of 1-mm and 2-mm plastic tubing glued onto it vertically. The result was PDMS with tubular holes that, once punched out with a 3.5-mm circular punch, will serve to deliver optical adhesive to the surface of a silicon nitride chip in a precise circular pattern to match the circular mica. I am still aligned with my projected timeline.

Week 3 (6/3/13-6/7/13)

I started my third week of the SURE program by punching 3.5-mm, circular stamps out of the PDMS that I had cast at the end of the previous week. The result was four small stamps with a 2-mm hole in the middle of them and four small stamps with a 1-mm hole in the middle of them. I also cut out four small, featureless, square stamps to be used in the experiment involving stamping thin mica flakes onto a silicon nitride chip. All of the stamps were cleaned through sonication in toluene and then isopropyl alcohol, each for five minutes. Second, I cleaned silicon nitride chips by sonicating them in ethyl alcohol, acetone, and then exposing them to UV light, each for five minutes. Third, I developed a quicker and more reliable way to produce thin, 3.5-mm circles of mica. As in my previous method, I used a metal punch to make deep circular impressions in the mica. The mica with deep punch marks was peeled once with scotch tape. A clean razor blade was then used to get underneath the circular mica pattern on the tape and cut it away from the tape. The rest of my week was spent developing a way to stamp optical adhesive onto surfaces. One of the stamps with a 1-mm hole was glued to a small piece of glass slide using a fast-setting epoxy. The stamp was pressed onto a Kimwipe saturated with UV-curable optical adhesive and then stamped onto a clean glass cover slip (cleaned using the same cleaning method as was used for silicon nitride chips). A punched mica circle was then laid onto the adhesive and pressed lightly with forceps, followed by a 200-second exposure in UV light. I am still aligned with my projected timeline.

Week 4 (6/10/13-6/17/13)

My fourth week of the SURE program was primarily spent drawing conclusions regarding the use of UV-curable optical adhesive to adhere a thin circle of mica to a silicon nitride chip. Most preliminary experimentation and prototyping was performed on clean, 10-mm x10-mm silicon chips (sonicated in ethyl alcohol and acetone and then UV-treated) as a substitute for silicon nitride chips because they are cheaper and more readily available. After many trials and much imaging with both our atomic force microscope (AFM) and its attached optical microscope, the following conclusions have been determined in regard to the optical adhesive procedure:

  1. Before the liquid adhesive has cured, it has the potential to spread out underneath the mica, filling the 1-mm hole intentionally left by the PDMS stamp. Furthermore, the adhesive occasionally leaks out from underneath the mica.
  2. Once the optical adhesive has cured, the mica stays in place very nicely. After peeling several times with Scotch tape, however, (in an effort to make the mica layer thinner) the mica flake begins to fragment, removing large pieces of mica.
  3. When observing said mica fragments under an optical microscope, differentiation between the mica and the optical adhesive can be very difficult. The same seems to be true for AFM imaging. It appears that the underlying, uneven adhesive layer is affecting the appearance of the mica laying on top of it.
  4. The layer of optical adhesive seems to be relatively thick, which makes any determination of mica thickness or its optical contrast challenging.

Because of these observations, I am going to begin exploring another possible method of adhering mica to a silicon/silicon nitride surface, using a type of molecular glue. The other task that I was able to accomplish in Week 4 was designing the staple sequences required for the formation of a structured, “braided tail” on a newer version of Cross DNA Origami. I am still on track with my projected timeline.

Week 5 (6/17/13-6/21/13)

My fifth week of the SURE Program was spent fully exploring the combination of APTES and GPS as a method to “glue” mica and a silicon nitride chip together on a molecular level. The first step was to aliquot 30-mL amounts of APTES and GPS out of their parent containers in an inert Argon atmosphere. The inert atmosphere was accomplished by the use of a glove bag in which room air was purged and replaced with Argon. The chemicals were then transferred to small glass bottles with self-healing septa. To remove small doses for experimentation, a syringe was filled Argon, inserted into the septum bottle, the gas was pushed into the bottle, and the resulting positive pressure pushed the chemical up into the syringe.

Two sets of experiments were conducted: those using chemicals in the liquid phase and those using chemicals in the vapor phase. Chemical vapor deposition was used in experiments performed in the vapor phase. 200 μL of APTES was mixed with 4 mL of toluene in a glass petri dish and set in a vacuum desiccator. A 3.5-mm circle of mica was placed on aluminum foil and also was set in the vacuum desiccator. A vacuum was then pulled and the system left alone for ninety minutes. The same process was then repeated for the silicon nitride chip and GPS. The two were then pressed together and put in an oven set to 110°C. After two trials this method proved to be unsuccessful at bonding the two surfaces. The experiment was also tried using neat APTES and GPS. While this did help the two surfaces bond together better, the mica was easily removed with a simple peel using Scotch tape.

There were two primary experiments performed in the liquid phase. The first involved depositing 1 μL of APTES on a 3.5-mm mica circle and 1 μL of GPS on a silicon chip, pressing them together, and putting them in an oven set to 110°C for thirty minutes. The other method involved mixing 1 μL of APTES and 1 μL of GPS on a silicon chip and then laying a 3.5-mm mica circle on top of the mixture, followed by the oven step. Though both experiments resulted in the two surfaces bonding together, they were separated after peeling a few times with Scotch tape.

I believe that the reason for the failure of mica and GPS to remain bound to each other is because the mica circles that I am using our so small in diameter. This greatly decreases the amount of surface area that can be covered with chemicals and, therefore, bind to the GPS present on the silicon chip. I will work to try to overcome this challenge. In addition to experimentation, I also began work on my SURE website. I am not quite aligned with my current timeline simply because adhering mica to SiN has taken so long.

Week 6 (6/24/13-6/28/13)

My primary focus during the sixth week of the SURE program was spent putting together an apparatus that will be used in an attempt to flatten a small silicon chip. An argon tank is connected to a regulator and valve, which run the gas through a flow meter and then through a quartz tube inside of a horizontal furnace. The gas then runs out into a mineral oil trap. The vertical furnace is capable of reaching very high temperatures. The silicon chip will placed on a small ceramic “boat” and placed inside of the quartz tube. The furnace will then be heated to 900 °C and then immediately cooled back down to room temperature. According to literature this temperature will be enough to flatten silicon atomically, which can be confirmed by AFM. This flat silicon, should the process work as described in literature, could be the perfect surface on which to place origami. It would be much flatter and smoother than the silicon nitride chips that we have currently.

Week 7 (7/1/13-7/5/13)

In my seventh week of the SURE program I first imaged a brand-new, extremely clean silicon chip with the AFM so that I would have a control to which experimental results could be compared. I also was able to perform the silicon flattening experiment twice. In both trials the silicon chip used was first cleaned with a dilute, 2% HF solution for two minutes (to remove native oxides), followed by thorough rinsing with ultra-pure Millipore water. It was then blown dry with nitrogen gas. At all times while using HF protective gear (goggles, face shield, multiple layers of gloves, and plastic sleeves) was worn and procedures were conducted within a fume hood. I also always made sure that someone was with me in case something were to go wrong. In the first trial the silicon chip was placed in a ceramic boat and pushed to the center of the quartz tube in the furnace with a long glass rod. However, the chip was inadvertently knocked down off of the boat upside down. It was left this way so as not to do any more damage than had already been done. The furnace was then heated to 900 °C and then cooled down immediately to room temperature. During the trial argon gas was flown through the tube at a rate of 5 mL/minute. In the second trial everything remained the same except for two factors. First, the flow rate was increased to 60 mL/minute. Second, the silicon chip was allowed to sit at 900 °C for twenty minutes before cooling down to room temperature. So far, the results of both trials seem to indicate that the experiment was unsuccessful, i.e., the silicon chips do not appear to be atomically flat. Further trials will be conducted. I am still a bit off of my projected timeline but my mentor is aware of this and we are working to continue steering my project in the right direction.

Week 8 (7/8/13-7/12/13)

In my eight week of the SURE program I first conducted a third trial of the HF/argon flattening procedure on a silicon chip. The experimental parameters were identical to those of the second trial except that the furnace was purged with argon for fifteen minutes before the experiment began. I then processed all of the AFM images obtained of my three silicon samples. Results showed that after undergoing HF treatment and argon flattening surface roughness decreased rather significantly. I then put 0D cross origami on the silicon chip from the third trial and imaged it. AFM imaging showed that origami can be resolved very well on the treated silicon, unlike on silicon nitride. It should be noted that, while surface roughness of silicon chips did indeed appear to decrease after the experiment, it is unknown whether the HF treatment, the furnace step, or both, are responsible for the decrease. Further experiments will be performed to resolve this uncertainty. The last experiment that I performed attempted to decrease the surface roughness of a silicon nitride chip in the same manner that silicon chips’ roughness was decreased (using the same experimental parameters as those set for third-trial silicon). AFM imaging revealed that that, unlike the silicon chips, the silicon nitride did not experience any significant decrease in surface roughness after undergoing the treatment.

Week 9 (7/15/13-7/19/13)

In my ninth week of the SURE program I first hydrated all of the anhydrous staple strands that I designed and ordered to create a long, structured “braided” tail for our newest origami construct. The twelve staple strands were each brought to 100 µM. A 0.5-µM working solution of all twelve staple strands was then mixed and the concentrated stock solutions placed in deep-freeze storage. The tail staple strand mix was then mixed with the regular origami components (single-stranded M13, buffer, etc.) and annealed for thirteen hours. AFM imaging shows that the tail construct forms well and as expected, though the concentration may need to be increased so that more origami crosses are affected. Next, I sonicated a Norcada silicon chip for ten minutes in acetone and then for ten minutes in ethanol. AFM imaging will reveal if this simple procedure is enough to decrease the surface roughness (white dots) of the silicon chips. Third, I mixed fresh PDMS, baked it overnight, and cut out a small, square stamp. I used this stamp to transfer small (and, hopefully, thin) mica flakes to a clean silicon surface. Literature reports obtaining atomically thin mica flakes using this method. Transferred flakes will be imaged and their optical contrasts calculated. The two measurements will then be correlated so that the thickness of mica can be obtained simply by calculating its optical contrast. This could eliminate the need for thinning mica after adhering it to silicon or silicon nitride and, also, the risk of removing the adhered mica. Lastly, I helped setup and install a new 3D printer that arrived in Dr. Norton’s lab. We are currently learning all of its capabilities and are developing many useful projects designed to make full use of those capabilities.

Week 10 (7/22/13-7/26/13)

In my tenth and final week of the SURE program, I first ran two small experiments and concluded a third small experiment from Week 9. The first simply involved dipping a Norcada silicon chip in HF as usual but not placing it in the furnace and heating it to 900ºC as usual. The second experiment involved dipping a Norcada silicon nitride chip in HF, placing it in the furnace, and then heating it to 900°C. In the previous trial involving silicon nitride, the chip was not dipped in HF before being heated in the furnace and showed no change in surface roughness. The purpose of the third experiment (that I wrapped up from last week) was to determine if simple sonication of a Norcada silicon chip in acetone and ethanol could decrease its surface rougness, i.e., remove a significant percentage of the “white dots” observed when imaging brand-new Norcada silicon chips. The results of the first experiment showed that, after HF treatment alone, a Norcada silicon chip undergoes the same 0.1-0.15-nm decrease in surface roughness as in trials where the chips were treated with HF and heated in the furnace. In other words, the current furnace conditions are having no effect on the silicon chips. The second experiment’s results showed a decrease in surface roughness of the silicon nitride chip of around 0.1 nm. This shows that, as with the silicon chips, the HF treatment, not the furnace treatment, seems to be responsible for the observed decreases in surface roughness. The third experiment’s results show that no noticeable decrease in surface roughness occurs when a Norcada silicon chip is sonicated in acetone and ethanol.

Next, I learned from Dr. Day a procedure that may be able to resolve two challenges that I am currently facing in my research. It involves depositing and polymerizing an alkane monolayer on silicon using UV light. According to him, this procedure has been shown to prevent oxidation of silicon. This could be of value to me because AFM imaging has shown that silicon chips actually seem to indicate increased surface roughness (even greater than its brand-new value) after sitting in room atmosphere for days after treatment. Secondly, Dr. Day mentioned that this alkane monolayer makes the silicon very hydrophobic and that he is not sure how well DNA would stick to it. This would actually be a desired effect for our work. Since DNA origami usually sticks very readily to mica and fairly readily to silicon (often clumping together), a surface that discourages the sticking of origami would ensure that only those crosses whose tail has been threaded through a nanopore would stick to the silicon surface.

Finally, I continued the experiment in which I am trying to correlate mica thickness to optical contrast. In an attempt to transfer very thin mica flakes to silicon, I used a fresh, clean PDMS stamp to cleave mica and then transfer it to the silicon. The literature from which this method was derived reports obtaining mica flakes as thin as one nanometer (a single layer of mica). So far, I have not come across any flakes thinner than 100 nm, so I will continue working on this experiment.

Previous Work

The link below accesses a PowerPoint version of a poster that I presented at Marshall's 2013 Sigma Xi Research Day and for which I won second place. It showcases the research that I conducted during the summer of 2012 as part of the High School Apprenticeship Program (HSAP).

2013 Marshall University Sigma Xi Research Day Poster

Informational Sites

Marshall University Molecular and Biological Imaging Center (MBIC)

Marshall University Chapter of Sigma Xi

Marshall University Summer Undergraduate Research Experience (SURE)

U.S. Army High School Apprenticeship Program (HSAP)

Paul W.K. Rothemund Caltech Home Page


1.) Rothemund, Paul W. K. Folding DNA to Create Nanoscale Shapes and Patterns. Nature. 2006, 440, 297-302. 

2.) Liu, Wenyan; Zhong, Hong; Wang, Risheng; Seeman, Nadrian C. Crystalline Two-Dimensional DNA-Origami Arrays. Angewandte Chemie International Edition. 2011, 50, 264-267.

3.) Kerenf, Kinneret; Berman, Rotem S.; Buchstab, Evgeny; Sivan, Uri; Braun, Erez. DNA-Templated Carbon Nanotube Field-Effect  
Transistor. Science. 2003, 302, 1380-1382.

4.) Sherman, William B.; Seeman, Nadrian C. A Precisely Controlled DNA Biped Walking Device. Nano Letters. 2004, 4, 1203-1207.

5.) Zhao, Yong-Xing; Shaw, Alan; Zeng, Xianghui; Benson, Erik; Nyström, Andreas M.; Högberg, Björn. DNA Origami Delivery System for Cancer Therapy with Tunable Release Properties. ACS Nano. 2012, 6 (10), 8684-8691.

6.) Azioune, Ammar; de Buyl, François; Pireaux, Jean-Jacques. Organosilane Molecular Glue For Bonding (Bio) Microchips and Sensors At Room Temperature. Dow Corning.

7.) Li, Xiang; Teramoto, Akinobu; Suwa Tomoyuki; Kuroda Rihito; Sugawa, Shigetoshi; Ohmi, Tadahiro. Formation Speed of Atomically Flat Surface On Si (100) In Ultra-Pure Argon. Microelectronic Engineering. 2011, 88, 3133-3139.


Last Updated: Friday 8/16/13 3:53 PM