Constructing a research plan

James J. De Yoreo

When I first meet with students to discuss research topics, they will often — almost always — talk about the type of measurement they want to make or, less often, what material they want to study.

For example, a student might say, “I want to use in situ TEM to study assembly of gold nanoparticles.”

At this point I will stop them and point to my whiteboard where I keep a series of questions in prominent view. They read as follows:

Notice that the two items on which most student focus when they think about their research are not addressed until the last two questions in this list. The point is that no one cares whether a particular measurement gets completed or whether some specific material is examined. What the world cares about is whether you significantly advance a scientific or technological field. In order to do so, you must understand what is currently scientifically known or technologically possible, have a vision of what would constitute an important leap forward, and be able to define the key scientific questions that keep you from accomplishing that leap. (From now on I will refer to these questions as the knowledge gaps.) Only after defining the vision of what you want to accomplish and what knowledge gaps you need to fill in order to do so, does it make sense to delineate how you will go about achieving your vision.

For example, my vision might be, “to create, at will, any 3D topology of semiconductors or metals with control over all topological features down to 5 nm by predictively assembling nanoparticles from solution.” There are many knowledge gaps associated with this vision, but one might be, “How can interaction potentials between particles be manipulated in order to drive attachment along specific crystallographic directions.” Given this knowledge gap, I can now go on to say what information I will need to fill it, how I will get that information, and what materials I should use to get it.

Answering the above questions is the first step to constructing a sound research plan. However, there are a few more pieces that need to be added. In what follows, I will focus on scientific research as opposed to technological research, but the basic approach applies to both. First, given your vision, what is the state-of-the-field relative to that vision?  By laying out this background on what is already known, you can logically arrive at your knowledge gaps.

Given the knowledge gaps, what are the scientific objectives of your research? The number of objectives should be small, typically two to four, but depends both on the number and diversity of the knowledge gaps and the magnitude of the research scope that you can reasonably tackle given your resources. Keep in mind that these objectives should be scientific, not logistical.

For example, “determining how interaction potentials between nanoparticles depend on particle separation and size, as well as solution conditions (i.e., pH, temperature, and electrolyte type and concentration)” is a scientific objective, while “performing in situ TEM to observe nanoparticle assembly” is a logistical objective.

To know what information you will need in order to achieve a given scientific objective, you will need to construct a scientific hypothesis that you can test. In some instances, constructing more than one hypothesis will make sense, but, as with the objectives, your goal should be to limit your hypotheses to the “minimum basis set” needed to achieve your objective. Here again, the hypothesis should be scientific, not logistical.

For example, the statement, “Variations in electrolyte type and concentration can alter the dynamics and outcomes of particle assembly, because electrolytes both screen electrostatic repulsion between particles and alter the structure of the hydration layers at the particle-solution interfaces, which in turn impacts the hydration barriers between the particles.” is a scientific hypothesis. In contrast, the statement “By collecting data on particle positions vs time using in situ TEM, I will be able to extract interparticle forces.” is a logistical hypothesis.

Once a scientific objective and related set of scientific hypotheses have been defined, the next logical step is to lay out your approach to testing your hypotheses and, thereby, achieving your objective. The approach consists of: 1) the information you will need to obtain, 2) the experimental and/or theoretical techniques you will use to obtain it, 3) the analyses you will apply to those results in order to obtain the needed information, and 4) the materials you will use in your research and why they are the right ones to use.  

For example, if I want to test the hypothesis stated above about the controls that electrolytes on interparticle forces, I might propose the following: “The interaction potentials between particles will be obtained both directly from AFM force-distance curves using single crystal probes interacting with single crystal surfaces, and indirectly using in situ TEM to track the positions of all particles in an ensemble as then undergo assembly. Interaction potentials will be extracted from the latter by determining the radial distribution function g(r) for every particle in the ensemble and taking U(r) = -kT ln[g(r)]. The hydration structure will be mapped in 3D using AFM-based fast force mapping. The surface charge of the particles will be determined from measurements of zeta potential. Molecular dynamics simulations of both a silica tip and a crystalline tip near a crystal surface will be used to extrapolate the experimental maps of hydration structure the structure that is created as the nanoparticle surfaces approach one another. These measurements and simulations will be applied to ZnO for which direction-specific oriented attachment of nanoparticles is well documented, Au, for which the directional dependence is expected to be weak, and mica for which cations that lie on surface lattice sites are both labile and exchangeable. The hydration structure, zeta potential and interparticle forces will be determined as a function of electrolyte type and concentration for monovalent, divalent and trivalent cations, and a range of pH that crosses the isoelectric point for each type of surface. By combining the data on interparticle forces, surface charge, and hydration structure as a function of pH nd electrolyte type and concentration, we will test the stated scientific hypothesis and thereby achieve the scientific objective.”

The final piece to completing your research plan is to explain what the impact of your results will be on your stated vision and on the broader scientific community for which your vision is relevant.  In many ways, this will be a restatement of what you said in your introduction, but it is important to relate back to your original vision and expound on the broader impacts of your work.

Dr. James J. De Yoreo is a Chief Scientist at Pacific Northwest National Lab and Affiliate Professor of Materials Science and Engineering at the University of Washington.

Related Content:

Expert Forum

What is a faculty search committee looking for?