Seeing molecules with AFM


Ever since his 2009 Science publication, Leo Gross and his IBM Zurich colleagues have captured our imagination by imaging the pentacene molecule with AFM and revealing clearly its five hexagonal carbon rings

Pentacene as imaged by Gross et al., Science p. 1110, 2009

Since that paper 8 years ago, fantastic images have continued to come in revealing structures of both natural and synthetic species in a wide array of applications ranging from biological to petroleum products.   One of the most interesting outcomes of this work is elucidations of unknown structures.  The ability to “just image” a molecular to figure out its structure is incredibly appealing, especially as it can be a critical component when added to more traditional methods such as NMR, mass spectroscopy, or vibrational spectroscopy  to unambiguously elucidate a molecule’s structure.    In an application close to my former life in the petroleum sector, this method was recently used to image asphaltenes, complex polycyclic aromatic hydrocabons typically found in the “bottom of the barrel” of crude oil, whose applications are primarily as paving materials for roads.  Beautiful images of various coal and petroleum-based asphaltenes emerged in 2015.

The asphaltene molecule as reported by Schuler, JACS, 2015, p. 9870

Even chemical reactions have been mapped including cyclization reactions and the famous organic reaction, the Bergman reaction. 

When these images came out, I would have my managers running to me with paper in hand asking when we could reproduce these results in our AFM lab with commercial, ambient systems on the third floor of our research building. 

These results are truly fantastic.  But they are done with fairly unique instrumentation and methods.  When I teach my AFM courses and cover the topic of resolution, I always include these results so that we know what is possible. But for most of us AFMers, these results are not possible in our own labs with commercial equipment – yet. 

Just to clear up the record, here are some of the customized features of the setups used to do these measurements. First, this work is done in ultra high vacuum (<10-8 torr) and ultra low temperatures (5K). This is done to have a pristine, stable, and clean surface which requires such a well-controlled environment.   I’m also guessing these systems are in the basement or equivalently quiet spot of their building.  Second, this setup uses incredibly stiff cantilevers. Whereas conventional silicon cantilevers have a spring constant of 40N/m at the high end, these cantilevers are actually tuning forks with a stiffness in the thousands of N/m. Such a stiff lever enables tiny oscillation amplitudes down to a few angstroms (compare with ambient operation of silicon cantilevers where the oscillation amplitude is typically a few to even a hundred nm).  The QPlus Sensor is a commercial example of this technology that was in fact used in the IBM Zurich publications and the sensors look like the one below:

A QPlus sensor

The tips are also specialized.  They are metallic tips (often copper) functionalized with an individual carbon monoxide molecule that easily binds to copper.  This kind of tip achieves many functions including a small radius, high aspect ratio, and inert to the sample.  Functionalization generally occurs by picking up CO from the surface with the tip.  Finally, the AFM operates in frequency modulated non-contact mode.  This means that the tip is oscillated at its resonance frequency where the amplitude is kept constant by a feedback circuit.  The resonance frequency shifts due to interaction of tip with sample, and this frequency shift is mapped during the scan.   Molecular structure images are also often collected in constant height where the tip scans over the surface at a fixed height while the frequency change is recorded. 

So where does this leave us?  For now, enjoy the spectacular results coming out of the few groups that have this capability. The latest one involved imaging triangulene, a molecule that chemists have long been dreaming about.

Triangulene as imaged by Pavlicek Nature Nanotech 2017

And in terms of doing it yourself? Well hopefully you have been armed with a number of reasons to show why it’s still not exactly readily available technology.  

Dalia Yablon, Ph.D.

SurfaceChar LLC

Website developed by S8080 Digital Media