In single molecule force spectroscopy experiments, force probes chemically modified with synthetic, single-stranded DNA oligomers produced characteristic steady-state forces connected by abrupt steps between plateaus, as the probes moved away from a graphite substrate. The force plateaus represent peeling of a small number of polymer molecules from the flat surface. The final force jump in the retraction region of the force–distance curves can be attributed to a single DNA molecule detaching from the graphite surface. Previously, Manohar et al. (Nano Lett. 2008, 8, 4365) reported the peeling forces of the pyrimidine oligomers as 85.3 ± 4.7 and 60.8 ± 5.5 pN for polythymine and polycytosine, respectively. We measured the force–distance curves for purine oligomers on a graphite surface and found the peeling forces to be 76.6 ± 3.0 and 66.4 ± 1.4 pN for polyadenine and polyguanine, respectively. Using a refined model for peeling a single freely jointed polymer chain from a frictionless substrate, we determined a ranking of the effective average binding energy per nucleotide for all four bases as T ≥ A > G ≥ C (11.3 ± 0.8, 9.9 ± 0.5, 8.3 ± 0.2, and 7.5 ± 0.8 k B T, respectively). The binding energy determined from the peeling force data did not scale with the size of the base. The distribution of peeling forces of polyguanine from the graphite surface was unusually broad in comparison to the other homopolymers, and often with inconsistent chain extensions, possibly indicating the presence of secondary structures (intra- or intermolecular) for this sequence.

We used single molecule force spectroscopy to measure the force required to remove single-stranded DNA (ssDNA) homopolymers from single-walled carbon nanotubes (SWCNTs) deposited on methyl-terminated self-assembled monolayers (SAMs). The peeling forces obtained from these experiments are bimodal in distribution. The cluster of low forces corresponds to peeling from the SAM surface, while the cluster of high forces corresponds to peeling from the SWCNTs. Using a simple equilibrium model of the single molecule peeling process, we calculated the free energy of binding per nucleotide. We found that the free energy of ssDNA binding to hydrophobic SAMs decreases as poly(A) > poly(G) ≈ poly(T) > poly(C) (16.9 ± 0.1; 9.7 ± 0.1; 9.5 ± 0.1; 8.7 ± 0.1 kBT, per nucleotide). The free energy of ssDNA binding to SWCNT adsorbed on this SAM also decreases in the same order poly(A) > poly(G) > poly(T) > poly(C), but its magnitude is significantly greater than that of DNA-SAM binding energy (38.1 ± 0.2; 33.9 ± 0.1; 23.3 ± 0.1; 17.1 ± 0.1 kBT, per nucleotide). An unexpected finding is that binding strength of ssDNA to the curved SWCNTs is much greater than to flat graphite, which also has a different ranking (poly(T) > poly(A) > poly(G) ≥ poly(C); 11.3 ± 0.8, 9.9 ± 0.5, 8.3 ± 0.2, and 7.5 ± 0.8 kBT, respectively, per nucleotide). Replica-exchange molecular dynamics simulations show that ssDNA binds preferentially to the curved SWCNT surface, leading us to conclude that the differences in ssDNA binding between graphite and nanotubes arise from the spontaneous curvature of ssDNA.

We used brownian dynamics to study the peeling of a polymer molecule, represented by a freely jointed chain, from a frictionless surface in an implicit solvent with parameters representative of single-stranded DNA adsorbed on graphite. For slow peeling rates, simulations match the predictions of an equilibrium statistical thermodynamic model. We show that deviations from equilibrium peeling forces are dominated by a combination of Stokes (viscous) drag forces acting on the desorbed section of the chain and a finite rate of hopping over a desorption barrier. Characteristic velocities separating equilibrium and nonequilibrium regimes are many orders of magnitude higher than values accessible in force spectroscopy experiments. Finite probe stiffness resulted in disappearance of force spikes due to desorption of individual links predicted by the statistical thermodynamic model under displacement control. Probe fluctuations also masked sharp transitions in peeling force between blocks of distinct sequences, indicating limitation in the ability of single-molecule force spectroscopy to distinguish small differences in homologous molecular structures.

No abstract