Creating Chemical and Biological Diamond Interfaces

J.N. Russell, Jr.,¹ J.E. Butler,¹ R.J. Hamers,² L.M. Smith,² and S.F. Bent^3
1Chemistry Division
²University of Wisconsin
³Stanford University

Overview: Developing organic synthetic routes for attaching organic and biological materials/structures to semiconductor substrates is important for the development of molecularly based electronic and sensing devices. We demonstrate several methods that promise to lead to molecular structures and functionality on diamond via thermal and photochemical means. We focus on diamond because of its chemical stablity and biocompatibility.

Surface Cycloaddition Reactions:^1,2 We draw our inspiration for chemical functionalization of the diamond surface from organic chemistry. To determine whether the hydrogen-free diamond (001) surface could be functionalized using cycloaddition chemistry, we used cyclopentene and 1,3-butadiene as chemical tests of the π-bonding nature of the hydrogen-free surface dimer bond. A type IIa natural diamond trapezoidal prism, oriented with the major faces exposing the (001) crystal planes, was mounted in an ultrahigh vacuum (UHV) chamber on a manipulator with a temperature range of 90 to 1473 K. Multiple internal reflection infrared spectroscopy (MIRIRS) was used to monitor the CH stretching region of the infrared spectrum. When the hydrogen-terminated diamond (100) surface was heated above 1323 K, the surface hydrogen recombined and H₂ desorbed. The resulting hydrogen-free surface was composed of partially π-bonded (partially di-_radical) surface dimers.

Neither cyclopentene nor 1,3-butadiene reacted with the hydrogen-terminated surface. The molecules physisorbed intact at 90 K and desorbed without reaction. In contrast, the hydrogen-free C(001) surface reacted with both molecules at room temperature. Figure 7(a) shows unpolarized MIRIRS for cyclopentene chemisorbed on the C(001) surface at room temperature compared to the spectrum of a physisorbed multilayer of cyclopentene at 90 K. Note the alkene C-H stretch at ~3050 cm^_1 disappeared upon chemisorption, indicating the removal of the C=C bond in the chemisorbed molecule. The spectrum is consistent with cyclopentene chemisorbing via the reaction mechanism shown in Fig. 7(b).

We also examined whether butadiene would react with the H-free surface dimer to yield a [2+2] (cis or trans conformation) and/or a [4+2] (Diels-Alder) reaction product (Fig. 7(c)). The room temperature MIRIR spectra for the perhydro-butadiene and 1,1,4,4-d₄-butadiene dosed H free C(001) are compared in Fig. 7(d). The asymmetric and symmetric alkane (-CH₂-) stretches are observed. Using deuterium labeling, the alkene C-H stretch in the reaction product is associated with H on the 2 and 3 positions of the butadiene molecule. Meanwhile, the vinylic =CH₂ stretch in perhydro-butadiene disappeared upon reaction with the surface. The data are consistent with predominately a concerted [4+2] cycloaddition reaction mechanism.

FIGURE 7
(a) MIRIRS of cyclopentene on a diamond (001) surface (note the loss of the alkene C-H stretch at ~3050 cm^_1); (b) Illustration of the [2+2] cycloaddition reaction; (c) Possible surface reaction mechanisms for 1,3-butadiene; (d) MIRIRS of butadiene on a diamond (001) surface.

The observation of a [2+2] reaction product for cyclopentene indicates a low symmetry reaction route, which accesses the di-σ-radical character of the dimer. In the case of butadiene, where the molecule could react to yield a product via the [2+2] or [4+2] reaction pathways, the [4+2] reaction product is favored. This illustrates the π-bonding character of the C(001) dimer bond and the higher probability of the high symmetry reaction when it is available. Thus, the surface is composed of partially π-bonded dimers. This work demonstrates the viability of organic synthetic routes for functionalizing diamond surfaces.

Photochemical Surface Attachment:^3,4 As noted above, the hydrogen-terminated diamond surface is unreactive at room temperature. By leveraging the UHV studies, we developed a UV photochemical process for reacting alkenes with the "chemically inert" hydrogen-terminated diamond surface under ambient conditions. Five to 10 μL of trifluoroacetamide-protected 10-aminodec-1-ene ("TFAAD") were placed directly on a freestanding (1-mm thick) hydrogen-terminated diamond film in a nitrogen-purged Teflon reaction chamber with a quartz window. The sample was exposed for about 12 hours to ultraviolet light from a low-pressure mercury vapor quartz grid lamp (λ_max = 254 nm, 0.35 milliwatt/cm²). The samples were rinsed in chloroform, then methanol, before X-ray photoelectron spectroscopy (XPS) analysis.

Figure 8 shows the C(1s), F(1s), and N(1s) XPS before and after photochemical surface modification with TFAAD. Before modification, the carbon spectrum has one major peak centered at 285.5 eV, corresponding to bulk carbon (off-scale on the right-hand side of the spectrum). After modification, the carbon spectrum developed two new peaks at 293.9 eV and 289.8 eV from the -CF₃ and carbonyl (C=O) groups, respectively. Fluorine (F(1s): 689.6 eV) and nitrogen (N1(s): 401.4 eV) peaks from the trifluoroacetamide (TFA) protecting group also verified TFAAD attachment. TFAAD did not adhere to the surface without UV irradiation.

By refluxing the TFAAD-modified surface in 2:5 MeOH:H₂O with 7% (w/w) K₂CO₃, the trifluoroacetamide protecting group was removed, leaving the surface terminated with chemically reactive amine groups. Figure 8 also shows XPS spectra of the same sample after deprotection. Based on the changes in the C(1s) and F(1s) data, ~85% of the -CF3 groups are removed by the deprotection process. The shift in the N1(s) binding energy is also consistent with a free amine.

DNA functionalization and stability:⁴ The amine-terminated diamond surface can be reacted with linker molecules that connect single-stranded DNA oligonucleotides to the surface. The stability of the oligonucleotide functionalized surface was investigated for repetitive cycles of hybridization and denaturation. In each cycle, the surface-bound DNA was hybridized with its fluorescently labeled complement and the fluorescence intensity was measured. Then the sample was denatured at room temperature in an aqueous solution of 8.3M urea and rinsed with distilled water. No significant fluorescence loss was measured for the hybridized DNA-modified diamond after repeating the hybridization/denaturation process 15 times (Fig. 9). However, similarly prepared DNA-modified silicon surfaces lost 1.8 ± 0.4 % fluorescence per hybridization cycle or an ~27% loss after 15 cycles. We conclude that diamond substrates are well-suited for reproducible DNA sensing applications.

Impact: Schemes for chemically functionalizing the hydrogen-free and hydrogen-terminated diamond surfaces that use thermal and photochemical processes were demonstrated. The modified diamond surfaces are chemically robust. In particular, the DNA-functionalized diamond surface is well-suited to repeated and reproducible sensing applications.

FIGURE 9
Direct comparison of the hybridization stability of DNA on diamond and silicon surfaces.

Acknowledgments: The cycloaddition work was performed in collaboration with the University of Wisconsin and Stanford University groups. The photochemical attachment and DNA work was in collaboration with the University of Wisconsin group. We thank Tanya Knickerbocker, Todd Strother, Michael Schwartz, Jennifer Hovis, and Sarah Coulter from the University of Wisconsin and George Wang from Stanford University for their valuable contributions.

[Sponsored by ONR, NSF, and NIH]

References

¹ J.S. Hovis, S.K. Coulter, R.J. Hamers, M.P. D'Evelyn, J.N. Russell, Jr., and J.E. Butler, "Cycloaddition Chemistry at Surfaces: Reaction of Alkenes with the Diamond (001)-2x1 Surface,"
J. Am. Chem. Soc. 112(4), 732-733 (2000).
² G.T. Wang, S.F. Bent, J.N. Russell, Jr., J.E. Butler, and M.P. D'Evelyn, "Functionalization of Diamond(100) by Diels-Alder Chemistry," J. Am. Chem. Soc. 112(4), 744-745 (2000).
³ T. Strother, T. Knickerbocker, J.N. Russell, Jr., J.E. Butler, L.M. Smith, and R.J. Hamers, "Photochemical Functionalization of Diamond Films," Langmuir 18(4), 968-971 (2002).
⁴ T. Knickerbocker, T. Strother, M.P. Schwartz, J.N. Russell, Jr., J.E. Butler, R.J. Hamers, and L.M. Smith, "DNA-Modified Diamond Films," Langmuir (2003), in press.