Limits...
Simple and effective graphene laser processing for neuron patterning application.

Lorenzoni M, Brandi F, Dante S, Giugni A, Torre B - Sci Rep (2013)

Bottom Line: Primary embryonic hippocampal neurons were cultured on our substrate, demonstrating an ordered interconnected neuron pattern mimicking the pattern design.Surprisingly, the functionalization is more effective on the SLG, resulting in notably higher alignment for neuron adhesion and growth.Therefore the proposed technique should be considered a valuable candidate to realize a new generation of highly specialized biosensors.

View Article: PubMed Central - PubMed

Affiliation: Nanophysics, Istituto Italiano di Tecnologia, Genova, Italy. matteo.lorenzoni@iit.it

ABSTRACT
A straightforward fabrication technique to obtain patterned substrates promoting ordered neuron growth is presented. Chemical vapor deposition (CVD) single layer graphene (SLG) was machined by means of single pulse UV laser ablation technique at the lowest effective laser fluence in order to minimize laser damage effects. Patterned substrates were then coated with poly-D-lysine by means of a simple immersion in solution. Primary embryonic hippocampal neurons were cultured on our substrate, demonstrating an ordered interconnected neuron pattern mimicking the pattern design. Surprisingly, the functionalization is more effective on the SLG, resulting in notably higher alignment for neuron adhesion and growth. Therefore the proposed technique should be considered a valuable candidate to realize a new generation of highly specialized biosensors.

Show MeSH

Related in: MedlinePlus

Raman characterization.(a) Comparison of Raman spectra collected on two pristine samples under the same conditions with labelled peaks: the Raman signal on SiO2 (black) is enhanced 5 times in respect to glass (red). (b) Raman component color map of the upper left edge of the square irradiated at 0.15 J/cm2 in type 1 sample; (c) same type of map on a region irradiated at 0.10 J/cm2. The maps have been generated by taking the fitted Raman spectra of unmodified SLG as component. The presence of the graphene Raman component along the horizontal section is plotted as a white line: at 0.15 J/cm2 graphene is absent but some residuals of graphitic material are still attached to the surface, as in spot B; at lower fluences (c) intact areas are clearly visible within the irradiated region. Circled points A, B, C are picked from the map and their spectra are reported in (d). For details about spectra see the main text. In (e) three different normalized spectra of the G area of type 1 sample, representing the intrinsic variability of CVD single layer graphene used.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3674432&req=5

f4: Raman characterization.(a) Comparison of Raman spectra collected on two pristine samples under the same conditions with labelled peaks: the Raman signal on SiO2 (black) is enhanced 5 times in respect to glass (red). (b) Raman component color map of the upper left edge of the square irradiated at 0.15 J/cm2 in type 1 sample; (c) same type of map on a region irradiated at 0.10 J/cm2. The maps have been generated by taking the fitted Raman spectra of unmodified SLG as component. The presence of the graphene Raman component along the horizontal section is plotted as a white line: at 0.15 J/cm2 graphene is absent but some residuals of graphitic material are still attached to the surface, as in spot B; at lower fluences (c) intact areas are clearly visible within the irradiated region. Circled points A, B, C are picked from the map and their spectra are reported in (d). For details about spectra see the main text. In (e) three different normalized spectra of the G area of type 1 sample, representing the intrinsic variability of CVD single layer graphene used.

Mentions: In addition to optical microscopy, we characterized the machined samples with Raman spectroscopy, atomic force microscopy (AFM) and Scanning Kelvin probe force microscopy (SKPM), thus evaluating the quality of the SLG and the ablation efficiency. Raman spectroscopy is a fundamental tool to characterize carbon-based materials and allotropes since carbon sp2 rings and chains give rise to particular Raman scattering. Following literature studies, in the case of graphene we used three major features as references22; the G band (~1580 cm−1), the D band (~1350 cm−1) and the 2D band (~2700 cm−1), as labelled in Fig. 4a, where we report two SLG spectra collected, respectively, on Si/SiO2 (black color) and on glass (red color). As shown in Fig. 4a, multi-reflection effects due to the SiO2 layer enhance the Raman intensity23. The G band is produced by in-plane vibration of C atoms and identifies the first-order Raman-allowed mode in graphene. The D band, which is absent in perfect SLG, provides valuable information on the presence of defects such as: graphene crystal edges, chemisorbed H and, in general, sp3 bond content arising from different sources. In fact, the D band, originating from the breathing modes of aromatic rings, requires a defect for its activation. The intensity ratio IG/ID is commonly used as a benchmark for the quality of commercial SLG. The 2D peak is the D-peak overtone22. Both the 2D peak position and the intensity ratio IG/I2D have been used to determine the number of graphene layers and other basic structural and electronic properties. In Fig. 4b and 4c we present intensity maps of the edges of sample regions irradiated at 0.15 J/cm2 and 0.10 J/cm2, respectively (SLG on SiO2/Si); after acquiring a detailed map by micro-Raman spectrometer we used the Raman spectra of pristine SLG in the region 1200–1800 cm−1 as component to generate the intensity map. Raman intensity maps averaged over a wide wavenumber region instead of a single peak intensity allows a mapping that is more sensitive to SLG alteration due to photo exposure. Circled points A, B, C, indicated in Fig. 4b, are representative spots on the map and their corresponding spectra are reported in Fig. 4d. In A the component belonging to pristine SLG is maximum, in B the spectra indicates a modification in SLG (i.e. a folded residual) while in C there is no significant Raman signal. Both maps clearly resolve the transition region between pristine and irradiated areas, with a lateral resolution limited to the Raman probe spot size (~0.7 μm). If exposed to lower energies (Fig. 4c) SLG remains intact in some regions. Within the irradiated area residuals of carbonaceous material are still attached to the surface but do not present the typical G peak of graphene. The residual's spectra (black line in Fig. 4d corresponding to position B) present a blue shift of the G peak and a raising of the D band, with an ID/IG value of 0.71. According to the three stage classification of disorder proposed by Ferrari et al.24 the Raman spectrum is considered to depend on the degree of amorphization, disorder, clustering of sp2 phase, presence of sp2 rings or chains, ratio between sp2 and sp3 bonds. In particular for disordered and amorphous carbon systems such a blue shift (10–20 nm) coupled with the contemporary presence of D and G features identify unequivocally a point in the so called “amorphization trajectory” thus indicating the graphitic nature of the residual. In the map of Fig. 4c the upper left edge of the square irradiated at 0.10 J/cm2 is presented. The component used for mapping is still the unmodified SLG Raman spectrum. In the black areas the SLG component is absent. It is important to stress how the reference G peak of the commercial SLG employed presents a certain degree of variability (Fig. 4e) that remains also after a pre-irradiation with a short pulse at low energy (0.05 J/cm2) in order to remove adsorbate layers and ambient hydrocarbons. In principle the position of the G band (GPOS) is invariant in all layered graphitic compounds. Although it presents several spots compatible with the SLG signature (GPOS at 1586 cm−1 and absence of D peak), the GPOS in our samples tends to be spread around this value, ranging from 1580 to 1600 cm−1. CVD graphene presents smooth surface regions of 0.5–2 μm surrounded by wrinkles rich in defects that generate a modest D band, as confirmed by the AFM topography given in Fig. 4a. These length scales are comparable with the optical resolution of the instrument (~0.7 μm). The result is a spread of GPOS over the mapped area with a predominant component around 1597 cm−1, consistent with a high degree of doping due to the fabrication process, most likely due to the Cu etchant involved8, together with an increase in the D peak when a defect rich region is probed. With regard to the 2D peak, in SLG it should have a symmetric shape centered at ~2680 cm−1; a FWHM of approx. 33 cm−1 and an IG/I2D intensity ratio of ~0.525. The spectra collected show instead a symmetric shaped peak with a FWHM of ~39 cm−1 blue shifted around 2695 cm−1. Similar Raman signature has been reported for CVD graphene grown on Cu at atmospheric pressure26 and could be explained by the presence of bi/multilayered regions with random orientation, due to a CVD process that is not properly limited once the SL is formed. In order to verify this, we performed an AFM topography of a 0.25 μm2 area (Fig. 5b) that shows sub 100 nm islands surrounded by thicker areas, regarded as partial second layer coverage of the lower SLG.


Simple and effective graphene laser processing for neuron patterning application.

Lorenzoni M, Brandi F, Dante S, Giugni A, Torre B - Sci Rep (2013)

Raman characterization.(a) Comparison of Raman spectra collected on two pristine samples under the same conditions with labelled peaks: the Raman signal on SiO2 (black) is enhanced 5 times in respect to glass (red). (b) Raman component color map of the upper left edge of the square irradiated at 0.15 J/cm2 in type 1 sample; (c) same type of map on a region irradiated at 0.10 J/cm2. The maps have been generated by taking the fitted Raman spectra of unmodified SLG as component. The presence of the graphene Raman component along the horizontal section is plotted as a white line: at 0.15 J/cm2 graphene is absent but some residuals of graphitic material are still attached to the surface, as in spot B; at lower fluences (c) intact areas are clearly visible within the irradiated region. Circled points A, B, C are picked from the map and their spectra are reported in (d). For details about spectra see the main text. In (e) three different normalized spectra of the G area of type 1 sample, representing the intrinsic variability of CVD single layer graphene used.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3674432&req=5

f4: Raman characterization.(a) Comparison of Raman spectra collected on two pristine samples under the same conditions with labelled peaks: the Raman signal on SiO2 (black) is enhanced 5 times in respect to glass (red). (b) Raman component color map of the upper left edge of the square irradiated at 0.15 J/cm2 in type 1 sample; (c) same type of map on a region irradiated at 0.10 J/cm2. The maps have been generated by taking the fitted Raman spectra of unmodified SLG as component. The presence of the graphene Raman component along the horizontal section is plotted as a white line: at 0.15 J/cm2 graphene is absent but some residuals of graphitic material are still attached to the surface, as in spot B; at lower fluences (c) intact areas are clearly visible within the irradiated region. Circled points A, B, C are picked from the map and their spectra are reported in (d). For details about spectra see the main text. In (e) three different normalized spectra of the G area of type 1 sample, representing the intrinsic variability of CVD single layer graphene used.
Mentions: In addition to optical microscopy, we characterized the machined samples with Raman spectroscopy, atomic force microscopy (AFM) and Scanning Kelvin probe force microscopy (SKPM), thus evaluating the quality of the SLG and the ablation efficiency. Raman spectroscopy is a fundamental tool to characterize carbon-based materials and allotropes since carbon sp2 rings and chains give rise to particular Raman scattering. Following literature studies, in the case of graphene we used three major features as references22; the G band (~1580 cm−1), the D band (~1350 cm−1) and the 2D band (~2700 cm−1), as labelled in Fig. 4a, where we report two SLG spectra collected, respectively, on Si/SiO2 (black color) and on glass (red color). As shown in Fig. 4a, multi-reflection effects due to the SiO2 layer enhance the Raman intensity23. The G band is produced by in-plane vibration of C atoms and identifies the first-order Raman-allowed mode in graphene. The D band, which is absent in perfect SLG, provides valuable information on the presence of defects such as: graphene crystal edges, chemisorbed H and, in general, sp3 bond content arising from different sources. In fact, the D band, originating from the breathing modes of aromatic rings, requires a defect for its activation. The intensity ratio IG/ID is commonly used as a benchmark for the quality of commercial SLG. The 2D peak is the D-peak overtone22. Both the 2D peak position and the intensity ratio IG/I2D have been used to determine the number of graphene layers and other basic structural and electronic properties. In Fig. 4b and 4c we present intensity maps of the edges of sample regions irradiated at 0.15 J/cm2 and 0.10 J/cm2, respectively (SLG on SiO2/Si); after acquiring a detailed map by micro-Raman spectrometer we used the Raman spectra of pristine SLG in the region 1200–1800 cm−1 as component to generate the intensity map. Raman intensity maps averaged over a wide wavenumber region instead of a single peak intensity allows a mapping that is more sensitive to SLG alteration due to photo exposure. Circled points A, B, C, indicated in Fig. 4b, are representative spots on the map and their corresponding spectra are reported in Fig. 4d. In A the component belonging to pristine SLG is maximum, in B the spectra indicates a modification in SLG (i.e. a folded residual) while in C there is no significant Raman signal. Both maps clearly resolve the transition region between pristine and irradiated areas, with a lateral resolution limited to the Raman probe spot size (~0.7 μm). If exposed to lower energies (Fig. 4c) SLG remains intact in some regions. Within the irradiated area residuals of carbonaceous material are still attached to the surface but do not present the typical G peak of graphene. The residual's spectra (black line in Fig. 4d corresponding to position B) present a blue shift of the G peak and a raising of the D band, with an ID/IG value of 0.71. According to the three stage classification of disorder proposed by Ferrari et al.24 the Raman spectrum is considered to depend on the degree of amorphization, disorder, clustering of sp2 phase, presence of sp2 rings or chains, ratio between sp2 and sp3 bonds. In particular for disordered and amorphous carbon systems such a blue shift (10–20 nm) coupled with the contemporary presence of D and G features identify unequivocally a point in the so called “amorphization trajectory” thus indicating the graphitic nature of the residual. In the map of Fig. 4c the upper left edge of the square irradiated at 0.10 J/cm2 is presented. The component used for mapping is still the unmodified SLG Raman spectrum. In the black areas the SLG component is absent. It is important to stress how the reference G peak of the commercial SLG employed presents a certain degree of variability (Fig. 4e) that remains also after a pre-irradiation with a short pulse at low energy (0.05 J/cm2) in order to remove adsorbate layers and ambient hydrocarbons. In principle the position of the G band (GPOS) is invariant in all layered graphitic compounds. Although it presents several spots compatible with the SLG signature (GPOS at 1586 cm−1 and absence of D peak), the GPOS in our samples tends to be spread around this value, ranging from 1580 to 1600 cm−1. CVD graphene presents smooth surface regions of 0.5–2 μm surrounded by wrinkles rich in defects that generate a modest D band, as confirmed by the AFM topography given in Fig. 4a. These length scales are comparable with the optical resolution of the instrument (~0.7 μm). The result is a spread of GPOS over the mapped area with a predominant component around 1597 cm−1, consistent with a high degree of doping due to the fabrication process, most likely due to the Cu etchant involved8, together with an increase in the D peak when a defect rich region is probed. With regard to the 2D peak, in SLG it should have a symmetric shape centered at ~2680 cm−1; a FWHM of approx. 33 cm−1 and an IG/I2D intensity ratio of ~0.525. The spectra collected show instead a symmetric shaped peak with a FWHM of ~39 cm−1 blue shifted around 2695 cm−1. Similar Raman signature has been reported for CVD graphene grown on Cu at atmospheric pressure26 and could be explained by the presence of bi/multilayered regions with random orientation, due to a CVD process that is not properly limited once the SL is formed. In order to verify this, we performed an AFM topography of a 0.25 μm2 area (Fig. 5b) that shows sub 100 nm islands surrounded by thicker areas, regarded as partial second layer coverage of the lower SLG.

Bottom Line: Primary embryonic hippocampal neurons were cultured on our substrate, demonstrating an ordered interconnected neuron pattern mimicking the pattern design.Surprisingly, the functionalization is more effective on the SLG, resulting in notably higher alignment for neuron adhesion and growth.Therefore the proposed technique should be considered a valuable candidate to realize a new generation of highly specialized biosensors.

View Article: PubMed Central - PubMed

Affiliation: Nanophysics, Istituto Italiano di Tecnologia, Genova, Italy. matteo.lorenzoni@iit.it

ABSTRACT
A straightforward fabrication technique to obtain patterned substrates promoting ordered neuron growth is presented. Chemical vapor deposition (CVD) single layer graphene (SLG) was machined by means of single pulse UV laser ablation technique at the lowest effective laser fluence in order to minimize laser damage effects. Patterned substrates were then coated with poly-D-lysine by means of a simple immersion in solution. Primary embryonic hippocampal neurons were cultured on our substrate, demonstrating an ordered interconnected neuron pattern mimicking the pattern design. Surprisingly, the functionalization is more effective on the SLG, resulting in notably higher alignment for neuron adhesion and growth. Therefore the proposed technique should be considered a valuable candidate to realize a new generation of highly specialized biosensors.

Show MeSH
Related in: MedlinePlus