Chemistry Reference
In-Depth Information
Fig. 1. (a) Original image of the graphene films on 300-nm-thick SiO
2
/Si substrate. The
image was taken using a CCD mounted on an optical microscope. The image can be
separated into the (b) red, (c) green, and (d) blue images by a conventional graphics
software.
intensity-levels, and combined in a picture image. In this way, the picture
image (Fig. 1(a)) can be resolved into R/G/B images (Figs. 1(b)-(d)) by
conventional graphic software. The residual glue from adhesive tape is
prominent and causes noise in the red image. The blue image has
relatively poor contrast in comparison with the green image. This
difference is consistent with the use of the green filter to obtain better
contrast in the optical-microscope observations. The green image taken
by the CCD has the advantage of direct and clear pixel-bit information
on the green intensity over other optical methods.
From the green image map (Fig. 2(a)), the pixel percentage histogram
in the dotted frame is obtained (the dotted line in Fig. 2(b)). Five peaks
with equal intensity intervals are separately observed (inset in Fig. 2(b)),
indicating that these are five pixel regions with different green-contrast.
The peak with the highest green-intensity (at ~133 in the intensity level)
is from the substrate, as confirmed by the image without graphene. Then,
the framed region in Fig. 2(a) contains four different thicknesses. Each
percentage of the observed pixels depends on the occupied area in the
dotted frame. This histogram is used as a reference to judge the number
of layers of a target sample (in a red frame in Fig. 2(c)). We also make a
green intensity histogram for the target sample (a red curve in Fig. 2(b)).
After normalizing the intensity of the sample by using the intensity from
the substrate, we compare the peak positions in the histogram of the
target with those in the reference (Fig. 2(d)) to judge the number of
layers in the target.
