Cross sections of material specimens for TEM analysis must be produced in the shortest time possible, contain few, if any, artifacts and have a large area available for analysis. The analyst must also be able to prepare these cross sections from specified areas of complex, heterogeneous structures on a routine, reproducible basis to meet the growing needs of the semiconductor industry for TEM analysis. The specimen preparation spatial resolution required for preparing precision cross sections is substantially less than one micron. Cross sections meeting these requirements can be prepared by mounting a specimen to the Tripod Polisher and mechanically polishing on one side of the specimen, using a sequence of progressively finer grit diamond lapping films, until the area of interest is reached. This polished surface is then very briefly polished on a cloth wheel with colloidal silica to attain the final polish on that side. The specimen is then flipped over on the Tripod Polisher and polished from the other side, using same sequence of diamond lapping films to reach the predefined area of interest. The Tripod Polisher is set at a slight angle, to produce a tapered, wedge-shaped specimen, which has the area of interest at the thinnest edge of the taper. The specimen is polished with the diamond lapping films and the colloidal silica until it is 1000 Angstroms or less in thickness. The specimen is removed from the polisher and mounted on a 2 × 1mm slotted grid with M-Bond 610 epoxy. After the epoxy is cured the specimen can be taken directly to the microscope for analysis. The need for ion milling has been eliminated or reduced to a few minutes in most of our work because of the thinness of the final specimen. The total specimen preparation time is between 2.5 and 4 hours, depending on the specimen and the size of the specified area. The area available for analysis ranges from 0.5mm up to the full size of the mounting grid opening. The wedge shape of the specimen provides the mechanical stability needed for a long thin specimen.

A grinding/polishing tool has been developed for preparing TEM samples. The hand-held tool is 2.50″ in diameter and 3.0″ high. Rough-cut samples, 300 to 600 microns thick, are routinely polished to 5 microns thick in four to six hours using this tool. As these 5 micron samples are so thin and uniform, a separate dimpling operation can be eliminated. Likewise, the time required to ion-mill the sample can be reduced to 0.5 to 2.0 hours – greatly reducing ion-milling artifacts and significantly increasing the area viewable by TEM. The process is equally effective for all classes of samples: Silicon devices, ceramics or metals – in either cross-section or planar views.

The procedures described in this paper allow both SEM and TEM analysis to be performed on the same, device specific, semiconductor cross section. In order to accomplish this, a number of tools and fixtures have been constructed that allow the user to polish into the sample to a predetermined plane-of-polish, bisecting the device or feature of interest for SEM analysis. After SEM examination, the specimen is prepared for TEM analysis by first affixing a grid to the just-examined surface, inverting the specimen and parallel-polishing the backside of the specimen until the specimen's total thickness is in the 0.5 to 1.0μm range using the described tools. A subsequent one to ten minute ion milling step cleans the specimen. A very considerable positive side-effectof this method is the nearelimination of artifacts arisingfrom the use of strong chemicals and lengthy ion milling. The method has been extended to the preparation of plan-view device samples and non-semiconductor specimens.

Electroplated Cu was found to have a fine as-plated microstructure, 0.05 ±0.03 μm, with multiple grains through the film thickness and evidence of twins and dislocations within grains. Over time at room temperature, the grains grew to greater than 1 μm in size. Studied as a function of annealing temperature, the recrystallized grains were shown to be 1.6 ± 1.0 μm in size, columnar and highly twinned. The grain growth was directly related to the time dependent decrease in sheet resistance. The initial grain structure was characterized using scanning transmission electron microscopy (STEM) from a cross-section sample prepared by a novel focused ion beam (FIB) and lift-out technique. The recrystallized grain structures were imaged using FIB secondary electron imaging. From these micrographs, the grain boundary structures were traced, and an image analysis program was used to measure the grain areas. A Gaussian fit of the log-normal distribution of grain areas was used to calculate the mean area and standard deviation. These values were converted to grain size diameters by assuming a circular grain geometry.

Combining substantial pre-thinning of semiconductor device specimens via tripod polishing, followed by multiple, selective material removal steps with a focussed ion beam (FIB) tool, offers some unique possibilities with regard to preparing plan-view specimens at precise locations in the material below, and in the interconnect layers above, the semiconductor surface. Extension of these methods to other types of specimens in different configurations allows for the fabrication of multiple specimens on the same TEM grid that may not be possible by any other means.