Procedure and results

From New IAC Wiki
Revision as of 02:11, 18 December 2011 by Abdehait (talk | contribs) (→‎Burned vs. Unburned Hole)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search


The design of the environmental scanning electron microscope (ESEM) is used to test the specimen properties that are less than nono scale. ESEM has two detectors that each gives a image for the specimen, the image quality is depending on the number electron scattered or back scattered detected by each detector. ESEM is supported by software package that device the monitor's screen into four small screens, two screens show the image of each detector and a third one shows the combination of both of the detectors' images, the last screen is showing a safety camera image which helps to avoid any direct contact between the sample and detectors the sample feeding process.

ESEM is supported by other software and hardware packages. A software package shows the chemical structure for the tested sample, the package shows on a screen the elements distribution in the sample, their percentages, and histograms that is supported with different statistical tools for advanced reports. ESEM has big variety of holders that are made from aluminium fit the samples shape and height as there are used for testing.

Specifically, in the case of applying ED7100 resistive paste on FR4 cards,the aim of using ESEM is to measure the paste thickness of samples that are prepared using different procedures and it provided different topographical images for the surface after applying and curing the paste in different thicknesses and curing temperatures.

Methods of applying ED7100

The resistive paste is applied using the following procedure:

Using a brush

Applying the resistive paste manually by a flat and liner painting brush. The flat brush is used to cover the paste over most of the card area but the liner brush is used to fill the closer areas to the copper frame. This procedure succeeded to give a uniform paste surface with a thickness of 5-10 microns. The paste is cured in an oven of 160 degrees Celsius for one hour.

Using a Squeegee

Sqweegee 1.jpg Sqweegee 2.jpg

Applying the paste using a mechanically fixed squeegee. In this procedure, a very flat squeegee is manufactured to be fixed on a the desired height (in micrometer) over a horizontally moving table. The paste thickness is controlled an uncertainty of [math] \pm 25 \mu m \,\,\, (1 mil) [/math], the set up is designed to have an thickness of 10-15 [math] \mu m [/math] in every single trip when the squeegee is scanning the card. The required thickness for the paste is 35 [math] \mu m [/math] in average which will be achieved by Assembling two layers of paste.

ESEM Images and Results

ESEM Images and Results Using the Brush

The images below show a very thin layer of the paste is applied varies from 2 -5 [math] \mu m [/math] provides a paste uniform surface on FR4 cards. The paste is cured using 160 degree Celsius oven for an hour without causing any cracks or humps on the surface.

Burnedcard 5um middle scale 2.jpg Burnedcard 5um surface 1.jpg

ESEM Images and Results Using the Squeegee

The images below show that by a single scanning trip for the squeegee over the FR4 card applies a paste of thickness of 12-22 [math] \mu m \pm 10 \mu m [/math] depending on the surface topography. The first layer provide a smooth surface for the second layer which increase the paste thickness to 26-31 [math] \mu m \pm 10 \mu m[/math]. In some area the paste thickness may reach up to 50 [math] \mu m [/math]. Applying the paste by assembling does not make any boundaries the layers, so the final result shows the two layers as one layer within the micro scale.

L scale 001.jpg 2LAYER THICKNESS scale 1.jpg 2layers 35um 2 L sample2 001.jpg 2layers 35um 1.jpg 2layers 35um 2 L sample2 surface.jpg

Using the squeegee helped to have a paste thickness of 35 [math] \mu m \pm 10 \mu m [/math], the process is essentially dependent on the FR4 card's surface which varies topographically from one point to another due to the chemical etching for the copper clad that covers both surfaces of each card.

The disadvantage of assembling the paste layer that surface smoothness decreases and humps are observed within 500 [math] \mu m [/math] scale as shown in the picture below

Boundary 20um 10um surfacemixture.jpg

The reason of having these humps is unknown, but it is observed experimentally that it is not dependent on the FR4 surface, but sudden change in the card's temperature may be a factor, In addition to applying the paste as the card's temperature is higher than the room temperature as shown in the images below.

Boundary 20um 10um surfacemixture.jpg 2LAYER THICKNESS surface 1.jpg 70um 15min surface paste surfuce coparison.jpg

Additional Tests and Images

Copper Clad and Resistive Paste

The copper clad thickness that covers the FR4 widely changes as shown in the figure below (it is supposed to be 17 [math] \mu m [/math] ).

Cu thickness bothsides 1.jpg

The picture below shows the copper frame surrounds the resistive paste covered area.

Cu paste combination 1.jpg

A chemical analysis shows the structure of the resistive paste as shown below, an interference between the copper frame and the resistive paste were included in the analysis.

Paste points.jpg Paste.jpg

Burned Hole vs. Unburned One

A sample from a damaged card by discharge is tested. The thickness of resistive paste 3.5[math] \mu m [/math] in average. The second image shows the area of damage on the rim of the hole where the nonconformity in paste and FR4 interference surface is expected, these areas accumulate a lot of charges. Cleaning the hole with distilled water removes the paste and any burn residues which affect the surface uniformity in the cleaned area. The last image does not show any change in the chemical structure in the burned area, but shows a lot of tiny cracks in that area are filled with air.

Burnedcard 5um scale 1.jpg Burnedhole vs hole.jpg Burnedhole 5um surface 001.jpg Burnedhole hole ch withbackground.jpg