Table of Contents Table of Contents
Previous Page  24 / 74 Next Page
Information
Show Menu
Previous Page 24 / 74 Next Page
Page Background

A D V A N C E D M A T E R I A L S & P R O C E S S E S | S E P T E M B E R 2 0 1 5

2 4

Fig. 5 —

Prototype SIRTF telescope under-

going random vibration testing

[5]

.

Fig. 6 —

James Webb Space Telescope and its mirrors

[6]

. Bottom right shows an artist’s concep-

tion of the telescope optics with 18 primary mirror segments. On the bottom row are the three

different mirror segments as seen from the rear showing the honeycomb structure that makes

themboth light and stiff.

a vacuum deposited gold coating,

1000 Å thick. The convex secondary

mirror, though much more difficult to

polish and measure, has a similar figure

accuracy.

BERYLLIUM POWDER:

HISTORICAL DEVELOPMENT

Beryllium processing traditionally

starts with vacuum melting and casting

as part of the refining process. Vacuum

melting removes residual magnesium

and oxides that might be carried over

from the reduction of beryllium hydrox-

ide into metallic Be. Newly refined as

well as recycled Be are input to the vac-

uum casting step. Statically cast beryl-

lium features a typical ingot structure,

with large columnar grains growing

inward from the ingot wall and resid-

ual porosity toward the center. This

results in property anisotropy, difficulty

in machining, and overall poor proper-

ties. However, machining the ingot into

chips prior to powder processing helps

improve properties.

Powder metallurgy was used in

the late 1940s to produce fine grained,

nearly random-structured material.

The more random grain and crystal ori-

entation improves isotropy, particularly

the coefficient of thermal expansion,

while the finer grain size also improves

mechanical properties. In the 1940s and

50s, beryllium powder was produced

by two different processes—ball and

attrition milling. Ball milling uses a

rotating canister filled with chips and

grinding media generally made of hard-

ened steel. Beryllium was crushed and

sheared into fine powder flakes, but

this was a costly batch operation and

resulted in iron contamination from the

grinding media. Attrition milling was a

continuous process in which beryllium

chip was fed into a device similar to old

flour mills. The chip passed between

rotating and stationary berylliumplates

and was sheared into powder particles.

The weakest orientation of the Be crys-

tal is the basal plane and this process

produced powder particles with a slight

flake-like structure by fracture on the

basal plane.

Impact grinding to produce beryl-

lium powder was introduced as a

production process in the late 1970s

and uses high velocity gas to propel

Be chips against a Be target. The air

is dried, and then cools as it expands

from the nozzle. The impact shatters

the chips into powder particles, which

exhibit a blockier shape than ball or

attrition milled powder. This process

was also semi-continuous because

oversized powder particles could be

separated by air elutriation and fed

back into the impact stream. This

process improvement reduced iron and

oxygen contamination, and produced a

powder shape less related to the crys-

tal structure than other processes. This

was important during vibratory loading

of containers in preparation for consol-

idation into a solid. The flat surfaces of

flakes are all basal planes. When flakes

are loaded into a press to consolidate

them, they can stack up like a deck

of cards or sheets of mica, with large

regions having the same crystal orien-

tation. Each of these regions behaves

differently than adjoining regions with

different orientations. When the solid,

polished mirror cools, the difference

between regions gives the surface an

orange-peel appearance, which means

that each region is reflecting a beam of

light at a slightly different angle, scat-

tering the beam rather than reflecting

it in one uniform direction. Each time

the mirror is warmed and cooled, these

regions actually deform one another

slightly. Scattering is unpredictable

and changes on each cooling cycle.

Therefore, additional polishing is not a

solution.

The process that made the most

sense for making powder was inert

gas atomization, developed in the late

1980s for beryllium. A stream of liquid

metal is blasted into small droplets by