Which is the sharpest Nikon lens ever made??

[rodeo_joe1]

I can't find the actual Kodak data for it, but it definitely does 632.8nm from HeNe lasers. (And is red sensitive.)

And at that wavelength 2000 lppmm, or cycles/mm is definitely impossible!

 

As the saying goes "Show me the money!" Anybody can write an unprovable claim on paper.

 

And holography film isn't the same as a Lippmann emulsion. I believe holography plates capture the interference pattern in the depth of the emulsion. Not as a planar set of lines.

[glen_h]

(snip)

And holography film isn't the same as a Lippmann emulsion. I believe holography plates capture the interference pattern in the depth of the emulsion. Not as a planar set of lines.

 

There are two kinds.

 

I believe the 649F and SO-253 are the planar lines type.

 

The thick emulsion ones work for white light holograms.

The interference patterns is in the depth of the (relatively) thick emulsion, such that given a point

source of white light, only light of the appropriate (or close enough) wavelength contributes.

However, the emulsion shrinks in processing. (Silver takes up less space than silver nitrate,

and even more, the parts removed in fixing take up even less space.)

 

So, the image comes out green.

 

I don't remember the number for those, though.

[rodeo_joe1]

However, the emulsion shrinks in processing. (Silver takes up less space than silver nitrate,

and even more, the parts removed in fixing take up even less space.)

An interesting factoid (according to a book written by one of Kodak's top photo-chemists) - the reduced silver doesn't occupy the same position as its parent halide crystal. Apparently the silver is ejected as a filament from the surface development site(s) of the crystal. Often curling up as it grows, into small balls.

Therefore, strictly speaking, we never see exactly the same spatial relationship between image-forming points and negative silver-deposits. The difference is microscopic (or more like Electron-microscopic), but it's there nonetheless.

So if two adjacent halide crystals eject their silver in opposite directions, a 0.25 micron gap (required for a resolution of 2000 lppmm) might easily be healed up or spread to twice that distance.

 

So I still think any claim of resolving 2000 line-pairs per millimetre is pure hooey!

[glen_h]

An interesting factoid (according to a book written by one of Kodak's top photo-chemists) - the reduced silver doesn't occupy the same position as its parent halide crystal. Apparently the silver is ejected as a filament from the surface development site(s) of the crystal. Often curling up as it grows, into small balls.

Therefore, strictly speaking, we never see exactly the same spatial relationship between image-forming points and negative silver-deposits. The difference is microscopic (or more like Electron-microscopic), but it's there nonetheless.

So if two adjacent halide crystals eject their silver in opposite directions, a 0.25 micron gap (required for a resolution of 2000 lppmm) might easily be healed up or spread to twice that distance.

 

So I still think any claim of resolving 2000 line-pairs per millimetre is pure hooey!

 

The 2000 came from a Google search, so maybe as good as any other web search.

 

I do know that it is the same emulsion as holography film, though.

 

If you put a sharp (enough) edge against the emulsion, and shine light on it, you can get sharper than the wavelength.

That is, if the emulsion is thin enough, the diffraction around the edge will (mostly) happen after it is through.

 

Or, more specifically, it is near field so diffraction doesn't really count.

But that doesn't help much for holography.

 

And I never did see a real data sheet for it.

[rodeo_joe1]

If you put a sharp (enough) edge against the emulsion, and shine light on it, you can get sharper than the wavelength.

So the semiconductor industry completely wasted their time and money using hard UV to get contact reproductions from their chrome masters?

I don't think so.

[mike_halliwell]

using hard UV to get contact reproductions

but how low can they go?

 

It's in a vacuum already to avoid uv and 'air' clashes...!

[glen_h]

So the semiconductor industry completely wasted their time and money using hard UV to get contact reproductions from their chrome masters?

I don't think so.

 

I suspect photoresist is thick enough that, yes you need hard UV.

 

As well as I know it, the emulsion for holography films, other than those designed for white light holograms (described above), is thin relative to the wavelength.

 

Here is the description of near-field microscopy, resolving much smaller than the wavelength:

 

Near-field scanning optical microscope - Wikipedia

[Brian1664876441]

' date=' post: 5915295, member: 10999298"']Agree highly with both thoughts. 50/f2 Nikkor in a multicoated version. And for close up the old 55/3.5 Micro-Nikkor with compensating aperture would be hard to top.

The Micro-Nikkor-P 55/3.5 has the same optics as the Micro-Nikkor 55/3.5. There was a multi-coated version of the Micro-Nikkor-P 55/3.5, but not the compensating lens. I was given the Compensating version not long ago, goes great with my Nikon F Photomic "Bullseye", where the compensating feature is a must.

[rodeo_joe1]

I suspect photoresist is thick enough that, yes you need hard UV.

Photoresist was only used to produce the chrome masters. All the production contacts were made - using hard UV - on thin emulsion Lippman plates. And the chrome-master photoresist was spun on using a fast centrifuge to get an extremely thin coating when dried. The resist was dissolved off after etching the chrome in any case.

 

I later tried to contact reproduce a chrome master without a UV source. Best I could rustle up was a deep blue filter. The result was hopeless by former standards, but just good enough to use as a scanner focussing/resolution test plate.

 

If you haven't been there and done that, then you have no idea how much trouble and expense needs to be spent getting micron-scale resolution. Let alone anything sub-micron.

 

Near-field microscopy? Yeah, I'm sure we've all got a suitable laser and 20nm aperture lying around just itching to be put to use.

[rodeo_joe1]

I was given the Compensating version not long ago, goes great with my Nikon F Photomic "Bullseye", where the compensating feature is a must.

FWIW, quite a few of the 'cheap' Praktica Jena lenses had a close-focusing compensation mechanism. It simply consisted of having a precise slant in the lever connecting the aperture ring to the iris. The effect was that the aperture was only able to close to the effective aperture as marked on the ring. Attempting to close-focus at maximum aperture resulted in the aperture ring being pushed round to show the effective aperture.

 

Surprisingly, this was a feature of some of the older 'zebra stripe' wide angles. The 35mm f/2.8 Flektogon for example.

AFAIK, it was never applied to the longer lenses, which I would have thought would have benefitted more.

[glen_h]

The actual production process exposes photoresist on a silicon wafer with an SiO2 layer on it.

The photoresist is then developed, the oxide dissolved with HF through the holes,

and then the photoresist is removed. Dopants are diffused through the holes, and not where

blocked by the oxide.

 

But as you note, there are similar steps for making the masks used to expose the silicon

wafers, also done with photoresist.

 

I had an undergrad class where we learned the technology, though not quite to the micron scale.

Well, that we way before industry was down to micron scale.

 

I made six transistors on a 1cm square wafer.

 

First we made our masks with Polaroid 146L film.

(The only time I ever used a Polaroid roll film camera. I suspect it was close to the

end of the production of such film.)

 

Coated our wafer with a spinner, and dropping one drop of photoresist on it while it was spinning,

expose through our Polaroid film masks with a mercury lamp, develop the photoresist,

and etch with HF. Then into tube ovens for whichever step was needed.

 

There is one oven with wet oxygen for the thicker oxide for masking, one with dry oxygen

for thin oxide for MOS devices, and ones for p and n doping.

 

This one seems to use similar technology, though a different school, and

now for microfluidics instead of electronics:

 

Shipley 1813 - Bennett Lab Wiki - Rice University Campus Wiki

 

Specifically, we did use the Shipley photoresist.

 

But yes, I don't know the details for sub-micron technology, other than having

read some about them. But as a lab, that was a lot of fun!

[glen_h]

Never know what you might find, or what Britney Spears does:

 

Semiconductor Fabrication: Photolithography

[rodeo_joe1]

there are similar steps for making the masks used to expose the silicon

wafers, also done with photoresist.

No! As explained, the sub-master masks are contact-printed directly onto thin emulsion Lippmann plates. It's the definition of these production plates that limits the feature size.

[glen_h]

No! As explained, the sub-master masks are contact-printed directly onto thin emulsion Lippmann plates. It's the definition of these production plates that limits the feature size.

 

So there are no photoresist coated silicon wafers anymore?

[mike_halliwell]

So there are no photoresist coated silicon wafers anymore?

 

Guys, you 2 are well off the topic of Nikon's sharpest by some margin!

 

...although I've learnt a lot.;)

[glen_h]

Guys, you 2 are well off the topic of Nikon's sharpest by some margin!

 

...although I've learnt a lot.;)

 

Maybe some Nikon optics are used in semiconductor processing.

 

Note that this forum only requires a Nikon product to be used somewhere in generating any image posted.

(No image here, so that doesn't apply.)

It doesn't have to be an important part, just some part.

[mike_halliwell]

Maybe....?

 

A certainty considering Nikon's a big player in Semiconductor Lithography......:)

 

Just not very handy for BIF, Portraits or Landscapes or........:(

[rodeo_joe1]

Maybe some Nikon optics are used in semiconductor processing

They certainly are. We paid Nikon about 100K in mid-1970s prices for a 10x reduction step-and-repeat 'camera'. It probably went for scrap 10 years later.

 

Anyway. The quest for the 'sharpest' Nikon lens is like hunting a Snark. And about as meaningful.

 

They're all pretty sharp if you ignore LoCa and other foibles that Nikon seemed loathe to correct.