All Screwed Up

http://sciam.com/article.cfm?chanID=...4F83414B7F0000

November 03, 2003

All Screwed Up
An obscure property of light puts a spin on astronomy
By George Musser


You'd think we'd have figured out light by now. Kids learn about
prisms and lenses in elementary school, people wear Maxwell's
equations on T-shirts, and the quantum version of those equations
is the most precise theory in science. Yet knotted up within the
theory is a phenomenon that physicists are still unraveling: an
unexplored property of light.

[...]

One peculiar aspect of twisted light could prove especially
endearing to astronomers. Just as Earth's North Pole sits in
every time zone, the central axis of the beam contains waves
of every phase. All those waves cancel one another out, leaving
utter blackness. As a result, a lens focuses twisted light to a
ring instead of a point. In 2001 physicist Grover Swartzlander
of the University of Arizona proposed using this feature to
look for extrasolar planets. Installed in a telescope, one of
the special diffraction gratings would smear starlight into a
ring, leaving a hole so dark that a nearby object millions or
billions of times as faint could become visible. "It's a
completely original idea," Padgett says. "When I first read
the paper, I said, 'Gosh, that's a cute idea.'" Contemporaries
of Newton probably thought it pretty cute that white light
could be split into a rainbow of colors. Maybe one day twisted
light will come to seem just as commonplace.

George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. Quantum mechanically, it is associated with the orbital
angular momentum of the photon and can have an infinite number of
discrete values. This is a somewhat obscure concept even among
physicists, so I'm not surprised there was confusion in the newsgroup.

Twisted waves can be oriented any which way; the question is where the
waves are in their oscillation cycle. In principle, you could have
twisted, unpolarized light or untwisted, polarized light. The
righthand diagram in the article shows twisted, polarized light. I
wish the diagram had turned out better -- in the available time, it
was a struggle to make it correct, let alone clear -- but if you study
it closely, the physics might become clearer.

The introduction of a twist produces a deep null for light along the
optical axis. In effect, on-axis light is bent away from the
axis. (The axis is a singularity where there is a superposition, and
therefore complete cancellation, of all wave phases.) Off-axis light,
including light from a planet or non-pointlike star, can be scattered
toward the optical axis. So, if you put a mask at the focal plane,
you can capture this scattered off-axis light and exclude the on-axis
light. In this way, the system can be used to observe exoplanets.
You won't get an image of the planet, at least not directly, but you
can study its light photometrically and spectroscopically.


(I was one of the people who had confused twisted and circularly
polarized light, so I should recant my post on this subject.)

--
Lt. Lazio, HTML police | e-mail:
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html

George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. Quantum mechanically, it is associated with the orbital
angular momentum of the photon and can have an infinite number of
discrete values. This is a somewhat obscure concept even among
physicists, so I'm not surprised there was confusion in the newsgroup.

Twisted waves can be oriented any which way; the question is where the
waves are in their oscillation cycle. In principle, you could have
twisted, unpolarized light or untwisted, polarized light. The
righthand diagram in the article shows twisted, polarized light. I
wish the diagram had turned out better -- in the available time, it
was a struggle to make it correct, let alone clear -- but if you study
it closely, the physics might become clearer.

The introduction of a twist produces a deep null for light along the
optical axis. In effect, on-axis light is bent away from the
axis. (The axis is a singularity where there is a superposition, and
therefore complete cancellation, of all wave phases.) Off-axis light,
including light from a planet or non-pointlike star, can be scattered
toward the optical axis. So, if you put a mask at the focal plane,
you can capture this scattered off-axis light and exclude the on-axis
light. In this way, the system can be used to observe exoplanets.
You won't get an image of the planet, at least not directly, but you
can study its light photometrically and spectroscopically.


(I was one of the people who had confused twisted and circularly
polarized light, so I should recant my post on this subject.)

--
Lt. Lazio, HTML police | e-mail:
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html

Joseph Lazio wrote:
George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. ...


Doing a reread and a quick web search and... I was rather too hasty in
my first misinterpretation. Instead, quite an interesting idea.

The key thing to note is that the description is of a _beam_ light.

Also, that the idea of focusing a twisted light beam to a ring to null
out the image of a central star involves using a helical lens or special
holographic grating /on the observing telescope/ to do the trick. This
raises the question of how the resolution of this compares with that of
doing the same kind of thing but with nulling interference between two
discrete telescopes. (See the recent post from Alfred A. Aburto Jr.
Direct Imaging of Extrasolar Planets.)


A few interesting links a

(Amusing)
http://habitat.igc.org/twisted-pairs/
http://www.cs.brandeis.edu/~xray/photos/1994/94g.html

(LCDs)
http://sharp-world.com/sc/library/lcd_e/s2_1_1e.htm

(The Sci Am article, watch out for line wrap)
http://sciam.com/article.cfm?chanID=...4F83414B7F0000

(PDF paper nicely describing twisted light and molecular chirality)
http://arxiv.org/pdf/physics/0305002



Our postings appear to be the most prominent of the Google offerings!

Regards,
Martin



--
---------- Give a man a fish and you have fed him for today.
- Martin - Teach him how to fish and he won't bother you for weeks!
- 53N 1W - - Anon
----------

Joseph Lazio wrote:
George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. ...


Doing a reread and a quick web search and... I was rather too hasty in
my first misinterpretation. Instead, quite an interesting idea.

The key thing to note is that the description is of a _beam_ light.

Also, that the idea of focusing a twisted light beam to a ring to null
out the image of a central star involves using a helical lens or special
holographic grating /on the observing telescope/ to do the trick. This
raises the question of how the resolution of this compares with that of
doing the same kind of thing but with nulling interference between two
discrete telescopes. (See the recent post from Alfred A. Aburto Jr.
Direct Imaging of Extrasolar Planets.)


A few interesting links a

(Amusing)
http://habitat.igc.org/twisted-pairs/
http://www.cs.brandeis.edu/~xray/photos/1994/94g.html

(LCDs)
http://sharp-world.com/sc/library/lcd_e/s2_1_1e.htm

(The Sci Am article, watch out for line wrap)
http://sciam.com/article.cfm?chanID=...4F83414B7F0000

(PDF paper nicely describing twisted light and molecular chirality)
http://arxiv.org/pdf/physics/0305002



Our postings appear to be the most prominent of the Google offerings!

Regards,
Martin



--
---------- Give a man a fish and you have fed him for today.
- Martin - Teach him how to fish and he won't bother you for weeks!
- 53N 1W - - Anon
----------

A good web site to visit regarding this phenomena of "twisted light":

http://departments.colgate.edu/physi...s/oamgp/gp.htm
(referenced in the SCIAM Nov 03 article, thanks George Musser &
Joseph Lazio, very interesting ...)

Al


"Joseph Lazio" wrote in message
...
George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. Quantum mechanically, it is associated with the orbital
angular momentum of the photon and can have an infinite number of
discrete values. This is a somewhat obscure concept even among
physicists, so I'm not surprised there was confusion in the newsgroup.

Twisted waves can be oriented any which way; the question is where the
waves are in their oscillation cycle. In principle, you could have
twisted, unpolarized light or untwisted, polarized light. The
righthand diagram in the article shows twisted, polarized light. I
wish the diagram had turned out better -- in the available time, it
was a struggle to make it correct, let alone clear -- but if you study
it closely, the physics might become clearer.

The introduction of a twist produces a deep null for light along the
optical axis. In effect, on-axis light is bent away from the
axis. (The axis is a singularity where there is a superposition, and
therefore complete cancellation, of all wave phases.) Off-axis light,
including light from a planet or non-pointlike star, can be scattered
toward the optical axis. So, if you put a mask at the focal plane,
you can capture this scattered off-axis light and exclude the on-axis
light. In this way, the system can be used to observe exoplanets.
You won't get an image of the planet, at least not directly, but you
can study its light photometrically and spectroscopically.


(I was one of the people who had confused twisted and circularly
polarized light, so I should recant my post on this subject.)

--
Lt. Lazio, HTML police | e-mail:
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html

A good web site to visit regarding this phenomena of "twisted light":

http://departments.colgate.edu/physi...s/oamgp/gp.htm
(referenced in the SCIAM Nov 03 article, thanks George Musser &
Joseph Lazio, very interesting ...)

Al


"Joseph Lazio" wrote in message
...
George Musser recently came across the discussion about "twisted
light" that occurred in this newsgroup. He's been having Usenet
access problems, so he asked me to post this:


Hi everyone,

I only just came across the discussion that took place several weeks
ago about my article on "twisted light" in the November issue of
Scientific American. To clarify: Twisted light is NOT the same as
circularly polarized light. Polarization refers to the orientation of
the electromagnetic wave in space. Quantum mechanically, it is
associated with the photon spin and has two senses, such as right- or
left-handed. Twisting, on the other hand, refers to the azimuthal
structure of a cylindrical light beam -- the phasing of the wave as a
function of angle on a plane perpendicular to the direction of
propagation. Quantum mechanically, it is associated with the orbital
angular momentum of the photon and can have an infinite number of
discrete values. This is a somewhat obscure concept even among
physicists, so I'm not surprised there was confusion in the newsgroup.

Twisted waves can be oriented any which way; the question is where the
waves are in their oscillation cycle. In principle, you could have
twisted, unpolarized light or untwisted, polarized light. The
righthand diagram in the article shows twisted, polarized light. I
wish the diagram had turned out better -- in the available time, it
was a struggle to make it correct, let alone clear -- but if you study
it closely, the physics might become clearer.

The introduction of a twist produces a deep null for light along the
optical axis. In effect, on-axis light is bent away from the
axis. (The axis is a singularity where there is a superposition, and
therefore complete cancellation, of all wave phases.) Off-axis light,
including light from a planet or non-pointlike star, can be scattered
toward the optical axis. So, if you put a mask at the focal plane,
you can capture this scattered off-axis light and exclude the on-axis
light. In this way, the system can be used to observe exoplanets.
You won't get an image of the planet, at least not directly, but you
can study its light photometrically and spectroscopically.


(I was one of the people who had confused twisted and circularly
polarized light, so I should recant my post on this subject.)

--
Lt. Lazio, HTML police | e-mail:
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html

Alfred A. Aburto Jr. wrote:
A good web site to visit regarding this phenomena of "twisted light":

http://departments.colgate.edu/physi...s/oamgp/gp.htm
(referenced in the SCIAM Nov 03 article, thanks George Musser &
Joseph Lazio, very interesting ...)

Al

That's a very good link with a good graphical explanation and
demonstrations. Almost spooky when you're used to seeing the more usual
interference patterns. (And not a single twisted in there... OK... Just
one twist (:-))

Thanks to George and Joseph for pulling the thread back into the correct
light.

Regards,
Martin


--
---------- Give a man a fish and you have fed him for today.
- Martin - Teach him how to fish and he won't bother you for weeks!
- 53N 1W - - Anon
----------

Alfred A. Aburto Jr. wrote:
A good web site to visit regarding this phenomena of "twisted light":

http://departments.colgate.edu/physi...s/oamgp/gp.htm
(referenced in the SCIAM Nov 03 article, thanks George Musser &
Joseph Lazio, very interesting ...)

Al

That's a very good link with a good graphical explanation and
demonstrations. Almost spooky when you're used to seeing the more usual
interference patterns. (And not a single twisted in there... OK... Just
one twist (:-))

Thanks to George and Joseph for pulling the thread back into the correct
light.

Regards,
Martin


--
---------- Give a man a fish and you have fed him for today.
- Martin - Teach him how to fish and he won't bother you for weeks!
- 53N 1W - - Anon
----------