View previous topic :: View next topic |
Author |
Message |
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 03, 2012 18:09 Post subject: Re: About cyclic twinning - (5) |
|
|
Why did I not remember this?!!!! Too many d*)&%^ things going through my head, apparently, and too few remaining in there, apparently! That was a great discussion, and I was very impressed with the technology that was used to examine the twinning.
Pete
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
Elise
Joined: 22 Dec 2009
Posts: 243
Location: New York State
|
Posted: Oct 03, 2012 18:28 Post subject: Re: About cyclic twinning - (5) |
|
|
Hi Pete,
I only remembered it having perked up at the SEM images you posted and Duncan's reference to Karl's alexandrite paper...my short-term memory needs a bread crumb trail ;-)
I have been getting some experience with the SEM, realizing how hard it is to capture such images as those!
Elise
_________________ Elise Skalwold |
|
Back to top |
|
|
Dean Allum
Joined: 27 Jan 2010
Posts: 17
Location: Colorado
|
Posted: Oct 04, 2012 21:57 Post subject: Re: About cyclic twinning - (5) |
|
|
Wow, I was just about post a question about the atomic origins of crystal twinning.
Elise, Do you have a pdf copy of that nanoLetters paper?
Regards,
Dean ALlum
|
|
Back to top |
|
|
Peter Farquhar
Site Admin
Joined: 14 May 2011
Posts: 51
Location: Virginia
|
Posted: Oct 07, 2012 16:20 Post subject: Re: About cyclic twinning - (5) |
|
|
So far our discussion of cyclic twinning in rutile has covered sixlings and eightlings with a central cavity. In response to Pete's earlier question, though, I cannot find any cyclic twinned rutiles with a hole clear through the center.
On the other hand, we have not yet discussed cyclic twinned rutiles without a central cavity. Below is another type of rutile eightling -- one without the familiar dimple in the center. This old rutile specimen is from Graves Mountain, Georgia; it can be compared with more common eightlings with central cavities from Magnet Cove, Arkansas as shown earlier in this thread.
Peter Farquhar
Claremont, California
USA
Description: |
Rutile eightling Graves Mountain, Georgia USA main crystal is 3.0 x 2.5 x 2.7 cm Rutile eightling (PF-2965) from Graves Mtn. without a re-entrant cavity. Likely collected more than 100 years ago. |
|
Viewed: |
36738 Time(s) |
|
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 13, 2012 14:27 Post subject: Re: About cyclic twinning - (5) |
|
|
This is a great thread, with some fascinating specimens, perhaps the latest one in particular.
Here attached is a very small, but lovely arsenopyrite cyclic twin from Canada that's fun to see.
- and, to support the comments above on interpenetration in trillings, an unusual Brazilian chrysoberyl in which one of the trilling elements is visible in continuity across the twin (this is a side view).
Description: |
Arsenopyrite, cyclic twin Rogers Mine, Madoc, Ontario 1 mm twin |
|
Viewed: |
37086 Time(s) |
|
Description: |
Chrysoberyl, side view of a trilling Minas Gerais, Brazil 22 mm |
|
Viewed: |
36558 Time(s) |
|
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 13, 2012 19:35 Post subject: Re: About cyclic twinning - (5) |
|
|
Back to rutile - here are some (more or less tidy) sixlings.
Description: |
Rutile Tetikanana, central Madagascar. 20 mm |
|
Viewed: |
36548 Time(s) |
|
Description: |
Rutile Kipushi, DR Congo 10 mm |
|
Viewed: |
36704 Time(s) |
|
Description: |
Rutile Kipushi, DR Congo 13 mm |
|
Viewed: |
36542 Time(s) |
|
Description: |
Rutile Kipushi, DR Congo 12 mm |
|
Viewed: |
36593 Time(s) |
|
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 14, 2012 12:07 Post subject: Re: About cyclic twinning - (5) |
|
|
And three more eightlings
Description: |
Rutile eightling Parkesburg, Pennsylvania, USA 2 cm across |
|
Viewed: |
36524 Time(s) |
|
Description: |
Rutile eightling Magnet Cove, Arkansas, USA 10 mm |
|
Viewed: |
36560 Time(s) |
|
Description: |
Rutile eightling Magnet Cove, Arkansas, USA 13 mm |
|
Viewed: |
36588 Time(s) |
|
|
|
Back to top |
|
|
Peter Farquhar
Site Admin
Joined: 14 May 2011
Posts: 51
Location: Virginia
|
Posted: Oct 16, 2012 19:09 Post subject: Re: About cyclic twinning - (5) |
|
|
I have been fascinated by Pete Richards' earlier comments about whether rutile cyclic twins exist with a "real hole" in the center.
Pete Richards wrote: |
As for the hollow "wheel twins", I think they are a figment of an old morphological crystallographer's imagination. ... I have never seen a hollow cyclic twin, though certainly ones with "dimples" in the middle are common. |
I've been searching my collection and photo archives to see if I could find an example of a "holey" cyclic rutile. I now have located a couple of candidates, but I'll let Pete be the judge if the specimen below qualifies. This rutile appears to be an incomplete sixling with a hole completely through what would be the center if it were complete.
This unusual rutile specimen is from the alluvial deposits in the eastern region of Oubangui-Chari of French Equatorial Africa (now the Central African Republic). It's the only cyclic twin I've seen from this location, although I've seen others from Cameroon (to the west) and DR Congo (to the south). The specimen was once in Jean Chervet's collection, and dates back to the 1930s or earlier.
I'm wondering if anyone has any further information about rutile specimen localities in Central Africa, and ideas about where this specimen might have originated. I'd also appreciate learning more about the French mineralogist Jean Chervet (1904-1962).
My thanks to Ru Smith for the great photos, and to Pete Richards and others for contributing to this ongoing thread.
Peter Farquhar
Claremont, California
USA
Description: |
Rutile - partial cyclic twin (PF-2011) Oubangui Oriental, French Equatorial Africa (now Central African Republic) 5.2 x 4.6 x 3.2 cm Large cyclic twinned rutile specimen from Deyrolle, Paris (ex. Jean Chervet collection #287). |
|
Viewed: |
36403 Time(s) |
|
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 16, 2012 20:42 Post subject: Re: About cyclic twinning - (5) |
|
|
How fascinating Pete. Jean Chervet (1904-62) has the mineral chervétite named after him and the mineral charmenite is said to honour him. He seems to have worked on minerals from Mounana near Franceville in Gabon (I lived in Gabon for 5 years).
I see he wrote about radioactive minerals: Les minerais uranifères français (1960).
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 17, 2012 21:08 Post subject: Re: About cyclic twinning - (5) |
|
|
Here's a 2 cm Cumbrian marcasite twin (my thanks to Phil Grounds) to go with Pete Richards' picture near the top of this thread. Only three elements of the twin (not the full five shown in the drawings) are present here though. Does anyone have a complete one?
Description: |
Marcasite Frizington, Cumbria, England 2 cm |
|
Viewed: |
36350 Time(s) |
|
|
|
Back to top |
|
|
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 18, 2012 09:04 Post subject: Re: About cyclic twinning - (5) |
|
|
RutileFox wrote: | I have been fascinated by Pete Richards' earlier comments about whether rutile cyclic twins exist with a "real hole" in the center.
I've been searching my collection and photo archives to see if I could find an example of a "holey" cyclic rutile. I now have located a couple of candidates, but I'll let Pete be the judge if the specimen below qualifies. This rutile appears to be an incomplete sixling with a hole completely through what would be the center if it were complete.
Peter Farquhar
Claremont, California
USA |
Well, I don't particularly want to be the only "judge" on this matter. I would suggest that the growth center of this twin is at the base of the notch and at the top (point) of the wedge-shaped individual at the bottom of the twin. I suggest that if the other members of the twin were present, they would fill the notch, and a hole would not result.
I also note that the top left individual in this twin group bends "the wrong way". To contribute to formation of a cyclic twin, it should twin to the right, but instead it twins toward the back.
This illustrates one difficulty with the repeated (independent) twinning model of cyclic twin formation. Each time rutile twins on {011} at one end of the crystal, it has four equivalent "choices" of direction available, only one of which continues the formation of a cyclic twin. Completion of a sixling requires five twinnings. The chance of all five of these being in the right direction is one in four to the fifth power, or 1 in 1024. Not good odds!
It is interesting to compare Ru Smith's sixling from Kipushi (rutile sixling 4) with Peter Farquhar's earlier image from Atglen, Pa (August 22). The first has parallel sides for the vertically-striated individuals opposite each other, while in the second, opposite individuals are obviously not parallel. What does this mean?
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 18, 2012 09:21 Post subject: Re: About cyclic twinning - (5) |
|
|
Ru Smith wrote: | This is a great thread, with some fascinating specimens, perhaps the latest one in particular.
Here attached is a very small, but lovely arsenopyrite cyclic twin from Canada that's fun to see.
- and, to support the comments above on interpenetration in trillings, an unusual Brazilian chrysoberyl in which one of the trilling elements is visible in continuity across the twin (this is a side view). |
A wonderful little arsenopyrite twin. The top, bottom, and left individuals seem to form a perfect hexagon, the top right one is at least trying to conform, but the last element on the bottom right fails to fit rather badly. The true angle formed by the outer edges of arsenopyrite twins on {101} (equivalent to the edges of this hexagon) should be 111°, not 120°, so one must wonder how this twin formed!
With chrysoberyl, the situation is different, because the twin operation produces angles that differ from 120° by only a few minutes of arc. Thus, repeated twinning five times, simultaneous nucleation of six separate individual crystals in twinned orientation, and penetration twinning involving three individuals would produce almost the same external geometry.
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 18, 2012 09:56 Post subject: Re: About cyclic twinning - (5) |
|
|
Ru Smith wrote: | Back to rutile - here are some (more or less tidy) sixlings. |
I suggest that "Rutile sixling 1" is not really a sixling. It is hard to tell how the bottom half relates to the top half, because of the region of apparent contact with the matrix that lacks well-formed crystal faces. However, I think the top half is a simple twin with the twin plane vertical. The faces on the left and right have a different appearance from the faces on the top. I think the left and right faces belong to the termination of the rutile crystal, whereas the faces on top belong to the prism, and have stronger rounding and more striations.
The cartoon below shows how a simply twinned crystal, if distorted, would produce such a shape. The blue lines represent extensions of the bottom half of the twin downward; the orange lines represent parts of the theoretical crystal that are not present because of contact with the bottom half of the aggregate.
Description: |
Rutile Hypothetical twin on {011}, with distortion. SHAPE drawing. |
|
Viewed: |
36185 Time(s) |
|
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
Peter Farquhar
Site Admin
Joined: 14 May 2011
Posts: 51
Location: Virginia
|
Posted: Oct 18, 2012 11:33 Post subject: Re: About cyclic twinning - (5) |
|
|
Pete, thanks for your thoughts on the Ubangi rutile specimen; I appreciate your expertise and will keep looking for more holes.
A question about the Kipushi locality: Jordi Fabre noted several months ago, "Although the first locality given in the market was D.R. of Congo, we can give the correct locality as Zambia, close to the D.R. of Congo border. Kinyanfumbe Zambia."
Does anyone know exactly where Kinyanfumbe is located in Zambia? I cannot find it on any map, even though this location is now used in place of Kipushi.
Peter Farquhar
Claremont, California
USA
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 18, 2012 18:55 Post subject: Re: About cyclic twinning - (5) |
|
|
Good question Peter. I can't find it on the map either. Kipushi seems to be right on the border.
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 18, 2012 18:59 Post subject: Re: About cyclic twinning - (5) |
|
|
Pete Richards wrote: | Ru Smith wrote: | Back to rutile - here are some (more or less tidy) sixlings. |
I suggest that "Rutile sixling 1" is not really a sixling. It is hard to tell how the bottom half relates to the top half, because of the region of apparent contact with the matrix that lacks well-formed crystal faces. However, I think the top half is a simple twin with the twin plane vertical. The faces on the left and right have a different appearance from the faces on the top. I think the left and right faces belong to the termination of the rutile crystal, whereas the faces on top belong to the prism, and have stronger rounding and more striations.
The cartoon below shows how a simply twinned crystal, if distorted, would produce such a shape. The blue lines represent extensions of the bottom half of the twin downward; the orange lines represent parts of the theoretical crystal that are not present because of contact with the bottom half of the aggregate. |
Beautiful cartoon Pete and yes the top two limbs of the twin do dominate the image. Here attached is a quick photo rotating the twin to the right so that you can see the adjacent two limbs. What do you use for the crystal sketches?
Description: |
|
Viewed: |
36149 Time(s) |
|
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 21, 2012 11:49 Post subject: Re: About cyclic twinning - (5) |
|
|
Pete Richards wrote: | It is interesting to compare Ru Smith's sixling from Kipushi (rutile sixling 4) with Peter Farquhar's earlier image from Atglen, Pa (August 22). The first has parallel sides for the vertically-striated individuals opposite each other, while in the second, opposite individuals are obviously not parallel. What does this mean? |
Does this not imply that the parallel-sided sixlings are not twinned on {101}? So, is there a different crystallographic direction (at about 60 degrees to the c-axis) that does the job? and are such parallel-sided sixlings known from elsewhere? - or is there another explanation?
|
|
Back to top |
|
|
Ru Smith
Joined: 13 Oct 2012
Posts: 362
|
Posted: Oct 22, 2012 12:15 Post subject: Re: About cyclic twinning - (5) |
|
|
See if this picture helps with the thinking.
The left hand three elements of the twin seem to result from a small "kink" near the centre of the twin which amplified as crystal growth progressed. I've "joined-the dots" of the kink axes. The left hand vertical limb of the twin gives the impression of being "extruded" to the left without any rotation, so of course the left hand vertical element is parallel to the right hand vertical element.
Now, the interlimb angles (measured with a protractor on the screen I'm embarrassed to say) are all less than 120 degrees (if my trig is good it should be 114.43, from 2xATAN(4.5937/2.9587)). So, we ARE looking at twinning on {101}. (I had a quick look for other crystallographic directions that would give 60 degrees but found 59.9 degrees for {10 0 9} and thought "No").
Now, look at the top left of the image and we can see the angular mismatch of that portion of the cycle grown from the right-hand and left-hand vertical limbs. It looks very close to the 16.71 degrees mismatch to be expected (if my trig is good).
So, that was fun. Apologies if I'm not using the right language (crystallography is a distant memory of a class from more than a couple of decades ago).
Description: |
|
Viewed: |
35873 Time(s) |
|
|
|
Back to top |
|
|
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 23, 2012 14:11 Post subject: Re: About cyclic twinning - (5) |
|
|
Ru Smith wrote: | Pete Richards wrote: | It is interesting to compare Ru Smith's sixling from Kipushi (rutile sixling 4) with Peter Farquhar's earlier image from Atglen, Pa (August 22). The first has parallel sides for the vertically-striated individuals opposite each other, while in the second, opposite individuals are obviously not parallel. What does this mean? |
Does this not imply that the parallel-sided sixlings are not twinned on {101}? So, is there a different crystallographic direction (at about 60 degrees to the c-axis) that does the job? and are such parallel-sided sixlings known from elsewhere? - or is there another explanation? |
Yes, strictly speaking, if the limbs are parallel sided they cannot belong to a cyclic twin on {101}. And no, there are no other twin laws that are even approximately right. The only other common one that I know of is {301}, which makes a roughly 60° angle between the limbs. In order to have opposite limbs be parallel (other than being part of the same individual) the misfit would have to somehow be equally split among the individuals, in which case the angle between them would not conform to the angle for the twin on {101}.
This is why I like the idea of an epitactic origin for some of these twins - use hematite or ilmenite as a template and you'll get perfect 120° angles, that approximate twinning on {101}.
In the picture, Ru has done a great job of marking off the individuals on the left half. I would say that all of the vertically-striated material is part of the same individual. The two wedges on the upper and lower left are twinned to it on two different faces of the form {101}, and then kink back to the vertically striated individual. On the right there are equivalent individuals top-right and bottom-right which are probably also {101} twins. There's even a second twin on the top right. These left and right individuals do not meet up, but if they did that's where the misfit would have to be accommodated.
A fun specimen!
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
Pete Richards
Site Admin
Joined: 29 Dec 2008
Posts: 828
Location: Northeast Ohio
|
Posted: Oct 25, 2012 15:01 Post subject: Re: About cyclic twinning - (5) |
|
|
To continue this thread, here are three pictures of cyclic twins of arsenopyrite and marcasite. The marcasite twin only has four of the five limbs visible, and it's not clear that the fifth one exists at all.
The arsenopyrite twin has all five limbs (labled I to V on the right), but in addition crystal I projects to the left as well as to the right, and has two related twins on the left side as well.
I've not actually measured the angles on these, but I think they conform to the theoretical ones.
Description: |
Arsenopyrite cyclic twin Plata, Val Medel, Graubunden, Switzerland Twin aggregate is 0.8 mm across |
|
Viewed: |
36921 Time(s) |
|
Description: |
Arsenopyrite cyclic twin Plata, Val Medel, Graubunden, Switzerland Twin aggregate is 0.8 mm across |
|
Viewed: |
35791 Time(s) |
|
Description: |
Marcasite cyclic twin Cedarville, Greene Co., Ohio, USA twin is 1.2 mm wide |
|
Viewed: |
35843 Time(s) |
|
_________________ Collecting and studying crystals with interesting habits, twinning, and epitaxy |
|
Back to top |
|
|
|