Sciencemadness Discussion Board - Arsenic from Skutterudite

Arsenic from Skutterudite

Pages: 1 2

elementcollector1 - 3-5-2014 at 14:54

Ah yes, the hydroxythingymejibs. A fine anion.

Anyway, took the remaining iron/cobalt solution and added tons more skutterudite. It's still reacting, but wouldn't react with the coarser, gray material left over from the first dissolution. I'm tempted to think that this is elemental arsenic, but the crystal grain size and color suggest coarser skutterudite that didn't dissolve as quickly.

Also: Upon diluting the wastewater (which is going to be contained in concrete when it's done), a lighter gray powder precipitated out. This is either also arsenic, a mixture of arsenic and arsenic trioxide, or the Philosopher's Stone.

Per the forensic test: Inconclusive, as so far the liquor is still too acidic to avoid copper dissolution.

[Edited on 5-4-2014 by elementcollector1]

elementcollector1 - 4-5-2014 at 20:43

So I filtered the cobalt-iron liquor through a bunch more skutterudite ore, in order to get as much in solution as I could. Then, I put a large piece of coiled up Cu wire in, and waited overnight. The results were interesting: Tons more hydroxythingymejibs (to the point where I could not distinguish the color of the solution, everything was opaque yellow) and no precipitation whatsoever on the copper, which appeared to be mildly etched. This pretty conclusively proves to me that copper is the cause of this precipitation, because at no point previously did it occur. In addition, this also presents conclusive evidence that all the arsenic is reacted away to insoluble arsenic trioxide (which is what the powder in the above photo probably is), and none is present in solution. Good!

Now, to distill some more nitric acid and dissolve all of the skutterudite. After that, I'm not sure what to do to separate the arsenic from the oxygen: Heat with powdered burnt-flour charcoal in a test tube to sublime off the arsenic? Re-dissolve with HCl, use hypophosphite?

I think I might go with the first one, as I can see it playing out well if I'm careful. What I would do is load the dried reactants into a standard test tube, evacuate it using my vacuum pump (it's not perfect vacuum, but it should work for this purpose), and then blowtorch the reactant mix, taking care to drive all of the arsenic into the topmost part of the tube. Once the reaction is done, I would smash the tube open and collect the arsenic.

However, I would probably have to design this so that the pressure when I'm finished is equal or close to 1 atm (it will increase due to the production of CO₂). Would it be possible to measure the volume of the test tube, work backwards from the ideal gas-molar relationship (22.4 L gas = 1 mol gas), and calculate the moles C / As₂O₃ to put in? Or would I have to calculate differently?

blogfast25 - 5-5-2014 at 04:29

Firstly, you need to establish whether the residue is metallic As or As2O3. I seem to recall that elemental As dissolves in hypochlorite.

A sublimation point of 615 C is quite high for a test tube.

Quote: Originally posted by elementcollector1

However, I would probably have to design this so that the pressure when I'm finished is equal or close to 1 atm (it will increase due to the production of CO₂). Would it be possible to measure the volume of the test tube, work backwards from the ideal gas-molar relationship (22.4 L gas = 1 mol gas), and calculate the moles C / As₂O₃ to put in? Or would I have to calculate differently?

You'll find that the amount of reagents needed when calculated that way will be very, very small.

I also very much doubt if a Bunsen burner can deliver the level of heat needed for the carbothermic reduction of As2O3. Without being an expert in any way, I think you will need red heat at least...

A carbothermic reduction of SnO2 I carried out years ago required intense charcoal heat and lots of time to complete.

[Edited on 5-5-2014 by blogfast25]

elementcollector1 - 5-5-2014 at 09:05

The residue is much closer to white than gray, leading me to believe it's just As₂O₃ with some leftover ore mixed in.

True enough - I was hoping for something significant, but once I think about it I shouldn't have expected much.

My blowtorch can melt glass, aluminum, and heat steel to orange temperatures, but you're right in that it can't deliver even heat for hours on end. Maybe a home furnace, same as used for melting aluminum? Or just go back to the wet route...

Also, you would think the BP of arsenic would be lower than 615 degrees C under vacuum, although I can't seem to find a single phase diagram.

blogfast25 - 5-5-2014 at 10:06

I'd say, 'go wet'. Dissolve the oxide in something and plate out the arsenic. On copper for instance.

I don't think it's worth potentially poisoning yourself with some ad hoc apparatus.

elementcollector1 - 5-5-2014 at 10:10

Heh. True that.
You think plating out will accomplish a decent yield, say with copper and AsCl₃?

blogfast25 - 5-5-2014 at 12:21

Quote: Originally posted by elementcollector1

Heh. True that.
You think plating out will accomplish a decent yield, say with copper and AsCl₃?

Should b near 100 %, I imagine.

blogfast25 - 6-5-2014 at 04:50

Or you could prepare an 'As mirror', based on the chemistry of the forensic Marsh test:

http://www.emsb.qc.ca/laurenhill/science/forensic.html

elementcollector1 - 6-5-2014 at 15:25

Uh... No thanks. I'll just stick with plating - no heat, no arsine...

blogfast25 - 7-5-2014 at 08:17

No arsine, no dead *rseholes...

elementcollector1 - 7-5-2014 at 08:57

True enough! Bought some more ammonium nitrate for distillation purposes, can't wait to try that again.
What am I going to do with all this (NH₄)₂SO₄ / NH₄HSO₄?

Although, if plating has such yields, why isn't it being used instead of messing around with hypophosphite / tin chlorides? I suspect purity might be less than satisfactory, even if yield is good...

blogfast25 - 8-5-2014 at 05:20

Quote: Originally posted by elementcollector1

Although, if plating has such yields, why isn't it being used instead of messing around with hypophosphite / tin chlorides?

There could be all kinds of reasons for this, such as economical. Brute force carbothermy with sublimation is simple and might be cheaper on a larger scale.

Plating can yield extraordinary purity: see electroplated copper for conductors.

elementcollector1 - 11-5-2014 at 21:50

Well, collected more of that stuff via an odd process: The whiter substance easily forms a very temporary suspension in the solution, so I stirred heavily, decanted off the mixture into a separate beaker, waited for it to settle, then decanted back. This got (by rough volume measurement) ~40g of grayish arsenic trioxide. (This is so much more than I actually need). There's still some skutterudite left in the reaction beaker, so I'll leave it to digest for a few more days. I ran out of nitric acid, and am hesitant to distill more because of the water costs of distilling under vacuum with my current setup. Fortunately, the liquid still appears to be dissolving the ore - just much more slowly.

So far, I've worked out a few reactions for this.
The initial step of dissolution:
2 Co + 6 HNO₃ -> 2 Co(NO₃)₃ + 3 H₂
2 Fe + 6 HNO₃ -> 2 Fe(NO₃)₃ + 3 H₂
Arsenic appears to be unaffected at this stage, but given that it's the same color as the inital ore I am not sure.

At higher concentrations of nitric acid / longer reaction times,

2 As + 6 HNO₃ -> As₂O₃ + 6 NO₂ + 3 H₂O

This was evidenced by copious release of NO₂, red fumes that quickly began spilling out of the beaker. In addition, the white powder that results is almost always partially floating on top and releasing gas bubbles. While these are not red (perhaps suggesting NO?), they could also be hydrogen or some other similar result.

Wiki notes that arsenic trioxide will dissolve in HCl - so all that's left to do is dissolve the oxide into concentrated, pure HCl, and plate out with copper?
I wonder what I should do with the leftover solution - maybe try to separate the cobalt from the iron, just for kicks?

blogfast25 - 12-5-2014 at 04:54

Quote: Originally posted by elementcollector1

Wiki notes that arsenic trioxide will dissolve in HCl - so all that's left to do is dissolve the oxide into concentrated, pure HCl, and plate out with copper?
I wonder what I should do with the leftover solution - maybe try to separate the cobalt from the iron, just for kicks?

It should dissolve in strong HCl but if it doesn't try strong NaOH, then neutralise and precipitate back to the oxide. 'Fresh' oxide should dissolve well.

Separating Fe and Co should be interesting. Co forms an ammine complex, if memory serves me.

Arsenic Confirmed

elementcollector1 - 14-5-2014 at 13:59

Welp. Plating worked perfectly... so far.

That there is a solution of AsCl₃, made by treating a small amount of the white powder with strong HCl. Odd olive-green color, but whatever. Put a heavy-gauge copper wire in, and the As immediately formed a beautiful mirror that I would love to put in Pretty Pictures if I had a decent camera.

And now, on to larger-scale plating... I just need about 1mL or so of this, to fill up the ampoule - judging from the initial speed of the plating reaction, this should be easy.

EDIT: Clearly I did not think this through - the plated layer of As blocks further deposition (no copper to react with), and shaking it isn't helping, indicating good adherence.

I wonder if I should electroplate the larger solution instead?

[Edited on 5-14-2014 by elementcollector1]

Mailinmypocket - 14-5-2014 at 14:39

Very cool EC1! I just saw this and wanted to try it myself since I have no elemental As sample. I dissolved a small amount of commercial As2O3 in HCl, the solution was clear but after adding the copper slowly became light green. I witnessed the same beautiful dark mirror as well, nice! Unfortunately it's very hard to make the camera focus on a piece of wire so that didn't work so well but if you could find some thin copper sheet or pipe it might be really neat

elementcollector1 - 14-5-2014 at 15:49

Curiously, mine went green *before* addition of copper... and given my HCl was water-clear, I'm guessing this might be an iron impurity from the original feed (Co, Fe, Ni)As₃.
On the larger scale, it was much the same - olive green solution.

Even more curiously, after most of the white powder dissolved into the HCl, a small amount of darker, crystalline powder was left behind. I'm having a hard time discerning if this itself is elemental arsenic or if it's just the skutterudite ore, as both are dark metallic powders. If I get a sample dried out, then I'll know - the skutterudite powder was roughly the color of a medium pencil mark on a paper, while the arsenic is naturally much darker (rather like a Taiwanese black pearl).

blogfast25 - 15-5-2014 at 04:01

Nice.

It could be worth doing the following mini-experiment. Take a small piece of the dried As plated wire and drop it into a clean test tube. Heat it strongly with Bunsen/blowtorch: the As should sublime and maybe deposit itself on the glass as a mirror.

Alexander Gettler (the famous US toxicologist) used this to distinguish As from Bi because Bi also plates out on copper but doesn't sublime in this test.

[Edited on 15-5-2014 by blogfast25]

elementcollector1 - 15-5-2014 at 18:14

Got a bunch of white, red, and green sublimates doing this... Not enough arsenic, or too much oxygen?
On a side note, electroplating doesn't do anything in terms of yield!

Mailinmypocket - 15-5-2014 at 18:28

Quote: Originally posted by blogfast25

Nice.

It could be worth doing the following mini-experiment. Take a small piece of the dried As plated wire and drop it into a clean test tube. Heat it strongly with Bunsen/blowtorch: the As should sublime and maybe deposit itself on the glass as a mirror.

Alexander Gettler (the famous US toxicologist) used this to distinguish As from Bi because Bi also plates out on copper but doesn't sublime in this test.

[Edited on 15-5-2014 by blogfast25]

I gave this a shot and it seems to have worked.

Here is the bit of wire that had As plated onto it. The test tube was clamped horizontally and heated strongly. Sorry for the dumb Mc Donalds ad... It's probably worse for you than the amount of As plated on the wire.

Eventually a black zone surrounded the wire. I stopped heating.

Viewed from underneath, a dark mirror.

elementcollector1 - 15-5-2014 at 19:12

Huh. Lot less contamination than mine - maybe something in the initial solution is messing it up?

[Edited on 5-16-2014 by elementcollector1]

blogfast25 - 16-5-2014 at 04:43

@MIMP:

Nice. Basically you've got a sample of relatively pure As.

@EC1:

What makes you say that?

You could purify by using As(III) 'acidity'. Add strong NaOH to precipitate iron (and others), Boil and As goes to sodium arsenite. Filter. Then cool and re-acidify filtrate to AsCl3 and plate out.

[Edited on 16-5-2014 by blogfast25]

elementcollector1 - 16-5-2014 at 15:17

Mainly due to the variety of colors produced upon sublimation, and that the solution was dark green to start with (well before addition of copper).

blogfast25 - 17-5-2014 at 04:34

Quote: Originally posted by elementcollector1

Mainly due to the variety of colors produced upon sublimation, [...]

These colours aren't by any chance caused by thin-film interference, now are they?

http://en.wikipedia.org/wiki/Thin-film_interference

elementcollector1 - 17-5-2014 at 10:06

I don't think so - they're pretty constant and well bounded. The red is where the copper wire sunk into the partially melted glass tube, and the green is all around that. The white is at the edges.

elementcollector1 - 17-5-2014 at 11:50

I now have fairly conclusive evidence that copper is the noted impurity.
For starters, the green solution from earlier was filtered and mixed with very concentrated NaOH solution. This produced a yellow precipitate at first, but upon addition of more NaOH this darkened to a green and then brown precipitant, finally settling at dark brown. (Likely CuO.)
Then, upon filtering the solution, it was first yellow, then brown (these are likely colloidal colors), then green (same olive green as starting), and now it's dark blue, leading me to suspect sodium cuprate. Another possibility is chromium, yet although being more common in minerals like skutterudite it doesn't explain the brown colors.
I suppose I'll have to reacidify this with HCl until the mess of arsenic oxide and copper hydroxide precipitates, then... do something to separate those two?

blogfast25 - 18-5-2014 at 05:26

If there really is copper present, take a small amount of the original solution and add an excess of strong ammonia to it. The deep blue of copper (II) ammine complex is hard to mistake.

Copper would initially precipitate with NaOH to blue Cu(OH)2, which in very strong NaOH will dissolve to deep blue cuprate. CuO would only form on heating of the Cu(OH)2. The green could be explained by tetrachlorocuprate (but only if there's a lot of chloride in the solution).

You should also test for iron with K (or NH4) thiocyanate.

elementcollector1 - 18-5-2014 at 09:24

The solution was extremely hot when precipitation was happening, likely due to NaOH neutralizing tons of excess HCl (of which there was assuredly plenty).
Don't have any thiocyanate, so that's a no-go.

elementcollector1 - 18-5-2014 at 14:43

Results are in - this is not copper. Ammonia had no effect on the brown precipitate (which, if CuO, should have reacted immediately). The only other candidate I can think of is chromium, but given the prevalence of blue and violet colors at high pH, this seems so unlikely. In addition, to my eye, the green solution and the ammonia made a yellow solution, possibly indicative of chromate. Uugh, it only gets worse!
Testing on whether the pH~7, yellow liquid resulting from the green solution and sodium hydroxide - HCl combinations is pending. In the meantime, I thought some of the insoluble residue from the initial nitric acid dissolution that was mixed in with the white arsenic oxide might be elemental arsenic, so I plan to attempt to sublime some of this powder.
I think I might have to restart this with a new sample of skutterudite - I have so many solutions resulting from the initial dissolution that I frankly don't even know what happened to the arsenic anymore!

EDIT: The insoluble residue tests positive for arsenic and has a density of ~5g/mL, within acceptable bounds for As's 5.727 g/mL given my sloppy techniques. The only thing left to do for this is test whether the initial ore has the same density (which it well might!) According to MinDat, skutterudite has a measured density of ~6.5g/mL, so I might have to use some more refined techniques... Given that I only have 0.5g of the unknown substance, and significantly more of the initial ore (a few g remaining), I'd like to know if there are any more accurate volume measurements than water displacement.

[Edited on 5-18-2014 by elementcollector1]

blogfast25 - 19-5-2014 at 04:52

Quote: Originally posted by elementcollector1

Results are in - this is not copper. Ammonia had no effect on the brown precipitate (which, if CuO, should have reacted immediately).

I'm not so sure about that: freshly precipitated CuO will be more sluggish towards NH3 than freshly prepared Cu(OH)2. Could cause a false negative. It would be better to test straight on the solution (neutralising the HCl with NH3 and then add excess).

Chromium: alkalise and add peroxide. A yellow/orange colour that causes precipitation with Pb(NO3)2/Pb(OAc)2 would mean Cr(III) (to Cr(VI)).

elementcollector1 - 19-5-2014 at 05:20

Per the test for copper: I did that too. Solution turned clear, then yellow, no precipitate whatsoever.

Will get back to you on the other stuff.

blogfast25 - 19-5-2014 at 09:09

Quote: Originally posted by elementcollector1

Per the test for copper: I did that too. Solution turned clear, then yellow, no precipitate whatsoever.

Will get back to you on the other stuff.

The yellow is bizarre. Seems to point to iron though. That would give a darkish precipitate when hot, as Fe(OH)3 also dehydrates.

Zyklon-A - 19-5-2014 at 09:36

Quote: Originally posted by elementcollector1

Results are in - this is not copper. Ammonia had no effect on the brown precipitate (which, if CuO, should have reacted immediately).

I just tested this: ~.1 g CuO was put in a vial, ~2 mL of 15% ammonia solution was dropped on it and it was shaken strongly. Nothing happened for about 5 minutes, then, after everything settled out, I could see the solution was faintly blue- green in color. Then I added a drop of con. Hydrochloric acid. Again nothing much happened for a while. About ten minutes after HCl (aq) addition, it is slightly more greenish - but still quite faint.
[EDIT] I'm not sure about the copper oxide purity, it was made by electrochemically oxidizing copper in aqueous solution. The ammonia is quite pure.

[Edited on 19-5-2014 by Zyklonb]

blogfast25 - 19-5-2014 at 11:48

Zb:

Nice work. But the solubility of CuO in strong ammonia is likely to depend on its history: a calcined grade isn't the same as a product freshly obtained by mild heating of Cu(OH)2 in watery suspension. Electrolytic copper oxide may yet again respond differently.

The Volatile Chemist - 19-5-2014 at 13:34

I don't know if this has been posted already, but this is a page on how to produce arsenic from "Arsenious Acid", I apologize if it is inapplicable.

http://books.google.com/books?id=TFAQAAAAYAAJ&lpg=PA378&...

Zyklon-A - 19-5-2014 at 14:46

Quote: Originally posted by blogfast25

Zb:

[...] A calcined grade isn't the same as a product freshly obtained by mild heating of Cu(OH)2 in watery suspension. Electrolytic copper oxide may yet again respond differently.

Yes, I did heat the CuO strongly in air for several minutes, but I'm still not entirely sure what differences that would make.

blogfast25 - 20-5-2014 at 04:55

Quote: Originally posted by Zyklonb

Yes, I did heat the CuO strongly in air for several minutes, but I'm still not entirely sure what differences that would make.

It's a fact that the 'reactivity' of oxides depends strongly on their thermal history. See e.g. fresh ferric hydroxide v. calcined Fe2O3 or hydrated alumina v. Bauxite etc etc.

elementcollector1 - 20-5-2014 at 11:25

Mine should have been fresh, but I digress.

News with the gray leftovers: Almost definitely residual skutterudite. The mystery material was dried and calcined with a blowtorch, causing various brown, white, and black/silver mirrors (for lack of a better word) to appear on the test tube. I then took some of the old ore, back from when I was trying H₂SO₄ and NaOCl to dissolve the skutterudite (update: Still nothing.). This gave the same results upon sublimation, and upon cooling had developed the same coloration and density, which seems pretty conclusive to me. Ah well!

An interesting note on the sublimation: It's too small to tell, but what appears to be a pure black As crystal roughly 1 mm in length grew in the middle of one of the colored patches. It has very defined crystalline boundaries... whatever it might be. In addition, some of the mirrors were, in fact, bright shiny silver in color before cooling (they dulled to gray-black afterwards). This is the same result as attempting to sublime the arsenic-coated thick-gauge copper wire (it turned silvery and this flowed towards the bottom before the iridescent copper oxides took over the coloration). In fact, I tried cutting a wire in half while it was coated and subliming: The arsenic appears to have soldered the two wires together!

Another thought occurred to me: What if this brown precipitate is elemental arsenic? The amount could make sense given the concentration of the starting solution, woelen has mentioned impure arsenic as being brown before, and the supernatant solution (once filtered) tested only weakly for arsenic, taking days to form anything on the wire (and it still is only barely darkened), indicating most of the arsenic has now left the solution. Of course, this fails to explain the variety of colorations I'm getting, but it holds more water than the copper hypothesis (as where would significant amounts of copper enter the equation?)

I guess the only way to find out is to attempt to sublime it once it's dry.

elementcollector1 - 22-5-2014 at 19:07

Have you been tired of the poor performance of your aluminium foil? Seeking something new and exciting? Well, try our brand new Arsenic Foil!

Anyway, this flaked off the test tube after I tried subliming some skutterudite ore. Sublimation works somewhat well - I get a mix of arsenic oxide (which fumes out of the test tube) and metallic arsenic, which forms a mirror on the tube walls. So, at least I have something...

Gonna try the following:
-Addition of carbon to reduce the As₂O₃ back to As
-Sublimation in a closed copper tube (just a length bought from the hardware store, hammered shut and possibly twisted at both ends)

Mailinmypocket - 23-5-2014 at 04:26

Nice! I tried scraping the stuff I had sublimed off the copper wire and there was so little that it became arsenic bits that were pathetically small. "See the dust in there? That's arsenic

" screw that!

elementcollector1 - 23-5-2014 at 17:03

Heh, yeah. Not dealing with that anymore.
Bought the copper tube, and decided to only hammer shut one end (which still needs some work before I decide that it's sealed). Unfortunately, the 1/2" cap is a rather loose fit - any ideas on tightening it, given that this is where the pure arsenic is supposed to collect? I really don't want to solder and unsolder it every time I wish to collect product, and can't exactly tape it shut because the pressure in there is going to be such that gas would escape easily...

elementcollector1 - 26-5-2014 at 18:31

Well, that was idiotic and then brilliant. In that order!
First, I mixed in an equal volume of charcoal, in the hopes that there would be no arsenic oxides from the oxygen in the tube. This ended up being true, but having weird side effects.

Using a cap on one end didn't go so well, as nothing I did could keep it shut enough for the gases to escape. So, I took that apart, hammered both ends shut, and heated for a very long time, being sure not to touch one end (as that was where I wanted the arsenic to end up). The end result, after I hacksawed it open at the middle, was pleasing!

(Everything is a lot more metallic than it appears in the photo, because my inside lighting is frankly terrible.)
The larger lumps are actually coalesced arsenic, that appeared to have melted together! I assume this is due to the carbon dioxide (from the charcoal, the oxygen, and heat) causing enough pressure inside the tube to mess with the phase diagram and cause the arsenic to melt rather than sublime. Works for me - now I didn't have to worry about scraping off foil!

I am very glad to have a sample of arsenic in my collection, and also to never have to touch this stuff again.

MrHomeScientist - 27-5-2014 at 06:25

Congratulations! I always love seeing these elemental isolations. So you simply ground up the ore, mixed with charcoal, and heated in a sealed pipe to sublime the As? I wonder what the rest of the products were.

Zyklon-A - 27-5-2014 at 06:31

Nice! I should buy some Skutterudite and try this, sounds like fun. And you're still alive, so that's a plus.

elementcollector1 - 27-5-2014 at 10:53

Quote: Originally posted by Zyklonb

Nice! I should buy some Skutterudite and try this, sounds like fun. And you're still alive, so that's a plus.

I think that was more luck than skill on my part - there were plenty of leaks from the tube, even after I triple-folded the flattened copper pipe!
In addition, make sure you do this outside, obviously.
I would honestly attempt the wet route for a sample of arsenic - for this, I got fed up and got lucky.

blogfast25 - 27-5-2014 at 13:08

Nice. Carbothermic reduction in action.

elementcollector1 - 28-5-2014 at 09:58

Quote: Originally posted by blogfast25

Nice. Carbothermic reduction in action.

To be fair, I should probably have used significantly less carbon. The stoichiometric amount for the amount of air in the tube would probably be milligrams, instead of the equal volume (to the skutterudite) I decided to add. This ended up being a drawback because the pressure of the CO₂ ended up causing it to work its way out of the tube through tiny cracks in the flattened ends, offgassing and scaring the pants off of me every time.

I wonder what would have to be done for a perfectly sealed tube? Likely it would have to be welded shut.

blogfast25 - 28-5-2014 at 11:36

Quote: Originally posted by elementcollector1

Quote: Originally posted by blogfast25

Nice. Carbothermic reduction in action.

No, no, EC1: excess is good here, trust me. You may have slightly overdone it but that's partly why it worked.

Stoichiometric would have been as slow as grass growing.

elementcollector1 - 15-7-2014 at 18:10

In recent news, it turns out the waste material had yet more arsenic to yield. This might need some purification and sublimation, but I was wondering if anyone wanted a sample of arsenic for their own purposes. Not sure on prices, as I am unsure of how much I actually have (some of the ore was dumped in that waste container too...)

elementcollector1 - 26-10-2015 at 18:06

Hmm. It seems all the pics in this thread died. I'll just re-upload what I have, then:

tumblr_n67nvvEUSV1rlgzvyo1_400.jpg - 42kB

tumblr_n60auidJ8g1rlgzvyo1_400.jpg - 42kB

tumblr_n5l3d9vZZM1rlgzvyo1_400.jpg - 57kB

tumblr_n5l3d9vZZM1rlgzvyo2_400.jpg - 67kB

tumblr_n50yjcpQet1rlgzvyo1_400.jpg - 54kB

tumblr_n50yjcpQet1rlgzvyo2_400.jpg - 43kB

tumblr_mypo4536nm1rlgzvyo1_400.jpg - 53kB

tumblr_myj9ieQ9hD1rlgzvyo1_400.jpg - 66kB

tumblr_myj9ieQ9hD1rlgzvyo2_250.jpg - 25kB

tumblr_myj9ieQ9hD1rlgzvyo4_400.jpg - 75kB

Pages: 1 2