I'm entering this into the Green Design Contest. If you like what you see, please vote for me!

This instructable ties in very closely with my previous instructable, Snapple, Steel and Green Fire in which the lantern that is the base and heat source for the samovar is built using re-purposed/up-cycled materials. I will give a high-level overview of the process of building the lantern, but for details, I encourage you to look at the instructable referenced above.

In Snapple, Steel and Green Fire, one of the comments (left by shizumadrive) asked if the lantern could be used to cook food. I said I thought yes, but as they say, the devil's in the details.

The first problem lies in creating a place for the food/drink to sit while it heats. Although from the top view picture, it would appear that you could just set a small pot on the grill of the lantern, in practice, doing so would cause the carbon dioxide exhaust from the lamp to smother the lamp flame. Drilling holes in the lantern top could solve this problem, but then part of the purpose of the lantern top was to reduce the level of particulates in the exhaust for the green fire version of the lamp (diboron trioxide). Holes in the top would limit this aspect of the top's effectiveness.

The second problem is in the efficiency (or lack thereof) of the arrangement. Although the lantern, being made in part of glass, is reasonably well insulated, the top is not, and that's where we need the efficiency improvements.

Warnings:

Making (and using) the Lantern and Samovar involves hot things and sharp things. Be careful! Cut metal is sharp metal. You will need to exercise caution when cutting and sanding the various pieces you make. Also, do not heat food or drink using the Green Fire version of this lantern. The particulate exhaust, (diboron trioxide), will almost certainly get into what you're heating. While boron is a suspected ultra-trace nutrient, you're probably getting plenty in your diet already. To quote Wikipedia "it is necessary in such small amounts that ultrapurified foods and dust filtration of air is necessary to induce boron deficiency." There is the possibility of large doses having unpleasant side-effects.

That said, I use the non-Green Fire version of the lamp and Samovar inside, and (as long as it it not left unattended), it should be no more dangerous than any other lamp or candle. Let me reiterate, however - do not leave unattended!

Other thoughts:

Boiling water + hot glass = nothing, apparently. I started this project with the idea that, worst case scenario, a boil-over could cause the glass lantern body, or even the glass of the alcohol lamp inside, to crack due to thermal stress. With that in mind, I made the first step of the project a spill tray of sufficient size to hold any alcohol that would spill in such an event. I then spent a fairly dull hour watching the water boil in the samovar, and repeatedly spurt onto the lantern top and the glass body, but nothing much happened. The water boiled off the hot surfaces immediately, but that was it.

My temperature assessing gear is not so sophisticated, but I decided to try to check the glass temperature with a standard digital food thermometer. The highest reading I got for the glass was 250˚ F. While this is definitely hot enough to burn your fingers, when you take into account the near-boiling temperature of the water that hits the glass during a spurt (or "bump"), the temperature difference is not really that great - about 40˚ F. As you will see when you read the glass cutting section of this instructable, this is not enough to propagate a crack through glass, even when a score line has been drawn across the surface. While the lid of the alcohol lamp inside certainly gets hotter than this, the glass doesn't appear to, as evidenced by the complete lack of charring, or even discoloration of the paper seal inside. I haven't had a boil over of such magnitude that water reaches the inside of the lantern, but I suspect that in this case as well, there would be no result other than to snuff out the lamp. Cold water is probably another story. Don't spill cold water on a hot lantern.

All this said, I would ask, nay implore, that you treat this as you would anything else that you have intentionally set on fire - do not leave unattended! Do not set water to boil, and then go out to (for example) mow the lawn, take a quick nap, run out to the store, finish off that last fifth of Jim Beam, chase the dragon, or engage in any other activity which absents you, physically or mentally, from the spirit samovar. This is because it is on fire, you see.

Step 1: Make the Lantern

This is the high-level overview of the lantern project. For details on building the lantern, and cutting the glass for the lantern body, analogous to the chimney on a kerosene lamp, see Glass, Steel and Green Fire and Bottle cutting: some thoughts.

Materials:
Snapple or Honest Tea Bottle (tea/juice removed) (or size equivalent e.g. some 10 oz Jelly/Jam jars)
Eden Organic Bean Can - 12 oz. (sans beans) (or size equivalent)
Slightly larger can from Chili/Soup - pull top e.g Campbell's Chunky (sans chili or stuff) (or size equivalent)
4 Socket (Allen-wrench) button-head bolts (M6 X 8mm or equivalent)
4 Hex nuts (M6 or equivalent)
3 6-32 thread X 1" machine screws (must fit standoffs below)
3 1/4" OD, aluminum, 6-32 thread, round, female/female standoffs/spacers (must fit machine screws above)
1 sampler-sized jelly jar (or equivalent)
lamp wick
methanol (Heet - yellow bottle from an auto supply shop)
or isopropyl alcohol - 91%
acetone/xylene/naptha/other non-polar solvent (optional)
lead-free solder

Tools:
Hacksaw
Utility knife
glass cutter
glass cutting jig
center punch
paper hole punch
electric drill
electric burner/toaster oven (not to be used for food again) and/or
propane or butane torch (optional, depending on patience)
cheapo files
sandpaper (120, 220 and 400 grit)

Process:

You will need to cut the glass for the body from an appropriately-sized bottle. I go into detail on how to do this (at least using my method - there are many others) in the above-referenced instructables. In essence, remove the top and bottom of the bottle, sand, and set aside. Although the samovar will be more efficient with a shorter body, I would still encourage a longer one to help preserve the glass during the heating and cooling the lantern undergoes - a longer body = a more gentle heat gradient between the metal base and top, both of which are excellent heat radiators.

Cut the base and top from the cans indicated above. You will need to make the optional two-tiered top in order for the samovar unit to rest stably on top. To connect the two can tops into one piece, I used lead-free, general metal solder, melted using a hotplate and a butane torch (for help). I encourage making the two-tiered base, also to increase stability.

If you are only making this for use as a samovar, the lid is unnecessary.

The spirit lamp portion is made from a small, sampler-sized jelly/jam jar. To fit in the lamp, and for safety reasons, I would not recommend going larger than a one-ounce size jar. The wick is standard, round lamp wick. As noted before, green fire is pretty, but it ain't fer cookin'. Don't do it!

Step 2: The Spill Pan

You now should have the completed lantern. One of the first things we'll need is a spill pan. The purpose is two-fold: boiling water (or cooking almost anything else) will cause water to periodically spurt/drip out from the heating vessel, and we would like to catch this, rather than letting it pool up on the table; because water will periodically drip onto the lantern glass itself, which could cause thermal stress, and breaking. Because a major fracture could cause the lamp inside to spill, we would want to catch any and all alcohol, flaming or otherwise. The chances of such a catastrophic failure seem vanishingly small; still, it can't hurt to be prepared - right?

We'll make the spill pan out of the base of a large (29 oz.) can. I use Eden Foods cans because the can linings (unlike most other cans) do not contain BPA, a probable endocrine disruptor. Remove the label of the can. The glue can be removed using Goo-Gone (xylene) or lighter fluid (naptha). A less toxic alternative would be vegetable oil, followed by dish soap. The oil serves as a non-polar solvent, and the dish soap allows the oil to be rinsed off with water. The takes longer than the petroleum products listed above, but is certainly a lot less nasty.

We will now cut the base (unopened end) of the can off from the rest of the can. I've found the most effective way to do this is to use a hacksaw, but not to try to saw straight through. Fit the hacksaw in the bottom groove of the corrugated section and draw the hacksaw towards you once. Then, rotate the can slightly, and repeat. Keep rotating the can, making one stroke with the saw until you start to wear through the steel. Resist the temptation to just saw straight through until there are little holes developing all around the groove you've cut. When you finally do cut through, the can should really just fall apart. Another benefit of this approach is that the little jags of steel that stick out after you've made the cut are paper-thin, and are less likely to cut you. Note I said "less likely" not "unable."

Clean up the cut with a file (I used a $1 file I got from a Sears bargain bin - nothin' fancy). Drag the file, gently, perpendicular to the edge of the can, bringing up the little metal jags that have gotten folded inside to a vertical position, then file them off by drawing the file parallel to the side of the can. This will go a long way towards cleaning-up the edge.

Finish off with a relatively fine grade of sandpaper - I used super cheap 220 grit. The edge should now be very smooth.

There! Spill tray complete!

Step 3: Problem One: Where the Food Goes

So how do you heat food on the lantern without snuffing the lamp flame inside? A rack would be one option, but that would require us to have a welder, which no one seems to want to buy for us, or actually buy the rack from a camping store, which defeats the purpose of upcycling this puppy! The solution I hit on was to use another can section with vents around the side for exhaust gases. Use a regular (15 oz) size soup can (BPA free if possible). Cut, file and sand the unopened end as described in the last step. Now mark a line using a marker or pencil around the entire circumference of can. I tend to rotate the can while holding the writing implement at a steady level. Mark off eight places for holes, ideally equidistant from each other. To make the holes you can use an electric drill, but as the steel is quite thin, I've found that a standard paper hole punch works well. It's not exactly easy to squeeze the hole punch shut, but it's very doable, and it makes an extremely neat hole with no burrs.

My first iteration had no vent holes in the can lid itself. My thinking was that this might save some soot accumulation on the "pan" (AKA measuring cup) - seen here in the background - as well as leave the metal in place to help support the load. This turned out to be massively inefficient. I didn't end up using the 8 vent configuration as used on the lantern top, and instead compromised on 4 vents. My hope was that there would be enough metal in place to continue to support a full cooking vessel. So far, it's worked fine. I'm including the paper template I used to trace the vent holes on the lid (using a sharpie). I then cut out the straight sections with a utility knife - carefully! The curved sections can be drilled and then filed round, or if you're feeling jaunty, you can try to cut them with the utility knife. I wouldn't use tin snips, as I think that these would tend to deform the lid. I suppose a plasma cutter would work too. I mean, if you have one lying around or whatever.

A note on metal: these little pieces of metal are sharp! Super sharp! Ditto the edges of the lid. Use lots of caution when doing this. I got yet another cut this time around. I was going to post a picture of my various red and white cells doing their respective jobs, but I thought that might not be appreciated by the more sensitive members of the community. You'll have to use your imaginations.

Also, I used the pictures from the lantern top here. I inexplicably don't have pictures of the four-holed samovar unit being fabricated. Fortunately, the process is the same as the eight-holed lantern top. I hope this doesn't cause any confusion.

When you've cut, drilled, filed and sanded the samovar unit, you'll need to burn off the oleoresin (if you used an Eden can) or nasty BPA liner (if you used pretty much anything else). I use a hotplate for this. Outside. Really, really don't do this inside. It's stinky and probably not terribly healthy to breathe, which is why we're burning it off prior to using the piece. I put a aluminum pie plate on top to keep the heat in. The hot plate was $20 years ago, and has since been appropriated for experiments. It's rated at 1100 watts. The intriguing spills you see are sulfur and lye, both of which will figure in later instructables (foreshadowing).

The end result will look like the last picture.

Step 4: Food Vessel: AKA the Pan

I would have liked to have made a food vessel for this out of a soup can or something similar, but since this is a food vessel, and I intend to eat or drink what I make on it, I need something food safe. I am not confident that anything I made from a can would be truly safe. As luck would have it, one of our steel measuring cups lost its handle while I was within striking distance of the trash can, and voila - food safe vessel! The one pictured here is a Crestware Stainless Steel Measure cup. It's most likely the one I have, and it costs about $1.50. The picture and link are to the Amazon page.

Step 5: Problem Two: the Details (Part One: Cutting Our Jacket to Suit to Our Cloth)

At this point you can warm items on your samovar. Note, I said warm, not heat or boil. This is because this is a ridiculously inefficient setup at this point. Although the air inside the samovar unit is in excess of 400˚ F (it could be hotter but this is where my cheapo thermometer maxed out), very little of that is radiated into the pan. I first tried to make a lid from a piece of a can, but ... well, you can see the result in the next step. To quote Neil Young, red water is icky. I then tried the top of a french press I had scored from a thrift shop. It looked very retro fashion-forward - almost steam punk. Alas, the set up was wicked inefficient. Using methanol, and heating the water for over an hour, the water temperature maxed out at about 150˚ F. This is enough to discomfit germs, but it's not getting us a nice cuppa anytime soon.

What we need is a way to concentrate the heat that is currently venting from the sides. Ease of use, as well as the ability to remove the attachment (so that we can go back to using this as a regular lamp) are also important. I chose an open metal jacket made from - you guessed it - a tin can. In this case, we are using the top part of the large can used for the spill pan. The height of the jacket will depend on what you would like to use for a boiling/heating vessel. My measuring cup vessel has no handle (which of course, is why it's no longer being used as a measuring cup), so I can make my jacket somewhat higher than the edge of the vessel. This will also improve efficiency.

Cut the can section as you did in the Spill Pan section above; then file and sand the edges. Now we need some means of attaching the jacket to the lantern. Again, we are looking for a non-permanent means so that we can go back to using our samovar in it's purely illuminative manifestation. However, it must also be stable. I chose to use four positioning screws that tighten onto the samovar attachment. Cut the holes (again equidistant from each other if possible) using the hole punch. The screws you use can be virtually any size that fit the holes, and when fully extended, are close to flush with the samovar attachment. Somewhat too short is fine. Too long is less good.

Because, without actually screwing the jacket to the samovar attachment, we're not going to get a very snug fit, I decided to augment the connection with some Sugru. For those without these delightful packets of DIY-goodness lying around, Oogoo is a reasonable substitute. You could also use straight silicone, although it is very sticky and kind of tough to work with. Why silicone-based? Because silicone is a fairly heat-resistant material, and can take well over 500˚ F without damage (Instructables member, Nepheron has used Sugru for metal casting of tin-based solder). To make the Sugru/Oogoo connectors, roll your silicone based clay into an oblong shape (sort of a Sugru-pill/plug ... plugcebo?). Cut these in half the along the shorter diameter, and screw them on to the positioning screws. Let these cure overnight.

Once the Sugru/Oogoo has cured, you can trim the connectors using a knife, or even the edge of the steel can/jacket. I would even go so far as to recommend the latter, because it's easier to get a curve that matches that of the outside of the can. To determine where they should be cut, press the samovar unit into the Sugru plugs, while they are still attached to the screws, gently. Then, remove the plugs and complete the cut along the line you scored. Reattach the plugs to the screws, and slide the jacket onto the samovar unit. The fit should now be very snug. So snug that you can lift the jacket and the samovar unit as though they were one.

Step 6: Problem Two: the Details (Part Two: Put a Lid on It)

Now to really let the heat build up, we need a lid. Sadly, my little measuring cup didn't come with one, so I had to improvise.

My first (and crappiest) thought was to try my hand at sinking a lid from the metal from the lid of our can. It's amazing how fast metal rusts. Literally half an hour into my first test, the water had turned a cheerful crimson. Obviously, this was a non-starter.

The choices then were a non-rusting metal or glass. I don't have sheets of heavy gauge aluminum lying around, and a positive dearth of stainless steel sheets. Galvanized yes, but then we'd be getting a bunch of zinc oxide in our water, which would taste funky, and could be unhealthy as well. So glass was the winner.

To make a glass lid, we need a nice, relatively-thick piece of glass - something like ... an applesauce jar! Actually, all we need is the base of a 24 ounce applesauce jar. To get this removed from the rest of the jar, however, is kind of fiddly. Basically, you'll use one of the bottle cutting methods described either in my instructable on bottle cutting or elsewhere on Instructables. The method I use is the glass cutter and hot water method. The trick here is in getting a nice score line with a glass cutter around the base. I tried a couple of different ways, but ultimately, I returned to my homemade jig. Because the picture is not so illustrative, should you wish to build a similar jig, I'm including my Autodesk 123d model. Generally, when I cut something new, I cut a new groove into the upright board of the jig, so I can hold the glass cutter steady while I rotate the bottle. However, instead of cutting a groove at an angle into upright board, I used four nails to hold the glass cutter at the angle I needed. It was just a better use of my limited upright-board-real-estate. I then turned the jar, while maintaining steady pressure with the cutter, to make the score line.

To get the lid to separate from the rest of the jar, heat a couple of cups of water to approximately 110˚ hotter than your cold tap water runs. My tap water runs at 70˚ F, so I heat my water to 180˚ F. Heat the score line with the hot water, then cool it with the tap water. Repeat this process until you begin to hire the tinkling sounds of a crack propagating around the jar. You will also be able to see the light refract through the glass in a different way when the crack propagates. If you see there is a place where the crack hasn't spread, heat some more water and concentrate on that area. The glass is thicker and less even at the bottom of the jar, and this process isn't as straightforward as the usual bottle cut. It is very doable however.

Once you have a crack propagated all the way around the jar. You will need to knock out the base. For reasons I don't understand, the base holds on like grim death even when the crack has formed all the way around. Use something wood (like the handle of a hammer or a dowel rod) and gently strike the base of the jar from the inside. It should pop out (ideally, onto something soft). Now sand the heck out of it, because it's going to be sharp, sharp sharp!

Now you have a lid!

Note: I had originally covered the edge of the lid with Sugru as well, and it make the fit really nice; however, I subsequently found out that Sugru is not food-safe, so I removed it. Plain silicone should be fine, but I haven't had a chance to experiment with it.

Step 7: Problem Two: the Details (Part 3 - (in)Efficiency!)

Okay, this is where the rubber hits the road ... or the OCD hits the brick wall. Is this a reasonable, environmentally friendly way of heating water? Here are my observations (with the following caveat - I'm not Mr. Physics-guy, so if you feel I've taken a wrong turn, please let me know ... kinda nicely):

With the jacket and the glass lid, water will come to a full boil in about 20 minutes. When filled with isopropyl alcohol, the lamp inside the Spirit Samovar will stay lit for roughly an hour and 20 minutes. This is enough for three cups of tea. Sounds reasonable to me. But is it eco-friendly?

To find out, I returned to my chemistry text for a refresher on enthalpy and heat capacity. The equation is a pretty straightforward one:

The energy required to change water from a liquid to a gas is equal to the grams of water, times the change in temperature in Celsius, times the heat capacity of water (4.186 J/g °C). Divide the result by 1000 to get kilojoules. Add the energy required to vaporize water, which is grams of water to be vaporized times the enthalpy of vaporization (2260 kJ/kG), et voila! So:

100 grams * (100 °C - 20 °C) * 4.186 J/g °C = 33488 J = 33.48kJ
100 g H2O *(2260 J/g)/1000 = 226 kJ
33.48kJ + 226 kJ = 259.048 kJ

Sort of. The problem is that whenever chem textbooks talk about this (and it took me a while - embarrassingly - to figure this out) they are talking about vaporizing all the water. While an interesting thing to do, converting all of the water in the pan to steam will not help with anything except clearing our sinuses. What we need to do is vaporize a specific amount of water, and measure how much energy that took. Here was my approach:

I measured out 20 ml of methanol into the lamp.
I measured out 100 ml (100g) of distilled water into the pan
I burned all of fuel with the pan of water on the samovar (lid off).
I measured how much water remained.
I then plugged that amount into the enthalpy of vaporization equation.

I used methanol because energy density figures for methanol are fairly easy to come by online. I boiled away 24 ml of water during my experiment, which is equal to 24g of water; I also heated all 100g water to roughly 100˚ C. plugging these values in to the above equations, we get:

(100*(100-20)*4.1855/1000) + (24*(2260/1000)) = 87.33 kJ

I then found a value for the amount of energy one could reasonably expect from burning methanol (http://en.wikipedia.org/wiki/Energy_density). Multiply this by the number of grams of fuel. With a density of .79 g/cc, I had 15.01 grams of fuel (the Heet brand methanol I used is 95% methanol, per MSDS). The above site lists the fuel value at roughly 19.7 MJ/kg, or 19.7 kJ/g, which means that my lamp was holding 295.7 kJ of energy - roughly. I divided the amount of energy that was actually transmitted to the pan (as indicated by the amount of boiled off water) by the maximum energy that could be expected from the amount of fuel I used.

87.33 kJ / 295.7 kJ = .295

This meant that my samovar was running at about 29.5% efficiency. I'm going to assume that the samovar's efficiency will be roughly the same when I use isopropyl alcohol, which is a more energy-dense fuel (who knew?). In my isopropyl alcohol test, I boiled away 34 ml of water. Using the same calculation above, the result is 109.94 kJ to boil that amount of liquid. Because I now know the samovar efficiency, I can divide the energy needed by the efficiency, and arrive at the total energy value for the 20 ml of 91% isopropyl alcohol: 109.94 kJ/.295=372.68 kJ. This amount divided by the mass of fuel:

20 ml * .91 = 18.2 ml. 18.2 ml * fuel density of .786 g/ml = 14.3g of fuel (I omitted the 9% H2O, which is not fuel here)

gives us the energy density for isopropyl alcohol:

372.68 kJ/14.3g = 26.06 kJ/g

How does this compare to a microwave? What's the carbon footprint? I found an interesting article on Treehugger that compared boiling water in a microwave vs. a kettle using a Kill-a-Watt meter (Ask Pablo: Electric Kettle, Stove, or Microwave Oven?). The author found that the microwave was about 47% efficient in turning electricity into heat (spoiler alert: the microwave beat the stove, but was beaten by the kettle). Not having a Kill-a-Watt, I can't estimate efficiency in the same way, but I can estimate carbon footprint. I did my test based on the following method:

I measured 100g distilled H2O into a measuring cup and timed how long it took to boil off 34 ml (34g) of water. I found the nominal wattage of my microwave (1100W), and then using the EPA carbon calculation (920.8 grams CO2/kWh) for states where electricity is generated using coal (as mine is), I calculated how much carbon was released from heating my water. With the microwave running for 165 seconds, it drew 50.14W. At 920.8g CO2/kWh, that means 46.42g CO2 was generated.

But how much was generated by my samovar? More chemistry!

The formula for isopropyl alcohol is C3H7OH. The chemical equation for the combustion of the fuel is:

2 C3H7OH (l) + 9 O2 (g) ----> 6 CO2 (g) + 8 H2O (g)

This means that for every 2 mol of C3H7OH, we generate 6 mol of CO2, which is a 1:3 ratio

I have 91% C3H7OH, which means that the 20 ml contains 18.2 ml of C3H7OH, or 14.31g C3H7OH, which at a molar mass of 60.10 g, gives us 0.238 mol of C3H7OH. The CO2 generated will be three times this many moles, and the molar mass of CO2 is 44g. This means that my samovar generated 31.42g CO2 to boil off the same amount of water. We have a winner!

Which is actually sort of sad. If a samovar assembled from scrap, running at 29.5% efficiency, beats a 1100 watt microwave, operating at 47% efficiency, pulling electricity from a multi-million dollar coal-fired generator, on their respective prorated carbon footprints, there's definitely a need for a change of plan. Maybe not so much with the coal, and not so much with the centralized power generation.

Quick note: Methanol vs. Isopropyl Alcohol

From the above, you can see that methanol is a much less energy dense fuel. It is also quite toxic. Isopropyl alcohol, however, tends to generate a fair amount of soot when it burns, as you can see from the rich, luxurious coat on the lantern top and samovar unit. Ethyl alcohol (drinkable) or denatured alcohol (ethyl with a small amount of methanol to render it undrinkable) tend to be a reasonable compromise, with better energy density than methanol, and a cleaner burn than isopropyl alcohol. I've also seen some anecdotal evidence online of people blending methanol and isopropyl alcohol to mitigate the soot and energy density problems. I haven't tried these options, so I can't comment on them other than to say "they're out there." If you have tried them, I'd love to hear about it though!