In Praise of Underloved Aluminum

I've been contemplating a couple of titanium and stainless steel lights, so I decided to do a little research into their properties. The charts below show the thermal conductivity, density, and price of eleven metals and alloys of potential interest:



I couldn't find this data in any one place, so it comes from multiple sources: CRC Handbook of Chemistry and Physics, London Metal Exchange, etc.

There are many other important properties too numerous to list: strength, hardness, toughness, ease of machining, corrosion resistance, susceptibility to hydrogen embrittlement, toxicity, and so on. Rather than categorize all of these, I've simply omitted materials that are clearly unusable such as lead (soft and toxic), iron (rusts), and depleted uranium.

Several things stand out:

(1) Ti and SS are terrible heat conductors. I knew they were not as good as copper or aluminum, but had no idea that they were 10-20 times worse! Making a small, high-output flashlight out of Ti or SS seems like a death wish for the poor LED. Why do we crave this? (Full disclosure: I totally did not get Fatal Attraction. His wife was HOT! He threw that away for skanky, chain-smoking Glenn Close? :green::thinking:)

(2) Why doesn't anyone make lights out of magnesium? It's cheap, feather-light, a decent conductor of heat, and has plenty of bling-cred (e.g. mag wheels). Is there some fatal weakness in one of its other properties?

And finally, the subject of this post:

(3) Why don't we show a little more love for humble aluminum? It ranks #2 in price (only tin is cheaper), #2 in density (only magnesium is lighter), and #4 in thermal conductivity (bested only by copper, silver, and gold). It is also durable, easy to machine, highly corrosion resistant, non-toxic, non-allergenic, and beautiful.

So for what it's worth, Aluminum, I for one think you're awesome. :thumbsup:

well done

what about a graph showing attractivness - visual and tactile. gold and brass might do well there.

copper is the really conductive one and thats the first thing the heat meets as it leaves the led. as the path lengthens the area/mass increases as we hit the SS or tit . but still pretty stark looking at a pictorial representation of it.




cheers

Interesting topic.

I would like to know how polymer/plastic would rank in the graphcs above. I think weight #1, thermal conductivity last, price #1

no surprising data at all, wonder that this is considered "difficult to get info" :thinking:
must be shown many times here.

in short: alum is the best compromise and thus mostly used
and I dont know if there is any equivalent to anodizing - at the same time easy, cheap and surprisingly good finish - with the other metals?

flashflood said:

(2) Why doesn't anyone make lights out of magnesium?

Click to expand...

VERY difficult + VERY dangerours to be formed

... there is aluminium present, at the moment no need to use more "fancy" materials - except for the users thinking, maybe.
Take a look into knives: Damascus, espesially the breeds that are made for looks, or show patterns, have not real advantage compared to modern steels. Its just for the eye and costs a lot more. Still the knives are sold like crazy (as are lights from tit, steel, ...)

flashflood said:

. . . . Several things stand out:
(1) Ti and SS are terrible heat conductors. I knew they were not as good as copper or aluminum, but had no idea that they were 10-20 times worse! Making a small, high-output flashlight out of Ti or SS seems like a death wish for the poor LED.

Click to expand...

The problem is that the facts are counterintuitive - all common metals are good conductors of heat, so people assume that Titanium or Stainless Steel will be reasonable thermal conductors !

Beryllium Oxide is a white ceramic - but thermally it's as good a conductore as aluminium, though it's an excellent electrical insulator - again counterintuitive. To save people asking why no-one uses it these days as a thermal pad - the dust is very toxic !

BBL said:

Interesting topic.

I would like to know how polymer/plastic would rank in the graphcs above. I think weight #1, thermal conductivity last, price #1

Click to expand...

Thanks. There's a pretty good list of thermal conductivity of common materials here:

http://en.wikipedia.org/wiki/List_of_thermal_conductivities

Plastics are all under 1.0 -- pretty good insulators. But certainly light and cheap.

At the other end of the spectrum, diamond is over five times the thermal conductivity of silver. Thin diamond films are actually used in some semiconductors as heat spreaders -- to move heat very quickly from tiny hot spots to a much larger surface area that a heat sink can dissipate.

yellow said:

VERY difficult + VERY dangerours to be formed

Click to expand...

Not quite sure why you say that. Magnesium alloys are used in many applications for decades. My guess is that using Mg-alloys just don't have any substantial benefit compared with Al, but require more unusual techniques and know-how. So why use something more difficult, if you can get very similar results easier?

beerwax said:

well done

what about a graph showing attractivness - visual and tactile. gold and brass might do well there.

copper is the really conductive one and thats the first thing the heat meets as it leaves the led. as the path lengthens the area/mass increases as we hit the SS or tit . but still pretty stark looking at a pictorial representation of it.




cheers

Click to expand...

That would be great. Mac, if you're listening -- it seems like you've made your EDC in just about every metal. Do you have pics of your handiwork you could share?

Appearance is tricky though. Aluminum alone can be brushed, polished, bead blasted, or anodized and look completely different. Copper can be new-penny bright to old-penny dull, or even green! Many forms of beauty out there.

Good thread!

It's a general interest thread, not specific to LED Flashlights, so I'm moving it to the General Flashlight Discussion section.

For members who are interested in the more technical aspects of materials, you'll find other threads in MMM.

Magnesium is even weaker in terms of structure and abrasive wear than aluminium, which is already quite soft. I am also not sure if it is as easily anodized or otherwise surface hardness treated as aluminium, so this would be a major downside.

In my opinion, the ideal flashlight material, all things considered (or ignored), is beryllium. Beryllium is lighter than aluminium, yet it is as strong and stiff as steel. It has excellent thermal *and* electrical conductivity, and is used in electrical connectors (alloyed with copper) for this reason.

It is, however, very expensive in raw form. In dust form it is also toxic and can cause a chronic lung disease without proper precautions, meaning that machining the material is yet another extra cost due to risks and precautions.

So - ain't gonna happen. It'd be awesome though.

I'm another Alu enthusiast. It is unfortunate that it isn't so popular for some of the "higher-end" lights where Ti seems to dominate. I don't like the threads on any of the Ti lights I've had (they ranged from almost acceptable to quite galling, even after work and lubricating), where my order of preference (for smooth thread operation) goes:
1 Copper,
2 SS,
3 Alu,

6 Ti.

Other people may have had better experiences with their Ti lights, I didn't.

drmaxx said:

Not quite sure why you say that. Magnesium alloys are used in many applications for decades. My guess is that using Mg-alloys just don't have any substantial benefit compared with Al, but require more unusual techniques and know-how. So why use something more difficult, if you can get very similar results easier?

Click to expand...

Well, alloys often have lower thermal conductivity than pure metals (http://www.physicsforums.com/showthread.php?t=97351), so the advantage might not be present in Mg-alloys.
BTW: The anodized part of Aluminum also has a lower thermal conductance than the metallic one (http://en.wikipedia.org/wiki/Anodizing).

As mentioned, there are other criteria like hardness and strength.

Regarding the price: Maybe for some applications a higher price is not seen as a disadvantage but an advantage, because then the product is more exclusive. So if for you regular diamond is not exclusive enogh, go for isotopically enriched 12C diamond - you could say it is because of the higher thermal conductivity. (http://link.aps.org/doi/10.1103/PhysRevLett.70.3764)

OK, here's one fundamental structural problem with magnesium (compared to aluminum):

Aluminum has an FCC (face-centered-cubic) crystal structure. There are many different crystal planes in an FCC lattice - this permits dislocations (atomic defects responsible for both strengthening mechanisms as well as ductility) to travel through the grain along multiple crystallographic planes without dislocation pileups. Taking a broader view, this permits a very ductile pure material to be alloyed extensively (for strengthening) while retaining some semblance of ductility - i.e. bending/deformation before ultimate fracture of the material.

Magnesium has an HCP (hexagonal-close-packed) crystal structure. There are relatively few crystal planes for dislocation motion in an HCP lattice, so dislocations tend to pile up in the lattice. The broader view of this is such that even pure magnesium is relatively brittle compared to aluminum, and cannot be alloyed (i.e. to add strength) nearly as extensively before the material will start to exhibit brittle failure.

So, conclusion: In pure form, both metals are relatively weak. However, aluminum can be alloyed to increase its strength dramatically (up to something like ~20x, give or take) and give us these relatively-thin-walled flashlights we like so much. Furthermore, damaging the flashlight body (via a bend or impact) could give us a slight (possibly undetectable) bend. However, magnesium is a different story - it cannot be alloyed much (i.e. strengthend) before that same flashlight body would fail without warning via fracture (which I guarantee is quite noticeable ).

We can talk about strength-to-weight ratios, but the ductility of the alloy must be taken into account - we can alloy Mg to have comparable strength-to-weight ratios to Al, but if the resultant Mg is far less ductile than Al at that strength, it is far less usable for our desired applications.


What I find facinating is that a relatively simple atomic property such as the atomic arrangement in a unit cell has such a huge influence on large-scale properties that we observe.

Al vs Ti was previously discussed in this thread. On page 2 js commented on the heat transfer and electrical conductivity of Ti.

One point in favour of Magnesium: all the high-end SLR cameras' bodies are made from Magnesium alloy, because it is very lightweight but tough. I haven't tried to bend my DSLR to see if it would fracture, however...

italianboy said:

One point in favour of Magnesium: all the high-end SLR cameras' bodies are made from Magnesium alloy, because it is very lightweight but tough. I haven't tried to bend my DSLR to see if it would fracture, however...

Click to expand...

I was thinking of this as well, I wonder why it is chosen over aluminum for this application? Perhaps the quality of the finish it takes? I also know of a few laptop computers produced that had a magnesium alloy chassis.

Kestrel said:

OK, here's one fundamental structural problem with magnesium (compared to aluminum):

. . . . The broader view of this is such that even pure magnesium is relatively brittle compared to aluminum, and cannot be alloyed (i.e. to add strength) nearly as extensively before the material will start to exhibit brittle failure. . . . .

Click to expand...

For many years Volkswagen Crankcases were made from Magnesium - what technique did they use ?

italianboy said:

One point in favour of Magnesium: all the high-end SLR cameras' bodies are made from Magnesium alloy, because it is very lightweight but tough. I haven't tried to bend my DSLR to see if it would fracture, however...

Click to expand...

Magnesium Ribbon is quite flexible and hard to break by flexing.

MikeAusC said:

For many years Volkswagen Crankcases were made from Magnesium - what technique did they use ?

Click to expand...

EMPI also marketed the Speedwell-BRM rim in the 60's, which were made with a high magnesium content as well. Popular among VW drag racers due to their light weight, but they suffered from corrosion problems if improperly cared for.

i understood the car parts are an alluminium magnesium alloy. mainly used for weight and strenght issues. magnesium in a torch might have corrosion issues too with the electric current .

diamond and some ceramic being a good heat conductor - totally counter intuitive.