| Pages:
1
2
3
4
5 |
The_Davster
A pnictogen
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
In the standard electrolysis demonstration, to be honest, I do not know why the sulfuric does not get reduced first, and then proceed to just
electrolyse the water. I had always been told the sulfuric is just there to increase the conductivity of the water. But this is at condidtions where
the sulfuric concentration is much below the standard 0.1M
Ah, I realize the discrepancy, the sulfate being reduced to sulfurous, came from a not-so-reputable book, a high school level text, wheras my
analytical text has no mention of this reaction. It was likely as you mentioned an oversimplification of a nacent hydrogen reduction. The 2H+ +2e-
<--> H2 reaction is next in line, which does support that you could drive off water by electrolysis, but at 0.1M. If I were so inclined, I
could dig up my old notes on how to convert potentials at various concentrations....but...I am not in the mood for all that mathemagic now.
But all sorts of other, less predictable side reactions will still occur, so I say that it is worth a shot to see what will happen, the elimination of
water may or may not be the primary reaction, but after you try it, be sure to do some analysis on it, and the gasses formed.
|
|
|
mericad193724
Hazard to Others
Posts: 121
Registered: 4-6-2006
Location: New Jersey, USA
Member Is Offline
Mood: why do you care?
|
|
I think I could probably try this out. I got a battery charger that can do 12V or 6V at 10 amps or 2 amps. The liquid will probably be very conductive
and get hot upon electrolysis. I think carbon rods would dissintergrate very quickly in these conditions.
What type of electrodes should I use for this test? I will start with a small scale electrolysis around 50ml 30% acid.
Mericad
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
I've electrolysed sodium sulfate with lead anode, don't know how graphite would fare. The lead oxidizes to a thick (strained, somewhat flaking) layer
of PbO2 so a heavy bar of lead would be prudent. Don't know about acid.
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Electrolysis has come up so often lately that I am
building a transformer based 40 ampere *continuous duty* variable 0-15v. DC supply from scrounged parts .
So far I have the variac , and the fixed step down transformer , the bridge rectifiers and heatsinks ,
a 4.5 inch 50 millivolt scale precision meter , which
will be across a .001 ohm solid 6 gauge 29.25 inch
copper shunt for reading amperes .
The plan is to use round jacketed three conductor 12 gauge extension cord for the added flexibility for the
output cables , and connect the three conductors
" three in hand " to give a shared conductor of
over 8 gauge and slightly smaller than 7 gauge size ,
to provide cool running and low loss output cables .
The terminal anode clamp contact is going to be silver , and heavy duty even if I have to fabricate it from bullion .
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
Just pick up 10-20 feet of stranded 8AWG at the hardware store... and for the shunt, use a few inches of bare #10 or #12, a miliohm is only
dissipating 1.6W at full rated current...
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Looked at the 8 gauge stuff but it is too stiff , meant for
permanent wiring and not for flexible hookup . I want something soft and flexible like welders cable . The specialty stuff is special order only
and it is a lot more expensive than than just getting a heavy duty extension cord and doing what I described going three in hand
on the conductors .
And as for the shunt , the size is common panel ground
~3/16" bare copper and has heavy machined copper T-connectors that slide for positioning like a split set collar to provide easy adjustment for
calibration and solid connection . I figured on just sleeving the shunt with
tygon tubing or split loom sleeve insulation and bending it into a wide U shape fastened to the base of the power supply . It was simply chosen
because it was the smallest size where all the hardware for needed connections was commonly available off the shelf , even though it is heavier than
necessary I know . There won't be any thermal drift on 1.6 watts due to conductor heating across 29.25 inches of 6 gauge copper .
The two transformers alone weight about 35 pounds ,
and everything is passive air cooled ...no fans .
I am going to add a polarity reversing switch for electrode preparation , voltmeter , self-resetting overcurrent breakers ( at least ) and possibly a
thermal breaker and fuses for backup protection ....since this thing will be run unattended for perhaps days at a time .
I have a 20 gallon aquarium which I keep thinking
looks a lot like a perchlorate cell to me
I have a couple of 12 liter pyrex jars with 100mm threaded ring and disc cover assemblies that could
be put to use too .
You know a cell running at low voltage like 3 or 4 volts
and high current , is really a scenario where a center
tapped transformer and a dual Schottky Diode bridge
is much more efficient , in comparison to a single output transformer and full wave bridge rectifiers having four
standard diodes . I could never seem to scrounge a
good center tapped transformer having a high current secondary at a cheap price .....so I have the cheaper and less efficient solution .
But I did save a link for some high current Schottky devices for cheap , in case anyone else might be interested in such a component for improving
rectification
efficiency greatly on low voltage high current DC power supply project .
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=29003...
Also a link here for a transformer that may be of interest
http://cgi.ebay.com/EGS-Hevi-Duty-Type-HS-Transformer-Ref-30...
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
I would use a couple (three) schottky diodes pulled from computer supplies. In the 300W range, these are typically 40A, 30V dual (common cathode) in
a TO-3P or TO-247 package.
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Yeah I really wanted to go with Schottkys , but I didn't want to buy extra heatsinks and micas , and having
no center tapped transformer meant I would need
four of the damn things ..... more than twice as efficient
as standard bridges and heatsinks I already had on hand .....so I stayed with what I got .
The four 35A monolithic full wave bridge rectifiers are
mounted in isolated bare metal square cases , which
I am mounting back to back in paralleled pairs in the
web of two separate 50W H-finned heatsinks , wiring
all four in parrallel to handle the heat dissipation and
avoiding having to use the expensive massive or fan cooled heatsinks ....by spreading the load comfortably
enough across what would be a 140A bridge array if it
was more massively heatsinked . Cost about half as much to afro engineer it the way I have got it
Heatsinks are damn expensive aren't they if you aren't
buying them by the ton .
|
|
|
BeerChloride
Harmless
Posts: 47
Registered: 17-9-2006
Location: Alabama, USA
Member Is Offline
Mood: Dunno
|
|
Interesting plan, Rosco. Yes, I often try to avoid all but the simplest heat sinks. My largest PS is a homebrew 12 v, ~10 A dual transformer,
pseudo-regulated with 2N3055's. I need to make a variable one for at least that kind of current.
I wanted to post about dry ice. Finally got some - more dry ice than I've ever seen. Now don't get me wrong, I used to play with liquid nitrogen all
the time, so CO2 pales in comparison, but it's neat too. Big 10 lb block. I used a wood chisel and a hammer to break it up. BTW, acetone with dry ice
is great - poor man's liquid nitrogen! (actually LN2 is just as cheap) It fizzles away the dry ice at first, then becomes totally stable with no
"clouds" like you always have with LN2, and almost no odor. After a while I decided to touch the cold acetone - it just looked so much like icewater!
My finger made a sizzling sound when I put the tip of it in - I had my finger out already before those auditory impulses even reached my brain. I DO
NOT recommend doing this! I am a trained mad scientist. Well, actually I am a trained scientist. So I shouldn't be doing such things, either.
I chilled several ml of the brown H2SO4. It became extremely viscous to the point of barely flowing, but didn't freeze after several hours. The 94%
seems to behave similar to sugar perhaps. Sulfuric acid rock candy???
|
|
|
unionised
International Hazard
Posts: 5103
Registered: 1-11-2003
Location: UK
Member Is Offline
Mood: No Mood
|
|
Err, is any of this chemistry?
|
|
|
not_important
International Hazard
Posts: 3873
Registered: 21-7-2006
Member Is Offline
Mood: No Mood
|
|
It relates to making strong sulfuric acid, so I think it's still on topic.
Low temperature freezing seems to readily give glasses or gels instead of crystals. The liquids get viscous enough that crystallisation is slowed
down, very slow cooling and extended times are needed for proper crystal growth. It might prove difficult to crystallise sulfuric acid of this
strength.
I use fairly dry ethanol or isopropanol instead of acetone with dry ice. A little less flammible, and less agressive at dissolving organics so I can
use plastic containers and tongs. However I believe that acetone gives you a bit lower temperature, I don't know why that would be.
The bath liquid serves several purposes. One is indead a heat transfer meadium, to get good contact. Another is temperature control, with liquids with
freezing points above the temperature of dry ice you can make a slush bath with a temperature near the freezing point of the liquid. There seems to
be another effect, as a number of books list different temperatures from dry ice in differing liquids that don't freeze at -70 C. Dissolution cooling,
perhaps?
|
|
|
mericad193724
Hazard to Others
Posts: 121
Registered: 4-6-2006
Location: New Jersey, USA
Member Is Offline
Mood: why do you care?
|
|
It turns out electrolysis of dilute H2SO4 will not concentrate the acid, it will form ozone as Davster said....
this is according to wikipedia:
Laboratory production
In the laboratory ozone can be produced by electrolysis using a 9 volt battery, a pencil graphite rod cathode, a platinum wire anode and a 3M sulfuric
acid electrolyte. The half cell reactions taking place are:
3 H2O → O3 + 6 H+ + 6 e− ΔEo = − 1.53 V
6 H+ + 6 e− → 3 H2 ΔEo = 0 V
2 H2O → O2 + 4 H+ + 4 e− ΔEo = −1. 23 V
So that in the net reaction three equivalents of water are converted into one equivalent of ozone and three equivalents of hydrogen. Oxygen formation
is a competing reaction.
This is only at 3M H2SO4.
Mericad
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
S'still removing oxygen and hydrogen...why would you say it wouldn't concentrate it?
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Could it possibly be forming peroxydisulfuric acid ?
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
Done hot it would tend to decompose, no?
How about distilling it when you're done?
Oooh, gaseous H2S2O8...toasty
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Know it is off topic , but regarding the linear power supply for an electrolytic cell or plating tank , I am
looking at doing some sort of basic voltage regulation
which is not terribly lossy nor requiring huge heatsinking
for any regulator dissipation , and I am thinking that
simply using a sizable capacitor filter may be adequate
for the task . Or if it doesn't do the whole job , it will
take a lot of the ripple out and ease the job for any regulator stage downstream . On the 40A transformer
output , I am considering using four 35,000 uF 25V caps paralleled across the output of the bridge for the main filter .
I haven't done any calculations on this filter capacitor value and am open to suggestions .
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
2000uF per amp is a rule of thumb. Best around 12V though. The rule of thumb should really say 24m F*V/A (that's m = mili), since ripple voltage is
independent of supply voltage but %ripple depends on supply voltage. This would suggest as much as 0.2F for your supply, which should be made of a
number of paralleled capacitors as you will draw a pretty good ripple current.
Instead of capacitance, you may consider inductance instead. Some 6-8AWG around a couple MOT cores, suitably gapped, should provide enough inductance
to filter your load with little or no capacitance (none needed on the upper slope of the cell's exponential curve, but low voltage or current loads
will need capacitance).
Tim
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
I am using a variac to vary the AC on the primary of the fixed transformer , so that I can drop the secondary
voltage without dissipative losses any more than
necessary across any regulator . It would be great
if I could get the ripple as low as possible with low loss
filtering , and not have any active regulation stage at
all , essentially flattening the 120Hz pulsed DC to
as close as possible to its steady state rms value ,
by a brute force filtering approach .
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
Well, with a 10H choke of 1mohm resistance followed by about 100F filtering will certainly get a wonderfully smooth DC output, if you don't mind the
energy storage. Indeed, the ripple attenuation approaches infinity as L and C approach infinity. That much is obvious, and it follows that ideal
attenuation cannot be had in a finite universe.
Thus, the question is not "can I get the ripple as low as possible?", but rather it must be "how low can I get ripple, economically?".
Interestingly, you can also use phase control, instead of expensive variacs, to vary the output voltage when choke-input filtering is used.
Electrical control can then be added, allowing for electronically regulated response.
Tim
|
|
|
unionised
International Hazard
Posts: 5103
Registered: 1-11-2003
Location: UK
Member Is Offline
Mood: No Mood
|
|
To get slightly closer to the topic again; how much does the ripple matter to electrolysis?
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Suppose you have an electrolytic reaction which
may produce any of several different products depending
upon the applied voltage to the electrodes , and there is
only a couple of tenths of a volt variation which will cause
different materials other than the one desired to be the product . The swinging value of the voltage from ripple
could cause all the possible products to be produced in
sequence at the frequency of the ripple voltage , so you end up with a mixture of products instead of the one product for which a steady state and
specific voltage is required .
For many electrolytic reactions , an unregulated DC may
not cause any problem other than perhaps increasing
evolution of gas from the electrodes at a lower average
current , thus decreasing the efficiency of the cell .
But for reactions which are voltage sensitive with regards to the products possible .....the effect of a varying voltage could be disastrous to yield
by producing
unwanted byproducts at the same time as the one
product desired . This not only wastes precursor material
but may then complicate isolation . And if the different
products are reactive with each other , it could render
the hoped for reaction impossible all together using an
unregulated supply .
From what I am seeing too in the transformer manufacturers design recommendations , most are
saying to derate the AC current specified for a
transformer to which a full wave bridge rectifier and large capacitor filter is added by a factor of 1.8  .........which
really stinks because what that means is that is the
rectified and filtered DC output current available for regulation before it is even regulated , is only 55.5%
of the AC current rating of the transformer and that is
a huge bite , when a 100 Amp AC secondary is basically
only good for about half that current after the rectifier , filter , and regulator are added . And it is a fair bet
that the " lost efficiency " is dumped as heat which
adds another problem of how to dissipate that wasted
energy .
Switching supplies do better , but still have quite a bit of loss in their circuitry , the main advantage seeming to be
reducing component size and weight for working at higher operating frequencies .
From what I have seen looking at low voltage high current supplies in terms of efficiency and bulk , for
any sizeable electrolytic cell like a commercial reactor ,
on site generation using specially wound DC generators
and brush commutated DC would absolutely rule in
terms of economy of operation . These would be
driven by electric motors , or diesel or steam or hydro .
From the looks of it , when more than a couple of hundred amperes of DC requirement are presented ,
the engine driven DC generator will rule as a source
of power . And next to that , multielement polyphase AC alternators driving Schottky rectifiers , having their elements outputs incrementally phased
, could provide
a synthetic DC having a high enough frequency and
low enough ripple voltage that filtering alone would
suffice .
If a serious high current and * high efficiency * DC source is required , and what I understand at this point is
correct , it really needs to be done from the actual power generation stage up , and be addressed as a power generation project , not as a power
supply project which is simply plugged into the existing AC commercial grid .
Anything in the way of a bench size linear supply with filter and regulator which can be easily enough built from off the shelf components is probably
going to reside in the range of
10 to 50A output , and weigh about a kilo per Ampere for
the size of the cores , case and other components .
[Edited on 5-11-2006 by Rosco Bodine]
|
|
|
unionised
International Hazard
Posts: 5103
Registered: 1-11-2003
Location: UK
Member Is Offline
Mood: No Mood
|
|
So, unless you set up a potentiostat which monitors the working electrode's potential with respect to a reference electrode (Ag/AgCl is popular) it
won't work properly.
Of course, all those reduction potentials are concentration dependent so, not only do they change as the reaction goes on, but the depletion of the
reactants (and the buildup of products) in the vicinity of the electrode also messes things up. At least you need a good stirrer.
Fortunately, for the reaction we are looking at 2 H2O--> 2 H2 and O2, this doesn't matter much. If you ramp up the current density far enough then
you can get ozone (though some of the patent data indicate it's bloody hard to get better than 50% yield of ozone).
When the voltage drops then the current falls dramatically- you get a lousy waveform but I don't think it's worth producing a regulated smoothed
supply for stripping water from dilute sulphuric acid.
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Long before you get to the point of having concentrated
H2SO4 from water loss by electrolysis , what you will be doing is making persulfuric acid .
|
|
|
unionised
International Hazard
Posts: 5103
Registered: 1-11-2003
Location: UK
Member Is Offline
Mood: No Mood
|
|
That's OK by me. I get rid of hydrogen by electrolysis; some of the excess oxygen gets incorporated into persulphuric acid and some leaves as
ozone/O2.
I add a little Pt as a catalyst to decompose the persulphuric acid and I've got conc H2SO4.
I think the really big problem here is the number of amp hours it takes to get rid of a litre of water.
A litre is about 56 Moles of H2O so I need to convert about 112 Moles of H+. The Faraday constant (IIRC) is about 100000 Coulombs per mole so I need
to get 11200000 Amp seconds of charge through the stuff. Something like 30 amps continuously for 4 days. (could someone check that please).
I'd be impressed by the electrodes that stood that.
Suddenly distilling out the water looks like a really good idea.
|
|
|
Rosco Bodine
Banned
Posts: 6370
Registered: 29-9-2004
Member Is Offline
Mood: analytical
|
|
Does the process go all the way to oleum ?
And what sort of voltage increase is experienced due to
the drastic increase of resistance as all water disappears
from the H2SO4 ? Isn't it something of a contest to see
which will happen first , electrolysis or actual boiling of the acid ?
[Edited on 5-11-2006 by Rosco Bodine]
|
|
|
| Pages:
1
2
3
4
5 |
![[*]](images/xpblue/default_icon.gif){kind=icon}
