[AndersHoveland]

What about this?
The interesting thing about this reaction is that the gold will reappear after the dissolved solution of gold is diluted in water! This reaction, however, involves much more concentrated acids, and the procedure is much more dangerous. So I do not suggest you try this reaction unless you have a good chemistry background and know about the proper safety precautions. This post is more for information purposes.
 
Dissolving Gold with Concentrated Nitric and Sulfuric Acids
 
A hot mixture of concentrated nitric and sulfuric acids can dissolve gold, with lower oxides of nitrogen forming. Addition of water caused the gold to precipitate back out in metallic form, but if a solution of permanganate is used instead, the gold remain dissolved.
 Reynolds, later by Spiller
 Chemical engineering, Volume 2, p316
 
The text also mentions that even concentrated mixtures of nitric and phosphoric acid attacks gold at room temperature, although the reaction is very slow unless heated.
 
The reaction is probably:
  3Au + 3NO₃^- + 18H⁺ --> 3Au⁺³ + 3NO⁺ + 6H₃O⁺
 (3)Au + (3)NO3[-] + (18)H[ +] --> (3)Au[+3] + (3)NO[ +] + (6)H3O[ +]
 
The nitrosyl ion, NO[ +], exists in the form of nitrosyl sulfuric acid, ONOSO3H.
 
The nitronium cations, NO2[ +], which form in equilibrium in concentrated nitric acid solutions, probably initially attack the gold, creating nitrogen dioxide. Basically,
 Au + (3)NO2[ +] --> Au[+3] + (3)NO2
 The nitrogen dioxide produced would likely remain in the concentrated acid,
 (2)NO2 + (3)H2SO4 --> NO[ +]HSO4[-] + NO2[ +]HSO4[-] + H3O[ +]HSO4[-]
 and the nitronium ions formed from the NO2 would then attack more gold.
 

Excess sulfuric acid needs to be used. This is an equilibrium reaction, and the gold is not going to dissolve easily. The mixture needs to be extremely acidic. Even a 1:1 ratio of 70% HNO3 to 95% H2SO4 is not going to be concentrated enough. For good results, use a 1:10 rato of 70% nitric acid to 98.5% concentrated sulfuric. Essentially, there can be no water in the reaction!
 
Even in the hot boiling mixed acids, the gold takes several minutes to dissolve.
 
The NO[ +] ion hydrolyzes (reacts with) water to form nitrous acid.
 
NO[ +] + (2)H2O --> HNO2 + H3O[ +]
 
Nitrous acid is fairly reactive, and can act as either a reducing or an oxidizing agent. It will reduce the dissolved gold (Au+3) to elemental form (Au). This explains why the gold precipitates back out when the reaction is diluted with water.
 
(2)Au[ +3] + (3)H2O (3)HNO2aq --> (2)Au + (6)H[ +]aq + (3)HNO3aq
 
(note that "aq", which stands for "aqueous", means it is dissolved in water)
 
If fuming nitric acid is added to the reaction containing the dissolved gold, the gold will solidify out as a purple solid. The gold is probably still in its elemental form, but small particle sizes of gold are known to exhibit strong colorations, from red to purple.
 
Nitrous acid is unstable, and only exists in the form of solutions which gradually degrade after several minutes. Solutions of nitrous acid exist in equilibrium with nitrogen dioxide and nitric oxide, the latter of which is an unstable radical which can either react with the oxygen in air to form more nitrogen dioxide, or if left on its own will disproportionate into nitrogen dioxide and nitrous oxide after several minutes.
 
(2)HNO2 <==> H2O + NO2 + NO
 
(3)NO --> N2O + NO2
 

In the reaction,
 (2)Au + (3)NO3[-] + (18)H[ +] --> (2)Au[ +3] + (3)NO[ +] + (6)H3O[ +],
 sulfate ions are not shown because they do not directly take place in the reaction. The literature even states that phosphoric acid can be used in place of the sulfuric acid.
 
The above reaction is in ionic form. Some of you may prefer to see it in the form:
 (2)Au + (3)HNO3 + (15)H2SO4 --> (2)Au(SO4H)3 + (3)NOSO4H + (6)H2SO4*H2O
 
Note that the "Au(SO4H)3" only exists in the solution, it cannot be isolated. Gold trinitrate, if it even exists, would also be nearly impossible to obtain as a pure solid. Gold trinitrate only exists in highly concentrated solutions of nitric acid. When these solutions are diluted with water, auric oxide precipitates out. Similarly, auric oxide only only dissolves in very concentrated acids, since it is only very weakly basic.
 
Au2O3 + (9)HNO3 <==> (2)Au(NO3)3 + (3)HNO3*H2O


The reaction is more interesting from a chemical perspective than a practical way to refine out gold. Nevertheless, the reaction may be useful to directly dissolve gold-silver alloys, without having to go to the trouble of inquartation, since aqua regia only dissolves such alloys with extreme difficulty.
 

Procedure and Precautions:
 
Yes, it is extremely dangerous. The dangers of using concentrated mixed acids are commonly taken for granted among those that frequently perform nitrations. Obviously those unfamiliar with such procedures should think twice before handling such high concentrations of acid.
 
More details about the reaction. The concentrated acid mix that contains the dissolved gold should be gradually transfered into the larger bowl of water using a 10mL glass transfer pipette. You will also need a rubber pipette suction bulb. For those of you unfamiliar with this tool, it is basically like a turkey baster that is used to suck up a small quantity of liquid, then move it to another container. The pipette can be bought here:
 http://www.pelletlab.com/pipette
 
Using the pipette to slowly add the acid mixture to the water is important for two reasons. First, safety. Water should never be added to concentrated acid, since this can result in the acid spraying up. Neither should the acid be poured into the water, because of the possibility of an accidental spill or splashing, and because it can be hard to control the rate that the liquid is poured in. Adding the acid in too fast can lead to overheating, which could result in boiling/splashing in the water. Second, it is important that each small portion of the acid quickly be diluted with as much excess water as possible. This will help prevent the gaseous nitrogen oxides (NO and NO2) from escaping. Although nitrosylsulfuric acid reacts with excess water to form a solution of nitrous acid, if not enough water is used nitrogen oxides will bubble out instead.
 
There will inevitibly be some loses of nitrogen oxides, in the form of some bubbling and some brown gas being given off. Unfortunately, when some of the nitrogen oxides escape, there will not be enough nitrous acid to completely reduce the gold. After neutralizing, all the gold will still precipitate out, but a small portion of it will be in the form of hydrated gold oxide, Au2O3. If the gold is going to later be melted, the gold oxide should not pose any problems, as the compound decomposes to the pure metal at 160°C, giving off oxygen gas.
 
One other note of warning, unless the gold oxide has been completely reduced, it should not be reacted with ammonia, as this will form the dangerous sensitive explosive known as "fulminating gold". In the event that the acid solution was previously boiled with ammonium sulfate to prevent precipitation of the gold, fulminating gold can result upon neutralization if too much ammonium sulfate was added.
 

More safety information:

Only use small quantities of mixed acids at a time. Be aware that with concentrated acids, even tiny drops can splash out and result in painful burns on exposed skin. To get some understanding of these dangers, try pouring cranberry into a glass, wearing a clean white long-sleaved shirt. Even with cautious pouring, you are likely to find one or two tiny little red stains on the sleeves afterwards, even though you were not aware of any splashing while the juice was being poured. If this was concentrated acid, painful burns would have been felt.
 
You may desire to cover your shoes with a plastic bags and a rubber band, so that if any of the acid spills onto the floor, it will not seap into your shoes. Protective shoe coverings can also be purchased:
 http://www.labsafety.com/search/shoe%2Bcovers/
 If you choose to wear rubber boots instead, it is advised that the top of the rubber be tied tight around your legs, so that if any of the acid is spilled on you, it will not drip down into the boots and collect in a puddle. If the acid is in contact with your skin for more than a few seconds, the burns will be much more severe. http://www.amazon.com/b?ie=UTF8&node=393294011
 
A boiling mixture of concentrated nitric and sulfuric acids is extremely dangerous, much more so than 70% concentrated sulfuric acid, for example. The chemistry of this mixture presents several unique hazards. Extremely concentrated sulfuric is a strong dehydrating agent, that will turn anything organic, such as a strip of paper or your skin, into black char immediately on contact. A note about treating concentrated nitric acid burns, after you immediately rinse the affected area with plenty of water, and neutralize with sodium bicarbonate solution, there is special recommendation for concentrated nitric acid burns. Use a swab dipped in chlorine bleach to gently scrub the affected area. Some of the yellow color from the burn should be absorbed onto the cotton swab. Continue to scrubbing with fresh swabs until no more yellow can be absorbed onto the cotton. Then rinse well in soapy water. Doing this will help remove some of the nitro compounds which have formed. These compounds act as allergens and greatly slow the healing process. In fact nitric acid burns take much longer to heal than sulfuric acid of the same concentration. The unique effects of concentrated nitric acid are due to the formation of nitronium ions, NO2[ +], in equilibrium in the solution. The addition of highly concentrated sulfuric acid greatly enhances this equilibrium, and so the special burn effect of nitric acid will be greatly exaggerated by the acid mixture. In other words, it would be very important to treat the burns in the way described above, and the healing time is likely to be much longer.
 
Practical Applications:
The extremely concentrated HNO3/H2SO4 mixture might be useful for dissolving gold-silver alloys, without the need for inquartation, since aqua regia only dissolves such alloys with extreme difficulty.
 

[AWK]

Coefficients before reagents should be written without brackets.
Since ^{AndersHoveland} suggests using acid solution (70 % and 98 % not anhydrous ones) then writing acid hydrates (H2SO4*H2O, HNO3*H2O) makes no sense.