Aqua regia (Latin: Royal Water) is one of the strongest acids known in Chemistry, and is capable of dissolving gold and platinum.

My copy of the Oxford science dictionary goes on to say (under the entry: Aqua regia) that metallic silver does not dissolve in aqua regia. Morever it does not mention any other examples of aqua regia-resistant metals. Further down, it mentions that silver's invulnerability to aqua regia is due to the formation of a protective silver chloride coating on the metal, which serves to protect the metal from further decomposition.

However, this Wikipedia article claims:

...aqua regia does not dissolve or corrode silver...

This, I find contradictory to the dictionary's "formation of silver chloride" claim.


  1. What metals (elemental, forget alloys) are neither attacked by nor dissolved in (freshly prepared) aqua regia?

  2. What makes those metals that don't dissolve or corrode in aqua regia so impervious to the acid?

  3. Does silver metal actually develop a silver chloride layer on exposure to aqua regia? If so, would that mean the Wikipedia article is incorrect?

  • 6
    $\begingroup$ The wikipedia article gives an answer to 1) above: "However, aqua regia does not dissolve or corrode silver, titanium, iridium, ruthenium, rhenium, tantalum, niobium, hafnium, osmium, or rhodium" i.e. the usual suspects +titanium and silver, -platinum $\endgroup$ – Ian Bush Jan 27 '17 at 14:32
  • 6
    $\begingroup$ Aqua regia isn't particularly strong acid. It's just nitric acid with HCl added, what may or may not help in dissolving metals. $\endgroup$ – Mithoron Jan 27 '17 at 15:55
  • 2
    $\begingroup$ @Mithoron yes, important point that the critical property is not its strength $\endgroup$ – Greg Jan 27 '17 at 16:16
  • $\begingroup$ TIL why rhodium is used to plate fountain pen nibs. $\endgroup$ – Charles Duffy Jan 27 '17 at 22:53

Keep in mind

The answer will depend upon the reaction conditions. Most importantly,

  • physical state of the metal: porosity, degree of comminution;
  • temperature;
  • mechanical aggravation of metal surface during reaction.

Often times a chemistry text mentions that no reaction occurs. The reaction might still happen. It is just that for the specified parameters the process is meaningless and negligible.

Short overview of aqua regia

Aqua regia is the $3:1$ volumetric mixture of $\ce{HCl}$ and $\ce{HNO3}$. Its additional reactive power draws from monochlorine created in situ.$^{[1]\ [2]}$

$$\ce{HNO3 + 3HCl -> Cl2 + NOCl + H2O\\ NOCl -> Cl + NO}$$

  1. Which metals are impervious to $3:1\ \ce{HCl/HNO3}$?

Almost every metal will react with aqua regia provided certain criteria are met.$^{[1]\ [2]}$ The closest you will probably get is ruthenium $\ce{Ru}$, and perhaps osmium $\ce{Os}$. To the best of my knowledge, $\ce{Ru}$ will not react with aqua regia in a meaningful way even if aqua regia is boiling.$^{[2]}$ The difference with $\ce{Os}$ is that powdered osmium is attacked by boiling aqua regia.$^{[1]\ [2]}$

$$\ce{Ru + HNO3 + HCl $\kern.6em\not\kern -.6em \longrightarrow$}$$

$$\ce{\underset{powder}{Os} + $\underbrace{\mathrm{HNO_3}}_{\text{boiling}}$ -> OsO4 + N_xO_y + H2O \\ OsO4 + 2H2O <=> H2[OsO4(OH)2] \\ OsO4 + HCl ->OsO2Cl2 + Cl2 + 2H2O\\ OsO2Cl2 + HCl ->OsCl4 + Cl2 + 2H2O\\ 2OsO2Cl2 + H2O <=> OsO2 + H2[OsO2Cl4] \\ 3OsCl4 + 2H2O <=> OsO2 + 2H2[OsCl6]\\ OsO2 + 6HCl <=> H2[OsCl6] + 2H2O}$$

Brief discussion about the list provided in the comments

Titanium $\ce{Ti}$ does react, and does so at room temperature.

$$\ce{3Ti + $\underbrace{\mathrm{12HCl + 4HNO_3}}_{\text{room temperature}}$ -> 3TiCl4 + 4NO + 8H2O}$$

Rhenium $\ce{Re}$ reacts slowly at room temperature $\ce{->HReO4}$. This will further react with $\ce{HCl -> ReCl4 + Cl2}$.$^{[2]}$

Hafnium $\ce{Hf}$ does react at room temperature. The reaction is slower than with titanium; overall equation is identical.$^{[2]}$

Tantalum $\ce{Ta}$ reacts when aqua regia is heated to $150\ ^{\circ}\mathrm{C}$. Rhodium $\ce{Rh}$ reacts in a grinded state. As a large compact piece, iridium $\ce{Ir}$ is affected over temperatures of $100\ ^{\circ}\mathrm{C}$. Niobium $\ce{Nb}$ is inert at room temperatures.$^{[2]}$

Summary: ruthenium $\ce{Ru}$ is your best bet.

  1. What makes metals $\ce{Ru}$ and $\ce{Os}$ so stable in aqua regia?

The nobility of these metals is not the best explanation. As you correctly pointed out, $\ce{Pt}$ and $\ce{Au}$ react fine. This is direct evidence that for other metals a protective layer should form. The layer varies from metal to metal, but usually is either an oxide (or oxide hydrate), or a chloride.

Effectiveness of mechanical aggravation also points to stable, non-reactive compound formation on the metal's surface.

For ruthenium, as of now I am unsure what this precipitate could be. If anyone has a reference, please edit or leave a comment.$^\text{[reference needed]}$

  1. What happens with silver?

Silver and aqua regia react very poorly, and for a short amount of time.$^{[2]}$ The culprit is $\ce{AgCl}$ ($K_s = 1.8 \cdot 10^{-10}$)$^{[2]}$. A slow reaction might still take place due to complexation.$^{[2]}$

Surprisingly, silver reacts with $\ce{HBr}$!$^{[2]}$ Its solubility product is even worse, $K_s = 5.0 \cdot 10^{-13}$.$^{[3]}$ My guess is that this layer is not as dense as $\ce{AgCl}$ but this still needs verifying.$^\text{[citation needed]}$


(In progress)

$[1]$ N. N. Ahmetov. Anorgaaniline keemia. (1974)

$[2]$ H. Karik, Kalle Truus. Elementide keemia. (2003)

$[3]$ Skoog, West, Holler, Crouch. Fundamentals of Analytical Chemistry. 9th edition. (2014)

  • $\begingroup$ Possibly bromide ion forms a soluble complex with silver, more easily than chloride ion? $\endgroup$ – Oscar Lanzi Jan 28 '17 at 3:58
  1. What metals (elemental, forget alloys) are neither attacked by nor dissolved in (freshly prepared) aqua regia?

At normal condition, most of the metals are impervious to aqua regia but if the conditions are changed like temperature, pressure, acid concentration or state of metal, they can be prone to attack.

The metals discussed falls in the category of noble metals. These metals resists corrosion and oxidation in moist air and most of the acids including aqua regia. But there are exceptions. More specifically:

Platinum, gold and mercury can be dissolved in aqua regia but iridium and silver cannot. Palladium and silver are, however, soluble in nitric acid. Ruthenium can be dissolved in aqua regia only when in the presence of oxygen (Why? will be asked in a separate question), while rhodium must be in a fine pulverized form. Niobium and tantalum are resistant to all acids, including aqua regia.

This handbook1 gives an overview of which noble metals are attacked by aqua regia:

  1. Niobium is reported a corrosion rate of 0.025mm/yr in aqua regia at 328.15 K.
  2. Gold is attacked by aqua regia.
  3. Iridium resist both cold and hot aqua regia but dissolves under pressure at 523-623 K.
  4. Platinum is attacked by aqua regia.
  5. Rhodium unattacked by aqua regia in either wrought or cast form.
  6. Ruthenium is by far the most chemically resistant of the noble metal, unattacked either in hot or cold aqua regia.
  7. Tantalum also not attacked by aqua regia even in elevated temperatures.
  8. Zirconium is attacked by aqua regia.
  9. Hafnium is dissolved in aqua regia only in the addition of some soluble fluoride salts.

So, the winners are ruthenium and tantalum.

  1. What makes those metals that don't dissolve or corrode in aqua regia so impervious to the acid?

Maybe there is some sort of protective layer like oxide or chloride layer that protects the metal from further attack. It can be explained by studying their potentials. Most of the noble metals sit right at the bottom of electrochemical series due to their high positive potential value. So, they are quite resistant to attacks.

  1. Does silver metal actually develop a silver chloride layer on exposure to aqua regia? If so, would that mean the Wikipedia article is incorrect?

No, silver actually forms a silver chloride layer that protects the metal from further attack. Aqua regia can only dissolve silver if the chloride layer is ruptured.


  1. https://en.wikipedia.org/wiki/Noble_metal
  2. https://en.wikipedia.org/wiki/Aqua_regia
  3. Handbook of Corrosion Data, Bruce D. Craig, David S. Anderson ASM International, 1994
  4. Update: Can aqua regia/royal water dissolve lead?
  • 2
    $\begingroup$ Regarding "No [the Wikipedia article is not incorrect], silver actually forms a silver chloride layer that protects the metal from further attack", I rather took @paracresol's meaning to be that the formation of a protective chloride layer is, in itself, a form of corrosion (albeit minimal, and self-limiting). Perhaps p-cresol could confirm/deny/clarify? $\endgroup$ – owjburnham Jan 27 '17 at 16:05
  • 1
    $\begingroup$ @owj That's precisely what I had in mind. The formation of a silver chloride layer does appear to be a form of corrosion. $\endgroup$ – paracetamol Jan 27 '17 at 16:09
  • 5
    $\begingroup$ I'd say corrosion necessarily includes the spalling of oxidised metal to expose more unreacted metal. Otherwise it's just passivation. $\endgroup$ – gsurfer04 Jan 27 '17 at 16:40
  • 6
    $\begingroup$ @gsurfer04 Concur with this terminological point: if the reaction is self-terminating, it's passivation; otherwise, it's corrosion. $\endgroup$ – hBy2Py Jan 27 '17 at 17:33
  • $\begingroup$ Definitions that I can find online for "corrosion" all seem to reference "gradual destruction". So perhaps we could argue that it needs to be an ongoing process to constitute "corrosion". Coming at it from a more intuitive, empirical, chemist's perspective, I don't imagine that any of us would argue that oxygen in the air corrodes aluminium. It does react and, in the process, forms a protective oxide layer which prevents corrosion. But it feels wrong to describe that passivation step as a kind of 'corrosion'. $\endgroup$ – owjburnham Jan 28 '17 at 13:45

Passivation does not influence potential which is thermodynamic, however it hinders or even completely stops dissolution by aqua regia as in Rh, Pt in the cold etc


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.