First - lets consider the well behaved compound. Sodium chloride comes in one form - NaCl. When you crystallize sodium chloride from a solution containing both chloride and hydroxide ions you get (nearly) pure crystals of sodium chloride. You don't get Na2Cl(OH) or Na3Cl(OH)2 Physical chemical experiment show that NaCl crystals contain one type of sodium ion and one type of chloride ion. Every sodium ion has the exact same environment as every other sodium ion. The crystals are regular. All crystals of sodium chloride are cubes (okay, there are a few rare twinned cubes and other defects but they can be explained as variations on cubical crystals) and have the same density and the same refractive index. They would all be transparent if they weren't filled with cracks from grinding. Solutions of sodium chloride are all the same color (clear), and have the same behavior. The behavior of sodium chloride is the same if you make it by reacting gaseous chlorine with sodium metal, or by neutralizing NaOH with HCl and evaporating off the water.
Copper is very different. The same compound of copper can come in many different forms, and there are often many different but closely related compounds with similar formulas.
Consider copper chloride. When copper chloride is dissolved in the minimum amount of water it forms a brown green solution. As more water is added the solution goes to dark green (50%), than green, then to a light green (25%) then to blue green, then to a greenish blue and finally to sky blue (16%). This is in direct contrast to most compounds which show the same color, and simply show changes in the amount of color. Another surprising feature is that the conductivity doesn't change linearly with concentration. Since ions conduct electricity in water solutions, the conductivity is a good measure of the number of ions in solution. If one starts with a dilute solution of CuCl2 in solution, and add more of the CuCl2 the conductivity first increases, and then decreases! It is at a maximum at 28%. That means that a 40% CuCl2 solution has fewer ions than a 28% solution! Some of the ions must be recombining to form covalent compounds as the concentration increases.
Next, consider copper oxychloride. Copper oxychloride has the nominal formula Cux(OH)yClz, where x, y and z can be whole or even fractional numbers. This is very different from a mixture! You can mix 3 NaOH with 1 NaCl to get a mixture that has the bulk composition corresponding to Na4(OH)3Cl, but the material behaves as if it were a physical mixture - it shows heat absorption characteristics of the two individual compounds, it is possible to leach out either the NaCl or NaOH leaving the other material behind, etc. It is not a true compound. But each of the 16 known copper oxychlorides each behave as a single compound distinct from the other 15! It is impossible to dissolve out the CuCl2 leaving insoluble Cu(OH)2 behind. Each compound has it's own crystal structure, it's own heat absorption, it's own magnetic properties, and crystal structure. How can this be?
However, the 6 waters are not all bound with equal strength - an opposing
pair of the waters (or ammonia molecules or chloride ions, etc.) are further
away and less strongly bound. Thus, the ion can be viewed as
a square planar complex with two loosely associated ligands.
In fact, this helps explain the common hydrated form of CuCl2
- which is CuCl2 • 2 H2O. The chlorines
occupy two of the corners, while the waters occupy two of the remaining
corners. Of course, something must happen when the waters are removed
to form anhydrous copper (II) chloride. That will be discussed
later, when discussing the structure of various copper compounds
in the solid phase.
The second feature that is important is the low electronegativity of
copper compared to (for example) sodium. When sodium reacts
with chlorine the electron is complete transferred to the chlorine,
affording Na+ and Cl-. While the two
ions stack in crystal of NaCl, they show little affinity for each
other in aqueous solutions. This isn't the case for copper.
The valence electrons in copper are held more tightly than in sodium.
This leads to a much greater degree of covalent character in copper compounds.
This affects the equilibrium behavior of copper in solution.
Consider the following reaction:
MX(solution) <- -> M+ (solution) X- (solution)
In a completely ionic compound the two ions are completely separated
and the equation proceeds to the right. In a completely covalent
compound (such as the C-H bond in hydrocarbons) the bond doesn't disassociate
to any practical degree, and the equation remains (proceeds) to the left.
However, a compound that shows mixed behavior has all the species present
and shows a sensitivity to the concentrations of the various components
of the solution. That's the case with copper since it is partially
covalent and partially ionic.
If CuCl2 is dissolved in water, then some of the compound
will exist as undisassociated CuCl2 while some of it will
exist as various ionic species. The ratio of the ions to the unionized
species is constant, and is called the equilibrium constant.
The concentrations of the species is represented by enclosing their names
in square brackets, so the equilibrium constant for the first ionization
of CuCl2 might seem to be
Keq1 = [CuCl+] [Cl-] / [CuCl2]
However, that clearly isn't correct. If this were the case, then doubling the amount of CuCl2 in solution would lead to an increase of the number of ions (otherwise the equilibrium wouldn't be constant). But remember that when the concentration of CuCl2 goes past 28%, the number of ions DECREASE. This is due to the fact that there are a vast number of competing reactions;
dissolution
CuCl2 (solid) + 4 H2O <- -> CuCl2 • 4H2O
(solution) starting from anhydrous
CuCl2 • 2 H2O (solid) + 2 H2O <- -> CuCl2
• 4H2O (solution) starting from non anhydrous
ionization
CuCl2 • 4H2O + H2O <- ->
CuCl+ • 5 H2O + Cl-
CuCl+ • 5 H2O + H2O <- -> Cu++ • 6 H2O +
Cl-
Lewis acid/base
Cu++ + H2O <- -> (CuOH)+
+ H+
Cu++ + 3 Cl- + H2O <- ->
(Cu(OH)Cl3)-
Ligand displacement
CuCl2 • 4 H2O + Cl- <- -> CuCl3- • 3 H2O
+ H2O
CuCl3- • 3 H2O <- -> CuCl4-- • 2 H2O
+ H2O
Condensation
CuCl2 • 4 H2O + CuCl2 • 4 H2O
<- -> Cu2Cl4 • 6 H2O + 2 H2O
Cu2Cl4 • 6 H2O + CuCl2 • 4 H2O <-
-> Cu3Cl6 • 8 H2O + 2 H2O
-- etc. ---
All of these reactions can happen simultaneously, but since the have different equilibrium constants they form a dynamic network, and the ratios of the products are exquisitely sensitive to concentrations of each of the many components. At least two features of the competing reactions are reducing the number of ions - lack of water and condensations.
Note that in the first two reactions each ionization requires a molecule of water to take the place of the chlorine in the octahedral complex. As the concentration of the CuCl2 increases beyond 28%, this starts to significantly reduce the amount of water in the solution (after all - every percent increase in CuCl2 is a percent decrease in the amount of water). Applying Le Chatelier's principal, we can see that reduction in the amount of water present will reduce the amount of ionization.
Secondly, in very high concentrations the actual nature of the copper species begins to change. Since the dissolution process consumes water, other reactions which afford free water become more favorable. The condensation reactions perform this exact feat. At any given moment the copper of CuCl2 can utilize any lone pair in the solution to satisfy the Lewis acid/base reactions. This would include the lone pairs of the water, or the lone pairs of the chlorines of another CuCl2 molecule. There is a slight preference for the use of water, but the copper can utilize those chlorines IF there isn't enough free water around. The copper chlorides begin to form a sort of inorganic polymer, as shown here (L represents a generalized ligand, which can be water (or as we will see later) other things).
Thus, as the solution becomes more concentrated, we see two simultaneous effects: the ionization becomes less preferred, and the condensation polymer becomes more preferred. This explains the reduction in the number of ions once the concentration of CuCl2 exceeds 28%.
This helps explain the unusual color changes on increasing the concentration of CuCl2 in solution. In very dilute solutions the copper exists as mostly Cu++ ions. Thus, the spectrum of dilute CuCl2, CuSO4, Cu(NO3)2 are all similar at high dilution. However, as CuCl2 is added the concentration of CuCl+ increases, then the concentration of solvated CuCl2, and finally the brown polymer is formed. This also explains the pH sensitivity of CuCl2 solutions. While addition of HO- gives various basic salts (see below), the addition of acids such as HCl gives new colors. When HCl is added, the ligand displacement reactions shown above occur, and the presence of yellow green CuCl3- and yellow CuCl4- - influences the color. Note that this can occur even with non-acids, such as NaCl.
Click here to see a large (250K) ray trace diagram of the crystal structure of CuCl2. The metallic red balls are the copper atoms, and the green balls are the chlorines. Note that there are 6 yellowish-green balls and 1 yellowish-red ball in the center of the diagram. These are highlighted to help identify one copper octahedron. There is also a raytraced drawing of a sideways view of 4 ribbons of poly CuCl2. It is perhaps easier to see the way the ribbons of poly CuCl2 stack in this view.
Consider a simple reaction;
MXn + n AY <-- --> MYn + n AX
Note that this reaction will be in equilibirum. If all four products are soluble then the reaction may in fact just be a complete disassociation to form a pool of all four ionic species. If one or more of the products is insoluble then that will drive the reaction to that side of the arrow. Thus, the reaction of AgNO3 with NaCl will form the insoluble AgCl, leaving soluble NaNO3 in solution.
We are used to thinking that this reaction goes cleanly. That
is due to two facts: 1) Examples where this reaction
goes cleanly abound. They may be more common than the messy examples,
although I've never done a survey. 2) The examples used in
chemistry classes tend to focus on the clean reactions.
However, the reaction can actually become very messy. Consider the
following hypothetical reaction;
CuCl2 + 2 NaOH --> Cu(OH)2 + 2
NaCl
(or CuSO4 + 2 NaOH --> Cu(OH)2 + Na2SO4,
etc.)
If you were to go into the lab and try this out you almost certainly wouldn't get Cu(OH)2. This is due to the fact that the species reacting with the HO- isn't Cu++, but a combination of various solvated halides, such as CuCl+ • 5 H2O, etc. If it were Cu++ • 6 H2O, then the following reaction sequence might occur;
Cu++ • 6H2O + HO- --> CuOH- • 5 H2O (sol)
CuOH- • 5 H2O (sol) + HO- --> Cu(OH)2
• 4H2O (insol)
However, at 'reasonable' concentrations of CuCl2 in water there is substnatial amounts of CuCl+ • 5 H2O. This causes a problem because the intermediate formed is not soluble - when the following reaction occurs
CuCl+ • 5 H2O + HO- --> Cu(OH)Cl • 4 H2O (insol) + H2O
the intermediate precipitates out of solution. This makes it harder for the secondary displacement with hydroxide to take place, and the second step occurs slowly. Thus, simple mixture of CuCl2 with NaOH will yield copper oxychloride, copper hydroxide, and various other species. A trick has to be used to make copper hydroxide cleanly. The reason the copper oxychloride precipitates is that the OH of one copper can easily hydrogen bond to the Cl of the next copper, forming a long chain which forms a colloidal precipitate (this can rearrange, as discussed later). Thus, the chloride has to be replaced with something that doesn't form hydrogen bonds so easily. This is accomplished by adding ammonia to the solution. The ammonia displaces the chloride and the water to form a tetraammine complex - [Cu(NH3)4]++ • 2H2O. When the displacement of the NH3 occurs, the intermediate is still soluble
[Cu(NH3)4]++ • 2H2O + HO- --> [Cu(OH)(NH3)3]+ • 2H2O (sol)
Once the second OH- is added, the copper hydroxide assembles itself
into long chains similar to the CuCl2 diagram above to give a fluffy colloidal
mass of hydrated Cu(OH)2 where L is water, and eventually loses the solvating
water to form a dense crystalline mass of poly-Cu(OH)2.
This can be converted to CuO by heating. The Cu(OH)2 can also
be used with acids to form various salts, such as perchloric acid to form
copper perchlorate, salicylic acid to form copper salicylate and 3,5-dinitrobenzoic
acid to form copper 3,5-dinitrobenzoate.
The lattice is very complicated. However, we can build up to an understanding of that lattice in a stepwise fashion. After all, a crystal is simply built up from a repeating network of small "unit cells". If you understand the unit cell and the way the cells align themselves side by side, then you understand the crystal.
The repeating unit may seem to be Cu(OH)Cl. However, remember that Cu wants to have 6 ligands - it wants to be in a distorted octahedron. A single CuOHCl molecule is bent. If another CuOHCl molecule 'backs into' the cavity formed by the bend the unshared electrons on the oxygen of the OH and the chlorine can "bite" onto the copper. This is called a 'bidentate' ligand. This satisfies part of the need of the copper to have 6 ligands. It now has four ligands. This process can occur over and over to create long ribbons of poly-CuOHCl. The question then arises how this polymerizes - top to top, or top to bottom (or randomly).
Because of various considerations (dipoles of the bonds, etc.) the form on the left is preferred. (Although there are compounds which contain the form on the right.) Okay, that gives us the unit cell. It is a long ribbon of repeating CuOHCl units. It has the shape of a thick ribbon. So how does these ribbons arrange themselves into crystals?
There are many factors that affect crystal packing - shape considerations,
charge attraction or repulsion, ligand sharing, etc.
If you remember the previous diagram you will remember that each
copper wants 6 ligands. The ribbon only has 4 ligands for each
copper. Thus, it needs two more. When Cu(OH)Cl initially
precipitates from solution it forms a delicate colloid that has an poorly
defined structure. That structure consists of numerous small lengths
of ribbon that has waters occupying the remaining two sites.
As the precipitate ages it loses that water, and the layers line up so
that the chlorines or hydroxyls of one ribbon are shared with the copper
of an overlying ribbon, etc... This is shown diagrammatically below,
and in a nicer ray traced drawing (207 K).
Note in the lower left hand side of the ray traced diagram that there are
3 chlorines that are more yellowing, and three hydroxyls that are plaer
blue, and a copper that is a fainter red. These have been colored
so that you can see how the octahedral coordination has been accomplished.
This is the simplest copper oxychloride, and has the formula Cu(OH)Cl. However, at this time I would like to introduce a shorthand to make the rest of the discussion easier. The copper oxychlorides will be given three numbers, representing the ratio of CuO, CuCl2 and water in the crystal. Note that this does not mean that the crystal is made up of CuO and CuCl2 repeating units... in fact the ratio of 1:1 CuO and H2O may indicate either a divalent oxygen of the CuO type along with a water of crystallization, or the water and CuO may have reacted to form Cu(OH)2. Thus, the number doesn't exactly specify the components of the crystal. It's just a convenient way to quickly organize the various forms. Thus, Cu(OH)Cl is categorized as 1:1:1. Please note that this is not a unique identifier, as several different crystal forms can have the same identifier - 1:1:1 can represent both the structure given above and a crystal of alternating ribbons of CuCl2 and Cu(OH)2. Both structures are known.
These leads to additional complexity. If we represent Cu(OH)Cl ribbons as A, CuCl2 ribbons as B, and Cu(OH)2 ribbons as C, we can then represent a crystal of pure CuCl2 as ...BBBBBB..., and a crystal of pure Cu(OH)2 as ...CCCCC...., but we find that we can make up crystals that have the bulk composition "Cu(OH)Cl" in many different ways;
When would such amorphous solids be produced? Unfortunately,
the answer is "in situations such as the industrial production of Cu(OH)Cl".
To understand this we need to take a brief glimpse at the process of crystallization.
Imagine a 'crystal' formed out of styrofoam spheres (atoms and atomic fragments
do act kind of sticky... this is called van der Walls forces).
If you pour a whole bunch of styrofoam spheres into a box it is very unlikely
they will acheive a perfect packing - there will probably be voids and
displaced spheres. If you want them to pack as closely as possible
then you would be well advised to pour them in slowly while shaking the
box. Rapid pouring into a still box is analogous to rapidly producing
a precipitate at a low temperature. This is ideal for making as much
solid as quickly as possible, but bad for making crystals!
It's even worse if you are trying to co-cyrstalize two different types
of molecules. Imagine trying to mix tennisballs and basketballs in
the box. The most dense packing is whee the basketballs form a regular
lattice with the tennisballs fitting into the gaps where 4 basketballs
come together. This will only be acheived with slow crystalization.
The same is true when trying to co-crystalize ribbons of Cu(OH)2 with CuCl2.
(The selection criterion isn't size, but electrostatic attractio, however
it gives tthe same overall effect). If the material is produced
slowly at an elevated temperature, then it can form perfect alternating
crystals. However, if it is "crashed out of solution" by rapid mixing
and colling, then the ribbons will just align themselves in anyway they
can and form a poorly defined amorphous compound.
Now, back to considering pure crystals of copper oxychlorides. We have discussed how crystals of the composition Cu(OH)Cl can be formed in many different ways. This alone gives a lot of complexity to the Cu(OH)Cl question. However, that's not the end - you can also get mixed crystals such as Cu(OH)Cl • CuCl2 (1:3:1) and Cu(OH)Cl • Cu(OH)2 (3:1:3). This gives compounds with the overall compositions Cu2OHCl3 and Cu3(OH)3Cl. In the latter compound there are alternating ribbons of Cu(OH)Cl and Cu(OH)2. A ray traced diagram of this shows how they alternate.
This explains several of the known copper(II) oxychloride structures - There is the alternative form of 1:1:0 as mentioned above. There is a 2:1:0 form that doesn't occur in nature, as well as 3:1:0 (mentioned above) which occurs in nature as paratacamite. Berthelot synthesized a 4:1:0 compound. Once again, these compounds are formed by the careful production and slow crystalization of a laboratory setting. However, in industrial production that seldom is the case, and amoprhous materials with some overall bulk analysis are generally produced.
There are even more complicated possibilities for the pure compounds. There is a class of crystals in which the lattice has a repeating hole. This can occur in partially dehydrated forms of copper oxychloride in which some of the Cu(OH)2 units have been replaced with a CuO unit. In such a case the ribbon has a repeating defect:
In the extreme cases, all of the oxygen can be in the form of divalent oxygen and one obtains copper oxychloride without any hydroxyl groups.
Note that the CRC Handbook lists the following types of copper (II)
chlorides and oxychlorides:
| basic copper (II) chloride | Cu(OH)2 • Cu(Cl)2 | yellow green hexagons | loses water at 250° to give copper (II) oxychloride
1:1:1 |
| copper (II) oxychloride
Nat. atacamite |
Cu2(OH)3Cl | green orthorhombic | 3:1:3 |
| copper (II) oxychloride
Brunswick Green |
CuCl2 • 3 Cu(OH)2 • 4 H2O | emerald green to green black | 3:1:7 |
| copper (II) trihydroxychlorid
paratacamite |
CuCl2 • 3 Cu(OH)2 | dark green | 3:1:0 |
| copper (II) chloride
anhydrous |
CuCl2 | brown-yellow powder | 0:1:0 |
| copper (II) chloride | CuCl2 • 2 H2O | blue green rhomohedrons | 0:1:1 |
| copper (II) hydroxide | Cu(OH)2 | sky blue precipitate | 1:0:1 |
Another type of crystal has an inclusion of a foreign body, such as water. In some crystals the water naturally fits into a dense crystal lattice. For example, in crystalline CuCl2•2 H2O, the water simply occupies two of the corners of the ribbon, and the structure of the CuCl2•2H2O crystal is very similar to that of the Cu(OH)Cl crystal. In other cases, the water has to disrupt an otherwise tight crystal packing and produces a less dense and mechanically weaker crystal.
The eighteen types of copper (II) oxychloride mentioned by Mellor are:
| 1 | 1 | 1 | |
| 2 | 1 | 0 | |
| 2 | 1 | 3 | ateline |
| 3 | 1 | 0 | |
| 6 | 2 | 3 | |
| 3 | 1 | 2 | |
| 3 | 1 | 3 | paratacamite, blue color |
| 6 | 2 | 7 | possibly 3.1.4 |
| 3 | 1 | 4 | atacamite, green |
| 6 | 2 | 9 | bolivian atacamite |
| 3 | 1 | 5 | |
| 3 | 1 | 6 | botallacite |
| 4 | 1 | 6 | |
| 4 | 1 | 8 | tallingite |
| 6 | 1 | 12 | unspecified cornish mineral |
| 8 | 1 | 12 | footeite |
| 3 | 1 | 4 | |
| 4 | 1 | 4 | made by Berthelot, dark green |
However, even this doesn't give the complete picture. Remember
that I said earlier that several different structures can correspond to
the same ratios. There are at least three structures with the
formula Cu(OH)Cl. One is made by adding base to a solution of CuCl2
in water. Another form is made by mixing CuCl2, CaCO3 and water
in a sealed vessel at 200° C for 48 hours. The third form
can be made by boiling CuCl2 and CuO in water for extended periods of time.
A paper published in 1949 claimed that there were 4 distinct forms of Cu2(OH)3Cl,
excluding differences in hydration. Another paper differentiated
between paratacamite (3 Cu(OH)2 • CuCl2) which has alternating layers
of Cu(OH)2 ribbons and Cu(OH)Cl ribbons, and a new form better represented
as 3 CuO • CuCl2 • 3 H2O in which ribbons of CuCl2 • CuO polymer are surrounded
by CuO polymer ribbons, with waters in the spaces left open by the lack
of octahedral bonding around the copper atoms. ...and so on and so
forth....
Laboratory methods include heating mixtures of CuO, CuCl2 and water in sealed tubes. Complete conversion of the mixture to a single pure compound can take days or weeks. This isn't surprising as the initially formed mass may not have well defined repeating units, and the individual atoms may have to migrate to their final positions. This is another demonstration that crystals can exhibit ion mobility when the temperature is above the van Tamman temperature (defined as a temperature where the individual atom or ion vibrational movement is so great that it can exchange places with it's neighbor). While these individual exchanges are not driven by a knowledge of the final "desired" crystal structure, the final structure is obtained when the ions arrive at the right location. This is due to the fact that the "right" structure is generally denser and gains substantial stability from close ion packing, etc. The amount of energy required to jiggle it out of the right structure is greater than that required to jiggle it into the right structure. It can be thought of as a sort of atomic pachinko machine, with the proper arrangement as the final resting state.
Other laboratory methods include:
So, what does Irving Snerd of No Name Chemical sell? I called
him up and asked his that question. His answer - "I don't know".
So he sent me to David at Service Chemical. They graciously provided
me with the typical analysis of several lots, and informed me that the
copper oxychloride was made by precipitation from a soluble copper salt.
Unfortunately David didn't know which salts and which alkali hydroxides,
but the analysis may give a clue - the values for sodium and sulfate are
both above trace amounts (0.3% for sodium, and 0.8% for sulfate).
The copper content of the two lots is also interesting.
Unfortunately the copper percentage for one lot is given as 57.0% minimum,
leaving open the possibility that there could be a higher percentage of
copper. However, the other analysis is given as 59.1%.
As can be seen from the table, the lot with the analysis given as 59.1
% is closer to either the atacamite structure or the paratacamite values
than to the values for any of the other formulas given in the CRC.
| basic copper (II) chloride | Cu(OH)2 • Cu(Cl)2 | 54.7 % |
| copper (II) oxychloride
Nat. atacamite |
Cu2(OH)3Cl | 59.4 % |
| copper (II) oxychloride
Brunswick Green |
CuCl2 • 3 Cu(OH)2 • 4 H2O | 57.1% |
| copper (II) trihydroxychloride
paratacamite |
CuCl2 • 3 Cu(OH)2 | 59.4 % |
| copper (II) chloride
anhydrous |
CuCl2 | 47.2 % |
| Copper (II) hydroxide | Cu(OH)2 | 65.1 % |
However, the material might not actually have either structure. The material does show a slight basic reaction in water, with a pH of 7.2 for a 5% suspension. This is not commensurate with the known values for the atacamites. However, there are two reasonable explanations. This could be indicative of a slight contamination of the material by sodium hydroxide. Alternatively, the material could consist of a mixture of basic copper chloride and copper hydroxide. This would help explain the color of the material - it's sold as a light aqua green or blue green powder. One of the samples I purchased from a different source is actually aqua with a slight greenish tinge. These colors are inconsistent with any of the reported colors for atacamite, paratacamite, or Brunswick green. However, it would be consistent with a mixture of copper hydroxide and basic copper chloride. This result was mentioned as the reason for the development of the methods of blowing air through ammoniacal baths of copper salts.
There is also a further question that arises at this time - who cares? Who cares what the exact scientific structure is as long as the material consistently gives the desired result in a fireworks formulation. However, that does have a pitfall... A fireworker might develop a really lovely blue with one type of "copper oxychloride" from one source. However, if that source dries up he might find that the replacement "copper oxychloride" gives a different blue color! This sort of thing has been known to happen, and we've probably all heard old fireworkers bemoaning the quality of the chemicals nowadays.
But lets face it... I've been typing this paper for quite a while,
and frankly, I'm not sure anybody is going to care enough to make it worth
my time to spend another day writing about more copper compounds.
Tell you what - if you really really want to learn more about the copper
carbonates, then send me a pretty postcard saying "I can't live without
learning more about copper carbonate". If I get 20 postcards
saying that (or a check for $20), then I'll type this section
Copper Carbonate Fan Club
PO Box 160
Amesville, OH 45711
