Classically, zeolites are defined as aluminosilicates with open 3-dimensional framework structures composed of corner-sharing TO4 tetrahedra, where T is Al or Si. Cations that balance the charge of the anionic framework are loosely associated with the framework oxygens, and the remaining pore volume is filled with water molecules. The non-framework cations are generally exchangable and the water molecules removable. This definition has since been expanded to include T-atoms other than Si and Al in the framework, and organic species (cationic or neutral) in the pores.

Each time a new zeolite framework structure is reported, the members of the Structure Commission of the International Zeolite Association (IZA-SC) check to see if it has a unique framework type. If it does and the structure has been established to the satisfaction of the members of the IZA-SC (see Rules for Framework Type Code Assignment), a unique 3-letter code is assigned (see Origin of zeolite names and codes).

Descriptions of all of the Framework Types that have been assigned codes by the IZA-SC are included in this database. A number of terms that are used in the database to describe zeolite framework structures are defined here.
   
  Following the rules set up by an IUPAC Commission on Zeolite Nomenclature in 1978 (R.M. Barrer, Pure Appl. Chem. 51, 1091 (1979)), a code consisting of three capital letters (in bold face type) is used to designate a Framework Type. These codes are generally derived from the names of the Type Materials (see also Origin of 3-Letter Codes) and do not include numbers and characters other than capital Roman letters. The assignment of Framework Type Codes is subject to review and clearance by the IZA Structure Commission. Codes are only assigned to established structures that satisfy the Rules for Framework Type Code Asssignment of the IZA Structure Commission. For interrupted frameworks, the 3-letter code is preceded by a hyphen. These mnemonic codes should not be confused or equated with actual materials. They only describe and define the network of the corner sharing tetrahedrally coordinated framework atoms (T-atoms). Thus, designations such as NaFAU are untenable. However, a material can be described using the IUPAC crystal chemical formula (L.B. McCusker, F. Liebau and G. Engelhardt, Pure Appl. Chem. 73, 381 (2001)), as |Na58| [Al58Si134 O384]-FAU or |Na-| [Al-Si-O]-FAU (Note that the chemical elements must be enclosed within the appropriate (boldface) brackets, i.e. | | for guest species and [ ] for the framework host). Framework Types do not depend on composition, distribution of the T-atoms (Si, Al, P, Ga, Ge, B, Be, etc.), cell dimensions or symmetry.

The Framework Types have been arranged in alphabetical order according to the Framework Type Code, because structural criteria alone do not provide an unambiguous classification scheme. This also facilitates later insertion of new codes and allows simple indexing.
   
  A Framework Type is independent of chemical composition. Therefore, idealized framework data (cell parameters, coordinates of T-atoms) were obtained from a DLS-refinement (Ch. Baerlocher, A. Hepp. and W.M Meier, DLS-76, a program for the simulation of crystal structures by geometric refinement, ETH Zurich (1978)) in the (highest possible) symmetry of the Framework Type. The refinement was carried out assuming a (sometimes hypothetical) SiO2 composition and with the following prescribed interatomic distances:
    dSi-O = 1.61 Å, weight = 2.0
dO-O = 2.629 Å, weight = 0.41
dSi-Si = 3.07 Å, weight = 0.23
 

In each case, the coordinates were first optimized within an approximate unit cell, and then the unit cell was also allowed to refine. The resulting DLS reliability index (i.e. how well the assumed geometric parameters are accommodated by the framework structure) is listed. The Framework Density for this SiO2 framework is given as FDSi.

The space group, the cell dimensions and the atomic coordinates of a real material will depend upon its chemical composition, but they will be related to the crystallographic data listed for the Framework Type. If the symmetry is different, it will be a subgroup of this space group, and the unit cell parameters will be related by relatively simple geometric considerations. The relationship between the unit cell parameters of the Type Material and those of the Framework Type are indicated where appropriate.
   
  The Framework Density is defined as the number of tetrahedrally coordinated atoms (T-atoms) per 1000 Å3. The figures given refer to the Type Materials. For non-zeolitic framework structures, values of at least 19 to 21 T/1000 Å3 are generally obtained, while for zeolites with fully crosslinked frameworks, the observed values range from about 12.1 for structures with the largest pore volume to around 20.6. To date, FD's of less than 12 have only been encountered for the interrupted framework cloverite (-CLO), for the silfide UCR-20 (RWY), and for hypothetical networks. The FD is obviously related to the pore volume but does not reflect the size of the pore openings. For some of the more flexible zeolite structures, the FD values can vary appreciably. In these cases (e.g., gismondine) values are given for the Type Material and for the framework in its most expanded state. The flexibility of a framework structure is, to some extent, revealed by the possible variation in the FD.
  FDSi
    FD values may also depend on composition. For all Framework Types, the Framework Density calculated for an idealized SiO2 composition in the highest possible space group (see idealized framework data) is also given (FDSi).
   
  The Coordination Sequence (CS) can be used to calculate a Topological Density (TD). As might be expected, the CS is a periodic function. This has been established for all observed framework topologies by R.W. Grosse-Kunstleve, G.O. Brunner and N.J.A. Sloane (Acta Crystallogr. A52, 879 (1996)). They showed that the CS of any T-atom can be described exactly by a set of p quadratic equations
    Nk = aik2 + bik + ci for k = i + np, n = 0,1,2,... and i = 1,2,3, ... p
  For example, the CS of ABW is exactly described by a set of three quatratic equations (p=3), namely
    Nk = 19/9 k2 + 1/9 k + 16/9 for k = 1 + 3n, n=0,1,2,...
Nk = 19/9 k2 - 1/9 k + 16/9 for k = 2 + 3n, n=0,1,2,...
Nk = 19/9 k2 - 0 k + 2 for k = 3 + 3n, n=0,1,2,...
  The number of equations p necessary to calculate all members of a particular coordination sequence varies from p=1 for SOD and p = 42 for FAU to p = 4,658,179,125,600 for STT and 79,357,853,975,400 (!) for IWW. With growing index k (the shell number of the CS), the linear and constant coefficients, bi and ci, respectively, become less and less important. Therefore we can define the exact Topological Density TD as the mean of all ai divided by the dimensionality of the topology (i.e. 3 for zeolites)
    TD = <ai>/3 = 1/(3p) Σ(ai) (i=1...p)
  This TD is the same for all T atoms in a structure. For some frameworks, this calculation can take quite a long time, so an approximation valid to ±0.001 has been used to calculate the values for each of the Framework Types.  The value for <ai> has been approximated as the mean of ai for the last 100 terms of a CS with 1000 terms (TD1000:100), weighted with the multiplicity of the atom position, and divided by three (dimensionality).  Click here for a list of these Topological Densities for all Framework Types. There is a simple relationship between TD and TD10: TD10 ~ TD *1155. Since TD10 is an approximation, i.e. it is 'arbitarily' terminated at N10, the values obtained by this formula deviate by 11% for -CLO and 5% for FAU but differences are generally below 3%. It seems that for very open framework structures, 10 steps are not sufficient for a satisfatory convergence. The correlation factor between the exact topological density TD and the framework density FD is 0.82.
   
  To provide more quantitative information about the size of the channel system, the following data is given in angstroms:

- the diameter of the largest possible included sphere
- the diameter of the largest-free-sphere that can diffuse along a
- the diameter of the largest-free-sphere that can diffuse along b
- the diameter of the largest-free-sphere that can diffuse along c

These maximum sphere diameters were computed geometrically by Delaunay triangulation with the following assumptions:

- both the framework T- and O-atoms are hard spheres of diameter 2.7 angstrom
- all extra-framework atoms (i.e. water, organics and cations) are ignored
- the interrupted frameworks are not terminated by hydrogen atoms, i.e. only T and O atoms are considered as hard spheres
- the calculations are based on the coordinates of ideal SiO2 frameworks in the highest possible symmetry, as given by the Atlas

Details of the calculations can be found in:
"A geometric solution to the largest-free-sphere problem in zeolite frameworks", M.D. Foster, I. Rivin, M.M.J. Treacy and O. Delgado Friedrichs, Micropor. Mesopor. Mat., 90:32-38, 2006.
   
  The concept of Coordination Sequences was originally introduced by G.O. Brunner and F. Laves (Wiss. Z. Techn. Univers. Dresden 20, 387 (1971) H2) and first applied to zeolite frameworks by W.M,. Meier and H.J. Moeck ( J. Solid State Chem. 27, 349 (1979)). In a typical zeolite framework, each T-atom is connected to N1 = 4 neighboring T-atoms through oxygen bridges. These neighboring T-atoms are then linked in the same manner to N2 T-atoms in the next shell. The latter are connected with N3 T-atoms etc. Each T-atoms is coounted only once. In this way, a Coordination Sequence can be determined for each T-atom of the 4-connected net of T-atoms. It follows that

N0 = 1 N1 ≤ 4 N2 ≤12 N3 ≤ 36... Nk ≤ 4 · 3k-1

CS's are listed from N1 up to N10 for each topologically distinct T-atom in the framework structure along with the vertex symbol. The CS and the vertex symbol together appear to be unique for a particular framework topology, i.e. they can be used to distinguish different zeolite Framework Types unambiguously. In this way, isotypic frameworks can be easily recognized.
   
  The Vertex Symbol was first used in connection with zeolite-type networks by M.O'Keeffe and S.T. Hyde (Zeolites 19, 370-374 (1997)). This symbol indicates the size of the smallest ring associated with each of the 6 angles of a tetrahedron (T-atom). The symbols for opposite pairs of angles are grouped together.  For FAU, the Vertex Symbol reads 4·4·4·6·6·12, indicating that one pair of opposing angles contains 4-rings, a second pair a 4-ring and a 6-ring, and the final pair a 6-ring and a 12-ring.  It is useful for determining the smallest rings in a framework.  In the case of DOH, for example, the vertex symbols for the four T-atoms are 5·5·5·5·5·6, 4·5·5·6·5·6, 5·5·5·5·5·6 and 5·5·5·5·5·5, so the smallest rings are 4- and 5-rings.  Sometimes more than one ring of the same size is found for a single angle.  This is indicated with a subscript like 62 or 82.  An asterisk in the vertex symbol indicates that no ring is formed for that angle. The Coordination Sequence and the Vertex Symbol together appear to be unique for a particular Framework Type.  That is, they can be used to distinguish different zeolite Framework Types unambiguously.  In this way, isotypic frameworks can be recognized easily.
   
  Natural tilings for periodic nets are defined and discussed by V. A. Blatov, O. Delgado-Friedrichs, M. O'Keeffe & D. M. Proserpio, Acta Crystallogr. A 63, 418-425 (2007). The tiling data for the Database were prepared by N. Anurova and V. A. Blatov with the TOPOS program (http://www.topos.ssu.samara.ru/) and illustrations made with 3dt (http://gavrog.org/). The data are discussed and analyzed by N. Anurova, V. A. Blatov, G. D. Ilyushin and D. M. Proserpio, J. Phys. Chem. C 114, 10160-10170 (2010). The four integers pqrs at the top right hand corner of each page indicate that there are p kinds of vertex, q kinds of edge, r kinds of face, and s kinds of tile in the tiling.
   
  The Loop Configuration is a simple graph showing how many 3- or 4-membered rings a given T-atom is involved in (List). Solid lines represent T-O-T linkages whereas dotted lines indicate non-connected T-O bonds found in interrupted frameworks. M. Sato (Proc. 6th IZC, Reno (Butterworth, 1984), p. 851) used the term "second coordination networks". Loop Configurations are likely to be of interest to spectroscopists. These data can also be used for classification purposes and for deducing rules relating to these structures which might be of predictive value (G.O. Brunner, Zeolites, 13, 88 (1993)). The information given in the loop configuration is a subset of the vertex symbol.
   
  Zeolite frameworks can be thought to consist of finite or infinite (i.e. chain- or layer-like) component units. These Secondary Building Unitsa, which contain up to 16 T-atoms, are derived assuming that the entire framework is made up of one type of SBU only. It should be noted that SBU’s are invariably non-chiralb. A unit cell always contains an integral number of SBU’s. As far as practicable, all possible SBU’s have been listed. The number of observed SBU’s has increased from 16 in 1992 to 23 at present. In some instances, combinations of SBU’s have been encountered. These have not been listed in extenso because this can arbitrary. The codes given below the drawings in Figure 2 are used in the data page to describe the SBU's. If more than one SBU is possible for a given framework type, all are listed.
Please note:The SBU's are only theoretical topological building units and should not be considered to be or equated with species that may be in the solution/gel during the crystallization of a zeolitic material.

a The primary building units are single TO4 tetrahedra.
b This means that SBU's in the isolated state of highest possible symmetry are neither left- nor right-handed.
   
Composite Building Units (CBU's)
  Some units (e.g. double 6-ring, cancrinite cage, alpha cavity, double crankshaft chain) appear in several different framework structures, and can be useful in identifying relationships between Framework Types.  Smith has compiled an exhaustive list of such units, not only for zeolite structures but also for hypothetical 3-dimensional 4-connected nets(Landolt-Boörnstein, Vol. 14, Subvolume A, Springer, Berlin, 2000).  In his Compendium of Zeolite Framework Types. Building Schemes and Type Characteristics, van Koningsveld has also included an extensive list of them (Elsevier, Amsterdam, 2007).  Here we have arbitrarily selected just 47 Composite Building Units and five chains that are found in at least two different Framework Types.  These are different from secondary building units in that they are not required to be achiral, and cannot necessarily be used to build the entire framework.  To facilitate communication, each unit has been assigned a lower case italic three-character designation.  With the exception of the double 4-, 6- and 8-rings (d4r, d6r and d8r, respectively), a code corresponding to one of the Framework Types containing the CBU has been used for this purpose.
   
  The type material is the species first used to establish the framework type.

The composition, expressed in terms of cell contents, has been idealized where necessary for simplicity. The chemical formula is given according to the new IUPAC rules(2). The space group and cell parameters listed for each type material are those taken from the reference cited. In many instances, further refinement of the structure taking into account ordering etc. would yield a lower symmetry. It should also be noted that the space group and other crystallographic data related to the type material structure do not necessarily apply to isotypes.

In some cases, the space group setting of the type material differs from that of the framework type. In these cases, the relationship of the unit cell orientation with respect to the framework type is listed. This relationship is important when comparing the orientation of the channel direction and the viewing direction of ring drawings (which are both given for the axis orientation of the type material) with that of the framework drawing.

Isotypic species are very frequent and are listed under ''Related materials".
   
  A shorthand notation has been adopted for the description of the channels in the various frameworks. Each system of equivalent channels has been characterized by
    the channel direction (relative to the axes of the Type Material structure),
the number of T-atoms (bold) forming the rings controlling diffusion through the channels, and
the crystallographic free diameters of the channels in Ångstroem units.
 

The number of asterisks in the notation indicates whether the channel system is one-, two- or three-dimensional. In most cases, the smaller openings simply form windows (rather than channels) connecting larger cavities. Interconnecting channel systems are separated by a double arrow (↔). A vertical bar (|) means that there is no direct access from one channel system to the other. The examples below have been selected to illustrate the use of these notations.

 
Cancrinite [001] 12 5.9 x 5.9*
Offretite [001] 12 6.7 x 6.8* ↔ [001] 8 3.6 x 4.9**
Mordenite [001] 12 6.5 x 7.0* ↔ {[010] 8 3.4 x 4.8 ↔ [001] 8 2.6 x 5.7}*
Zeolite Rho <100> 8 3.6 x 3.6*** | 100> 8 3.6 x 3.6***
Gismondine {[100] 8 3.1 x 4.5 ↔ [010] 8 2.8 x 4.8}***
 
Cancrinite is characterized by a 1-dimensional system of channels parallel to [001] or c with circular 12-ring apertures. In offretite the main channels are similar but they are interconnected at right angles by a 2-dimensional system of 8-ring channels, and thus form a 3-dimensional channel system. The channel system in mordenite is essentially 2-dimensional with somewhat elliptical 12-ring apertures. The 8-ring limiting diffusion in the [001] direction is an example of a highly puckered aperture. Zeolite rho is an example of a Framework Type containing two non-interconnecting 3-dimensional channel systems which are displaced with respect to one another (<100> means there are channels parallel to all crystallographically equivalent axes of the cubic structure, i.e., along x, y and z.). In gismondine, the channels parallel to [100] together with those parallel to [010] give rise to a 3-dimensional channel system which can be pictured as an array of partially overlapping tubes.

Please note: The channel direction is given for the axis orientation of the type material. This orientation may be different from the orientation given in the framework drawing (see the cell relationship given under "crystal chemical data" for these cases.

A summary of the channel systems, ordered by decreasing number of T-atoms in the largest rings, is given in Table 3. The free diameter values given in these channel descriptions and on the ring drawings are based upon the atomic coordinates of the type materials and an oxygen radius of 1.35Å. Both minimum and maximum free diameter values are given for noncircular apertures. In some instances, the corresponding interatomic distance vectors are only approximately coplanar, in other cases the plane of the ring is not normal to the direction of the channel. Close inspection of the framework and ring drawings should provide qualitative evidence of these factors. Some ring openings are defined by a very complex arrangement of oxygen atoms, so in these cases other short interatomic distances that are not listed may also be observed. It should be noted that crystallographic free diameters may depend upon the hydration state of the zeolite, particularly for the more flexible frameworks. It should also be borne in mind that effective free diameters can be affected by non-framework cations and may also be temperature dependent.
   
  Under the heading "Related Materials" as-synthesized materials that have the same framework type but different chemical composition or have a different laboratory code are listed. Materials obtained by post synthesis treatment (e.g. ion exchange, dealumination, etc.) are generally not included. The type material is given first and marked with an asterisk. Isotypic species, which have sometimes been termed "homeotypic"(11), are very frequent and can also be searched for by clicking on the "Framework type " tab.

Zeolite-type silicates and phosphates apparently constitute two distinctive categories of microporous materials. Table 1 (which is based on the isotypes listed in the 5th edition of the Atlas) shows, however, that there are three, rather than two distinct groups of framework types. Apart from those associated with silicates and phosphates there is a sizable group of structure types which have been found to occur both in silicates and phosphates.