Microporous and Mesoporous Materials 195 (2014) 311–318

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier .com/locate /micromeso

Intergrowth structure modelling in silicoaluminophosphate SAPO-18/34family

http://dx.doi.org/10.1016/j.micromeso.2014.04.0241387-1811/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding author. Tel.: +47 22857014.E-mail addresses: w.a.slawinski@smn.uio.no, wojciech@fuw.edu.pl (W.A.

Sławinski).1 On leave from Institute of Experimental Physics, University of Warsaw, Ho _za 69,

00-681 Warsaw, Poland.

Wojciech A. Sławinski a,⇑,1, David S. Wragg a, Duncan Akporiaye b, Helmer Fjellvåg a

a Centre for Materials Science and Nanotechnology, Department of Chemistry, University of Oslo, PO Box 1126, 0315 Oslo, Norwayb SINTEF Materials and Chemistry, Forskningsvn 1, 0314 Oslo, Norway

a r t i c l e i n f o a b s t r a c t

Article history:Received 20 January 2014Received in revised form 9 April 2014Accepted 10 April 2014Available online 19 April 2014

Keywords:SAPOIntergrowthXRDStacking faultsCrystal structure modelling

We present a consistent model of the crystal structure of SAPO-18/34 family members. The model utilisestwo types of stacking fault: Displacement and Growth which have significantly different effects on the dif-fraction pattern. A series of powder diffraction patterns is calculated using the Discus software package.Changes in the level of intergrowth and stacking fault type strongly affect the calculated pattern. A seriesof patterns has been calculated to illustrate this. The structure of an intergrown SAPO-34 sample with4.8% Si content is modelled and refined using Displacement stacking faults. An example of ‘‘defect-free’’AlPO-18 (0% Si content member of SAPO-18/34 family) is then presented. Refinement of the model showsthat even this contains a small amount of stacking faults. Finally, a simple method for defect level esti-mation is proposed based on FWHM ratios for selected Bragg reflections.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

Silicoaluminophosphate framework structures have beenwidely studied because of their many technological applications[1]. Probably their most important application is in catalysis [2]in which chemical composition, crystal structure and sample mor-phology [3–5] are all important. The most significant application ofsilicoaluminophosphate type framework catalysts is in the metha-nol-to-olefin (MTO) conversion process [6], catalysed by SAPO-34(the silicoaluminophosphate form of the chabazite (CHA) zeoliteframework with silicon substituted into its structure). The effec-tiveness of SAPO-34 in the MTO process is due to both the shapeselective properties of the framework and the concentration andstrengths of the acid sites created by silicon substitution [7]. Akey structural feature of SAPO-34 which makes it a good MTO cat-alyst is its 8-ring channels. These are defined by rings in the frame-work with 8 tetrahedrally coordinated Al/Si/P atoms and 8 oxygenatoms. SAPO-34 is not the only 8-ring silicoaluminophosphateframework which can act as an MTO catalyst, the most studiedof the others is AlPO-18 (zeolite framework type AEI), which hasa very closely related structure and can form intergrowths withSAPO-34 [8]. It has been suggested that the level of intergrowth

can affect the efficiency of the MTO process [9]. It was alreadyshown that the level of intergrowth strongly affects selectivity ofa molecular sieve of ZSM-5/11 [10]. To better understand the cat-alytic properties of SAPO-18/34 intergrowths it is crucial todescribe their crystal structure in a precise way.

Like many other zeolite families (described in detail at [8]) [11],CHA/AEI shows intergrowth structure which strongly depends onthe synthesis route and doping level [9]. In this case, by inter-growth we mean one-dimensional stacking faults in the sequenceof two-dimensional layers perpendicular to the stacking direction.This kind of 1D disorder is also observed in other intergrown zeo-lite structures, e.g. MTT/TON [12], Zeolite beta [13,14], SSZ-33/SSZ-26 [15], EMT/FAU [16], RTH/ITE [17] or STF/SFF [18]. There arealso zeolites with higher dimension disorders like inclusions ornano domains or more complicated polymorphs: UDT-1 [19],SSZ-31 [20,21] or ZSM-48 [22,23]. All of these disordered struc-tures give rise to diffuse scattering in diffraction experiments.Standard crystallographic representations cannot be used todescribe the details of the SAPO-18/34 intergrowth structure.

In this paper we present a consistent way of modelling the crys-tal structure of the SAPO-18/34 intergrowth family. Two possiblemodels of 1D stacking faulted structure are presented: Displace-ment and Growth. They have different effects on the powder dif-fraction pattern. Also a series of calculated synchrotron radiationbased X-ray powder diffraction patterns are shown. Two examplesof SAPO-34 (4.8% Si content) and AlPO-18 (the 0% Si content mem-ber of the SAPO-18/34 family) are described showing different

312 W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318

stacking fault types and their refinement by a differential evolu-tionary algorithm method. In addition a simple method based onthe full width at half maximum (FWHM) of selected Bragg reflec-tions is presented which allows a quick estimate of the defect levelin SAPO-34 and AlPO-18 with near perfect structures.

The crystal structure of the SAPO-18/34 intergrowth was mod-elled in [24]. In this model powder diffraction patterns were calcu-lated using the program DIFFaX [25]. A model of Growth stackingfaults was used in [24] and the percentages of the two simulatedphases were refined using a simple least squares calculation tofit the observed data. The DIFFaX program [25] allows both Dis-placement and Growth stacking faults but can not refine the levelof intergrowth nor apply multiphase models. A refinement package(DIFFaX+ [26]) is available, however, this only allows a single phaserefinement which is not suitable for real samples of the SAPO-18/34 system which have been found in this work and [24] to consistof a mixture of highly intergrown and near-defect-free phases.

We have used the DISCUS package to create a more developedmodel of the SAPO-18/34 crystal structure based on two stackingfault types, Displacement and Growth. The new model can describethe experimental data in detail. The crystal structure model ofintergrowth SAPO-34 (4.8% Si, presented here), which consists oftwo phases with different levels of intergrowth (Displacementstacking fault type), is different to the one presented in [24](Growth stacking fault type). Furthermore we have been able tofit our model against real data using a genetic algorithm methodto refine not only the fractions of the two phases but also the levelsof intergrowth in each phase.

2. Modelling of stacking faults in the SAPO-18/34 family

The crystal structures of the defect-free members of the SAPO-18/34 family are known [27,28]. The framework of SAPO-34 (CHA) isdescribed in the space group R�3m while AlPO-18 (AEI) crystallizes

Fig. 1. Crystallographic unit cells of SAPO-34 (left panel, projection along ½110�) and AlPlayered unit cells on top of which the following layered unit cell is marked in grey. A a

in the monoclinic space group C2=c. It is also possible to build thestructures using a layer-based model which allows the developmentof stacking fault models. The basic building block for both structurescan be derived from either the SAPO-34 or AlPO-18 structures. Fig. 1shows the crystallographic unit cells for SAPO-34 (left panel, projec-tion along ½110�) and AlPO-18 (right, projection along ½100�). Blackrectangles show the corresponding layered unit cells on top of whichthe following layered unit cell is marked in grey. The layered unit cellis set to have orthogonal axes, as shown on Fig. 2. The building blockoutlined in black will be denoted as an A-type layer. In order to con-struct the AlPO-18 structure a second layer type is required. The B-type layer is a mirror (parallel to the ab plane) transformation ofthe A-layer. In all the SAPO-18/34 intergrowth calculations only Al,P and O atoms were used. No Si atoms were introduced. This isbecause Al and Si are adjacent to each other in the periodic tableand therefore have very similar X-ray scattering factors and negligi-ble contrast in X-ray diffraction.

The perfect crystal of SAPO-34 can be built by stacking asequence of identical AAAA. . . or BBBB. . . layers in the z direction(= the stacking direction). It is important to note that there is a shiftof ð0d1Þ where d � �1=12 between two stacked A-type layers (seeFig. 1 (left)). The same structure can be obtained by stacking B-typelayers with a shift by ð0 � d1Þ.

The defect-free crystal of AlPO-18 can be described as asequence of ABAB. . . layers shifted by ð001Þ in the z direction.

A stacking fault crystal structure occurs when there is any devi-ation from the perfect sequence of AAAA. . . (SAPO-34) orABAB. . . (AlPO-18). The imperfections can be randomly distributedin the crystal (both models described in this paper) or form morecomplicated subsequences (i.e. short range order based on stackingfaults, not discussed here). In this paper we present two differentways of selecting a sequence of stacked A and B layers which leadto two types of stacking faults Displacement and Growth. Thesehave significantly different effects on the diffraction pattern.

O-18 (right panel, projection along ½100�). Black rectangles show the correspondingnd B indicate the type of the layered unit cell.

Fig. 2. Crystal structure of faulted SAPO-18/34 family members. The left panelshows a Displacement type stacking fault. Two dashed lines indicate an introducedB-type layer which causes a stacking fault. The right panel shows a Growth typestacking fault. The dashed line indicates a change in the direction of crystal growth -the stacking fault.

W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318 313

2.1. Displacement stacking faults

Displacement stacking faults (also known as Deformation stack-ing faults) occur when the sequence of AAAA. . . layers is disturbedby a dislocation of the layers or the occurrence of a different layertype. In the SAPO-18/34 system this type of defect is always asso-ciated with the insertion of additional B type layers into the start-ing sequence. This means that a perfect AAAA. . . sequence ischanged into AABA. . . This change in the sequence causes a shiftof the layer stack by �2d along the y axis. The same defect struc-ture can be obtained with the BBAB. . . sequence. Fig. 2 (left) pre-sents an example of a Displacement type stacking fault in the

crystal structure (10 layered unit cells stacked along z). The perfectsequence of 10 A-type layers is modified by single layer change (Ainto B).

In order to describe the level of intergrowth in a given structure,we will define a quantity called Stacking Fault Rate (SFR). The SFR isdefined as a probability of finding B type layer in an AAAA. . .

sequence for SFR 2 ½0;0:5� (or finding A type layer in anBBBB. . . sequence respectively). SFR ¼ 0 therefore leads to a perfectSAPO-34 structure (sequence of AAAA. . . or BBBB. . .). ForSFR 2 ½0;0:5� any intergrowth structure can be obtained from a par-ent structure of SAPO-34.

In the case of AlPO-18 related structures SFR will be defined as1� p, where p is a probability of exchanging, A into B or B into Alayers, in the ABAB. . . sequence. This means that SFR ¼ 1:0 givesa perfect AlPO-18 structure (ABAB. . . sequence). ForSFR 2 ½0:5;1:0� any intergrowth structure can be obtained from aparent structure of AlPO-18.

2.2. Growth stacking faults

Growth type stacking faults occur when the direction of perfectcrystal growth is changed at some point. From this point anotherdomain of a perfect structure grows in the other direction. In thiscase, this means the starting sequence of AAAA. . . layers is fol-lowed by a sequence of BBBB. . . layers, forming the sequence ofAAAABBBB. . . Fig. 2 (right) shows an example of a crystal wherethe direction of crystal growth is changed from AAAA (bottom part)into BBBB (top part).

In the case of Growth type stacking faults SFR corresponds to theprobability of finding that the following layer is of a different typeto the preceding layer (from A to B or from B to A). The probabilityof finding no change of layer type (from A to A or from B to B) isthen equal to 1� SFR. SFR ¼ 0 gives a perfect SAPO-34 structure(AAAAA. . . or BBBBB. . . sequence) and SFR ¼ 1 leads to perfectAlPO-18 structure (sequence of ABABAB. . .). For SFR 2 ½0;1� a seriesof intergrowth structures can be obtained.

3. Experimental

The AlPO-18 and intergrown SAPO-34 samples were synthe-sized by modifying the synthesis procedure presented by Mertenset al. [29]. All samples were calcined for 15 h at 600 �C.

Synchrotron radiation (SR) powder diffraction patterns werecollected at the European Synchrotron Radiation Facility in Gre-noble (France). The Swiss-Norwegian beamline (BM01B) wasused in high resolution powder diffraction mode. Powder sam-ples were placed in 0.7 mm borosilicate glass, open capillaries.The wavelength was k ¼ 0:50575 Å and the scattering vectorrange, Q, covered by the experiment was 0.217 Å�1 -6 Q 6 5.387 Å�1. All measurements were made at 250 �C to keepthe sample dehydrated. Only a limited range of Q from the mea-sured diffraction patterns was used in the calculations i.e. the2h ¼ 3:115 deg diffraction peak was not included in the refine-ment due to its overwhelming intensity which makes it hardto see the subtle effects of intergrowth on the selected regionof the pattern and its sensitivity to water or organic materialsin the porous structure.

The computer package Discus has been used to calculate pow-der diffraction patterns [30,31]. Each single calculated powderdiffraction pattern is based on a random selection of stacked lay-ers. Each iteration of a calculation gives a different stackingsequence, therefore averaging over a number of calculated pat-terns is required. All refinements used an evolutionary algorithm[32] implemented in the Diffev program. Standard deviation inthe final population members was taken as a measure of statis-

314 W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318

tical error in all parameter estimation. Note that this method cangive errors which are very dependent on the population size andnot directly comparable with the esds obtained from methodssuch as Rietveld refinement.

4. Results

4.1. Calculation of powder patterns from intergrowth structures ofSAPO-18/34

Two different models of possible stacking fault types aredescribed above. Depending on the model used, the calculatedpowder diffraction pattern (averaged over 100 simulations)changes significantly. In order to analyse the structure of real sam-ples a series of powder patterns has been calculated. The followingassumptions have been made:

� Wavelength k = 0.50487 Å.� Layered unit cells a = 13.634748 Å, b = 12.685199 Å,

c = 9.247540 Å, d ¼ 0:079647;Biso ¼ 1:0Å2

(obtained from anapproximate Rietveld refinement of the low intergrowth levelSAPO-34 phase).� Peak width parameters: U ¼ �0:2057;V ¼ 0:3654;W ¼ �0:0077,

peak shape parameter g ¼ 0:6 for Pseudo-Voight peakshape for peaks broadening perpendicular to the stackingdirection.� A model crystal was built with 1000 layers along the z direction

which corresponds to a crystal size of about 1 lm.� Each final pattern was calculated as an average of 100 individ-

ual patterns to obtain good statistical averaging.

Fig. 3. Series of calculated synchrotron radiation powder diffraction patterns of (k = 0.50Displacement (left) and Growth (right) stacking fault type structures. The top patterns corrthe right correspond to Stacking Fault Rate (see text).

Fig. 3 presents two series of calculated diffraction patterns. Thetwo panels show patterns calculated assuming Displacement (left)and Growth (right) type stacking faults. Numbers on the right arestacking fault rates (SFR). SFR ¼ 0 means ’’defect-free’’ SAPO-34and SFR ¼ 1:0 means ‘‘defect-free’’ AlPO-18. One can see significantchanges in the diffraction patterns for increasing intergrowth rate.These changes are visible as both peak position shifts and broaden-ing. At SFR ¼ 0:5 both types of stacking faults give a totally randomdistribution of layers and both calculated patterns are exactly thesame. When comparing calculated powder patterns for structureswith the same SFR (not equal to 0.5) but different stacking faulttypes one can see differences in peaks shapes and intensity ratios.This means that it is possible to distinguish the type of stackingfaults for the whole range of SFR excluding the region where it isclose to 0.5.

4.2. Estimation of defect level in ‘‘defect-free’’ SAPO-18/34 structures

Calculation of powder diffraction patterns is also a very sensi-tive tool to determine the level of defects in ‘‘pure’’ SAPO-34 andAlPO-18 caused by stacking faults. For very low defect-levels,0 6 SFR 6 0:1 and 0:9 6 SFR 6 1:0 for both Displacement andGrowth type stacking faults calculated powder diffraction patternsshow a very strong dependence on the SFR.

Fig. 4 shows selected parts of the powder diffraction patternsfor low SFR (0 6 SFR 6 0:1 and 0:9 6 SFR 6 1:0) for Displacement(left panels) and Growth (right panels) stacking faults. The topand bottom panels present SAPO-34 and AlPO-18 based patternsrespectively. Arrows below the patterns indicate the peaks whichare strongly affected by the increasing level of intergrowth.

487 Å) SAPO-18/34 intergrowth family. The two panels show patterns calculated forespond to ‘‘defect-free’’ SAPO-34 and the bottom ‘‘defect-free’’ AlPO-18. Numbers on

Fig. 4. Series of calculated SR powder diffraction patterns of the SAPO-18/34 intergrowth family for low stacking fault rate values. The top panels show patterns calculated forlow SFR based on SAPO-34. Two bottom patterns show patterns based on AlPO-18. The left and right panels show patterns calculated for Displacement and Growth stackingfault type structures respectively. Numbers on the right correspond to stacking fault rate (see text). Arrows indicate diffraction maxima strongly affected by stacking faults.Diffraction maxima indicated by ‘‘Ref.’’, L and R are used for estimation of defect level in ‘‘defect-free’’ SAPO-18/34 structures (see text).

Fig. 5. Synchrotron radiation powder diffraction pattern (k ¼ 0:50575Å) for the 4.8% Si content member of the SAPO-18/34 intergrowth family fitted with Diffev against ourstructural model. The calculated profile (straight line) assumes a two phase model consisting of virtually defect free SAPO-34 phase (SFR ¼ 0:020ð6Þ and a very highlyintergrown phase with SFR ¼ 0:411ð37Þ, both with Displacement stacking faults. Bragg peak positions are indicated by tickmarks below the profile for pure SAPO-34 (top) andAlPO-18 (bottom). The difference curve is shown below the tickmarks.

W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318 315

Table 1Summary of the crystal structure description obtained from powder diffractioncalculation and refinement for SAPO-34 4.8% Si and AlPO-18 0.0 % Si content samples.

SAPO-34 4.8% Si AlPO-18 0.0% SiRw [%] 0.0588(20) 0.0904(4)

a [Å] 13.6856(14) 13.6673(7)b [Å] 12.7290(20) 12.7126(5)c [Å] 9.2892(12) 9.2672(3)d [b unit] 0.0835(5) 0.0775(6)

Stacking fault type Displacement Growth

SFR (phase 1) 0.020(6) 0.9904(8)SFR (phase 2) 0.411(38)Fraction (phase 1) [%] 29.4(2.2) 100.0Fraction (phase 2) [%] 70.6(2.2)

316 W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318

In the case of SAPO-34 based structures, Displacement stackingfaults cause changes in characteristic peak positions and widths(see peak marked S in Fig. 4). The second stacking fault type,Growth, mainly affects peak widths (see peaks marked with S). It

Fig. 6. Synchrotron radiation powder diffraction pattern (k = 0.50575 Å) for an AlPO-18 sline) is from a single phase model (AlPO-18 with SFR ¼ 0:9904ð8Þ and Growth stacking

Fig. 7. Calculated ratios of the FWHM of selected Bragg reflections (marked in Fig. 4 for SA(right panel) with respect to the reference reflection.

is also worth noting that when SFR increases, the calculated peakshapes are described by a Lorentzian function rather thanPseudo-Voigt function.

The most important difference between calculated powder pat-terns based on the AlPO-18 structure for Displacement and Growthstacking faults is the peak width ratio of some characteristic peaks.For Displacement the reflection marked with L is broader then theone marked with R. In the case of Growth stacking faults weobserve the opposite relationship.

To illustrate the characteristic features of selected Bragg reflec-tions we calculated the peak width ratios for different stackingfault types and SFR. Fig. 7 presents calculated FWHM ratio for theselected peaks in Fig. 4 with S for SAPO-34 faulted structures (leftpanel) and those denoted with L and R for AlPO-18 based struc-tures (right panel) with respect to the reference reflection marked‘‘ref’’. As a reference peak the one which does no change withstacking fault level was chosen. One can see a significant differencein calculated values depending on the stacking fault type. This fea-ture can be used to estimate the SFR of a given sample without fullpattern calculations. It is important to mention that the peak width

ample fitted with Diffev against our structural model. The calculated profile (straightfaults). Bragg peak positions are shown below for defect free AlPO-18.

PO-34 based structures with S (left panel), and L and R for AlPO-18 based structures

W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318 317

ratio is weakly dependent on the peak width parameters used inthe calculation.

4.3. Displacement stacking faults in SAPO-34

Powder pattern calculations as described in Section 4.1 wereused to model PXRD data from a SAPO-34 sample with a Si contentof 4.8% measured by Energy Dispersive Spectroscopy (EDS) in anFEI scanning electron microscope with EDAX EDS attachment.The structure model assumes two crystalline phases with differentSFR values. Displacement type stacking faults were used. All otherstructural and instrumental parameters were constrained duringrefinement to be equal for the two phases. Fig. 5 shows measuredand calculated powder diffraction pattern for the 4.8% Si SAPO-34sample.

The 2h range shown in the figure was used in the calculations.Below the pattern Bragg reflection positions for pure SAPO-34and AlPO-18 structures are marked. One can see the existence ofa slightly faulted SAPO-34 phase (narrow Bragg reflections inSAPO-34 positions) and a very intergrown phase (very broadreflections present close to AlPO-18 Bragg peaks positions). Thecalculations allowed us to estimate SFR equal to 0.025(10) and0.535(90) for the SAPO-34 and AlPO-18 based phases respectively.Table 1 (left column) shows a summary of the refinement results.

In this case the differences between fits with Displacement(Rw ¼ 0:0909) and Growth (Rw ¼ 0:0993) are too small to see butthe statistical evidence is clear.

4.4. Growth stacking faults in ‘‘defect-free’’ AlPO-18

The growth stacking fault model was used to calculate theAlPO-18 powder diffraction pattern. The diffraction pattern of theAlPO-18 sample was fitted using single phase model. Fig. 6 showsmeasured and calculated powder diffraction patterns. Positions ofBragg reflections for AlPO-18 and the difference curve are givenbelow. Table 1 (right column) presents crystal structure parame-ters obtained from the refinement. The SFR is equal to 0.9904(8).This means that crystal structure of AlPO-18 (0.0 % Si) is a near per-fect AlPO-18 structure with the probability of stacking faults equalto 0.9908(18). The agreement factors calculated for Growth(Rw ¼ 0:1078) and Displacement (Rw ¼ 0:1207) stacking faultsclearly show the agreement of the Growth type model with theexperimental data.

The defect estimation method described in Section 4.2 wasapplied to the AlPO-18 sample. Table 2 shows calculated FWHMratios for selected reflections: experimental data, Displacementand Growth stacking fault model (both for SFR ¼ 0:99). The refinedvalue for SFR taken from the whole pattern refinement on AlPO-18was 0.9904(18). The values in Table 2 indicate that even for verylow levels of intergrowth the FWHM comparison does providesome indication of the defect type in the sample.

Table 2Calculated FWHM ratios for selected Bragg reflections (as shown in Fig. 4) forexperimental data taken for an AlPO-18 sample (refined SFR ¼ 0:9908ð18Þ) andDisplacement and Growth stacking fault models with SFR ¼ 0:99.

FWHM ratioL/Ref.

FWHM ratioR/Ref.

Experimental data for AlPO-18 1.22(4) 1.53(4)SFR ¼ 0:9904ð8Þ

Calculation for Displacement stacking fault type 1.32(2) 1.22(2)SFR ¼ 0:99

Calculation for Growth stacking fault type 1.14(2) 1.64(2)SFR ¼ 0:99

The estimated FWHM ratios for experimental data and Growthstacking fault type calculations are in agreement within a 3r test.This proves again that the Growth stacking fault model is valid forAlPO-18 with a low defect level.

5. Conclusion

In this paper we have presented a precise description of twostacking fault types: Displacement and Growth in the SAPO-34/18system. These stacking fault types can be obtained from differentstacking sequences of two layer types. The different types of layersequence lead to significant differences in the calculated powderdiffraction patterns. Refinement of measured powder diffractionpatterns can reveal the stacking fault type present in the real sam-ple. The broadening ratios of selected peaks can also be used as afinger print for stacking fault type at low levels of SFR. The Displace-ment stacking fault model was used to describe the two phasestructure of SAPO-34 containing 4.8% Si. The method was alsoapplied to describe the structure of AlPO-18 with a low defectlevel. The measured defect level (Growth defects) was approxi-mately 1%, which is enough to have a significant influence on themeasured powder diffraction pattern and to indicate the defecttype from the FWHM ratio ’’finger print’’.

Acknowledgements

The work presented in this paper was part of the InterCat Pro-ject (Intergrowth Materials for Improved Methanol-To-Olefin Cat-alysts) funded by the GASSMAX program of the Research Councilof Norway, Grant No. 208325. It was also supported by the Syn-chrotron and neutron travel Grant, No. 216625. We also acknowl-edge Anna Lind (SINTEF Materials and Chemistry) for developingthe synthesis of AEI/CHA intergrowth materials and the AlPO-18sample.

References

[1] H. Pastore, S. Coluccia, L. Marchese, Ann. Rev. Mater. Res. 35 (1) (2005) 351–395.

[2] M. Stöcker, Microporous Mesoporous Mater. 29 (1–2) (1999) 3–48.[3] D. Chen, K. Moljord, T. Fuglerud, A. Holmen, Microporous Mesoporous Mater.

29 (1–2) (1999) 191–203.[4] S. Wilson, P. Barger, Microporous Mesoporous Mater. 29 (1–2) (1999) 117–

126.[5] S. Li, J.L. Falconer, R.D. Noble, J. Membr. Sci. 241 (1) (2004) 121–135.[6] J.Q. Chen, A. Bozzano, B. Glover, T. Fuglerud, S. Kvisle, Catal. Today 106 (1–4)

(2005) 103–107.[7] F. Bleken, M. Bjrgen, L. Palumbo, S. Bordiga, S. Svelle, K. Lillerud, U. Olsbye, Top.

Catal. 52 (3) (2009) 218–228.[8] C. Baerlocher, L. McCusker, International Zeolite Association: <http://www.iza-

structure.org/>[9] M. Janssen, A. Verberckmoes, M. Mertens, A. Bons, W. Mortier, ExxonMobile

Chemical Europe Inc. patent no. EP 1 365 992 B1 (2007).[10] M. Conte, B. Xu, T.E. Davies, J.K. Bartley, A.F. Carley, S.H. Taylor, K. Khalid, G.J.

Hutchings, Microporous Mesoporous Mater. 164 (2012) 207–213.[11] T. Willhammar, X. Zou, Z. Kristallogr. 228 (1) (2013) 11–27.[12] A. Burton, S. Zones, T. Rea, I. Chan, Microporous Mesoporous Mater. 132 (1–2)

(2010) 54–59.[13] J.M. Newsam, M.M.J. Treacy, W.T. Koetsier, C.B. De Gruyter, Proc. R. Soc.

London Ser., A: Math. Phys. Sci. 420 (1859) (1988) 375–405.[14] J.B. Higgins, R.B. LaPierre, J.L. Schlenker, A.C. Rohrman, J.D. Wood, G.T. Kerr,

W.J. Rohrbaugh, Zeolites 8 (6) (1988) 446–452.[15] R.F. Lobo, M. Pan, I. Chan, R.C. Medrud, S.I. Zones, P.A. Crozier, M.E. Davis, J.

Phys. Chem. 98 (46) (1994) 12040–12052.[16] J.M. Newsam, M.M.J. Treacy, D.E.W. Vaughan, K.G. Strohmaier, W.J. Mortier, J.

Chem. Soc., Chem. Commun. (1989) 493–495.[17] P. Wagner, Y. Nakagawa, G.S. Lee, M.E. Davis, S. Elomari, R.C. Medrud, S.I.

Zones, J. Am. Chem. Soc. 122 (2) (2000) 263–273.[18] L.A. Villaescusa, W. Zhou, R.E. Morris, P.A. Barrett, J. Mater. Chem. 14 (13)

(2004) 1982–1987.[19] R.F. Lobo, M. Tsapatsis, C.C. Freyhardt, S. Khodabandeh, P. Wagner, C. Chen, K.J.

Balkus Jr., S.I. Zones, M.E. Davis, J. Am. Chem. Soc. 119 (36) (1997) 8474–8484.[20] R.F. Lobo, M. Tsapatsis, C.C. Freyhardt, I. Chan, C. Chen, S.I. Zones, M.E. Davis, J.

Am. Chem. Soc. 119 (16) (1997) 3732–3744.

318 W.A. Sławinski et al. / Microporous and Mesoporous Materials 195 (2014) 311–318

[21] H. Van Koningsveld, R.F. Lobo, J. Phys. Chem. B 107 (40) (2003) 10983–10989.[22] R.F. Lobo, H. Van Koningsveld, J. Am. Chem. Soc. 124 (44) (2002)

13222–13230.[23] J.L. Schlenker, W.J. Rohrbaugh, P. Chu, E.W. Valyocsik, G.T. Kokotailo, Zeolites 5

(6) (1985) 355–358.[24] M. Mertens, A. Verberckmoes, M. Janssen, Y.F. Chang, L. Martens, S. Vaughn, K.

Clem, W. Mortier patant no. US 7,914,760 B2 (2010).[25] M.M.J. Treacy, J.M. Newsam, M.W. Deem, Proc. R. Soc. London A 433 (1889)

(1991) 499–520.[26] M. Leoni, A.F. Gualtieri, N. Roveri, J. Appl. Crystallogr. 37 (1) (2004)

166–173.

[27] J. Chen, J.M. Thomas, P.A. Wright, R.P. Townsend, Catal. Lett. 28 (2–4) (1994)241–248.

[28] A.M. Prakash, S. Unnikrishnan, J. Chem. Soc., Faraday Trans. 90 (15) (1994)2291–2296.

[29] M. Mertens, Synthesis and use of aei structure-type molecular sieves, patentnumber WO2009117186 A1 (Sep. 24 2009).

[30] T. Proffen, R.B. Neder, J. Appl. Crystallogr. 30 (2) (1997) 171–175.[31] R.B. Neder, T. Proffen, Diffuse Scattering and Defect Structure Simulations,

Oxford University Press, 2008.[32] K. Price, R.M. Storn, J.A. Lampinen, Differential Evolution: A Practical Approach

to Global Optimization, Springer, 2005.