'Solving Crystal Structures of Inorganic, Organic and ... crystal structures of inorganic, organic, and coordination compounds using synchrotron powder data james a. kaduk bp amoco

SOLVING CRYSTAL STRUCTURES OF INORGANIC, ORGANIC, AND COORDINATION COMPOUNDS USING SYNCHROTRON POWDER

DATA

James A. Kaduk BP Amoco p.l.c., P. 0. Box 3011 MC F-9, Naperville IL 60.566

INTRODUCTION

Many technologically-important materials do not form the single crystals “necessary” for determining their solid state structures using conventional crystallographic techniques. Knowledge of the crystal structure of a material makes it possible to explain many other properties, and facilitates the use of diffraction techniques to obtain morphological and mesostructural information from powder data. Having the structure “in hand” when it is needed saves time and money; the effort of determining crystal structures thus is a sort of “intellectual capital investment”, and can be justified as such.

In industry, we do not have the option of sending the graduate student back to the lab to make better material, but must characterize the materials of commerce. I am therefore called upon to solve crystal structures using only powder diffraction data. Although it is possible to solve structures using laboratory powder data, the use of synchrotron powder data makes the process much easier, and thus more efficient. Each of these crystal structures was solved in a different way; the solutions illustrate different aspects of the process.

POTASSIUM ALUMINUM BORATE, K,Al,B,O,

Only one ternary phase, K,AlB,O,, [ 11, in the K,O-Al, Q -B Q system has been characterized structurally. During exploration of this ternary system, we discovered a black amorphous semiconducting phase of the approximate composition lK,O : 1 A&O, : 2B,O,. From preparations having compositions near lK,O : 1 A&O, : lB,O,, a phase having the X-ray powder diffraction pattern of K,Al,B,O, [2] was produced.

The powder pattern of K2AlZB207 can be indexed on a very high quality trigonal/hexagonal unit cell having a = 8.5597(4) and c = 8.4597(4) A. MASNMR indicated that the Al coordination is tetrahedral, and that the borons are trigonal. The observed density indicated that the cell contents were K,A16B60,, . Because K,Al,B,O, is formed near the semiconducting phase in the phase diagram, we believed that knowledge of the crystal structure, and comparison of the structure to those of nearby crystalline phases, would give some insight into the structure of the amorphous phase and the mechanisms of conductivity.

The sample was mixed with 4.87 wt% NIST 640b Si internal standard, and ground in a McCrone micronising mill using ethanol as the milling liquid. The X-ray powder diffraction pattern was measured on beamline X3B 1 at the National Synchrotron Light Source at BrookhavenNational

Copyright(C)JCPDS-International Centre for Diffraction Data 2000, Advances in X-ray Analysis, Vol.42 333Copyright(C)JCPDS-International Centre for Diffraction Data 2000, Advances in X-ray Analysis, Vol.42 333ISSN 1097-0002

http://www.icdd.com

This document was presented at the Denver X-ray Conference (DXC) on Applications of X-ray Analysis.

Sponsored by the International Centre for Diffraction Data (ICDD).

This document is provided by ICDD in cooperation with the authors and presenters of the DXC for the express purpose of educating the scientific community.

All copyrights for the document are retained by ICDD.

Usage is restricted for the purposes of education and scientific research.

DXC Website – www.dxcicdd.com

ICDD Website - www.icdd.com

ISSN 1097-0002

http://www.dxcicdd.com/

http://www.icdd.com/

Laboratory, using a wavelengthof 0.89978 A. The pattern was measured from 6-74” 28 in 0.004” steps, counting for 4 set/step. The sample was rocked from o-l to o-t1 ’ during each data point. An o-scan of the (112) peak at 17.206’ 28 indicated significant granularity. The degree of powder averaging obtained by a 2 o rocking would not be expected to affect solution of the structure, but may affect the agreement of the observed and calculated patterns.

No systematic absences were observed in the powder pattern. Together with the trigonal/hexagonal unit cell (no evidence was observed for lower, especially orthorhombic, symmetry), this “limits” the possible space groups to: P3, Pz P321, P3m1, PTml, P312, P3lm, P?lm, P6, Pg P6/m, P622, P6mm, P62m, P6m2, and P6/mmm. In all of the Laue classes but d/mmm, there are inequivalentreflections which overlap exactly. This overlap makes the extraction of accurate structure factors from the powder pattern more difficult. Many reflections, which would be absent for a rhombohedral cell, are absent or very weak in the observed pattern. These weak reflections suggested the likely presence of pseudosymmetry, and that the structure solution would be difficult.

Consideration of the coordination requirements of the cations and the approximate number of atoms in the unit cell suggested that the presence of a 6-fold rotation axis was unlikely. Attempts to solve the structure from laboratory powder data in P62m and P5m2 suggested a hexagonal array of heavy atoms at 000, OO%, %%O, %%$A, %%O, and %W. A calculated pattern with K atoms at these sites had many features in common with the observed pattern, and suggested that there was some truth to this model. All attempts to locate more atoms in 6/mmm space groups were unsuccessful.

The strategy followed, therefore, was to try to solve and refine the structure in the lowest possible symmetry space group, P3, and to seek additional symmetry elements to define the true space group. Initial powder data processing was carried out using GSAS [3]. A 3-term cosine Fourier series background function, scale factors for each of the four phases, the lattice parameters, and profile terms were refined. The structure model used the six K atoms at the above coordinates. The LeBail extraction procedure incorporated into GSAS was used to extract 426 individual structure factors for 28 < 56” (d > 0.96 A).

These structure factors were used to create a SHELXTL Plus [4] data file. Attempts to solve the structure using direct methods were unsuccessful. A Patterson synthesis suggested the presence of a K atom at the origin, and a second K at %,%,0.5148. The z coordinate of the first atom was fixed to define the origin. Successive cycles of refinement and difference Fourier synthesis yielded the positions of some other reasonable atoms, but not the complete structure. The second-best heavy atom position suggested by the Patterson was %%O - a general position in P3. Successive cycles of least squares refinement and difference Fourier maps yielded another K on a general position, and six Al positions on the 3-fold axes. Further refinement of this model gradually yielded the positions of the 0 and B atoms. The B and Al were distinguished by the coordination number and the M-O distances.

Final refinement of the structure was carried out using GSAS. The atom coordinates were refined subject to soft constraints of 1.74( 1) A on the Al-O bonds and 1.37( 1) A on the B-O bonds.


The weight of these constraints was gradually decreased through the course of refinement. The atoms were refined with a common isotropic displacement coefficient for each atom type. Included in the final refinements were scale factors for each phase, and the lattice parameters for K,Al,B,O, and an Al,O, impurity. The size broadening terms X and ptec were refined for K,Al,B,O,. The profiles for the other phases were described using fixed X and Y terms determined early in the refinement. The background indicated the presence of an amorphous component, and was described by a 6-term real space pair correlation function, using two fixed distances of 2.80 and 0.50 A.

The final refinement in P3 of 56 variables using 17029 observations yielded the residuals Rp = 0.1467 and wRp = 0.1821. The final reduced x2 was 6.388. The Bragg R(F) was 0.1489. At this point, the asymmetric unit contained two independent K, six Al on 3-fold axes, two B on general positions, and nine 0 (six on general positions, and three on 3-fold axes). One of the boron atoms was decidedly nonplanar (the sum of the O-B-O angles was 330”). The Al-O-Al angles of the Al,O, units were 180 ‘; this large angle is uncommon (but not unprecedented), and larger than the usual 130-160’ Al-O-Al angle. Although the framework topology was almost certainly correct, these structural features, and the difficulty of convergence, pointed to a space group error.

The structure was examined for the presence of additional symmetry elements by application of the MISSYM program of the NRCVAX software system [5]. A number of additional symmetry elements were found to relate the K, Al, and B atoms, but when the default tolerances were applied to all atoms, no additional symmetry was detected. When the tolerances were increased slightly, additional 2-fold axes parallel to [Ol 01, [ 1001, and [ 1 lo] were suggested. Space group P321 (#150) therefore seemed likely. The 2-fold axes intersected at %ZO. A coordinate transformation was therefore applied to the P3 atom coordinates, and the unique atoms in P321 retained. The new asymmetric unit contained two K atoms on 2-fold axes, three Al on 3-fold axes, one B at a general position, three 0 at general positions, one 0 on a 3-fold axis, and one 0 at the origin, a position of 32 symmetry. Other coordinate transformations lead to wrong combinations of atoms on special positions, which led to the wrong stoichiometry.

Refinement using common isotropic displacement coefficients by atom type proceeded smoothly. The Al-O-Al are still linear in this model. The constraints on Uiso for 04 and 05 were removed, to test the possibility of disorder off the 3-fold axes. The Uiso of 04 essentially did not change, but that of 05 increased greatly. (The largest hole in the AF map was at this atom - the origin.) 05 was thus moved off the origin, to a position x00 on a 2-fold axis, and refined. The resulting Uiso is comparable to those of the other oxygens, and the A13-05-A13 angle is 151’ - a more typical value. There is no evidence that 04 is displaced off the 3-fold axis, and thus that the All-04-A12 angle is not 180”. The B is planar to within experimental error. In general, this structure is chemically more satisfactory.

The final refinement in P321 of 39 variables using 170 14 observations yielded the residuals wRp = 0.1814andRp = 0.1463. The finalreducedx* was 5.800. The BraggR(F) was 0.1451. The agreement of observed and calculated patterns is slightly better, with fewer variables, in P321, suggesting that it is the correct space group. Applicationof MISSYM suggested the presence of no additional symmetry elements, even with very large tolerances. The agreement between F, and F, is poorest for the highest-angle reflections. The observed, calculated, and difference patterns are


illustrated in Figure 1. The largest differences occur at K2A12B207 reflections, and probably indicate that a true powder average was not achieved. The largest peak in a difference Fourier map was 2.0 electrons, and the largest hole was - 1.6 electrons. The slope of the final Wilson plot suggests that the standard uncertainties are underestimated by a factor of 2.1. The final structural parameters are reported in Table 1.

Table 1. Crystal Structure of K,Al,B,O,

Space group P321, a = 8.55802(2), c = 8.45576(3) A

Name Kl K2 All Al2 A13 Bl 01 02 03 04 05

X Y 0 -0.3589(5) 0.6889(5) 0 ‘ii - ‘L3 ‘h - ‘/i 0 0 0.3388(14) 0.0070(16) 0.2077(7) 0.0393(9) 0.5090(8) O-1168(7) O-2446(10) -0.1994(8) ‘h - ‘h

0.0501(27) 0

Z 0 % 0.3145(9) 0.7243(8) 0.1994(7) 0.7612(20) 0.7252(8) O-7533(8) 0.7879(8) 0.5192(11) 0

Ui*lOO 2.86(5) 2.86(5) 2.25(7) 2.25(7) 2.25(7) 2.33 (26) 2.10(8) 2.10(8) 2.10(8) 2.13(32) 1.68(85)

Description of the Structure

The crystal structure of K2A12B207 consists of a 3-dimensionalnetwork composed of corner- sharing BOX triangles and A1207 units. The K cations reside in channels parallel to the c-axis formed by this framework. The cation coordination spheres and the atom numbering scheme are illustrated in Figure 2. The boron coordination is trigonal and very nearly planar; the sum of the O-B-O angles is 359”. The B atom is +0.05 A, and the oxygens are -0.01-0.02 A from the mean BO, plane.

There are two independent Al,O, units. Both of them occupy 3-fold axes, and consequently the average Al-O-Al angles are 180’. Although this configuration is uncommon in high-quality structures, it is not unprecedented. The Al-O bond distances and angles fall within the normal range. The deviations from ideal tetrahedral geometry are small. The configurations of the two Al,O, units are different. A convenient measure is the O-Al-Al-O “torsion angle”. For the Al 1 -A12 group, this angle is 39”, indicating that the tetrahedra are nearly in the staggered configuration (60”). For the A13-A13A group, the angle is 20”, indicating that this unit is closer to the eclipsed configuration.

The occurrence of two distinct Al,O, units, both with imposed 3-fold symmetry, reflects a subtle feature of space group P321. There are two different types of 3-fold axes in this space group. One type, passing though the origin and occupied by the A13-A13A units, is intersected by the 2-fold axes in the ab plane. The other type, of which there are 2/tell, pass through %GfI and %%O, and do not intersect the perpendicular 2-fold axes. These axes are occupied by the All-Al2 units.


9523-i38-C K/Al/B + 4.87% Si, X381, plate, 24 June 93 Lambda 0.8998 A, L-S cycle 334 Obsd. and Diff. Profiles

I I I I I I 0

4 -- OD

w

2 O- . XW

0 .-

v

0 .-

cy

0 .

0 I

X5

0 .-

$7

;0 .- u-r

I 1.0 2-Theta, deg

I I

2.0 3.0 4. XlOE 1

9523-138-C K/Al/B + 4.87% Si, X3B1, plate, 24 June 93 Lambda 0.8998 A, L-S cycle 334 Obsd. and Diff. Profill

I I I r-4

W 0 rl x

C

c

0

0

:: co .

Sd u ’

x 20

I I I I 4.0 5.0 6.0 7.0

a-Theta, deg ------- ~~

XlOE 1

es

Fig. 1. Observed, calculated, and difference diffraction patterns of K2A&B20,. points, and the solid line the calculatedpattern.

The crosses represent the measured data The difference pattern is plotted at the same scale as the other curves.

The bottom row of tick marks represent the calculated reflection positions for K,Al,B,O,. Successively higher rows of ticks indicate the Si, Al,O,, and K,B,O,(OH),. 2H,O peak positions. The vertical scales have been multiplied by factors of 5 and 20.


Fig. 2. The coordinationof the B, Al, and K atoms in K2AlzB207, with the atom numbering scheme. K-O distances < 3.2 A are illustrated by thin solid lines, and those > 3.2 A by dashed lines. 50% probability ellipsoids. Only one of the three disordered 05 positions is shown.


A consequence of the symmetry is that there are two All-Al2 units for every A13-A13A unit. The central oxygens also have different site symmetries. The oxygen 04 of the Al 1 -A12 unit has site symmetry 3, but the 05 of the A13-A13A unit has the higher symmetry 32 at its average position.

There are two independent K ions, each residing on a site of 2-fold symmetry. Considering only K-O distances < 3.2 A, Kl is 5-coordinate, and the coordinationtrigonal bipyramidal. Two 02 and one 05 make up the equatorial plane, and two 03 are the axial ligands. Two additional 01 are 3.32 A from Kl in the X-ray refinement, but only 3.19 A away in the neutron results. If 05 is located at its average position (the origin), the Kl-05 distance is 3.07 A (3.08 A X-ray and 3.06 A neutron). Threefold disorder of 05 off the origin leads to shorter distances to two different Kl (2.88 A X-ray and 3.00 A neutron). Because Kl is relatively coordinatively unsaturated, such distortion seems to be reasonable to meet the bonding requirements of Kl . It is most reasonable to consider K2 to be 8-coordinate (two 01, two 02, two 03, and two 04). The geometry is irregular. Two additional 01 are 3.23 8, from K2.

A convenient way of assessing the strength and reasonableness of a K-O bond is to calculate the bond valence [6]. Considering the nearest oxygens, Kl is only 5-coordinate, and all the bonds are relatively short. For the X-ray refinement, the sum of these bond valences is 1.09, close to the expected K valence of t-1. The K2-0 bonds are in general longer than those to Kl . The K2 valence derived from the 8 shortest distances is only 0.90, but increases to 1 .OO when the two additional 3.23 A K2-01 distances are included.

The crystal structure is most easily visualized when viewed down the c-axis (Figure 3). The B03 triangles lie approximately parallel to the ab plane, and planes containing them are joined by A&O7 “pillars” parallel to c. Each B is connected to two All-Al2 pillars pointing “down” and one A13-A13A pillar pointing “up” (or vice versa). The K ions lie within 6-ring channels parallel to c. The K alternate in layers. The 7-coordinate Kl lie in the A13-A13A layers, where the atom density is relatively lower. The 8-coordinate K2 lie in the All-Al2 layers, in which there are more atoms. The K do not lie in the centers of the channels, but are displaced toward the walls.

The structure has the same topology as, but different symmetry than, Rb,Be,Si,O, [7], which crystallizes in the orthorhombic space group P2nn. Normally such an isostructural relationship would be detected by similarities in the lattice constants or the powder diffraction patterns. In this case, the 40% difference in the atomic volumes of K and Rb lead to cell volumes which differ by 40%. This difference is large enough that the relation between these two structures was not detected with the usual software tools. The structure similarity was found because both compounds have the relatively unusual A,B&X, stoichiometry.


Fig. 3. A polyhedral rendering of the K,Al,B,07 structure, viewed in projection down the c-axis. The open triangles represent the B03 units, and the shaded tetrahedra indicate the AlO, subunits.


DIMETHYL 2,7-NAPHTHALENEDICARBOXYLATE

Dimethyl 2,6-naphthalenedicarboxylate(NDC) is one of the monomers used to manufacture poly(ethylene 2,6-naphthalenedicarboxylate) (PEN), which shows promise in packaging and electronics applications requiring better strength and dimensional stability than is provided by poly(ethylene terephthalate) (PET). A common engineering practice is to modify polymer properties by incorporating a co-monomer to alter the packing of the polymer chains, and thus the physical properties of the polymer. Dimethyl 2,7-naphthalenedicarboxylate(27NDC, CAS Registry Number 2549-47-5) is of interest as a potential co-monomer in PEN applications.

A sample of 27NDC was ground in a McCrone micronising mill using corundum grinding media and hexane as the milling liquid. The powder was packed into a 1 mm fused silica capillary. The X-ray powder pattern was measured at beamline X3Bl at the National Synchrotron Light Source at BrookhavenNational Laboratory, using a wavelength of 1.149985( 14) A. The pattern was measured from 5-35’ 28 in 0.004” steps, counting for 1.2 set/step. The sample was rocked from o-l to o+l’ during each data point. The pattern could be indexed [8] on a very high-quality primitivemonoclinic unit cell having a = 24.4846, b = 6.0652, c = 3.9549 A, ,0= 92.4845”, and V = 586.78 A3. All observed peaks were indexed. To obtain a reasonable density of 1.382 g/cc for Cr4Hi204, Z = 2. The systematic absences were consistent with space groups P2, and P2Jm. A pattern measured from 2.50-3.00’ 28 revealed a very weak (100) peak at 2.688 ‘.

The structure was solved using a laboratory pattern, measured from a separate flat plate sample of 27NDC mixed with 40.04% NIST 640b silicon internal standard on a Scintag PAD V diffractometer equipped with an Ortec Ge detector. The 7-28” 28 portion of the pattern was extracted, and used for the structure solution. A 27NDC molecule was built using Cerius2 TM [9]. The ester groups were oriented in the 2 conformation. This molecule was imported into the observed unit cell, and space group P2, was assumed. The planar molecule was rotated and translated manually to obtain a position and orientation for which the molecules did not overlap; the long a and short c cell dimensions put severe constraints on the orientation of the molecule.

This structure was used as input to the STRUCTURE_SOLVE module of Insight11 [lo]. This module implements a Monte Carlo simulated annealing procedure, in which the metric is not energy, but the agreement of the observed and calculated powder diffraction patterns. The 7-28 ’ pattern was interpolated to contain 801 points, and scaled to O-l 00 counts. In all runs, the lattice parameters were fixed at the experimentalvalues. A pseudo-Voigtprofile shape (80% Cauchy), with profile coefficients U = 0.150, V= -0.075, and W = 0.020, was used for all simulations. The initial runs used a rigid planar molecule, to obtain an approximate position in the cell. Later runs allowed the C2-Cl1 and C7-Cl4 torsion angles to vary, as well as the molecular position and orientation. The best solution was used as input to a Rietveld refinement, using the synchrotron pattern.

All subsequent data processing was carried out using GSAS [3]. Only the 1 O-35 ’ portion of the pattern was included in the refinements. The naphthalene core was modeled as a planar rigid body with C-C = 1.40 A. Extensive use was made of soft constraints to restrain the geometry of the ester linkages: C2-Cl l/C7-C14= 1.42(l), Cll-012/C14-015= 1.25(l), 013-Cl l/013-C17/016- C14/016-Cl8 = 1.30(1),C2-013/C7-016=2.40(2),C2-012/C7-015=2.32(2),012-013/015-016


= 2.17(2), Cl l-Cl/Cl l-C3/C14-C6/C14-C8 = 2.44(2), Cl l-C17/C14-Cl8 = 2.30(2), and 012- C 17/O 15-C 18 = 2.5 8(2) A. The weight given to the soft constraints was decreased during the course of refinement. The y-coordinate of Cl 1 was fixed to define the origin. A common isotropic displacement coefficient was refined for the naphthalene carbons, and another isotropic thermal parameter was refined for the substituent atoms. The hydrogens were included in calculated positions [4], which were updated during the course of refinement, but which were not refined. To check the orientations of the ester groups, Cl 7 and Cl 8 were removed from the model, and a difference Fourier map calculated. Large peaks appeared at the positions of these two carbon atoms; this confirmed the conformations as 2 rather than E.

Included in the refinement were a scale factor and the lattice parameters. A March-Dollase preferred orientation ratio (unique axis 100) was refined. The peak profiles were described using a pseudo-Voigt function; refined were the Gaussian U, Cauchy Yand stec (unique axis OOl), and asym coefficients. The background from the capillary was described by a 6-term real space pair correlation function with two characteristic distances of 0.952(2) and 4.771(14) A.

The final refinement of 47 variables using 6270 observations yielded the residuals wRp = 0.1123, Rp = 0.0838, x2 = 5.393, R(F2) = 0.0961, and R(F) = 0.069. The soft constraintscontributed only 2% to the final x 2. A normal probability plot indicated that the standard uncertainties were underestimated by a factor of 1.7. The agreement of the observed and calculated patterns (Figure 4) is excellent. The largest errors occur in describing the asymmetry of the low-angle peaks, and at the (1 li) peak at 28 = 20.09 ‘; this peak contains a spike which seems to indicate incomplete powder averaging. The largest peak in a difference Fourier map was 0.37 eAs3, 0.50 A from H8; the largest difference hole was -0.29 eAe3, midway between 016 and H8. Application of the NRCVAX program MISSYM [5] indicated no additional symmetry elements, confirming the space group as P2,. The refined coordinates and isotropic displacements coefficients are reported in Table 2.


Since a rigid body and soft constraints were used, the bond distances and angles fall within the expected ranges. The molecular structure and atom numbering scheme are illustrated in Figure 5. The agreement among chemically-equivalent quantities is not as good as is typically observed in such refinements, probably because of the limited range of observations. The C-O and C=O distances are not as different as expected, but the conformations of the ester groups are unambiguous. The sums of the angles around C2, C7, Cl 1, and Cl4 are 360.0,359.8, 359.0, and 358.9”, respectively.

The angle between the naphthalene plane and the mean plane of the ester group on C2 is 11’) and the comparable angle for the ester group on C7 is 6”. Both ester groups are oriented in the 2 conformation observed in NDC [ 111, but are not rotated out of the ring plane as much as they are in NDC (20”). The displacement of the C7 ester group from the naphthalene plane is not so much a rotation around the C7-Cl4 bond, but a bending of the entire ester out of the ring plane.


\c I- c\;

ii?

=1y -4

u

%uY

u

I I4

I-r’ In< m

i : ; 4 4 ; i I, I 3 ; i ) b I- I I , 7 I .

I 1

I I

A.L

--- -

-- W

V

u-z 0’0

O’Z-

z 30TX

squnoa

In

4 rl W

0 rl SC

0 4

Ln

c;

0 r.i

n 4

F a L

z

:2

+E;’ N

Copyright(C

)JCPD

S-International Centre for D

iffraction Data 2000, A

dvances in X-ray A

nalysis, Vol.42

343C

opyright(C)JC

PDS-International C

entre for Diffraction D

ata 2000, Advances in X

-ray Analysis, V

ol.42343

ISSN 1097-0002

s2-

e. 0

6’

$2

Ll

g-1

z x

%!

ZFY

5

0 $2

Q

-K

s V

J &

nF:

02

q -En

0 0

5 ),

%m

E

2

qu

g .o

g-

+..

g F

r=L$

n

2 P

-’

Fi.2

59

F-

z Y+

a0

a

9 %

g R

B

g z

v: 3

. 3c

z y&

m

0%

5 5

E?

%,1

%

T S

F j$

Cop

yrig

ht(C

)JC

PDS-

Inte

rnat

iona

l Cen

tre

for

Dif

frac

tion

Dat

a 20

00, A

dvan

ces

in X

-ray

Ana

lysi

s, V

ol.4

234

4C

opyr

ight

(C)J

CPD

S-In

tern

atio

nal C

entr

e fo

r D

iffr

actio

n D

ata

2000

, Adv

ance

s in

X-r

ay A

naly

sis,

Vol

.42

344

ISSN 1097-0002

Table 2. Crystal Structure of dimethyl 2,7-naphthalenedicarboxylate

Space Group P2,, a = 24.4968(3), b = 6.06949(5), c = 3.95662(3)A,

P 0

Name X

Cl 0.2010(4) c2 0.1517 c3 0.1500 c4 0.1977 c5 0.2951 C6 0.3444 c7 0.3461 C8 0.2984 c9 0.2489 Cl0 0.2472 Cl1 0.1024(5) 012 0.0615(6) 013 0.1006(6) Cl4 0.3955(5) 015 0.4400(5) 016 0.3977(5) Cl7 0.0561(7) Cl8 0.4412(7) Hl 0.202120 H3 0.116150 H4 0.196350 H5 0.293990 H6 0.377120 H8 0.299650 H17a 0.060880 H17b 0.051530 H17c 0.024310 H18a 0.430820 H18b 0.459010 H18c 0.465310

Y Z Ui*lOO

0.6613(22) 0.5079(13) 2.50(29) 0.5458 0.5108 2.50 0.3371 0.6589 2.50 0.2444 0.8039 2.50 0.2672 0.9466 2.50 0.3827 0.9438 2.50 0.5914 0.7956 2.50 0.6841 0.6506 2.50 0.5689 0.6530 2.50 0.3595 0.8016 2.50 0.647187 0.3692(58) 6.60(28) O-5320(27) 0.3222(31) 6.60 0.8467(23) 0.2904(35) 6.60 0.7195 (29) 0.7568(58) 6.60 O-6251(26) O-8559(32) 6.60 0.9018(27) 0.6210(36) 6.60 0.9287(34) 0.1200(49) 6.60 1.0197(33) 0.5766(43) 6.60 0.804010 0.406300 5.00 0.256930 0.660570 5.00 0.100060 0.905310 5.00 0.122700 1.048650 5.00 0.317420 1.043810 5.00 0.826520 0.549510 5.00 1.082720 0.079850 8.00 0.853100 -0.094640 8.00 0.905080 0.247730 8.00 1.153740 0.463820 8.00 1.050110 0.789690 8.00 0.937350 0.437050 8.00

Our previous conformational analyses demonstrate that the planar ester conformation has the lowest energy, but that the sum of these two observed rotations results in an energy penalty of only 0.4 Kcal/mol. This energy difference is less than kT (0.6 Kcal/mol). Significant distortions of the molecule are readily possible to accommodate packing constraints.


The crystal structure of 27NDC consists of stacks of parallel molecules perpendicular to a (Figure 6). In an individual layer, all the 27NDC molecules are parallel, but the “tilt” differs in - adjacent layers. The ring planes are approximately 1 li and 111.

DIAMMONIUM TEREPHTHALATE

BP Amoco technology produces over 8 x 1 O9 pounds of purified terephthalic acid (TA) per year. From time to time, we isolate terephthalate complexes from process streams. Understanding the natures of terephthalate complexes of catalyst and corrosion metals will lead to process insights and improvements. Nothing, however, is known about the solid-state structures of such complexes, because they are intractable solids.

We have undertaken a program to prepare and understand the solid state structures of complexes of aromatic carboxylates. Since TA is so insoluble, the typical methods of preparation involve reaction with base to prepare soluble Na or K terephthalates. These preparations involve excess base, and thus complexes are prepared at high pH. Since the oxidation ofp-xylene to TA takes place in acetic acid/water, such high-pH complexes may not be relevant to process chemistry, although they are interesting in themselves.

In an attempt to prepare a reagent that would permit lower-pH syntheses, I carried out a gas- solid reaction between TA and aqueous ammonia to produce diammonium terephthalate. Surprisingly, neither this compound nor its crystal structure have been reported. This structure is of interest as a model for other structures, but also because I’ve solved it three times - in three different unit cells and space groups!

I’ve glossed over the determinationof the unit cell in the previous structures (IT0 [8] and/or DICVOL [12]), but Armel Le Bail is right when he emphasizes the importance of getting the right cell! In this case, it was only the use of synchrotron data that permitted determination of the correct cell, and refinement of the structure.

The synchrotron pattern (Figure 7) can be indexed on a primitive orthorhombic cell. Laboratory patterns yielded only monoclinic or orthorhombic subcells, which permitted solution (but not refinement) of the structure. The systematic absences limited the space groups to Pbc2, or Pbcm. With Z = 4, the natural assumption is to place the TA anion at a center in Pbcm, and the one ammonium ion at a general position. The structure, however, could not be solved in Pbcm. Only when the space group was lowered to Pbc2, could the structure be solved. No additional symmetry elements were detected in the refined structure.

This structure was also solved using the Insight11 STRUCTURE_SOLVEmodule [lo]. The positions and orientations of a TA and two ammonium ions were allowed to vary/anneal, as well as the carboxyl torsion angles. After a few thousand cycles (and some false minima), a refinable model was identified.


r

n

Copyright(C

)JCPD



dvances in X-ray A

nalysis, Vol.42

347C

opyright(C)JC

PDS-International C



-ray Analysis, V

ol.42347

ISSN 1097-0002

19436-148-1 (NH4)2TA, X3B1, cap., lam=1.15008, 13 Jul 97 Scan no. = 1 Lambdal,lambda2 = 1.150 Observed Profile

N

w z” c-4

0

m

C

I--

z c3 2C

G

I I I I I

1.0 2.0 3.0 4.0 5.0 2-Theta, deg XlOE 1

Fig. 7. The synchrotron powder pattern of diammonium terephthalate.


Since the synchrotron pattern showed signs of incomplete powder averaging, the structure was refined using laboratory data, collected on a freshly-prepared sample of diammonium terephthalate. The pattern was measured from 3-70” in 0.02” steps for 1.2 set/step on a Scintag PAD V diffractometerequipped with an Ortec intrinsic Ge detector. All data processing was carried out using GSAS [3]. Only the 8-70” 20 portion of the pattern was included. The benzene ring carbon and hydrogen atoms, and the carboxyl carbons C7 and C8 were described as a rigid body. Extensive use was made of soft constraints to restrain the bonded and non-bonded distances in the two independent carboxyl groups. The z-coordinate of 09 was fixed to determine the origin. In initial refinements, the hydrogen atoms of the ammonium ions were omitted. Trial positions for these hydrogens were obtained by using computational chemistry techniques. The heavy atom positions were fixed, and the ammonium ions were allowed to rotate, minimizing the energy using the Dreiding II force field as implemented in Cerius2 [9]. In the final refinement, the ammonium cations were described as rigid bodies.

Included in the refinement were a scale factor, the lattice parameters, and a March-Dollase preferred orientationratio (unique axis [OOl]). The peak profiles were described by a pseudo-Voigt function; only the Cauchy Y and stec (unique axis 00 1) strain broadening terms were refined, along with a sample displacement term. The background was described by a 6-term real space pair correlation function with two characteristic distances of 0.70 and 4.40 A.


Table 3. Crystal Structure of Diammonium Terephthalate Sp. Gr. PbcZ,, a = 4.0053(5), b = 11.8136(21), c = 20.1857(24) A

Name X Cl -0.157(5) c2 -0.088 c3 -0.171 c4 -0.323 c5 -0.392 C6 -0.309 c7 -0.067 C8 -0.413 H2 0.017 H3 -0.123 H5 -0.497 H6 -0.357 09 -0.070(5) 010 0.001(5) 011 -0.549(5) 012 -O-483(5) N13 0.504(4) H13a 0.445 H13b 0.329 H13c 0.674 H13d 0.569 N14 0.933(6) H14a 1.084 H14b 0.911 H14c 1.001 H14d 0.735

Y Z Ui*lOO -O-0853(11) 0.0389(13) 4.27(56) -0.1934 0.0152 4.27 -0.2222 -0.0496 4.27 -0.1429 -0.0907 4.27 -0.0348 -0.0669 4.27 -0.0059 -0.0021 4.27 -0.0541 0.1089 0.37(26) -0.1740 -0.1606 0.37(26) -0.2481 0.0436 6.00 -0.2968 -0.0659 6.00

0.0199 -0.0953 6.00 0.0686 0.0142 6.00

-0.1366(19) 0.151685 0.37 0.0464(17) 0.1269(24) 0.37

-0.1072(22) -O-1998(8) 0.37 -0.2859(16) -O-1680(24) 0.37

0.1902(26) 0.1778(20) 4.00 0.2635 0.1747 6.00 0.1462 0.1671 6.00 0.1763 0.1498 6.00 0.1750 0.2195 6.00 O-9226(25) 0.2818(21) 4.00 0.8667 0.2784 6.00 0.9428 0.3246 6.00 0.9827 0.2580 6.00 0.8984 0.2662 6.00

The final refinement of 38 variables using 3 108 observations yielded the residuals wRp = 0.1790,Rp=0.1335,~2=7.365,R(F2)=0.1339,andR(F)=0.0724. Theagreementoftheobserved and calculated patterns (Figure 8) is acceptable. The largest peak in a difference Fourier map was 0.40 eAm3 (0.72 8, f rom 012), and the largest difference hole was -0.35 eAm3. A normal probability plot indicated that the standard uncertainties were underestimated by a factor of 2.2. The refined structural parameters are reported in Table 3.


; r- -l- Lc E

pi u-i u-l -i-l n a 2

” 8

;-^d -Y

cN

mm

2 2

Q:

-j zc E

-ix -

I P

a; G

X2 z -u: C

l-i-3

I Lr

$7 l-l

G

2’

4-I

+- I

b i - - t .

O'P

0'2

0'0 E

3O

TX

squno3

3 . .

d w

4: 3x i 0 tfi

0 4 0 . m

0 4

5 a .

92 4

E:

CJ as

%

wq

8 ; $1

c?zm

b C

dA

aEd

$a E

6Y

&

360 $g -3 ‘3 cY

’c( 38

; E

QO

‘3

w

5 88 ‘85

.

g $&2

$8 .z 0 crj 32 :

22 E

SO

gz.2 ca-

+ 0

am

gj

333

8-d z

&EP .M

E

a v;

$

1 3

Q

cd 09

,a *cd

zi Y

gj* 2ag g-2

2 0 g.g

“ag g

z s

$2 &

P+o O

S0

ok%,

8 Q

Q

) bil

&tz *G

g

3

Copyright(C

)JCPD



dvances in X-ray A

nalysis, Vol.42

351C

opyright(C)JC

PDS-International C



-ray Analysis, V

ol.42351

ISSN 1097-0002


The crystal structure of diammoniumterephthalatereported here is an average structure. The appearance of the pattern (particularly between 20-30’ 28) indicates that diammonium terephthalate is heavily faulted. We have not yet described the detailed nature of the faulting. The fact that the displacement coefficients of the benzene ring carbon atoms are larger than those of the peripheral atoms indicates some disorder of the stacking. It is difficult to prepare diammonium terephthalate phase-pure; most preparations contain a few percent of ammonium hydrogen terephthalate, the powder pattern of which is similar to that of diammonium terephthalate. Not including this impurity phase (if present; the sample of Figure 8 is phase-pure) in refinement leads to distorted structural results.

Because rigid bodies and soft constraints were used in the refinement, bond distances and angles fall within the normal ranges. It has not proved possible to decide unambiguously whether the C-O bonds of the carboxylate groups are resonant; the single crystal structure of potassium terephthalate [ 13,141 leads us to suspect that they are. The carboxyl group C7-09-O 10 is apparently rotated 30 O out of the ring plane, and the C8-0 11-O 12 carboxyl exhibits a pyramidal distortion. The data quality supports only a determination of the average crystal structure, not details of the molecular structure.

As might be expected, this is also a layered structure (Figure 9), with alternating hydrophobic and hydrophilic layers perpendicularto a. Each nitrogen is surrounded by four oxygens in a roughly tetrahedral arrangement; each proton of each ammonium ion participates in hydrogen bonding. In the symmetry-constrained molecular mechanics refinement with the positions of the heavy atoms fixed, the hydrogens of the ammonium ions rotate to form bifurcated hydrogen bonds. It seems that the potential energy surface is fairly flat.

Within a layer, the tilts of the TA anions alternate. This structure is similar in some ways to those of potassium terephthalate [ 13,141 and sodium terephthalate [ 141, and provides a model for the TA layers in the solid state structures of complexes of more interesting metals, as well as for TA as an intercalant in clays.

We are now applying these simulated annealing methods to the structures of transition metal terephthalates, but have encountered two factors which make this procedure difficult. The metals so dominate the scattering that it becomes more difficult to place the terephthalates reliably. Although large numbers of metal carboxylate structure have not been reported, the observed structures exhibit a great variety of metal-carboxylate coordinations. Improvements in the force fields are clearly necessary before these tools can be used more widely.


2 + d 32 2 %

3 3 5 ‘a & i

.s ‘13r3 C

Hk

O,o

2-o 3s t; .z gg

Copyright(C

)JCPD



dvances in X-ray A

nalysis, Vol.42

353C

opyright(C)JC

PDS-International C



-ray Analysis, V

ol.42353

ISSN 1097-0002

ACKNOWLEDGMENTS

I thank Ying-Mei Chen, Maria Kaminsky, and Simine Short for technical support, and John Faber and Shiyou Pei for scientific encouragement. Myle Nguyen and B.L. Meyers performed TGA analyses. C.H. Cullen, C. Price, and G.J. Ray carried out the NMR studies. A. Bhattacharyya, L. C. Satek, W. P. Schammel, and J. J. Harper provided the materials upon which these studies were based. G. E. Kuhlmann provided partial financial support. J. T. Golab carried out the quantum chemical studies.

P. W. Stephens and his colleagues were most gracious and helpful during the collection of the synchrotron data. This work represents research carried out in part at the National Synchrotron Light Source at Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Division of Materials Sciences, and Division of Chemical Sciences. The SUNY X3 beamline at NSLS is supported by the Division of Basic Energy Sciences of the U.S. Department of Energy (DE-FG02-86ER4523 1).

REFERENCES 1.

2.

3.

4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14.

Y. Tanaka, J. Fukunaga, M. Setoguchi, T. Higashi, and M. Ihara, J. Ceram. Sot. Japan, 90, 458-463 (1982); ICSD collection code 201351. I. I. Kozhina, E. E. Kornilova, G. T. Petrovskii, and S. A. Stepanov, Vestn. Leningr. Univ., Fiz., Khim., 40-46( 1983). A. C. Larson and R. B. Von Dreele, GSAS, The General Structure Analysis System, Los Alamos National Laboratory, version of February 1993. G. M. Sheldrick, SHELXTL Plus, Version 3.4, Nicolet Instrument Corporation, 1988. A. C. Larson, F. L. Lee, Y. LePage, M. Webster, J. P. Charland, and E. J. Gabe, The NRCVAX Crystal Structure System, Chemistry Division, NRC, Canada, 1988. I. D. Brown and D. Altermatt, Acta Cryst., B41,244-247 (1985). R. Howie and A. West, Acta Cryst., B33,38 l-385 (1977); ICSD collection code 828. J. W. Visser, J Appl. Cryst., 2, 89-95 (1969). Molecular Simulations, Inc., Cerius’ TM, Version 2.1 (1996). Molecular Simulations, Inc., Insight11 TM, Version 4.0.0 (1996). J. A. Kaduk and J. T. Golab, accepted for publication in Acta Cryst. Section B (1999). A. Boultif and D. Louer, J Appl. Cryst., 24,987-993 (1991). N. Ebara and S. Furuyama, Sci. Pap. C. Gen. Educ. U Tokyo, 23,29 (1973). J. A. Kaduk, unpublished results.


'Solving Crystal Structures of Inorganic, Organic and ... crystal structures of inorganic, organic, and coordination compounds using synchrotron powder data james a. kaduk bp amoco

Documents