Synthesis and characterization of Pt(CN-p-(C2H5) C6H4)2(CN)2, a crystalline vapoluminescent compound that detects vapor-phase aromatic hydrocarbons

Article abstract of DOI:10.1021/ja011986v

The vapochromic and vapoluminescent compound Pt(CN-p-(C₂H₅) C₆H₄)₂(CN)₂ (abbreviated PtC₂) is conveniently synthesized by the thermal rearrangement of [Pt(CN-p-(C₂H₅) C₆H₄)₄][Pt(CN)₄]. Recrystallization of PtC₂ gives a crystalline orange morph (O-PtC₂) and an amorphous purple morph (P-PtC₂) that both contain the cis isomer and differ only in their solid-state packing arrangements. Both compounds were fully characterized by elemental analysis, X-ray powder diffraction, NMR, IR, and TGA. O-PtC₂ is vapochromic and vapoluminescent: it reversibly sorbs aromatic hydrocarbons and ethanol from air or nitrogen with resulting color changes and shifts in the emission spectrum. The λ_max values (nm) for emission from O-PtC₂·(guest) are as follows: no guest, 611; toluene, 565; benzene, 586; chlorobenzene, 592; p-xylene, 585; mesitylene, 565; ethanol, 587. In the case of toluene and mesitylene, intermediate emitting phases are observed. Recrystallization of O-PtC₂ from dichloromethane/toluene gave single crystals of PtC₂·0.5(toluene). The single-crystal X-ray structure of PtC₂·0.5(toluene) contains infinite stacks of cis geometry and planar molecules with chains of platinum atoms parallel to the c axis of the monoclinic unit cell. The average Pt-Pt separation in PtC₂·0.5(toluene) is c/4 = 3.288(2) A. There are solvent channels parallel with the c axis that contain the toluene molecule guests. Thin films of O-PtC₂ rapidly sorb toluene from the gas phase to form PtC₂·0.25(toluene) and PtC₂·0.5(toluene). Long-term exposure gives PtC₂· 1.0(toluene). Removal of the toluene source causes rapid desorption to PtC₂·0.5(toluene) and then to PtC₂·0.25(toluene). The remaining 0.25(toluene) lattice guests require heating for rapid removal. X-ray powder diffraction identified the PtC₂, PtC₂·0.25(toluene), and PtC₂·0.5(toluene) phases and showed that the sorption of toluene is accompanied by small changes in the unit cell dimensions that include lengthening the Pt-Pt distances in the structure. The sorption process improves the packing in the structure by utilizing some of the free volume for the toluene lattice guests.

Full text of DOI:10.1021/ja011986v

Published on Web 01/17/2002
Synthesis and Characterization of Pt(CN-p-(C2H5)C6H4)2(CN)2,
a Crystalline Vapoluminescent Compound That Detects
Vapor-Phase Aromatic Hydrocarbons
Carrie E. Buss and Kent R. Mann*
Contribution from the Department of Chemistry, UniVersity of Minnesota,
Minneapolis, Minnesota 55455
Received August 16, 2001
Abstract: The vapochromic and vapoluminescent compound Pt(CN-p-(C2H5)C6H4)2(CN)2 (abbreviated PtC2)
is conveniently synthesized by the thermal rearrangement of [Pt(CN-p-(C2H5)C6H4)4][Pt(CN)4]. Recrystal-
lization of PtC2 gives a crystalline orange morph (O-PtC2) and an amorphous purple morph (P-PtC2) that
both contain the cis isomer and differ only in their solid-state packing arrangements. Both compounds
were fully characterized by elemental analysis, X-ray powder diffraction, NMR, IR, and TGA. O-PtC2 is
vapochromic and vapoluminescent: it reversibly sorbs aromatic hydrocarbons and ethanol from air or
nitrogen with resulting color changes and shifts in the emission spectrum. The λmax values (nm) for emission
from O-PtC2‚(guest) are as follows: no guest, 611; toluene, 565; benzene, 586; chlorobenzene, 592;
p-xylene, 585; mesitylene, 565; ethanol, 587. In the case of toluene and mesitylene, intermediate emitting
phases are observed. Recrystallization of O-PtC2 from dichloromethane/toluene gave single crystals of
PtC2‚0.5(toluene). The single-crystal X-ray structure of PtC2‚0.5(toluene) contains infinite stacks of cis
geometry and planar molecules with chains of platinum atoms parallel to the c axis of the monoclinic unit
cell. The average Pt-Pt separation in PtC2‚0.5(toluene) is c/4 ) 3.288(2) Å. There are solvent channels
parallel with the c axis that contain the toluene molecule guests. Thin films of O-PtC2 rapidly sorb toluene
from the gas phase to form PtC2‚0.25(toluene) and PtC2‚0.5(toluene). Long-term exposure gives PtC2‚
1.0(toluene). Removal of the toluene source causes rapid desorption to PtC2‚0.5(toluene) and then to PtC2‚
0.25(toluene). The remaining 0.25(toluene) lattice guests require heating for rapid removal. X-ray powder
diffraction identified the PtC2, PtC2‚0.25(toluene), and PtC2‚0.5(toluene) phases and showed that the sorption
of toluene is accompanied by small changes in the unit cell dimensions that include lengthening the Pt-Pt
distances in the structure. The sorption process improves the packing in the structure by utilizing some of
the free volume for the toluene lattice guests.
sensitive layer for such a chemical sensor system.10-14 In
general, the complexes are robust and form intensely colored
solid-state materials that reversibly sorb solvent molecules from
Introduction
The development of reversible chemical sensor materials with
high stability has received increasing attention.1 Materials that
show dramatic and reversible color changes in the visible or
near-infrared (near-IR) spectral regions upon exposure to volatile
organic compounds (VOCs) are of particular interest.2-5 To be
useful, such materials should not only detect VOCs at low levels
but should also show a unique response for each VOC. Several
compounds of this type could also be used as a sensor array in
an “electronic nose”6 or as the detecting or emitting layer in a
photodiode or in a LED.7-9
(6) (a) Drew, S. M.; Janzen, D. E.; Buss, C. E.; MacEwan, D. I.; Dublin, K.
I.; Mann, K. R. J. Am. Chem. Soc. 2001, 123, 8414. (b) Gardner, J. W.;
Bartlett, P. N. Electronic Noses: Principles and Applications; Oxford
University Press: New York, 1999. (c) Persaud, K.; Dodd, G. H. Nature
(London) 1982, 299, 352. (d) Albert, K. J.; Lewis, N. S.; Schauer, C. L.;
Sotzing, G. A.; Stitzel, S. E.; Vaid, T. P.; Walt, D. R. Chem. ReV. 2000,
100, 2595-2626. (e) Grate, J. W. Chem. ReV. 2000, 100, 2627-2648. (f)
Lundstrom, I.; Hedborg, E.; Spetz, A.; Sundgren, H.; Winquist, F. Electronic
Nose Based on Field Effect Structures. In Sensors and Sensory Systems
for an Electronic Nose; Gardner, J. W., Bartlett, P. N., Eds.; NATO ASI
Series 212; Kluwer: Dordrecht, The Netherlands, 1992; pp 303-319. (g)
Miller, L. L.; Boyd, D. C.; Schmidt, A. J.; Nitzkowski, S. C.; Rigaut, S.
Chem. Mater. 2001, 13, 9-11. (h) Dickinson, T. A.; White, J.; Kauer, J.
S.; Walt, D. R. Nature 1996, 382, 697-700. (i) Rakow, N. A.; Suslick, K.
S. Nature 2000, 406, 710-713. (j) Seker, F.; Meeker, K.; Kuech, T. F.;
Ellis, A. B. Chem. ReV. 2000, 100, 2505-2536. (k) Sohn, H.; Letant, S.;
Sailor, M. J.; Trogler, W. C. J. Am. Chem. Soc. 2000, 122, 5399-5400. (l)
Additional references are given in the Supporting Information.
(7) Kunugi, Y.; Mann, K. R.; Miller, L. L.; Pomije, M. K. (Regents of the
University of Minnesota) U.S. Patent 6,137,118, 2000.
Our previous studies of square-planar d8 platinum compounds
of the form [Pt(CNR)4][M(CN)4] (M ) Pt, Pd; R ) aryl, alkyl;
“double salts”) have shown that they can form the chemically
* To whom correspondence should be addressed.
(1) A literature search produced several thousand references in the general
area of sensors published between 1996 and 2000.
(8) Kunugi, Y.; Mann, K. R.; Miller, L. L.; Exstrom, C. L. (Regents of the
University of Minnesota) U.S. Patent 6,160,267, 2000.
(2) Mansour, M. A.; Connick, W. B.; Lachicotte, R. J.; Gysling, H. J.;
Eisenberg, R. J. Am. Chem. Soc. 1998, 120, 1329.
(9) Kunugi, Y.; Miller, L. L.; Mann, K. R.; Pomije, M. K. Chem. Mater. 1998,
10, 1487.
(3) Beauvais, L. G.; Shores, M. P.; Long, J. R. J. Am. Chem. Soc. 2000, 122,
2763.
(10) Daws, C. A.; Exstrom, C. L.; Sowa, J. R.; Mann, K. R. Chem. Mater.
1997, 9, 363.
(4) Cariati, E.; Bu, X.; Ford, P. C. Chem. Mater. 2000, 12, 3385.
(5) Evju, J. K.; Mann, K. R. Chem. Mater. 1999, 11, 1425.
(11) Exstrom, C. L.; Pomije, M. K.; Mann, K. R. Chem. Mater. 1998, 10, 942.
9
10.1021/ja011986v CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002 1031

A R T I C L E S
Buss and Mann
slight modification of the method of Ugi and Meyr.19 Triethylamine
was dried over 3 Å molecular sieves and passed through a short column
of activated alumina immediately prior to use. A mixture of p-(C2H5)-
C6H4-NHCHO (5.695 g, 38.17 mmol), triethylamine (26.6 mL, 190.8
mmol), and CH2Cl2 (100 mL) was cooled to 0 °C in an ice bath. With
stirring, POCl3 (3.91 mL, 41.9 mmol) was added to the mixture
dropwise over a 30 min period. The mixture was stirred at 0 °C for 30
min, during which the solution color gradually turned from pale yellow
to orange-brown. The ice-water bath was removed, and 100 mL of a
phosphate buffer (pH 6.3) solution was added dropwise over a period
of 30 min. The organic layer was removed in a separatory funnel,
washed (3 × 100 mL) with a saturated NaCl solution, and dried over
anhydrous CaCl2. The CaCl2 was removed via filtration, and the solvent
was removed in vacuo to yield a dark brown liquid (4.724 g, 94%). 1H
NMR (CD2Cl2): δ 7.29 (d, 2 H, Ph), 7.21 (d, 2 H, Ph), 2.66 (q, 2 H,
CH2), 1.21 (t, 3 H, CH3). 13C NMR (CD2Cl2): δ 164.28 (CNR), 146.66
(Ph), 129.34 (Ph), 126.77 (Ph), 124.91 (Ph), 29.13 (CH2), 15.60 (CH3);
IR (ATR ZnSe crystal) νCN (cm-1) 2122 vs.
the gas phase, a process that is coupled with a shift in the
absorption spectrum. This process has been termed vapo-
chromism15 andisaccompaniedbyreversiblephysicalchanges.10-14
The vapochromic phenomenon has been studied with a variety
of techniques that include FT-IR, UV-vis absorption and
emission spectroscopy, solid-state NMR, powder and single-
crystal X-ray diffraction, and thermal gravimetric analysis.
Vapochromism in crystalline [Pt(CNR)4][Pt(CN)4] salts arises
from highly anisotropic packing forces that enable solvent
vapors to reversibly penetrate the interior of the material to form
a new crystalline phase with precisely determined solvent-
chromophore interactions.10-14 The [Pt(CNR)4][Pt(CN)4] solids
2+
consist of infinite stacks of alternating Pt(CNR)4 dications
and Pt(CN)42- dianions. The interionic metal-metal interactions
produce the chromophore.10-14 The vapor inclusion causes color
changes that result from a combination of chemical interactions
with the chromophore, including changes in the dielectric
constant near the chromophore, hydrogen bonding between the
solvent and coordinated cyanide, and expansion or contraction
of the unit cell that is coupled to the Pt-Pt distance.
[Pt(p-CN-C6H4C2H5)4][Pt(CN)4]20 (4C2). This compound was previ-
ously made by Keller and Lorentz by an alternate method.21 A better
procedure described for preparing [Pt(p-CN-C6H4C10H21)4][Pt(CN)4]19
was followed: Pt(CH3CN)2Cl2 (1.19 g, 3.43 mmol), [(n-C4H9)4N]2[Pt-
(CN)4] (2.69 g, 3.43 mmol), and p-CN-C6H4C2H5 (2.24 g, 17.1 mmol)
in acetonitrile (150 mL) were stirred for 1.0 h to yield a blue precipitate
that was isolated by filtration (3.01 g, 86% yield). Mp: 201 °C dec.
IR (ATR ZnSe crystal): νCNR 2258 cm-1 (vs); νCN 2126 cm-1 (vs).
Near-IR (ATR ZnSe crystal): λmax 876 nm. Anal. Calcd for C20H18N4-
Pt: C, 47.15; H, 3.56; N, 11.00. Found: C, 46.40; H, 3.32; N, 10.59.
Purple Pt(p-CN-C6H4C2H5)2(CN)2 (P-PtC2). The neutral compound
was synthesized by heating [Pt(p-CN-C6H4C2H5)4][Pt(CN)4] (0.772,
0.758 mmol) under an argon atmosphere to the melting point. When
the mixture was cooled, a glassy orange-red solid that is soluble in
CH2Cl2 was obtained. Recrystallization from CH2Cl2 and excess hexanes
When the double-salt compounds [Pt(CNR)4][Pt(CN)4] are
exposed to liquid chlorinated solvents for long periods of time
(days) or are heated to the melting point in the absence of
solvent, a ligand rearrangement reaction forms the isomeric
neutral compounds Pt(CNR)2(CN)2.16 These neutral compounds
are interesting because they also exhibit vapochromic properties,
have enhanced thermal stability, and have favorable solubility
properties. This paper reports a well-characterized example (Pt-
(CN-p-(C2H5)C6H4)2(CN)2) of this neutral compound class that
illustrates some aspects of their sensing capabilities.
1
led to a purple powder (0.589 g, 1.16 mmol, 76%). H NMR (CD2-
Experimental Section
Cl2): δ 7.51 (d, 2 H, Ph), 7.36 (d, 2 H, Ph), 2.74 (q, 2 H, CH2), 1.25
(t, 3 H, CH3). IR (ATR ZnSe crystal): νCNR 2251 (vs), 2231 cm-1
(vs); νCN 2157 (vs), 2150 cm-1 (vs). UV-vis (ATR, cubic zirconia
crystal): λmax 570 nm. Mp: 143 °C. Anal. Calcd for C20H18N4Pt: C,
47.15; H, 3.56; N, 11.00. Found: C, 46.99; H, 3.52; N, 10.75.
Orange Pt(p-CN-C6H4C2H5)2(CN)2 (O-PtC2). O-PtC2 was obtained
by the slow recrystallization of P-PtC2 from acetone/ether but the
procedure was prone to produce material contaminated with P-PtC2.
A better method was to slurry a sample of P-PtC2 (obtained by the
method above) with THF to give a yellow solid. Removal of the THF
solvent followed by heating to 100 °C yielded the orange product upon
cooling. Both procedures gave products that are identical by NMR and
FTIR. 1H NMR (CD2Cl2): δ 7.52 (d, 2 H, Ph), 7.36 (d, 2 H, Ph), 2.74
(q, 2 H, CH2), 1.25 (t, 3 H, CH3). 13C NMR (CD2Cl2): δ 171.61 (CNR),
149.50 (Ph), 129.75 (Ph), 127.61 (Ph), 123.28 (Ph), 110.67 (CN) with
satellite peaks at δ 118.14 and 103.22 (J(195Pt-13C) ) 562.8 Hz), 29.31
(CH2), 15.38 (CH3). IR (ATR ZnSe crystal): νCNR 2247 (vs), 2229
cm-1 (vs); νCN 2157 (vs), 2150 cm-1 (vs). UV-vis (thin film
transmission): λmax 515 nm. Mp: 174 °C. Anal. Calcd for C20H18N4-
Pt: C, 47.15; H, 3.56; N, 11.00. Found: C, 47.09; H, 3.58; N, 10.76.
Gravimetric Measurements. Solvent uptake/release studies were
conducted by thermal gravimetric analysis (TGA) with a Perkin-Elmer
TGA 7 instrument or by gravimetric analysis (Acculab LA-60 electronic
balance). TGA measurements were conducted with both P-PtC2 and
O-PtC2 in the unexposed state to determine if any lattice solvent was
present. A 5-10 mg sample was purged with nitrogen gas while heating.
The balance experiments were conducted in a 100 mL three-neck flask
fitted with stopcocks and a 10 cm ground-glass tube on the center neck.
A stainless steel wire was hung from the hook arm underneath the
General Considerations. 4-Ethylaniline was purchased from Aldrich
Chemical Co. Elemental analyses were performed by Quantitative
Technologies Inc. Analytical Laboratories. Pt(CH3CN)2Cl2 was prepared
from K2[PtCl4] (Colonial Metals) as previously reported.17 [(n-C4H9)4N]2-
[Pt(CN)4] was prepared from [(n-C4H9)4N]Br (Aldrich) and K2[Pt(CN)4]
(Colonial Metals) as previously reported.18 All solvents used in the IR
and UV-vis studies were ACS reagent grade and dried over molecular
sieves prior to use.
Materials. p-(C2H5)C6H4-NHCHO. A mixture of p-(C2H5)C6H4-
NH2 (10.3 mL, 0.0825 mol), 88% formic acid (30 mL, 0.82 mol), and
toluene (100 mL) was stirred at reflux for 9 h in a round-bottom flask
equipped with a condenser and a Dean-Stark water separator. Cooling
the mixture to room temperature afforded white flakes of p-(C2H5)-
C6H4-NHCHO that were separated from the mixture by filtration. A
second crop of the formamide was obtained by concentrating the filtrate
under vacuum to give a total of 10.749 g (87% yield). 1H NMR
1
1
(CDCl3): δ 8.63 (d, E isomer, /2 H, CHO), 8.48 (br d, Z isomer, /2
1
1
H, NH), 8.32 (d, Z isomer, /2 H, CHO), 7.58 (br s, E isomer, /2 H,
NH), 7.43 (m, 1H, Ph), 7.15 (m, 2H, Ph), 7.00 (m, 1H, Ph), 2.60 (m,
2H, CH2), 1.20 (m, 3H, CH3). IR (ATR ZnSe crystal): νCO (cm-1
)
1679 vs.
p-(C2H5)C6H4-NC. The aryl isocyanide was prepared from the
corresponding amine by dehydrating the formamide compound via a
(12) Buss, C. E.; Anderson, C. E.; Pomije, M. K.; Lutz, C. M.; Britton, D.;
Mann, K. R. J. Am. Chem. Soc. 1998, 120, 7783.
(13) Exstrom, C. L.; Sowa, J. R., Jr.; Daws, C. A.; Janzen, D. E.; Mann, D. R.
Chem. Mater. 1995, 7, 15.
(14) Mann, K. R.; Daws, C. A.; Exstrom, C. L.; Janzen, D. E.; Pomije, M. K.
(Regents of the University of Minnesota) U.S. Patent 5,766,952, 1998.
(15) Nagel, C. C. U.S. Patent 4,834,909, 1989.
(16) Exstrom, C. L. Ph.D. Dissertation, University of Minnesota, 1995.
(17) Fanizzi, F. P.; Intini, L.; Maresca, L.; Natile, G. J. Chem. Soc., Dalton
Trans. 1990, 199.
(19) Ugi, I.; Meyr, R. Org. Synth. 1961, 41, 101.
(20) Anderson, C. E. Masters Thesis, University of Minnesota, 1997.
(21) Keller, H. J.; Lorentz, R. J. Organomet. Chem. 1975, 102, 119.
(18) Mason, W. R.; Gray, H. B. J. Am. Chem. Soc. 1968, 90, 5721.
9
1032 J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002

A Crystalline Vapoluminescent Compound
A R T I C L E S
Table 1. Crystal Data, Data Collection Details, and Refinement
Parameters for Pt(CN-p-(C2H5)C6H4)2(CN)2‚0.5(toluene) (1) and
Pt(CN-p-(C2H5)C6H4)2(CN)2‚x(hexanes) (2)
Table 2. Bond Lengths (Å) and Selected Bond Angles (deg)a for
Pt(CN-p-(C2H5)C6H4)2(CN)2‚0.5(toluene) (1) and
Pt(CN-p-(C2H5)C6H4)2(CN)2‚x(hexanes) (2)
1
2
1
2
formula
habit, color
size, mm
lattice type
space group
a, Å
C23.5H22N4Pt
needle, yellow
C20H18N4Pta
needle, yellow
Pt(1)-C(3)
Pt(1)-C(12)
Pt(1)-C(1)
Pt(1)-C(2)
N(1)-C(1)
C(2)-N(2)
C(3)-N(3)
N(3)-C(4)
N(4)-C(12)
N(4)-C(13)
C(4)-C(5)
1.992(7)
1.977(6)
2.001(7)
2.010(7)
1.144(7)
1.138(7)
1.129(7)
1.404(8)
1.143(7)
1.410(7)
1.386(8)
1.378(8)
1.389(8)
1.387(8)
1.376(8)
1.508(8)
1.398(8)
1.501(9)
1.391(9)
1.373(8)
1.378(8)
1.389(8)
1.381(9)
1.505(8)
1.376(8)
1.507(9)
1.983(10)
1.989(11)
1.998(10)
2.014(11)
1.129(11)
1.136(12)
1.137(11)
1.411(11)
1.143(12)
1.398(11)
1.362(13)
1.403(12)
1.357(13)
1.382(13)
1.369(13)
1.558(12)
1.385(12)
1.448(14)
1.355(13)
1.384(12)
1.358(13)
1.419(12)
1.375(14)
1.532(13)
1.392(13)
1.519(13)
0.30 × 0.10 × 0.10 0.35 × 0.13 × 0.07
monoclinic
C2/c
monoclinic
C2/c
19.893(1)
18.365(1)
13.150(1)
118.127(1)
4237.1(4)
8
555.54
1.742
6.639
2152
1.61-25.01
-23 e h e 20,
0 e k e 21,
0 e l e 15
9918
3687 (Rint ) 0.0274) 3683 (Rint ) 0.0357)
0.0307, 3.4971
0.000 37(3)
20.094(1)
18.243(1)
13.141(1)
117.813(1)
4260.6(3)
8
b, Å
c, Å
â, deg
V, Å3
Z
fw
509.47a
C(4)-C(9)
Dc, g cm-3
µ, mm-1
F(000)
1.589a
C(5)-C(6)
6.594a
C(6)-C(7)
1952a
C(7)-C(8)
θ range, deg
index ranges
1.60-25.05
-23 e h e 21,
0 e k e 21,
0 e l e 15
9183
C(7)-C(10)
C(8)-C(9)
C(10)-C(11)
C(13)-C(14)
C(13)-C(18)
C(14)-C(15)
C(15)-C(16)
C(16)-C(17)
C(16)-C(19)
C(17)-C(18)
C(19)-C(20)
no. of rflns collected
no. of unique rflns
weighting factors:b a, b
extinction coeff
max, min transmissn
no. of data, restraints, params 3687/41/252
0.0406, 0.0000
0.000 10(3)
0.630, 0.099
3683/30/229
0.0420, 0.0848
0.1055, 0.1023
0.932
0.514, 0.136
R1, wR2 (I > 2σ(I))
0.0318, 0.0631
0.0703, 0.0732
1.032
R1, wR2 (all data)
C(3)-Pt(1)-C(12)
C(3)-Pt(1)-C(1)
C(12)-Pt(1)-C(1)
C(3)-Pt(1)-C(2)
C(12)-Pt(1)-C(2)
C(1)-Pt(1)-C(2)
91.0(4)
88.9(4)
177.7(4)
177.5(4)
91.4(4)
88.6(4)
91.3(2)
89.5(2)
177.6(2)
177.2(2)
90.8(2)
88.3(2)
goodness of fit (on F2)
largest diff peak, hole, e Å-3 0.802, -0.485
1.753, -1.179
a These items do not take into account the electron density of the dis-
2
ordered solvent molecule. b w ) [σ2(Fo ) + (aP)2 + (bP)]-1, where P )
2
2
(Fo + 2Fc )/3.
a Estimated standard deviations in the least significant figure are given
in parentheses.
balance and attached to a holder designed for a 1 cm diameter aluminum
pan. Samples (15-30 mg) were placed in the pan, set on the holder,
and suspended down the tube into the flask. After the sample was
purged with pure nitrogen, the nitrogen stream was bubbled through
the liquid VOC. Sample mass was recorded as a function of time.
X-ray Structure Determination of Pt(p-CN-C6H4C2H5)2(CN)2‚
x(hexanes). Crystals were grown by slow evaporation from a CH2Cl2/
hexanes mixture. A crystal was attached to a glass fiber and mounted
on the Siemens SMART CCD platform system22 fitted with a graphite
monochromator for data collection at 173(2) K with Mo KR (R )
0.710 73 Å) radiation. Crystal and X-ray collection and refinement data
are summarized in Table 1.
The space group C2/c was determined on the basis of systematic
absences and intensity statistics. A direct-methods solution provided
the positions of most non-hydrogen atoms. Several full-matrix least-
squares/difference Fourier cycles were performed to locate the positions
of the remaining light, non-hydrogen atoms. The structure was found
to contain channels filled with disordered solvent molecules. Attempts
to model the disordered hexanes failed; the R1 value of the structure
without modeling the solvent was R1 ) 5.00%. Additional information
about the solvent channels was obtained using the Platon/Squeeze
programs.23 The channels were found to occupy 796.0 Å3, or 17.8% of
the total unit cell volume. A total electron count of 304.2 e (ap-
proximately six hexanes) was found to occupy the channels. The
original data were corrected for the disordered solvent, and the structure
refined as riding atoms with isotropic relative displacement parameters.
Additionally, the phenyl rings of the isocyanide ligands were refined
as rigid bodies. Several atoms had large anisotropic displacement
parameter maximum/minimum ratios: in particular, N3 and several
carbon atoms on the phenyl rings.
The asymmetric unit is one formula unit. The platinum atom was
found to occupy a general position. The two largest peaks in the
difference Fourier map with heights greater than 1 e/Å3 are near the
platinum atom: height 1.75 e/Å3, 1.02 Å from Pt1; height 1.35 e/Å3,
1.14 Å from Pt1. All calculations were performed with the SHELXTL
V5.0 suite of programs.24 Selected bond lengths and bond angles are
given in Table 2.
X-ray Structure Determination of Pt(p-CN-C6H4C2H5)2(CN)2‚
0.5(toluene). Crystals were grown by slow evaporation from a CH2-
Cl2/toluene mixture. A crystal was attached to a glass fiber and mounted
on the Siemens SMART CCD platform system.22 Crystal and X-ray
collection and refinement data are summarized in Table 1. Initially a
1
subcell with /4 the size of the c axis was indexed. Subsequently the
data were reindexed and it was found that the cell conformed to the
current C-centered monoclinic cell. The space group C2/c was
determined on the basis of systematic absences and intensity statistics.
The structure initially could not be solved in C2/c and was solved in
Cc. A 2-fold axis was found, and the structure was transformed to the
C2/c space group.
A direct-methods solution provided the positions of most non-
hydrogen atoms. Several full-matrix least-squares/difference Fourier
cycles were performed to locate the positions of the remaining light,
non-hydrogen atoms. The structure was found to contain channels filled
with disordered toluene molecules. The channels were found to occupy
was refined to give a final R1 ) 4.20%. The empirical formula, F000
,
formula weight, and absorption coefficient are all known to be incorrect,
since no solvent was taken into consideration in those calculations.
All non-hydrogen atoms were refined with anisotropic displacement
parameters. Hydrogen atoms were placed in idealized positions and
(22) Siemens SMART Platform CCD; Siemens Industrial Automation, Inc.,
Madison, WI.
(23) Spek, A. L. Acta Crystallogr. 1990, A46, C34.
(24) Sheldrick, G. M. SHELXTL, version 5.0; 1994.
9
J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002 1033

A R T I C L E S
Buss and Mann
773.3 Å3, or 18.2% of the total unit cell volume. The toluene molecule
found occupying the channel was disordered and was refined as a rigid
body. A total of 41 restraints were used to refine the toluene molecule.
The toluene molecule is disordered over an inversion center with each
position having an occupancy of 0.5. Final refinement gave R1 ) 3.17%.
All non-hydrogen atoms were refined with anisotropic displacement
parameters. Hydrogen atoms were placed in idealized positions and
refined as riding atoms with isotropic relative displacement parameters.
The asymmetric unit is one formula unit. The platinum atom is located
on a general position. There are no remaining electron density peaks
greater than 1 e/Å3 in the Fourier map. All calculations were performed
with the SHELXTL V5.0 suite of programs.24 Selected bond lengths
and bond angles are given in Table 2.
Powder X-ray Diffraction Data Collection. Room temperature (23
°C) X-ray powder diffraction patterns were obtained with a Siemens
D-5005 diffractometer with a 2.2 kW sealed copper source equipped
with a single-crystal graphite monochromator, scintillation counter, and
slits of [1°, 1°, 1°, 0.6°, 0.6°]. Powder diffraction patterns were collected
from 2θ ) 2-60° with a 2θ stepping angle of 0.02° and an angle
dwell of 2 s.
Infrared Spectroscopy. Infrared spectra were obtained by the
attenuated total reflectance (ATR) method with a Nicolet Magna-FTIR
System 550 spectrometer equipped with a ZnSe trough HATR cell from
PIKE Technologies. Data were processed with OMNIC ESP v. 4.1
software. Sample films were coated on the ZnSe crystal from either an
ether suspension or a CH2Cl2 solution. For vapochromic experiments,
a beaker filled with glass wool saturated with several milliliters of VOC
was placed on the ZnSe mount and the ATR cell was then covered.
Spectra were recorded as the equilibrium vapor pressure of the VOC
was established, after the VOC source was removed, and after heating
of the sample in a 100 °C oven.
Emission Spectra. Emission spectra were recorded at room tem-
perature with a custom Princeton Instruments ST138 UV-vis/near-IR
sensitive LN/CCD spectrometer. The Winspec v2.3.3 software package
was used for acquiring the emission spectra. The emission spectra (front
face detection from solids) were collected from samples irradiated at
435.8 nm with the interference-filtered output of an Oriel 100 W Hg
arc lamp. The emission spectra were corrected for grating efficiency
and detector response.26 The emission efficiency was estimated to be
similar to Ru(bpy)3Cl2 (bpy ) 2,2′-bipyridine) by measuring the
scattered excitation light and emitted light from a film of O-PtC2 and
then from an optically dense solution of Ru(bpy)3Cl2 in methanol.27
A film of the O-PtC2 compound was coated onto a calcium fluoride
window from an ether slurry and fitted into a flow-through cell. Heating
tape was wrapped around the cell, and the temperature of the sample
was monitored via a thermocouple that was placed to contact the sample
film without interfering with the emission of the compound. A gas
inlet and outlet located on the back of the cell were used to flow dry
N2 or VOC-saturated N2 over the sample. During the exposure of the
sample to the VOC a continuous series of spectra were collected.
SPECFIT,28 a spectral data analysis software package, was used to fit
the spectral changes and to calculate the eigenspectra of the different
uptake phases.
The sample cell was constructed by gluing four glass microscope
slide covers to a 45 mm × 45 mm × 3.2 mm glass plate. The slide
covers were arranged to produce a 7.8 mm × 23.3 mm indentation 0.4
mm deep. This arrangement forms a thin sample well that requires a
small sample size (∼20 mg) and allows rapid equilibration of the
compound with the solvent vapor. The sample powder was packed into
the sample well and leveled off with a microscope slide.
Vapochromic X-ray experiments were carried out by enclosing the
sample holder in a small, resealable plastic bag. The powder patterns
collected with the sample inside the plastic bag show a slight increase
in the baseline intensity but do not contain additional reflections in the
2θ ) 2-60° data collection range. The samples were exposed to VOCs
by placing glass wool saturated with 5-10 mL of toluene in the bag,
which was then sealed to create a room-temperature atmosphere
saturated with the VOC vapors. The sample, toluene vapor, and liquid
toluene were allowed to come to equilibrium. The equilibration period
was monitored by collecting fast “snapshot” scans until the position of
the reflections became constant. These “snapshot” scans (∼10 min)
were typically collected for 2θ ) 2-20° with slits of [1°, 1°, 1°, 0.6°,
0.6°], a 2θ stepping rate of 0.05°, and an angle dwell of 1 s. As the
VOC equilibrated with the sample, the reflections became increasingly
broad because the sample surface buckles due to the increase in volume
of the unit cell. Once the peak positions were stable, the sample was
quickly removed from the bag and the surface of the sample was
flattened. The sample was again placed in the bag with the VOC,
“snapshot” scans were used to again check for equilibration, and a long
scan was then collected. The sample was then removed from the bag
and allowed to equilibrate with the room atmosphere. Again snapshot
scans were used to determine when the positions of the reflections had
become constant, and a long scan was collected. The sample was then
allowed to sit in the laboratory atmosphere for approximately 1 week,
and a final scan was then obtained.
Results
Synthesis and Characterization. The neutral PtC2 com-
pound is synthesized in high yield by heating the precursor
double-salt compound 4C2 to the 201 °C melting point.
Recrystallization of the resulting material results in either purple
(λmax 570 nm) or orange (λmax 515 nm) products that are identical
by elemental analysis and solution NMR. The purple product
is easily obtained by rapidly adding excess hexanes to a CH2-
Cl2 solution of PtC2. The orange product (O-PtC2) is obtained
in pure form by slow crystallization from CH2Cl2 solution, but
it was more conveniently produced by slurrying P-PtC2 with
THF to give a yellow solid. Heating this yellow compound to
110 °C and subsequent cooling to room temperature results in
an orange product that NMR, FTIR, and X-ray powder
diffraction show is identical with O-PtC2 obtained by crystal-
lization from CH2Cl2.
Both P-PtC2 and O-PtC2 are stable in the laboratory
atmosphere; thus, no special precautions were taken prior to
any experiments. The possibility that P-PtC2 and O-PtC2
correspond to a cis/trans isomer pair was ruled out by thin-
layer chromatography and subsequent physical studies (vide
infra). Qualitative vapochromic tests indicated that P-PtC2 is
not appreciably vapochromic, while O-PtC2 responded visually
Variable-temperature X-ray powder diffraction scans were obtained
on a Scintag XDS200Theta/Theta diffractometer equipped with a high-
and low-temperature apparatus and slits of [1°, 2°, 0.5°, 0.3°] moving
from the source to the detector. Patterns were collected in the step-
scanning mode from 2θ ) 2 to 2θ ) 60° with a 2θ step angle of 0.02°
and an angle dwell time of 2 s. The sample material was exposed to
the toluene vapor for several hours, packed into the sample holder,
and allowed to equilibrate with the laboratory atmosphere. A long scan
was collected, and then the sample was heated to 100 °C. Once the
temperature set point was reached, the sample was allowed to equilibrate
at that temperature for 30 min before the data collection was begun. A
second postheating room-temperature long scan was collected. The
sample holder consisted of a copper plate with a circular sample well.
All powder diffraction data were processed using Jade v.3.0 software.25
(25) Jade V3.0 software; Materials Data Inc., Livermore, CA.
(26) The emission spectra were corrected to arbitrary units proportional to
photons per energy. For this calculation see: Blasse, G.; Grabmier, B. C.
Luminescent Materials; Springer-Verlag: Heidelberg, Germany, 1994; p
A225.
(27) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583.
(28) SPECFIT: Binstead, R. A.; Zuberbu¨hler, A. D. Specfit, version 2.09;
Spectrum Software Associates, Chapel Hill, NC, 1995.
9
1034 J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002

A Crystalline Vapoluminescent Compound
A R T I C L E S
Table 3. Emission λmax, Vapochromic Shifts, and VOC
Concentrations for Emission Studies of O-PtC2
vapochromic
shift (nm)a
VOC concn
mole fractionb
VOC
λmax
none
toluene
benzene
chlorobenzene
p-xylene
mesitylene
ethanol
611
565
586
592
585
565
587
-46
-25
-19
-26
-46
-24
0.0337
0.114
0.0142
0.0100
0.0028
0.0689
a Vapochromic shift ) λmax(VOC) - λmax(none). b The concentrations
of each VOC in the N2 stream calculated from the partial pressures at
296 K.
to a variety of solvents; consequently, studies focused on
O-PtC2.
Vapochromic Luminescence Studies of O-PtC2. The solid-
state emission spectrum of O-PtC2 in the dry state under a
nitrogen atmosphere at room temperature shows an intense
emission band at 611 nm. The emission spectra in air were in
all respects identical with those measured under nitrogen. Solid
O-PtC2 gave an emission intensity (after correcting for scatter)
similar to an optically dense solution of Ru(bpy)3Cl2 in
methanol. By analogy to the linear chain M2Pt(CN)4 and Pt-
(CNR)4Pt(CN)4 compounds, we assign this emission band to a
component of the dσ*-pσ one-electron excitation.10,11,13,29,30
Exposure of the compound to the VOC vapors studied here
resulted in a shift of the emission band to higher energy and a
decrease in the band intensity (Table 3). In all cases heating to
110 °C while purging the sample with N2 completely removed
the solvent and regenerated the original spectrum of dry O-PtC2.
The data for each solvent were taken after two or three exposure/
heat/cool cycles. In all cases the process was found to be
completely reversible and no detectable degradation or rear-
rangement of the compound to P-PtC2 occurred after many
cycles. The same film was used for multiple solvent exposures.
The exposure to toluene and mesitylene occurred with the
formation of an observable intermediate, while exposure to
benzene, xylene, chlorobenzene, and ethanol did not.
Figure 1. Emission spectra excited at λ 435.8 nm of a thin film of O-PtC2
as toluene vapor is admitted. The arrow shows the direction of intensity
change. The first nine spectra are taken 0.6 s apart, the next five spectra
are taken 2.0 s apart, and the final six spectra are taken 10 s apart.
Figure 2. Calculated pure component spectra from the data in Figure 1:
(A) unexposed O-PtC2; (B) O-PtC2‚0.25(toluene); (C) O-PtC2‚0.5(toluene).
Table 4. FT-IR Data for P-PtC2 and O-PtC2 Exposed to Toluene
Changes in the emission spectrum of O-PtC2 during the
exposure to toluene clearly show the two sorption processes.
During the first sorption the emission spectrum shifts rapidly
(ca. 5 s) to higher energy by 16 nm and the emission intensity
at the maximum decreases by approximately 45%. The second
sorption occurs on a slower time scale (ca. 2 min); the spectrum
shifts to higher energy by 28 nm and the intensity decreases an
additional 12%. Both shifts occur isosbestically (Figure 1). The
two isosbestic points occur at approximately 535 and 562 nm.
Specfit28 was used to obtain emission spectra of the three
different pure phases (Figure 2) involved in the toluene
exposure. Similar results were obtained for the mesitylene
exposure, but the emission maxima occur at different wave-
lengths (Table 3). The concentrations of the VOCs in the
nitrogen stream were calculated from the partial pressure of each
VOC at 296 K.31 Results are summarized in Table 3.
ν
CNR (cm-1
)
ν )
CN (cm-1
compd
P-PtC2
O-PtC2
O-PtC2
2251
2247
2246
2244
2244
2231
2229
2229
2229
2229
2157
2150
2150
2153
2153
2152
a
2157
2159
2159
2159
b
O-PtC2‚0.25(toluene)
O-PtC2‚0.5(toluene)
a Original preexposure sample. b After exposure to toluene.
the intermediate phase but reversal to the original spectrum
requires heating or long-term (days) N2 purging. In the case of
benzene, xylene, chlorobenzene, and ethanol the original
spectrum is obtained with heating.
Vapochromic Infrared Studies. We investigated the interac-
tions between the VOC guest and the O-PtC2 host with ATR
FTIR spectroscopy (Table 4). Previously, the CtN stretching
region (2100-2300 cm-1) of the mid-IR has been shown to be
sensitive to the presence of VOC guest molecules.10-12 A film
of O-PtC2 (deposited on the ATR crystal from an ether slurry)
showed νCN stretches at 2157 and 2150 cm-1 and νCNR stretches
at 2247 and 2229 cm-1. The observation of two bands each for
νCN and νCNR is consistent with a cis geometry about the Pt
center (confirmed by crystallography; vide infra). The orange
In the case of toluene and mesitylene exposures, removal of
the VOC causes the spectrum to quickly (minutes) reverse to
(29) Gliemann, G.; Yersin, H. Struct. Bonding 1985, 62, 87.
(30) Miskowski, V. M.; Houlding, V. H. Inorg. Chem. 1991, 30, 4446.
(31) Nelson, G. O. Gas Mixtures-Preparation and Control; Lewis Publishers:
Chelsea, MI, 1992.
9
J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002 1035

A R T I C L E S
Buss and Mann
film cast from ether was dissolved on the ATR plate with CH2-
Cl2, and when the solvent evaporated, a purple film remained.
After the film was purged with N2 for several minutes, the
spectrum of the purple film exhibited νCN stretches at 2157 and
2150 cm-1 and νCNR stretches at 2251 and 2231 cm-1. The IR
spectra of O-PtC2 and P-PtC2 are nearly identical. A second
spectrum of the purple material was obtained from a film of
P-PtC2 cast from an ether slurry. This spectrum was identical
with that of the purple film resulting from the O-PtC2 film
redissolved in CH2Cl2 and allowed to evaporate to dryness.
Again, both the P-PtC2 and the O-PtC2 IR spectra are consistent
with a cis geometry about the Pt metal center.
The first exposure of the dry O-PtC2 film to toluene vapor
results in a rapid change at room temperature of the CtN
stretching region (Table 4). The full exposure to toluene occurs
quickly, within approximately 2 min. Peaks for this species are
at νCN 2159 and 2152 cm-1 and νCNR 2244 and 2229 cm-1
.
Figure 3. ORTEP diagram showing one molecule of O-PtC2 from the
O-PtC2‚0.5(toluene) crystal structure. Ellipsoids are drawn at the 50% level.
The atom-numbering scheme for the O-PtC2‚x(hexanes) crystal is identical.
The peak area of the CNR region increased by 28%, and the
peak area of the CN region increased by 40% upon exposure
of the dry sample to toluene vapors. Additional equilibration
time with the toluene vapor for 1 h gave no further changes.
Removal of the toluene vapor gave a rapid spectral change (2
min) to an intermediate phase that is stable under nitrogen for
more than 1 h. This intermediate phase has peaks at νCN 2159
and 2153 cm-1 and νCNR 2229 cm-1. The peak area of the CNR
stretching region decreased 28% as compared to the toluene-
exposed sample, returning to the intensity of the original dry
sample. However, the peak area of the CN region decreased by
only 20%, or half of the original increase. Cycling between the
intermediate and full exposure spectra occurs rapidly and very
reversibly. The complete removal of toluene from the O-PtC2
film requires heating the sample to 100 °C with an N2 purge to
regenerate the spectrum of the dry sample (peaks at νCN 2159
and 2153 cm-1 and νCNR 2246 and 2229 cm-1). The complete
cycle was repeated several times with identical results and with
no noticeable degradation of the compound.
Figure 4. Solid-state packing diagram of O-PtC2‚0.5(toluene) showing
the Pt-Pt interactions.
Single-Crystal Structures of Pt(CN-p-(C2H5)C6H4)2(CN)2.
Single-crystal data for PtC2‚0.5(toluene) and PtC2‚x(hexanes)
are given in Table 1; selected bond lengths and angles are given
in Table 2. The two crystal structures of PtC2 have nearly
identical packing arrangements. Both structures consist of
infinite stacks of cis geometry and planar molecules with chains
of platinum atoms that are parallel to the c axis (Figure 3).
Within the unit cell there are two crystallographically equivalent
chains, one centered near a ) 0, b ) 0 and the other related by
the c face centering, near a ) 0.5 and b ) 0.5. In each chain
four molecules stack along the c axis per unit cell (Figure 4).
The platinum atoms are all on general positions. The second
PtC2 molecule in the chain is related to the first PtC2 molecule
by a 2-fold axis parallel with the b axis and located at 1/4 along
the c axis, the third PtC2 molecule in the chain is related to the
first PtC2 molecule by a c glide plane, and the fourth PtC2
molecule is related to the first PtC2 molecule by an inversion
of platinum atoms are slightly zigzagged, giving Pt-Pt-Pt
angles of 175.5° in PtC2‚0.5(toluene) and 176.1° in PtC2‚
x(hexanes). The location of the isocyanide ligands allows solvent
channels parallel with the c axis. These channels are located at
approximately a ) 0.5, b ) 0 and a ) 0, b ) 0.5 (Figure 5).
Each channel occupies 387.5 Å3 within one unit cell length.
The total solvent-accessible volume occupies 18.2% of the total
volume of the unit cell (or 193 Å3 per toluene molecule for the
PtC2‚0.5(toluene) structure).
TGA and Gravimetric Studies. TGA studies with both the
purple and orange morphs of PtC2 showed that neither of these
compounds lost significant mass or decomposed when heated
to 150 °C. This result is consistent with no lattice solvent present
in either compound. The purple morph was not investigated
further.
Gravimetric studies with bulk samples (ca. 10 mg) show that
exposure of O-PtC2 to toluene results initially in a very fast
mass gain equivalent to 0.5(1) toluene molecules per O-PtC2
molecule over approximately 6 min. Additional exposure for
long periods of time (hours) results in additional mass gain with
a final mass equivalent to 0.9(1) toluene molecules per O-PtC2.
After the VOC source is removed from the O-PtC2 environment
and the sample is purged with N2, it quickly (5 min) loses weight
1
center positioned at a ) 0, b ) 0, c ) /2. The average Pt-Pt
separation in PtC2‚0.5(toluene) is 3.2905(2) Å, nearly identical
with c/4 ) 3.288(2) Å. This structure packs with an alternating
long and short Pt-Pt separation (3.281 and 3.300 Å). The
average Pt-Pt separation in PtC2‚x(hexanes) is 3.287(2) Å (with
c/4 ) 3.285(2) Å); this structure also packs with an alternating
long and short Pt-Pt separation (3.278 and 3.296 Å). The chains
9
1036 J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002

A Crystalline Vapoluminescent Compound
A R T I C L E S
Cerius2 software32 program, and the cell constants were varied
until its predicted powder pattern matched that of the experi-
mental pattern. Final indexing and cell refinement was done
with Jade V 3.0 software,25 and the results are given in Table
5. The resulting cell is monoclinic with Z ) 8 and dimensions
a ) 18.95(1) Å, b ) 17.89(1) Å, c ) 12.80(1) Å, and â )
107.9(1)°. Exposure of the dry O-PtC2 to toluene results in a
color change from orange to yellow and changes in the powder
diffraction pattern. The toluene-exposed O-PtC2 compound
powder pattern was indexed to dimensions a ) 20.22(1) Å, b
) 18.29(1) Å, c ) 13.15(1) Å, and â ) 118.13 (1)°. This cell
is nearly identical with that obtained for the single-crystal study
and is assigned to the PtC2‚0.5(toluene) phase. Removal of the
toluene vapor from the compound results in a third powder
pattern but retention of the yellow color. The powder diffraction
pattern of this intermediate was indexed to a cell with dimen-
sions a ) 20.22(1) Å, b ) 18.29(1) Å, c ) 12.80(1) Å, â )
118.13(1)°. We suggest this powder pattern corresponds to the
O-PtC2‚0.25(toluene) phase that is observed in the gravimetric
studies. Finally, an exposed sample was heated to 100 °C for
30 min, and cooled, and a final powder pattern of the resulting
orange material was taken. The powder pattern of the cooled
sample is identical with that of the original preexposure sample.
The unit cell data for the different phases of toluene exposure
as well as for the single-crystal structures are found in Table 5.
Figure 5. Solid-state packing diagram of O-PtC2‚0.5(toluene). The view
is down the c axis, showing the solvent channels parallel with the c axis.
The toluene molecules were removed for clarity.
Discussion
Previous studies of the platinum double salts of the general
formula [Pt(CNR)4][Pt(CN)4] (R ) aryl, alkyl) have shown that
the color as well as the unit cell changes that occur upon uptake
of VOC vapor molecules can be dramatic.7-16 It is interesting
that these ionic double salts are precursors for isomeric neutral
compounds such as PtC2, which also exhibit vapochromic
behavior. The broad implication is that the packing of the
insoluble salt and the corresponding soluble neutral species are
similar enough to allow sorption but have significantly different
behavior. Although the neutral compound studied here does not
exhibit color (orange to yellow) or unit cell changes associated
with the solvent vapor uptake as dramatic as the double salts,
O-PtC2 has several important advantages as a potential sensor
material. The neutral compound is thermodynamically stable
to at least 120 °C, it does not respond appreciably to water or
air, it is a crystalline phase so all of the solvent guest sites are
identical crystallographically, and it is soluble in many common
solvents, from which it can be cast as films.
A complicating factor for PtC2 is its existence as two different
solid-state morphs. Purple P-PtC2 and orange O-PtC2 have
equivalent solution NMR spectra, are identical by TLC, and
have IR spectra indicating a cis geometry about the platinum
center, eliminating the possibility that the two forms are a cis/
trans isomer pair. The literature has several examples of similar
neutral platinum compounds that exist in two differently colored
forms.30,33 Specifically, Pt(2,2′-bipyrimidine)(CN)2, a compound
that also forms linear chains of platinum atoms in the solid state,
was found to have both a purple and a yellow form.34 In this
case the authors concluded that the color difference results from
differing numbers of water molecules in the lattice. However,
Figure 6. X-ray powder diffraction patterns for O-PtC2‚x(toluene): (A)
unexposed O-PtC2; (B) O-PtC2‚0.25(toluene); (C) O-PtC2‚0.5(toluene).
until it reaches 0.5 equiv of toluene; at this point the rate of
mass loss becomes dramatically slower, and with an additional
3 h of purging the sample reaches 0.25 equiv. Additional long-
term N2 purging is required to reach the original “dry” mass;
heating speeds up the process significantly.
Vapochromic Powder Diffraction Studies. Repeated at-
tempts to grow crystals of the “dry” orange and purple forms
failed; therefore, these materials and the toluene-exposed
materials were studied by X-ray powder diffraction. The powder
pattern of P-PtC2 exhibits a single weak diffraction peak at
approximately 2θ ) 7°, indicating that this material has very
little crystalline character. The 2θ ) 7° peak is most likely due
to a small impurity of O-PtC2, as it is very crystalline and
exhibits its strongest diffraction peak near 2θ ) 7° (vide infra).
Powder diffraction data for O-PtC2 (Figure 6) were collected
between 2θ ) 2° and 2θ ) 60°. The resulting pattern is similar
to the pattern generated from the single-crystal study of PtC2‚
0.5(toluene); attempts to reindex the pattern in a higher crystal
system failed. The single-crystal unit cell was input into the
(32) Cerius2 V3.0 software; Molecular Simulations Inc., San Diego, CA.
(33) Miskowski, V. M.; Houlding, V. H.; Che, C. M.; Wang, Y. Inorg. Chem.
1993, 32, 2518.
(34) Kiernan, P. M.; Ludi, A. J. Chem. Soc., Dalton Trans. 1978, 1127.
9
J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002 1037

A R T I C L E S
Buss and Mann
Table 5. Unit Cell Parameters and Packing Coefficients for O-PtC2
3
compd
a (Å)
b (Å)
c (Å)
â (deg)
V (Å )
C c
k
PtC2‚x(hexanes)a
20.094(1)
19.893(1)
18.95(1)
20.22(1)
20.22(1)
18.243(1)
18.365(1)
17.89(1)
18.29(1)
18.29(1)
13.141(1)
13.150(1)
12.80(1)
12.80(1)
13.15(1)
117.813(1)
118.127(1)
107.9(1)
118.13(1)
118.13(1)
4260.6(3)
4237.1(4)
4129(1)
4175(1)
4289(1)
PtC2‚0.5(toluene)a
0.692
0.620
0.658
0.684
b
O-PtC2
O-PtC2‚0.25(toluene)b
O-PtC2‚0.5(toluene)b
a Data obtained from single-crystal X-ray diffraction studies at 173 K. b Room-temperature powder diffraction data. c Kitaigorodsky packing coefficients
are calculated from [Z (mol vol)]/V.
in the case of P-PtC2 and O-PtC2, TGA experiments and
elemental analyses indicate that neither compound includes
lattice solvent. In addition to their electronic absorption and
emission spectra, the most significant difference between P-PtC2
and O-PtC2 is in their level of crystallinity. Powder diffraction
experiments show that P-PtC2 is amorphous, while O-PtC2 is
crystalline. It is proposed that the difference in color between
P-PtC2 and O-PtC2 is due to different solid-state packing
arrangements, including what must be large differences in the
Pt-Pt distance. Fortuitously, it was the crystalline O-PtC2
compound that was found to be vapochromic, and further efforts
were focused accordingly.
Gravimetric studies show that even bulk samples of micro-
crystalline O-PtC2 quickly sorb 0.5(1) molar equiv of toluene
when exposed to toluene vapor. Samples continue to pick up
additional mass but at a much slower rate until, at very long
exposure times, 0.9(1) equiv of toluene is sorbed. The binding
of toluene in this phase is very weak, and the phase was not
studied further. When the toluene vapor source is removed, the
mass loss is rapid and reaches a metastable state at 0.5 equiv
of toluene. With continued N2 purging, the sample slowly
continues to lose weight until it reaches ca. 0.25 equiv. The
remaining 0.25 equiv of toluene is held tenaciously and requires
heating and purging for complete removal.
The powder diffraction studies show that the unit cell of
O-PtC2 expands slightly upon exposure to toluene vapor and,
upon removal of the toluene source, the unit cell contracts again.
During the unit cell expansion all three axes increase signifi-
cantly in length; however, this expansion is counteracted
somewhat by a significant increase in the monoclinic angle. In
harmony with the gravimetric experiments, this expansion occurs
in two distinct steps from the initial dry phase; the O-PtC2‚
0.25(toluene) phase is formed first and then the O-PtC2‚0.5-
(toluene) phase is formed (Figure 6). The 0.25(toluene) phase
is not observed by the relatively slow diffraction experiments
in the forward direction, but it is observed upon removal of the
toluene vapor source. Release of toluene from O-PtC2‚0.25-
(toluene) to regenerate the original unit cell requires either
heating or several days of purging.
Analysis of the three monoclinic unit cells involved in the
sorption process reveal that the exposed material has the same
unit cell dimensions as the single-crystal structure of O-PtC2‚
0.5(toluene). A Connelly analysis35 of the dry O-PtC2 unit
cell indicates that there is 1569 Å3 of free volume within the
unit cell and a packing coefficient (Ck) of 0.620 (Table 5). The
Ck value of dry O-PtC2 is at the low end of the normal range
for packing coefficients of organic molecules (0.595-0.887).36
As determined by gravimetric analysis, four toluene molecules,
with a volume of 372 Å3, enter the unit cell to form O-PtC2‚
0.5(toluene), but the unit cell volume increases by only 160
Å3, or 43% of the required volume. The packing coefficient
calculated for O-PtC2‚0.5(toluene) is 0.684, indicating that the
uptake of toluene vapor gives a 10% increase in packing
efficiency. An increase in the packing efficiency is the primary
method that provides space for the incorporated solvent
molecules. In sharp contrast, the vapochromic [Pt(CN-iso-
C3H7)4][Pt(CN)4] (PtiC3) double-salt compound previously
reported is also poorly packed (Ck ) 0.622), and the uptake of
solvent molecules also results in a small (6% or less) change in
packing efficiency.12 However, in the PtiC3 case, the small
increase in packing efficiency accompanies a very dramatic
change in the unit cell volume (54% to 95% increases in volume,
depending on the solvent). In summary, O-PtC2 sorbs vapors
by allowing guest solvent molecules into preexisting pockets
while PtiC3 responds by allowing the guest to form new volume.
A possible explanation for the disparity in expansion behavior
between PtiC3 and O-PtC2 is a difference in guest-host
interactions. Consistent with the low value of R, Abraham’s
H-bonding parameter,37 and our previous study,11 the sorption
of toluene causes only 1 cm-1 shifts in the νCN stretching
frequency. In contrast, the crystal structure of PtC2‚0.5(toluene)
also shows that there are no direct interactions between the guest
toluene molecules and the cyanide ligands (the closest Ncyanide‚‚‚
Htoluene distance is 5.50(1) Å). The single-crystal structure of
PtiC3‚16H2O clearly shows a network of hydrogen bonding with
Ncyanide‚‚‚Hwater distances of approximately 2.8 Å. It is possible
that the special molecular arrangements required for hydrogen-
bonding networks between the anion and the included solvent
allow the large unit cell expansions observed for PtiC3.
The lack of important H-bonding interactions between the
host and guest suggested that changes in the Pt-Pt distance
might explain the vapoluminescent shifts observed for O-PtC2.
Powder diffraction studies reveal that, upon exposure to toluene
vapor, the c axis of the O-PtC2 unit cell increases by 0.35 Å to
give a Pt-Pt separation increase of about 0.09 Å. The increase
2
in distance produces weaker interactions between the dz and pz
orbitals of neighboring Pt atoms that result in a greater HOMO
to LUMO energy gap.30 The shifts to higher energy observed
for the emission spectra of O-PtC2‚0.25(toluene) and O-PtC2‚
0.5(toluene) are consistent with the Pt-Pt distance lengthening.
Emission spectra for several additional O-PtC2‚VOC adducts
(mesitylene, chlorobenzene, benzene, ethanol, and p-xylene) also
show reversible changes. It is interesting that the two VOCs
that cause the largest shifts in the emission spectrum of O-PtC2
(toluene and mesitylene) are not the largest, most polar, or most
volatile VOCs. There is no obvious relationship between the
magnitude of the vapochromic shift and any one VOC molecular
(35) Connelly volume calculations were completed using Cerius2 software.
(36) Kitaigorodskii, A. I. Organic Chemical Crystallography; Consultants Bureau
Enterprises: New York, 1961; p 107.
(37) Abraham, M. H. Chem. Soc. ReV. 1993, 73.
9
1038 J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002

A Crystalline Vapoluminescent Compound
A R T I C L E S
parameter. We cautiously suggest that the shifts caused by the
VOC guest may couple the molecular shape to the Pt-Pt
distance. More experiments with a larger variety of VOCs will
be needed to elucidate the vapochromic mechanism for this
compound.
shifts that also occur for benzene, chlorobenzene, mesitylene,
p-xylene, and ethanol are consistent with a lengthening of the
average Pt-Pt distance. Overall, Pt(CN-p-(C2H5)C6H4)2(CN)2
is a very stable, vapoluminescent compound that has highly
reversible and reproducible emission spectra.
Acknowledgment. K.R.M. acknowledges the financial
support of the National Science Foundation under Grant No.
CHE-9307837. We also acknowledge the single-crystal X-ray
diffraction facility of the Chemistry Department at the University
of Minnesota for the use of their Siemens SMART CCD
platform system.
Conclusions
The single-crystal structure of O-PtC2‚0.5(toluene) shows that
the cis geometry neutral complex crystallizes in a monoclinic
space group and has a structure related to the double-salt
materials [Pt(CNR)4][Pt(CN)4]. The structure packs with linear
chains of Pt atoms and solvent channels parallel with the c axis.
Powder diffraction studies show that solvent-free O-PtC2 also
crystallizes in a monoclinic space group and the unit cell
parameters of the exposed material are almost identical with
those obtained from the single-crystal study. Gravimetric studies
indicate that when O-PtC2 is exposed to toluene vapor 0.5 equiv
is easily incorporated into the unit cell, but the reverse process
occurs slowly, in two stages. The unit cell changes that occur
when O-PtC2 is exposed to toluene vapor are accompanied by
large, reversible blue shifts in the emission spectrum. Similar
Supporting Information Available: X-ray crystallographic
files in CIF format for Pt(CN-p-(C2H5)C6H4)2(CN)2‚x(hexanes)
and Pt(CN-p-(C2H5)C6H4)2(CN)2‚0.5(toluene), additional refer-
ences for footnote 6, emission spectra for O-PtC2 exposures to
benzene, p-xylene, mesitylene, chlorobenzene, and ethanol, and
powder diffraction data for four patterns. This material is
available free of charge via the Internet at http://pubs.acs.org.
JA011986V
9
J. AM. CHEM. SOC. VOL. 124, NO. 6, 2002 1039