COMMUNICATIONS Diisocyanate** Giinther Maier,* Matthias Naumann, Hans Peter Reisenauer, and Jiirgen Eckwert After the first synthesis of cyanogen (1) by Gay-Lussac in 1815['1 173 years passed until Bickelhaupt and van der Does''' described the formation of a second C,N, isomer (2).[31It took only four more years until diisocyanogen (3) was detected by matrix spectro~copy.[~1 NZC-C=ST NIC-N=C C=N-N=C 1 2 3 In analogy to the C N radical the ambivalent fragments C N O and NCO can formally be combined to give structurally different isomers of the formula C,N,O,. Three of a total of ten typical atomic arrangements are compounds 4-6. The first species of this kind, cyanogendi-N-oxide (4), was obtained by Grundmann in O+N~C-C=N+O O+N=C-N=C=O 1963.[51 Here we report on the identifica4 5 tion of the structural isomer diisocyanate O=C=N-N=C=O O=N-CEN (6). 6 7 Our aim was not only to prove the existence of 6. In addition, we assumed that upon matrix-photolysis of 6 in argon at 10 K two CO molecules would be formed in the same rare gas matrix cage without interfering by-products. A singlet/triplet-photoexcitation of one of the 'CO molecules followed by a ['CO + 3CO]addition might provide an access to the still unknown carbon monoxide dimer C,O, (triplet ground stater6'). Matrix-irradiationr7I of 4 does not lead to the desired diisocyanate 6 but yields-presumably via the primary, singly rearranged product 5-nitrosyl cyanide 7. Therefore, we had to find another route to 6 . A promising precursor for 6 is oxalic acid diazide (9). We first tried to obtain matrix-isolated diazide 9 by passing oxalic azid dibromide over solid sodium azide.18] This reaction turned out to be poorly reproducible. Hence, we synthesized 9 in a reaction of oxalic azid dichloride (8) with tri-n- butyltin azide.[" However, in this case the formation of two by-products, azidoformyl isocyanate (10) and the fairly explosive carbonyl diazide, is disadvantageous. In the meantime we learned that it is more convenient and safer to synthesize 9, in a modified procedure of Roesky and Glemser,['ol by the reaction of oxalic acid dihydrazide (11) with nitrous acid. Utilizing this approach we isolated pure. crystalline oxalyl diazide and deposited it on a matrix window. In solution (CH,CI,) at room temperature 9 loses one molecule of nitrogen, and 10 is formed. Removal of the solvent yields 10 as a colorless, pure liquid that can be safely handled, easily evaporated, and condensed on a spectroscopic window. Irradiation of 9 and 10 in argon at 10 K with light of the wavelength 254 nm furnishes 6. Compound 6 is also formed upon high-vacuum flash pyrolysis of 9 and 10 (quartz tube filled with quartz wool, 900 "C). The proof of the identity of 6 is based on the following findings: Independently of the applied method of generation the IR spectrum exhibits a very intense absorption at 2201 cm-', two much weaker bands at 661 and 534 cm- and six additional, very weak bands above 2000 cm-'. All absorptions diminish simultaneously upon irradiation with 254 or 193 nm light. and C O is produced. A comparison with the spectrum of 6 CdiCUlated by a b initio methods unequivocally shows that all bands belong to diisocyanate (Fig. 1, Table 1 ) . ', 2l A !I- 0-, * **- 4000 3500 I 3000 2500 2000 1500 1000 500 t \;/cm-' 9C-C0 N=C=O O=C=N' H ~ N - N ~NH-NH, hv O=C + C = O + N2 Fig. 1 . IR spectrum of6. Top: Spectrum calculated hy ;ib initio methods (BLYP 631 l G * ) . Bottom: Experimentally obaerved spectrum (Ar matrix. I S K. from flash pyrolysis of 10) The experimental spectrum is ii difference spectrum of the photofrngmentation of 6 i n t o CO and N,. *: NCO radical. A = absorbance. 6 11 I E,CzOzC-HN-NH-COzCzH, [*I I*'[ 12 Prof. Dr. G Maier, Dip1 -Chem. M. Naumann. Dr. H. P. Reisenauer. Dr. J Eckwert lnstitut fur Organische Chemie der Universitat Heinrich-Buff-Ring 58. D-35392 Giessen (Germany) Fax: In!. code +(641)702-5712 In thiscommunication in analogy to the trivid names used forcompounds 1-3 (cpanogen. isocyanosen. and diisocyanogen) . the title compound is referred to as diisocyanate inatcad of isocyanato isocyanate. The geometry optimization of 6 at the BLYP/6-311G* level["] results in a minimum for the s-tr-ans-conformatiSn (d(C-0) 4 . 1 7 8 , d(C-N) ~ 1 . 2 3 3 , d(N-N) d . 3 8 7 A, ~ ( 0 - C - N= ) 168.8', K(C-N-N) = 125.9') with C,, symmetry (only real frequencies). In contrast, the C,,-symmetrical molecule with s-cis-conformation is calculated to be a transition state (one imaginary frequency). Due to the high symmetry only six of the twelve fundamental vibrations belong to the IR-active species a, and b,, and only four should be found in the investigated spectral range (5000-400 cm-'). The three most intense ones are readily assigned to the measured bands at 2201 (NCO Table 1 Calculated (BLYP.6-31lG*) and experimentally observed vibrational spectral data of 6 (Ar matrix, 15 K. from pyrolysis of 10; relative intensities in parentheics) 1'1 21, NCO str I.y b,, NCO > t r I.? il, L',,, h,, 1.3 :I NCO str NCO sir YiY str ip bend [c] ip bend oop bend [d] oop bend ip bend ip bend oop bend /1 1% b, a b, r<, % "1 it,. )',I )'a Y,? h,, r- Li,, 2265 (La.) "C. 2248 2209 ( 100.0)[a] "C: 2165 'SO: 2198 1451 (La.) 1269 (0.5) 786 ( I a ) 625 (3.8) 595 (i.8.) 506 (1.a.) 494 ( I 7) 271 (1.a.) 133 (0.5) 72 (0 0) Received: February 26. 1996 [288601E] German version: A n p v Cl~eni 1996. 108. 1800- 1x01 (ca. 2235) [b] 2225.7 Keywords: a b initio calculations density functional theory * IR spectroscopy * isocyanates . matrix isolation 2200.6 (100.0) 2164.9 2184.4 (ca. 1460) [b] 840) [b] 661 .0 (4.2) (ca. 533.5 ( 1 7) Combinalion band5 (plausible assignment). 3660.9 ( v 2 + v 9 ) . 3488 7 3040.6 ( v 3 + r,,). 2895 6 ( v , + r , , ) , 2124.3 (I.> + I ' , ~ ) . 2089.0 (Y, via) + (Y, With respect to the aforementioned C,O, problem we would like to add that so far a photoinduced dimerization of the two CO molecules trapped in the same matrix cage could not be achieved. + v,~), [a] Absolute intensit?. 1731 kmmol- I . [b] Calculated from combination bands. [c] ip = in-plane [d] oop = out-or-plane. stretching vibration, i s g , bu), 661 (in-plane bending, v l l , bJ, and 534 cm (out-of-plane bending, Y ~ a,). , The missing fourth fundamental absorption (NCO stretching vibration, vl0, b,) is very weak in accordance with the calculation. Besides these fundamental vibrations six other weak bands of 6 were identified, which can, aided by the calculation, be interpreted as combination bands of IR-inactive and IR-active transitions (Table 1). The band y g possesses an intensity high enough to enable the detection of the corresponding absorptions of the 3C and I8O isotopomers in natural abundance. The measured intensity of the 13Cband (2.2%) confirms that the molecule contains two symmetry-equivalent carbon atoms. Due to the breakdown of symmetry in the "C;"C isotopomer the band, which is inactive in the parent 6. can be detected. The photolability of 6 towards light with short (.; = 254 or 193 nm). but not towards longer wavelengths ( i >310 nm) demands an absorption in the UV range. Indeed, an electronic transition with a i.,,, = 220 nm can be registered in the UV spectrum. This band position shows that the bathochromic shift from an isolated monoisocyanate (HCNO: i,,, = 167 nmC1'I) to the conjugated diisocyanate ( 6 )is of the same order of magnitude as in the pair ethene (i,,,= 165 nm)/butadiene (i,,, = 217 nm). The correctness of the identification of 6 is further supported by its reaction with ethanol: We cocondensed 6 and ethanol on a cold finger at 77 K and obtained after thawing exclusively diethyl 1.2-hydrazinedicarboxylate(12). How stable is 6? As mentioned above, 6 survives the conditions of flash pyrolysis. In this case cleavage into NCO radic a l ~ [ "was ~ observed only to a small extent. If the thermal products of the precursor 10 are condensed without argon on a spectroscopic wJindow at 80 K, the IR spectra reveal that 6 is viable under these conditions. Subsequent annealing results in the onset of a reaction at about 130 K. At 190 K all IR bands of 6 have vanished. While 6 does not exist in pure form at room temperature its lifetime in diluted form in the gas phase is much higher. We mixed 10 and argon (1 : 1000) and irradiated this sample at room temperature in an IR gas cell with 254 nm light. Spectra measured immediately thereafter exhibited the strongest absorption of 6 as a weak band with clearly visible Pand R-branches (centered at 2208 cm- '). The intensity of this band decreased with a half-life of about 10 minutes. ' [I] L. J. Gay-Lussac. Anii. Cliini. I P N I ' I J )1815. 95. 175  T. van der Does, F. Bickelhaupt. AnReiv. Chein. 1988. / M I . Y98-1000: Angrit'. C/7Plil. In!. €d € J T ~ / . 1988, 27. 936-938.  Structural proofs: a) Microwave spectroscopy: F. Srroh. M. Winnewisser. Cheiii P/ii..s. Lrrr. 1989. 155. 21 -26: h) matrix 1R spectroxmpy: F. Stroh. M. Winnewisser. 8 . P Winnewisser. H. P. Reisenauer. G. Maier. S. J. Goede. F. Bickelhaupt. ;bid 1989. 1611, 105-112  G. Maier. H. P Reisenauer. J. Eck\*,ei-t,C. Sierakowski. T. Stumpf. Angeii.. C h i w 1992. 104. 1287 1289; A i i p . C h i i . hi.Ed E1ii.l. 1992, 31, 1218 1220.  a ) C. Grundmann. A n p i t . Ciinii. 1963. 75. 450; An,~cw C'/iiw In!. E d D7gI. 1963.2. 760: b) C Grundmanii. V Mini. J. M. Dean. H.-D Fromineld. Jii.\tuc Liebig.\ Ann. Choni. 1965. 687. 191-214.  G. Maier. H. P. Reisenauer. B. Rother, 1. Eckwert. Liehi:.\ A i m 1996, 303 306. and references therein.  G M a w . J. H. Teles. A n g m . Chein. 1987. YY. 152 153: An,qi,ii. C/wni. /n!. €11. €fig/. 1987. 26. 155- 156.  J Eckwerr. Diplomarheic. Universit'it Giessen. 1990. [Y] H. R Kricheldorf, E. Leppert. S~nrhe.\i.\,1976, 229. 230. [lo] H. Roesky. 0 . Glemser. Ciiein. Ber. 1964. Y7. 1710-1712. [ I I ] Gaussian 94, Revision B.I. M. J. Frisch. G. W. Trucks. H. B. Schlegel. P. M. W. Gill. B G. Johnson. M. A Robb. J. R. Cheeseman. T. Keith. G. A. Petersson. J. A. Montgomery. K. Raghavachari. M. A. Al-aham. V G . Zakrzewski. J. V. Ortiz. J. B. Foreman, J Cioslowski. B. B. Stefanob. A Nanayakkara. M. Challacombe. C. Y. Peng, P. Y . Ayah. W. Chen. M. W Wong. J L. Andres. E. S. Replogle. R. Gomperts, R. L. Martin. D. J. Fox. J. S Binkley. D 1. Defrees. J. Baker. J. P Stewart. M. Head-Gordon. C Gonralr/. J. A Pople. Gaussian. Inc.. Pittsburgh PA. 1995. [I21 H. Okabe, J Ciien?. P/ij..s. 1970. 53. 3507 3515 [I31 D. E. Milligan, M. E. Jacox, J Ciieni. Phi..\. 1967. 47. 5 1 5 7 - 5168. ill Prenylation of Benzoic Acid Derivatives Catalyzed by a Transferase from Escherichiu coli Overproduction: Method Development and Substrate Specificity** Ludger Wessjohann* and Bernd Sontag Enzyme-catalyzed reactions offer possibilities that extend beyond their obvious advantages of chemo-, regio-, and stereoselectivity, such as especially mild conditions, use in water and reproducibility on a microscale basis." -41 This is especially important for those reactions for which classical methods are not available. Due to their easy accessibility, hydrolases (peptidases, lipases, and esterases) now belong to the standard repertoire for organic synthesis,['b.dl whereas enzyme-catalyzed C-C bond formations"'. have not yet found widespread application. To [*I Dr. L. A Wessjohann. B. Sontag Institut fur Orginische Chemie der Universitiit Karlstrasse 23. D-80333 Munchen (Germany) Fax: Int. code +(89)5902-483 e-mail. l i i w ~oi;e.chemie.uni-muenchen i de [**I This work was supported by the Fonds der Chemischeii Industrie. We thank the Bayer AG and the BASF AG for the donation ofchemicals and equipment. We also thank Prof. Dr. L. Heide for providing us with thc bacteria strains. Dr. K. Secerin, Dr. M Melzer. and Dr. T. Kutchan for their asaistance i n molecular hiological and microbiological discussions. as well >is Prof. Dr W Steglich. Prof. Dr M. Zenk. and Dr. G. Greull for the use o f their instruments and fdCilltieS.