Charge-transfer excited states have frequently been studied by using 4-dimethylaminobenzonitrile (DMABN) as a model. In nonpolar solvents, a single fluorescence band is observed from a locally excited (LE) state. In polar solvents, the initially populated LE state reacts further to produce a stable intramolecular charge-transfer (ICT) state, which gives rise to a second fluorescence band that overlaps with, but is abnormally red-shifted from, the LE emission.[1] Results of experiments using aprotic solvents are well described by models in which polarity is the only solvent property that affects the charge transfer reaction activation energy and the relative stabilization of the ICT and LE states.[2] Whilst much work continues to concentrate on determining the structures of the LE and ICT states, [3][4][5][6][7] the precise nature of the difference between the properties of the excited state in protic and aprotic solvents is little understood. For example, the fluorescence quantum yield of DMABN in protic solvents is lower and the fluorescence spectrum is further red-shifted and broadened, relative to measurements in aprotic solvents of the same polarity, [8,9] and the fluorescence decay kinetics are difficult to interpret.[2] Hydrogen bonding in protic solvents can lead to complicated interactions [10] but although specific solute-solvent and solute-solute interactions have been discussed, [8,[11][12][13][14] there is no generally accepted explanation. There are similar problems in other cases of dual fluorescence.[15]The time-resolved infrared (TRIR) absorption spectra presented here demonstrate and monitor the formation of a hydrogen-bonded charge-transfer state of photoexcited DMABN in the protic solvent methanol (MeOH), through the development of the CN IR absorption band from an initial singlet into a doublet. The initial single band is interpreted as belonging to an ICT state like that created in aprotic acetonitrile (MeCN), where only one absorption band is observed at all delay times. The second component is interpreted as being due to the hydrogen-bonded chargetransfer state; the kinetics show the populations of the free and hydrogen-bonded species coming to dynamic equilibrium. We designate the hydrogen-bonded state as HICT. This is the first direct observation of hydrogen bonding in an excited state. Since the populations in the LE state and the two charge-transfer states coexist, the fluorescence will be triple, not dual in character. Neglect of this major factor is considered to account for much of the difficulty in interpreting the fluorescence results. [2,8,[11][12][13] A mechanism of this kind has not to our knowledge been proposed before. We believe this interpretation is applicable to other molecules with solvent-dependent dual fluorescence. Figure 1 shows TRIR spectra of DMABN in MeCN (a) and MeOH (b) recorded with sub-picosecond time resolution at pump-probe delays from 2 to 3000 ps after excitation; Figure 2 gives the time-dependence of the absorption band areas. Kinetics parameters were dete...