We report the detection of the Pa \alpha emission line in the z = 2.515 galaxy SMM J163554.2+661225 using Spitzer spectroscopy .
SMM J163554.2+661225 is a sub-millimeter–selected infrared ( IR ) –luminous galaxy maintaining a high star–formation rate ( SFR ) , with no evidence of an AGN from optical or infrared spectroscopy , nor X-ray emission .
This galaxy is lensed gravitationally by the cluster Abell 2218 , making it accessible to Spitzer spectroscopy .
We measure a line luminosity , L ( \hbox { Pa$ \alpha$ } ) = ( 2.05 \pm 0.33 ) \times 10 ^ { 42 } erg s ^ { -1 } , corrected for gravitational lensing .
Comparing the H \alpha and Pa \alpha luminosities , we derive a nebular extinction , A ( V ) = 3.6 \pm 0.4 mag .
The dust–corrected luminosity , L ( \hbox { Pa$ \alpha$ } ) = ( 2.57 \pm 0.43 ) \times 10 ^ { 42 } erg s ^ { -1 } , corresponds to an ionization rate , Q _ { 0 } = ( 1.6 \pm 0.3 ) \times 10 ^ { 55 } \gamma s ^ { -1 } .
The instantaneous SFR is \psi = 171 \pm 28 M _ { \odot } yr ^ { -1 } , assuming a Salpeter–like initial mass function from 0.1 to 100 M _ { \odot } yr ^ { -1 } .
The total IR luminosity derived using 70 , 450 , and 850 µm data is \hbox { $L _ { \mathrm { IR } } $ } = ( 5 - 10 ) \times 10 ^ { 11 } L _ { \odot } , corrected for gravitational lensing .
This corresponds to \psi = 90 - 180 M _ { \odot } yr ^ { -1 } , where the upper range is consistent with that derived from the Pa \alpha luminosity .
While the L ( 8 \micron ) / L ( \hbox { Pa$ \alpha$ } ) ratio is consistent with the extrapolated relation observed in local galaxies and star–forming regions , the rest–frame 24 µm luminosity is significantly lower with respect to local galaxies of comparable Pa \alpha luminosity .
Thus , SMM J163554.2+661225 arguably lacks a warmer dust component ( T _ { D } \sim 70 K ) , which is associated with deeply embedded star formation , and which contrasts with local galaxies with comparable SFRs .
Rather , the starburst in SMM J163554.2+661225 is consistent with star–forming local galaxies with intrinsic luminosities , \hbox { $L _ { \mathrm { IR } } $ } \approx 10 ^ { 10 } L _ { \odot } , but “ scaled–up ” by a factor of \sim 10–100 .