Two deep level defects (2.25 and 2.03 eV) associated with oxygen vacancies (V o ) were identified in ZnO nanorods (NRs) grown by low cost chemical bath deposition. A transient behaviour in the photoluminescence (PL) intensity of the two V o states was found to be sensitive to the ambient environment and to NR postgrowth treatment. The largest transient was found in samples dried on a hot plate with a PL intensity decay time, in air only, of 23 and 80 s for the 2.25 and 2.03 eV peaks, respectively. Resistance measurements under UV exposure exhibited a transient behaviour in full agreement with the PL transient indicating a clear role of atmospheric O 2 on the surface defect states. A model for surface defect transient behaviour due to band bending with respect to the Fermi level is proposed. The results have implications for a variety of sensing and photovoltaic applications of ZnO NRs.PACS numbers: 68.43. Fg, 68.43.Tj, 73.20.Hb, 73.20.At Zinc oxide is a wide band gap (∼ 3.2→3.4 eV) ntype semiconductor with a large exciton binding energy (60 meV) and is a promising material for a range of applications 1 . However, studies on low cost ZnO nanostructures (NSs) and thin films are unclear as to the source of n-type conductivity and persistent photoconductivity (PPC), the UV sensing mechanism, and the defect landscape 2,3 . In particular, the correlation between the defect landscape and sensing (whether gas, pH, or UV) responsivity in low cost ZnO NRs is debated. In one model, the neutral oxygen vacancy (V o ) is an n-type donor state 3 , atmospheric O 2 absorbs at this site 2 , and a depletion region forms beneath the surface 4 . UV excitation creates electron-hole pairs, holes migrate to the depletion region and O 2 desorption occurs, thus reducing the depletion region and increasing the conductivity 4,5 . Therefore, the kinetics of O 2 desorption determine the response time of a UV sensor. In a second model, V o is reported to be a deep level state (DLS) The quality, diameter, and length of the NRs was measured using a Gemini field emission scanning electron microscopy (SEM) Carl Zeiss SUPRATM 25. CBD introduces water based absorbates 13-15 , we controlled this parameter by comparing samples as-prepared not dried (ND), dried at 100 o C for 20 min on a hot plate, and left to dry over two weeks. Additionally, we annealed a sample at 600 o C for 30 min in O 2 to understand the role of V o .Resistance measurements were performed under exposure to 364 nm UV light. The samples were biased to force a current of 1 nA between two probes 1 mm apart to extract the resistance. This value was used to avoid compliance in the samples and gave a good overall response under UV exposure. PL measurements were performed by pumping at 1.5 mW the 325 nm line of a He-Cd laser chopped through an acousto-optic modulator at a frequency of 55 Hz. The PL signal was analyzed by a single grating monochromator, detected with a Hamamatsu visible photomultiplier, and recorded with a lock-in am-