Graphene’s inherent nonselectivity
and strong atmospheric
doping render most graphene-based sensors unsuitable for atmospheric
applications in environmental monitoring of pollutants and breath
detection of biomarkers for noninvasive medical diagnosis. Hence,
demonstrations of graphene’s gas sensitivity are often in inert
environments such as nitrogen, consequently of little practical relevance.
Herein, target gas sensing at the graphene–activated carbon
interface of a graphene-nanopored activated carbon molecular-sieve
sensor obtained via the postlithographic pyrolysis of Novolac resin
residues on graphene nanoribbons is shown to simultaneously induce
ammonia selectivity and atmospheric passivation of graphene. Consequently,
500 parts per trillion (ppt) ammonia sensitivity in atmospheric air
is achieved with a response time of ∼3 s. The similar graphene
and a-C workfunctions ensure that the ambipolar and gas-adsorption-induced
charge transfer characteristics of pristine graphene are retained.
Harnessing the van der Waals bonding memory and electrically tunable
charge-transfer characteristics of the adsorbed molecules on the graphene
channel, a molecular identification technique (charge neutrality point
disparity) is developed and demonstrated to be suitable even at parts
per billion (ppb) gas concentrations. The selectivity and atmospheric
passivation induced by the graphene–activated carbon interface
enable atmospheric applications of graphene sensors in environmental
monitoring and noninvasive medical diagnosis.