Khartes (χάρτης) is where we get the modern word chart and translates, according to the Oxford dictionary’s etymology of chart, as “papyrus leaf.” According to Wegner’s The Journey from Texts to Translations: the Origin and Development of the Bible, it is likely that khartes “probably refers to a single piece of papyrus.” According to Liddell & Scott A Greek-English Lexicon it referred to papyrus or a scroll thereof, but their An Intermediate Greek-English Lexicon suggests that it was a single sheet. As well, Anatole Bailley’s work ascertains it’s meaning as being a single sheet of papyrus. It’s worth noting here that a scroll wasn’t a contiguous piece of papyrus, rather it was a series of connected sheets; that’s not particularly important for the discussion, but it is something to bear in mind.

For an informative description of extant papyri, it’s worthwhile to look at An Introduction to Latin and Greek Palaeography by Thomas Maunde Thompson. In the book, he details the production and dimensions of papyrus. Firstly, he address the height: 6 inches among the earliest Egyptian papyri, later extending to 9, 11, and even over 15 inches; the papyri of the British Museum from the Homeric period ranged from 9-12 inches. In terms of width, he addresses it from the Roman perspective: charta Augusta and charta Livia were 9.5 inches wide; charta hieratica was 8 inches; charta amphitheatrica was 6.5 inches; charta Fanniana was 7.5 inches; charta Saitica was 5.75 inches; charta Taeniotica was 5 inches; and macrocollon was 16 inches wide. He lists some other minor variations, as well. As a note here, charta hieratica seemed to have once been the largest and highest quality, but its name was deprecated in favor of Augusta and Livia, which were invented to flatter Emperor Augustus. We might then speculate, being the oldest dimension of papyrus on list, the 9.5 inch variant was perhaps drawn from the standard width of papyrus in some other Mediterranean culture (Greek or Egyptian being the most likely candidate.)

As to your question of cost, the evidence would seem to suggest it was per sheet, but that depends on how pieces were attached into scrolls. If it was done by the purchaser or user, than we can assume the price was per sheet. If, however, sheets were attached by the manufacturer, than whether it was price per sheet or scroll depends entirely on our confidence in the translation of χάρτης.