The unique aspects of x-ray lithography that make it attractive for th
e sub-100 nm domain include: a highly localized, sharply peaked point-
spread function, leading to minimal proximity effects, absence of spur
ious scattering; an intrinsic resolution below 30 nm; compatibility wi
th all pattern geometries; and parallel exposure (i.e., compatibility
with volume production). The major problem areas are: the mask-sample
gap (less than 5 mu m for linewidths below 70 nm), and absorber stress
, which must be near zero to avoid mask distortion. Nanometer-level pa
ttern placement and alignment are considered achievable by means of sp
atial-phase-locked e-beam lithography and interferometric-broad-band i
maging, respectively. The efficacy of x-ray nanolithography has been d
emonstrated via the fabrication of a variety of sub-100 nm-featured qu
antum-effect devices, Si MOSFETs, and grating-based optoelectronic dev
ices. In the event that the small gaps required of proximity x-ray nan
olithography prove unacceptable in manufacturing, x-ray projection usi
ng arrays of zone plates appears to be the only approach that can empl
oy the optimal wavelengths (i.e., similar to 1 nm or 4.5 nm) and achie
ve deep sub-100 nm resolution. A scheme is proposed that employs an ar
ray of zone plates in a pattern generator mode.