Air pollution may increase spontaneous miscarriage risk, potentially through inflammation. Prior studies are heterogeneous, and none have used a mixtures approach. We used data from participants who conceived in a prospective time-to-pregnancy study (N = 446) in North Carolina to examine spontaneous miscarriage, defined as a positive home pregnancy test and a self-reported pregnancy loss before gestational week 20 (N = 101). We characterized average and peak exposure to PM, PM, CO, NO, NO, SO, and O through linked residential addresses with fusion and chemical transport models. We used single pollutant and exposure mixtures models (quantile-based g-computation) to estimate associations in six exposure windows including spermatogenesis, early follicle development, and the follicular and luteal phases of the conception cycle. Sensitivity analyses stratified by vitamin D level (an anti-inflammatory). Multivariable Cox proportional hazards models estimated adjusted hazard ratios (HR) and 95 % confidence intervals (CI) per interquartile range increase in pollutant concentration. In exposure mixtures models, while the confidence intervals were wide, the magnitude and direction of several estimates were consistent with increased spontaneous miscarriage risk with increasing air pollutant exposure: spermatogenesis (HR [CI]: 1.2 [0.80, 1.8]), early follicle development (1.2 [0.80, 1.8]), and luteal phase (1.2 [0.80, 1.9]). Associations were stronger among those with low vitamin D, for example, increasing ozone was associated with increased spontaneous miscarriage only among those with low vitamin D (follicular phase HR [CI]: 3.1 [1.3, 7.4] vs. 0.84 [0.46, 1.5] for high vitamin D, p = 0.002). Air pollutants may be associated with small increases in miscarriage risk, but larger mixtures studies are needed. Further study of low vitamin D and air pollution risk is important for understanding the public health implications of vitamin D supplementation.