While numerous thermal conductivity investigations of amorphous dielectrics have been reported, relatively few have attempted to correlate to the influence of processing conditions and the resulting atomic structure. In this regard, we have investigated the influence of growth conditions, post deposition curing, elemental composition, atomic structure, and nano-porosity on the thermal conductivity for a series of organo-silicate (SiOCH) thin films. Time-domain thermoreflectance (TDTR) was specifically utilized to measure thermal conductivity while the influence of growth conditions and post deposition curing on composition, mass density, atomic structure, and porosity were examined using nuclear reaction analysis (NRA), Rutherford backscattering spectroscopy (RBS), Fourier-transform infrared (FTIR) spectroscopy, nuclear magnetic resonance (NMR), ellipsometric porosimetry (EP), and positronium annihilation lifetime spectroscopy (PALS). Analytical models describing the thermal conductivity dependence on mass density and vol% porosity were found to generally over-predict the measured thermal conductivity, but improved agreement was obtained when considering only the heat carrying network density determined by FTIR. Ashby’s semi-empirical relation, which assumes only 1/3 of the heat carrying bonds are aligned to the heat transport direction, was also found to reasonably describe the observed trends. However, the thermal conductivity results were best described via a model proposed by Sumirat (J Porous Mater 9:439 (2006)) which considers the effect of both vol% porosity and phonon scattering by nanometer sized pores. Post-deposition curing was additionally observed to increase thermal conductivity despite an increase in nano-porosity. This effect was attributed to an increase in the Si–O–Si network bonding produced by the cure.

中文翻译：

---

非晶纳米多孔有机硅酸盐薄膜的热导率-结构-处理关系

尽管已经报道了无定形电介质的大量导热系数研究，但很少有人试图将其与加工条件和所产生的原子结构的影响相关联。在这方面，我们研究了一系列有机硅酸盐（SiOCH）薄膜的生长条件，沉积后固化，元素组成，原子结构和纳米孔隙度对热导率的影响。时域热反射率（TDTR）用于测量热导率，同时使用核反应分析（NRA），卢瑟福背散射光谱（RBS）检查了生长条件和沉积后固化对成分，质量密度，原子结构和孔隙率的影响。 ），傅立叶变换红外（FTIR）光谱，核磁共振（NMR），椭圆孔隙率法（EP）和正电子an没寿命谱（PALS）。人们发现，描述热导率对质量密度和体积百分比孔隙率的依赖性的分析模型通常会高估测得的热导率，但是当仅考虑由FTIR确定的载热网络密度时，可以获得更好的一致性。阿什比的半经验关系（假设只有1/3的载热键与传热方向对齐）也被发现可以合理地描述观察到的趋势。然而，通过Sumirat（J Porous Mater 9：439（2006））提出的模型可以最好地描述热导率结果，该模型考虑了纳米级孔隙的体积百分比孔隙率和声子散射的影响。尽管纳米孔隙率增加，但另外观察到沉积后固化可提高热导率。这种作用归因于固化产生的Si–O–Si网络键合的增加。