近年来，科研论文数据的快速增长激发了科学学研究的动机。科研基金作为支持研究进展的基础部分，其数据能够反映科学家的研究行为。我们从美国物理学会（America Physical Society，简称APS）发表论文的致谢信息中搜集数据，关注已经成功立项科研基金的运行模式：包括人员组织、研究领域以及合作模式，探讨其对科研基金最后产出的影响。 我们在这些数据中发现了大量的幂律特征，譬如基金项目发表文章数、基金项目发表PRL文章数、基金项目包含作者数的分布就是典型的幂指数在2-3之间的幂律分布。进一步的，我们研究科研基金的人员组织结构和不同类型的参与者对科研基金产出水平的影响。我们发现，科研基金项目中参与者和参与者团队的参与度是不均匀的，并且主要的2-3个参与者团队发表了基金项目90%以上的文章。此外，我们发现科研基金项目的主要参与人员的学术水平相差无几。我们发现核心团队和核心参与者的学术水平在一定程度上与科研基金项目的产出水平呈现正相关，但达到一定程度之后，核心团队和核心参与者的学术水平不再是影响基金项目产出的主要因素。 之后，我们研究了科研基金研究领域的分类和不同类型研究领域在科研产出水平的差异。通过建立科研项目的文献耦合网络，并进行社团划分，我们定义了科研基金项目研究领域的三种类型：核心领域、次要领域和孤立散点。我们发现，在科研基金项目规模比较小的时候，核心领域发表了更高比例的高水平文章；而当科研基金项目规模达到一定程度，次要领域发表高水平文章的比例超过了核心领域。 最后，我们把眼光由科研基金项目的局部扩展到全局，研究国家间的科研项目合作模式。我们定义了合作强度来表征国家与国家之间科研基金项目的合作强度。研究表明，同一个国家的基金更倾向于与自己国家的基金合作。此外，我们研究了国家经济发展水平与发表文章水平之间的关系。我们发现一个国家的国民生产总值（GDP）越高，其发表论文总数越多；而一个国家的人均国民生产总值（人均GDP）越高，其发表高水平文章的比例越高。这个结果表明一个国家的科研实力与其经济实力密不可分。 本文的创新点主要有：第一，数据具有独创性，前人或是利用基金申请书的部分数据，或是使用各种基金委提供的数据，从已发表论文致谢部分信息提取成功立项基金项目的发表记录数据尚属首次 。第二，前人的研究大多集中讨论影响基金项目成功立项的因素，本工作着眼于基金成功立项后影响其科研产出水平的因素。第三，研究方法具有创新性，之前评价基金或者资助机构多用线性或非线性模型或者计算基金的成功率，本工作从基金的网络结构和基金资助文章的网络结构入手。

In recent years, the rapid growth of scientific papers have stimulated the research of Science of Science. Grant data, as a fundamental part of supporting research progress, can reflect the behavior of scientists. We collected data from the acknowledgement of the papers published in APS, focused on the operation mode of the successful research grants, including personnel organization, research field and cooperation mode, and discussed their effect on the final output of the research grants. First, we found plenty of power law characteristics, such as the number of articles published by research grant, the number of PRL articles published by research grant projects, and the distribution of the number of authors included in research grant projects, which are typical power law distributions with power exponent between 2 and 3. Furthermore, we discuss the mode of personnel organization of scientific research project and its influence on scientific research output. We found that the participation of the authors and the author team in the research grant projects was uneven, and over 90% papers of the grant project were published by only 2-3 major author teams.In addition, we found that the academic ability of the main participants in the grant projects was similar.In addition, we found that the academic standards of the main participants in the research fund programs were similar. We found that the academic ability of core teams and core participants is positively correlated with the output level of scientific grants to some extent, but after reaching a certain level, the academic ability of core teams and core participants is no longer the main factor affecting the output of scientific grants. After that, we discuss the classification of research fields in grant projects and their different performances of scientific output. Through the establishment of bibliographic coupling network of scientific research projects and the detection of communities, three types of research fields of grant projects are defined: core field, secondary field and isolated scatter. We found that when the scale of grant projects was relatively small, a higher proportion of high-level articles were published in core fields. However, when the scale of grant projects reaches a certain level, the proportion of high-level articles published in secondary fields exceeds that in core fields. Finally, we expand our vision from the grant projects itself to the overall, and explore the cooperation mode between countries in scientific research projects. We define the cooperation intensity to represent the cooperation intensity between countries. Research shows that grants from the same country are more likely to cooperate with each other. In addition, we explored the relationship between the level of national economic development and the publication of high-level articles. We found that the higher the GDP of a country, the more papers it published. In addition, the higher the per capita GDP of a country, the higher the proportion of high-level articles. It shows that a country's research capabilities is inseparable from its economic capabilities. The innovation points of this paper mainly include: First, the data is original. It is the first time to extract the grants information from the acknowledgement of published papers, either by using partial data of fund application or data provided by various fund committees. Second, most previous studies focus on detecting how a research project apply for fund successfully, this work focuses on the factors that affect the level of scientific grants output after the successful establishment of fund projects. Third, the research method is innovative. Linear or nonlinear models are used to evaluate the grants or fundations or calculate the success rate of the grants. This work starts from the network structure of the grants, such as author networks and paper networks.