Collaboration has become an essential paradigm in sustainable development research and in strategies for meeting the United Nations Sustainable Development Goals (SDGs). This study uses bibliometric methods and network analysis to examine research output and collaboration supporting the SDGs and explores means to detect and analyze research collaboration beyond the traditional definition of multiple, one-time co-authorship. We employed two additional lenses of collaboration: repeat collaboration and collaboration time point to quantify and visualize co-authorship data sourced from Microsoft Academic Graph. Our results show an increased collaboration rate over time at the author and institutional levels; however they also indicate that the majority of collaborations in SDG-related research only happened once. We also found out that on average, repeat collaboration happens more frequently, but after a longer duration, at the institutional level than at the author level. For this reason, we further analyzed institutions and identified core institutions that could help influence more consistent collaboration and sustain or grow the SDG-related research network. Our results have implications for understanding sustainable partnerships in research related to SDGs and other global challenges.