In a scenario where you are working with a large dataset in Azure Databricks that contains a 'product_description' column with text descriptions of products, you are tasked with extracting the product name from each description. The product name is always the first word in the description. Considering the need for efficiency and accuracy in processing large datasets, which of the following Spark SQL queries would you use to achieve this task? Choose the best option that correctly extracts the product name while ensuring optimal performance.