数据帧python中的格式时差

我的数据中有两行需要时差

        request           Req_Created       Req_Closed    
       0 REQ0079455     15/05/2019  16:51   23/05/2019 20:53
     1 REQ0079455     15/05/2019 16:51    23/05/2019 20:53

使用数据帧:

df['Req_time_taken'] = pd.to_datetime(df['Req_Closed'], format ='%d/%m/%Y %H:%M') - pd.to_datetime(df['Req_Created'], format ='%d/%m/%Y %H:%M') 
df.to_csv ("102.csv")

问题是输出格式不可读

          request     Req_Created          Req_Closed         Req_time_taken
       0  REQ0079455  15/05/2019 16:51    23/05/2019 20:53    8 days 04:02:00.000000000
       1  REQ0079455  15/05/2019 16:51    23/05/2019 20:53    8 days 04:02:00.000000000

如何以以下两种格式显示结果:

          request     Req_Created          Req_Closed         Req_time_taken_1    Req_time_taken_2
       0  REQ0079455  15/05/2019 16:51    23/05/2019 20:53    8 days 04:02        196 hours 02 minutes
       1  REQ0079455  15/05/2019 16:51    23/05/2019 20:53    8 days 04:02        196 hours 02 minutes

请注意申请时间和申请时间的格式差异