40 



To assess the analytical precision of the laboratory in 1997, sample 1 1 A was 

 homogenized, divided, and sent to the laboratory as two samples: 1 1 A and NLDS QCA. In 

 addition, the laboratory analyzed the duplicate matrix spike samples (Section 2.4.4.2) for 

 additional precision analyses. 



The RPD for metals in the submitted blind duplicates (1 1 A and QCA) ranged from 

 ±2% to 19%, indicating good precision for metals. In addition, the laboratory reported RPDs 

 of two samples, QCA and 10A (0.75-1.0 m) for both non-spiked and spiked matrices. For 

 metal samples with values greater than ten times the instrument detection limit, the control 

 limit is ±20% RPD. The spiked sample or spiked duplicate sample recovery must be within 

 ±25% of the actual value or within the documented historical acceptance limits. The RPDs 

 were between 0.7-14% for the QCA duplicates, and 0.6% for Zn for 10A. The RPDs of the 

 QCA matrix spiked duplicates ranged from 0-5.5% for the listed metals, excluding Zn which 

 was 20% and therefore within the limit for spiked samples. For the Zn only sample 10A 

 (0.75-1.0 m), the spiked duplicates had an RPD of 1.6%. 



For TOC analyses of samples QC and 1 1 A, the RPDs were ±4.6% and 18.5% for the 

 matrix spike duplicates, and ±5.6% for the duplicate RPD, indicating acceptable precision. 

 For TOC method quality control, the laboratory selected samples NLDS7A and NLDS 12A 

 for triplicate analyses. When there are more than two sample values to consider, the relative 

 standard deviation (RSD) is used to assess precision. The RSD is calculated using the 

 following equation: 



%RSD = ( standard deviation ) xl0Q 

 v average of samples 7 



The RSD was 12.4% for sample 7 A and 8.3% for 4A, which also indicated an acceptable 

 level of precision. 



The PAH MS/MSD sample was NLDS 1 A. The RPD values ranged from 12% to 

 47%. Of the 16 values calculated, only one compound was outside the required range, 

 pyrene. As stated previously, all other analytes met both accuracy and precision limits. 

 Therefore, the analytical system was assessed to be in control and no further corrective 

 action was implemented. 



The analytical precision results for the 1998 samples indicated the RPD for metals were 

 reported for both non-spiked and spiked matrices. The precision criteria are ±20% RPD. The 

 metal MS/MSD RPDs were between 0-19% for NL-14B for all metals except Cd, which was 



Monitoring Cruise at the New London Disposal Site, Seawolf Mound 1995 - 1998 



