am I able to determine Ca in the protein sample with ICP-MS 7800 ( Include Helium collision and High Matrix)?

my sample is expected to contain  Calcium at the level of 20ppm.  we are trying to determine it with 7800 ICP-MS (includes Helium collision and High Matrix). 

my questions are following: 

1. can we reduce FP from 1550 W to 900W, so we can use cold plasma to test 40Ca? 

2. how do we define  "total dissolved solid" to the digested sample?

for example, sample weight 0.5g, after digested , then diluted to 50mL. is TDS 1%?

we need to evaporate digested liquid sample , then get some solid left, dilute to 50mL. Then  we calculate TDS by solid weight divided by 50. then TDS is much less than 1%. 

  • Hello ,

    1) I would not reduce the FP to try 40Ca, keep it to 1500W and use 44Ca, you will have enough signal since Ca ionizes easily in the plasma.

    2) The total dissolved solid of a sample is normally defined as the residual mass from the evaporation of the solvent (in the case of waters). In your case, 0,5 g of a sample, when digested and diluted to 50 mL, will have a maximum TDS of 1%. I say maximum because you may have compounds that will be mineralized... organic matter, for instance, will be taken to C and H (or simpler derivatives, depending on how aggressive is your digestion). If you digest 0,5 g of NaCl and take to 50 mL, you will have 1% of NaCl in the sample as Na+ and Cl-, along with the counter anions from the acids used.

    Lets say you have 20 mg/Kg of Ca in your sample, and with this sample preparation, your final Ca concentration will be around 0.20 mg/L. That is more than enough to run on the 7800. I would even suggest using 0.50 L/min as the neb gas and 0.50 L/min as dilution gas to further dilute the tertiary aerosol and reduce the Ca signal, plus reducing the interface of the TDS in your sample. Please note you have to apply aerosol dilution in your blank, standards and solution. Using He at 4.3 - 5.0 mL/min will also give good results. Please check if this works with other elements that might be determined along Ca.

    If determining 20 mg/L of Ca is your actual need, you can also use AD at a higher dilution proportion and force the detector to work in analog mode. To do this, you have to go to "options", then "system settings" and tick "force analog mode". In the element table, you can mark Ca to be detected only in analog mode, so you don't overcharge your EM with too much signal.

    Or maybe just dilute the sample to a lower concentration - bare in mind that successively dilutions will increase your method's uncertainty. HMI (AD) helps in this way since it is an online process and the dilution error is minimized.

    Hope this helps!


  • Hi Rodolfo, 

    Thank you for offering these suggestions. We tried your suggestions in two ways to see the sensitivity at 0.2ppm of 44Ca.  one is with Helium along with HMI ,and the other is only Helium. The results showed Helium mode without HMI improved sensitivity, but it did not achieve our target of S/N>10. In addition,  after injecting our unspiked sample, we assumed that our Limit of Quantitation possibly is  0.1ppm. 

    I assumed that maybe some instrument setting in our ICP-MS setting needs to be optimized, because sensitivity at this level should be good according to what you told me. 

    Could you please help me figure out what I could modify to increase sensitivity? We really appreciate your assistance.  

    The following table are counts from our two trials.  And the instrument setting for two trials were attached on this mail. 

    Helium without HIM (AD)

    Sample Name



    Average of three Blank



    Average of three 0.2ppm Ca STD

    5292 (S/N=7)


    Unspiked Sample (only one)



    Helium with HIM(AD)

    Sample Name



    Average of three Blank



    Average of three 0.2ppm Ca STD



    Unspiked Sample (only one)



    Thank you so much 

    Di Xu

  • Hello Di Xu,

    I think you are getting this low SN because 44Ca has only 2% abundance. For sure using HMI will reduce your sensitivity (less sample being introduced).

    Let me ask - why you need to work with SN>10? Is this a quality norm?

    You can change the sample depth to 8.0 mm, that will give some more signal to you.

    Are you doing the autotune before starting the tests?

Reply Children
  • Hi Rodolfo, 

    Thank you for  offering suggestions. I will try sample depth 8.0mm. 

    But could you tell me why shortening sample depth to 8.0mm can increase signals ? 

    Yes. we did autotune before starting the tests. 

    I am not sure if our client wants S/N >10. But USP requires sensitivity SN>10 for the quantitative limit test. 

    Let me ask my boss if the client wants S/N>10. 

    In addition, I also noticed that Helium mode reduces the signals of Ca44, since it removes some Ca44 when removing polyatomic interference. 

    Thank you 

    Best regards, 

    Di Xu

  • Hello Di Xu,

    You will increase the signal with 8.0 mm because the ion transfer will be more effective, in a general way. On the other hand, if the sample has too much TDS, the cones will get dirt faster. It is a compromise situation when you have to choose between sensibility vs productivity.

    In general, the introduction of He as the collision gas will reduce sensitivity because there will be collisions of your analyte with the He, hence diminishing its kinetic energy and reducing the ion transfer efficiency. The kinetic energy discriminator (KED) (set as "Energy Discriminator") will make a barrier, in which species with higher kinetic energy (KE) will pass through this barrier and species with lower KE will be blocked. If you set the Energy Discriminator too high, nothing will pass, if it is too low, your analyte and interferences (such as ArHe and SiO) will likely pass through the barrier and go to the MS. Generally we use between 3-5 V in normal conditions. Although there are these drawbacks, KED with He has been proven to be the best choice when working with ICP-MS (H2 and other reactive gases can give you good results, but in a single quadrupole the reaction is not always selective and you might get inaccurate results).

    Which USP are you following?

  • Hi Rodolfo, 

    Thank you for telling use this. We really learned a lot from you about ICP-MS 

    USP is  United States Pharmacopeia, which sets  guide or requirements to the analytical methods for pharmaceutical companies  which produce medicines and want to sale these product in the North American Area. Typically, USP requires QL (Quantitation Limit)should be S/N>=10 , see below snap shot)

    we will try sample depth 0.8mm to  see if it will help increase signals.(possibly today)

    For KED, do you suggest we can modify energy between 3-5 v to see if it can help us improve signals but not make big impact on accuracy of signals? From our trials, we noticed it was set 5 V. 

    Thank you 

    Best regards, 

    DI Xu

  • Hi Rodolfo, 

    Thank you for suggesting us to adjust sample depth and KE. 

    WE set sample depth to 8.0mm and 4V as discrimination energy . The signals increased, and S/N is around 8. the instrument setting was attached. 

      CPS CPS RSD Average  S/N
    Blank  1654.57 5.2 1615 8.020433
    Blank  1617.89 3.5
    Blank  1646.78 5.9
    Blank  1587.88 7.9
    Blank  1570.11 1.9
    0.2ppm 12750.42 1.1 12953
    0.2ppm 12990.66 1.8
    0.2ppm 13032.91 1.6
    0.2ppm 13036.23 1.9

    There are several questions I would like to ask: 

    1. Can we use "cold plasma" by reducing RF power to determine Ca 40. , so we can obtain ppt level signals? 

    2. Is ICP-MS 7800 allowed to use cold plasma mode?  

    3. Are  there any drawbacks for the instrument or data accuracy if we use cold plasma .

    4. we can increase signals by increasing amount of sample to be digested. if we increase digested sample amount, do you think some carbon will accumulate in the instruments which will decrease the instrument sensitivity? 

    we currently digest 500 mg collagen sample with microwave. if we increase to 1 g of sample, do you think carbon in the digested solution possibly result in deposit on the detector? 

    5. as you mentioned before, if we put sample depth8.0mm, this will make instrument dirty faster. could you tell us which part could get dirty faster( skimmer cone or sample cone) , so we can prepare to clean. 

    I know I asked a lot of questions. hope you would not get annoyed.  I am a new technician for this instrument and learning staff. 

    Thank you 

    Best regards. 

    Di Xu

Was this helpful?