Calculate the difference between all values in a group and a reference value for the group in R

I have a dataframe like this:

df1 <- data.frame(ID = c(1,1,1,1,1,2,2,2,2,2,2),
                  Var1 = c(23, 41, 32, 58, 60,12,34,55,49,60,64),
                  Var2 = c(4,5,2,1,7,3,4,5,5,6,9))
   ID Var1 Var2
1   1   23    4
2   1   41    5
3   1   32    2
4   1   58    1
5   1   60    7
6   2   12    3
7   2   34    4
8   2   55    5
9   2   49    5
10  2   60    6
11  2   64    9

and another one like this:

df2 <- data.frame(ID = c(1,2),
                  Var1 = c(20,67),
                  Var2 = c(5,3))

  ID Var1 Var2
1  1   20    5
2  2   67    3

I want to add a column to the first dataframe with the absolute difference between the columns of the two dataframes for every ID group. That is 20 (i.e., the Var1 value with ID equal to 1 in the second dataframe) should be subtracted from every Var1 value with an ID equal to 1 in the first dataframe and then the absolute value should be given in an added column.

I figured out one solution which is rather cumbersome as I have a much larger dataframe with many variables:
First, I would merge the two dataframes by the ID column and then add another column calculating the absolute difference:

df3 <- merge(df1,df2,by="ID")
df3$Var1_Diff <- abs(df3$Var1.x - df3$Var1.y)
print(df3)

So that I end up with a ugly dataframe like this:

       ID Var1.x Var2.x Var1.y Var2.y Var1_Diff
1   1     23      4     20      5         3
2   1     41      5     20      5        21
3   1     32      2     20      5        12
4   1     58      1     20      5        38
5   1     60      7     20      5        40
6   2     12      3     67      3        55
7   2     34      4     67      3        33
8   2     55      5     67      3        12
9   2     49      5     67      3        18
10  2     60      6     67      3         7
11  2     64      9     67      3         3

Can anyone suggest a simpler solution where I just add a column to the first dataframe?

Also, as I have do this process for many Var colums it would be awesome if the solution could also take that into consideration.

Thank you!

This should scale up nicely:

library(dplyr)
bind_rows(list(keep = df1, ref = df2), .id = "source") |>
  mutate(
    across(starts_with("Var"), \(x) abs(x - x[source == "ref"]),
           .names = "{.col}_diff"),
    .by = ID
  ) |>
  filter(source == "keep") |>
  select(-source)
#    ID Var1 Var2 Var1_diff Var2_diff
# 1   1   23    4         3         1
# 2   1   41    5        21         0
# 3   1   32    2        12         3
# 4   1   58    1        38         4
# 5   1   60    7        40         2
# 6   2   12    3        55         0
# 7   2   34    4        33         1
# 8   2   55    5        12         2
# 9   2   49    5        18         2
# 10  2   60    6         7         3
# 11  2   64    9         3         6

You can skip some of the intermediates with a data.table join:

library(data.table)

setDT(df1)[
  setDT(df2),
  `:=`(diff1 = abs(i.Var1 - Var1), diff2 = abs(i.Var2 - Var2)),
  on = "ID"
][]
#>     ID Var1 Var2 diff1 diff2
#>  1:  1   23    4     3     1
#>  2:  1   41    5    21     0
#>  3:  1   32    2    12     3
#>  4:  1   58    1    38     4
#>  5:  1   60    7    40     2
#>  6:  2   12    3    55     0
#>  7:  2   34    4    33     1
#>  8:  2   55    5    12     2
#>  9:  2   49    5    18     2
#> 10:  2   60    6     7     3
#> 11:  2   64    9     3     6