Introduction
Both quicksort and mergesort algorithms are based on the divide and conquer approach. The quick sort is an internal sorting algorithm where the data is sorted in the main memory. In contrast, merge sort is an external sorting algorithm in which the data cannot be stored in the main memory and thus requires an Auxiliary memory for Sorting the algorithm.
In quick sort, the arrays are divided into any ratio as there is no compulsion to divide them into equal parts. Whereas in merge sort, the array is divided into two halves only.
Merge sort is more stable than quick sort, as two elements having equal values appear in the same order in the sorted output as they were in the unsorted input. Quick sort becomes unstable for this scenario. But we can make it stable by making some changes in the code.
The worstcase complexity of quick sort is O(n^2), as there is a need for many comparisons in the worst condition. In merge sort, worst and average cases have the same complexities O(n*log n).
See, Mergesort in C and Rabin Karp Algorithm
Why is Quick Sort preferred for Arrays?
No other sorting algorithm performs better than Quick sort on Arrays because of the following reasons:

Quick sort is an inplace sorting algorithm, i.e. which means it does not require any additional space, whereas Merge sort does, which can be rather costly. In merge sort, the allocation and deallocation of the excess space increase the execution time of the algorithm.

The locality of reference is one of the critical reasons for quick sortâ€™s efficiency. It allows the computer system to access memory locations close to each other, which is faster than memory locations distributed throughout the memory, as in the case of merge sort.

Quick sort is most commonly implemented using a randomized version with anticipated time complexity of O(NlogN). Although the worst case is possible in the randomized version, it does not occur for a specific pattern (such as a sorted array). Hence the randomized quick sort works well in practice.
Recommended Topic, Floyds Algorithm